University of Southern California Dissertations and Theses

Relationship Between Perceptual Style, Achievement, And Childrearing Practices In Elementary-School Boys And Girls

(USC Thesis Other)

Relationship Between Perceptual Style, Achievement, And Childrearing Practices In Elementary-School Boys And Girls

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content RELATIONSHIP BETWEEN PERCEPTUAL STYLE, ACHIEVEMENT, AND
CHILDREARING PRACTICES IN ELEMENTARY-
SCHOOL BOYS AND GIRLS
by
Patricia Kirby Johnston
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Psychology)
February 1974
INFORMATION TO USERS
This material was produced from a microfilm copy of the original document. While
the most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the original
submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. If necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific pages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print.- Filmed a s
received.
Xerox University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
74-9070
JOHNSTON, Patricia Kirby, 1936-
RELATIONSHIP BETWEEN PERCEPTUAL STYLE,
ACHIEVEMENT, AND CHILDREARING PRACTICES IN
ELEMENTARY-SCHOOL BOYS AND GIRLS.
University of Southern California, Ph.D., 1974
Psychology, general
University Microfilms, A XEROX Company, Ann Arbor, Michigan
© 1974
PATRICIA KIRBY JOHNSTON
ALL RIGHTS RESERVED
UNIVERSITY O F SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
U N IVER SITY PARK
LOS ANGELES. C ALI FORNIA 9 0 0 0 7
This dissertation, written by
under the direction of hex.... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
Dean
DISSERTATION COMMITTEE
Chairman
ACKNOWLEDGEMENT
A study of this size could not have been done
without the active support of many people, all of whom
I want to thank. I sun particularly indebted to:
— the fifty families who participated.
-- Marengo School officials and staff:
Dr. Forest M. Fouts, Superintendent of Schools
in South Pasadena;
Mr. H. Curtis Eastwood, Principal of Marengo
School;
Mrs. Joan Minard, Secretary of Marengo School;
The teachers of Marengo School.
— those who assisted in the data collection
and processing:
Diana Smith
Debby Manning
Earl Cooper, Programmer at USC Computer Center
The Computer Center Staff.
-- my Dissertation Committee Members:
Dr. Albert R. Marston, Chairman;
Dr. David Rigler
Dr. Perry London.
— Dr. David Rigler and Childrens Hospital of
Los Angeles for furnishing the perceptual
style testing materials and apparatus.
-- NIMH Grant Number MH 20552-03 and Dr. Perry
London for support, in part, of this research.
-- Dr. James Kahan for his many hours of help in
analyzing the data and introducing me to
the computer.
Finally, my special thanks to Irma Love, to friends, and
to my family — Peggy, Rowly, Jim, and especially my
husband, Bill, — for their patience and understanding
through what has seemed to us all to be an endless
endeavor.
LIST OF ABBREVIATIONS
AIS = Age-Independence Scale
BST = Block Stacking Task
EFT = Embedded Figures Test
FD = field dependent
FI = field independent
RFT = Rod-and-Frame Test
RTT = Ring Toss Task
TAT = Thematic Apperception Test
TABLE OF CONTENTS
ACKNOWLEDGEMENTS................................... ii
LIST OF ABBREVIATIONS............................. iv
LIST OF FIGURES................................... x
LIST OF TABLES ................................... xi
Sections
INTRODUCTION..................................... 1
Perceptual Style
The Field Dependence—
Field Independence Dimension ................ 1
Definition and Measurement ................... 1
Pervasiveness ............................... 6
Stability ......................... ..... 10
Sex Differences............................. 15
Origin....................................... 20
Relation between Perceptual Style and
Achievement Motives and/or Behavior ........ 21
Sex Differences in Independence Training .... 25
Statement of the Problem....................... 27
METHOD............................................ 34
Subject Selection and Classification .......... 35
Measuring Instruments ......................... 42
The Measures of Perceptual Style ....... 42
The Embedded Figures Test (EFT) .......... 42
The Rod-and-Frame Test (RFT).............. 44
The Measures of Parental Expectancies .... 46
The Age-Independence Scale (AIS) .......... 46
v
Sections
Page
Description . . 46
Construction and Revisions .............. 47
Administration ........................... 50
Reliability............................. 51
Scoring................................. 51
The Block Stacking Task (BST)............ 52
The Measure of Achievement Motivation .... 56
The Ring Toss Task (RTT)................... 56
Procedure..................................... 59
Data Treatment and Analysis................... 63
RESULTS AND DISCUSSION........................... 67
Hypothesis la................................. 68
Hypothesis lb................................. 71
Hypothesis lc................................. 75
Discussion of General Hypothesis 1 ............ 77
Hypothesis 2a................................. 86
Hypothesis 2b............ 88
Hypothesis 2c................................. 89
Discussion of General Hypothesis 2 ............ 89
Relation between Hypothesis 1
and Hypothesis 2 ............................. 96
Hypothesis 3 ................................... 101
Hypothesis 3a................................. 102
Hypothesis 3b................................. 106
Hypothesis 3c................................. 108
Discussion of General Hypothesis 3 ............ 109
The AIS as a Measure of Parental Expectancies. . 114
The BST as a Measure of Parental Expectancies. . 124
Hypothesis 4 ................................... 128
Relation between Measures of
Children's Achievement ....................... 140
BST and School Achievement Tests ............ 140
RTT and School Achievement Tests............ 142
BST and R T T ................................. 145
vi
Sections Page
OVERVIEW........................................... 148
BIBLIOGRAPHY ....................................... 158
APPENDICES......................................... 169
vii
TABLE OF CONTENTS
Page
APPENDIX
A FIGURES....................................... 169
B TABLES....................................... 172
C FORMS ....................................... 220
Initial Letter to Parents ................... 221
Information Sheet ........................... 222
Authorization for Participation
in USC S t u d y ............................. 223
Final Letter to Parents..................... 224
Research Summary ........................... 225
D MEASURES OF PERCEPTUAL STYLE ................ 228
EFT Simplified Directions ................. 229
EFT Procedural N o t e s ................ 232
EFT Record Sheet ..... 233
Portable RFT - Directions for Children . . . 234
RFT Record Sheet ................... 237
E THE AGE-INDEPENDENCE S C A L E .................. 238
Verbal Directions ........................... 239
Direction Sheet ............................. 241
The Age-Independence Scale ................. 242
Summary P a g e ............................... 245
AIS Items Listed by Subscale Category .... 246
F AGE-INDEPENDENCE SCALE DATA.................. 249
Parental AIS Item Means..................... 250
Parental AIS Subscale and Overall Means . . . 253
Tabulation of Parental "Especially
Important to Me" AIS Responses.......... 254
Tabulation of Parental "Not Appropriate"
AIS Responses............................. 257
viii
Page
APPENDIX
G MEASURES OF ACHIEVEMENT....................... 259
BST Directions............................. 260
BST Record S h e e t ........................... 263
RTT Directions............................. 264
RTT Record S h e e t ........................... 266
LIST OF FIGURES
Figure Page
1. School Achievement Relative to the Group
and I.Q. for Children in the Subject Pool . 170
2. School Achievement Relative to Individual
Ability and I.Q. for Children in
the Subject Pool........................... 171
x
LIST OF TABLES
Table Page
1. Background Information on Children
Used in S t u d y ............................ 173
2. Background Information on Mothers
Used in S t u d y ............................. 175
3. Background Information on Fathers
Used in S t u d y ............................ 176
4. Age Independence Scale Means for Parents
of High and Low Achievers................ 177
5. Analysis of Variance of Age Independence
Scale Overall Means for Parents of
High and Low Achievers and Boys and Girls. . 178
6. Multivariate Analysis of Variance of Age
Independence Scale Subscale Means for
Parents of High and Low Achievers
and Boys and Girls......................... 179
7. Mean Block Stacking Task Scores for
Parents of High and Low Achievers........ 180
8. Multivariate Analysis of Variance and
Covariance of Block Stacking Task
Scores for Parents of High and Low
Achievers and Boys and Girls.............. 181
9. Mean Block Stacking Task Estimates for
Trials # 2 + 3 of Mothers and Fathers
of High and Low Achievers................. 183
10. Correlations between Age Independence
Scale and Block Stacking Scale Scores
for Parents of Boys and G i r l s............ 184
xi
Table Page
11. Number of Age Independence Scale Items
Marked "Especially Important to Me" by
Parents of High and Low Achievers........ 185
12. Number of Age Independence Scale Items
Marked "Not Appropriate" by Parents of
High and Low Achievers..................... 186
13. Tabulation of Responses to Age
Independence Scale Summary Page for
Parents of High and Low Achievers........ 187
14. Differences between Fathers and Mothers
Age Independence Scale Means for
Parents of High and Low Achievers........ 188
15. Analysis of Variance of Age Independence
Scale Differences between Fathers and
Mothers Overall Means for Parents of
High and Low Achievers and Boys and Girls. . 189
16. Multivariate Analysis of Variance of
Differences between Fathers and Mothers
Age Independence Scale Subscale Means
for Parents of High and Low Achievers
and Boys and Girls......................... 190
17. Age Independence Scale Means for Parents
of Boys and G i r l s ......................... 191
18. Mean Block Stacking Task Scores for
Parents of Boys and G i r l s ................. 192
19. Number of Age Independence Scale Items
Marked "Especially Important to Me" by
Parents of Boys and G i r l s ................. 193
20. Number of Age Independence Scale Items
Marked "Not Appropriate" by Parents
of Boys and G i r l s ........................ 194
21. Tabulation of Responses to Age
Independence Scale Summary Page for
Parents of Boys and G i r l s ................. 195
xii
Table Page
22. Differences between Fathers and Mothers
Age Independence Scale Means for
Parents of Boys and G i r l s ................. 196
23. Correlations between Parental Age
Independence Scale Means and Their
Children's Perceptual Style Scores ........ 197
24. Correlations between Parental Block
Stacking Scale Scores and Their
Children’s Perceptual Style Scores ........ 198
25. Correlations between Children's Perceptual
Style Scores and the Differences between
Their Fathers and Mothers AIS Means .... 199
26. Analysis of Variance of Differences between
Fathers and Mothers on Age Independence
Scale Overall Grand Mean................... 200
27. Multivariate Analysis of Variance of
Differences between Fathers and Mothers
on Age Independence Scale
Subscale Grand M e a n ....................... 201
28. Correlations between Parental Age
Independence Scale Means and Their
Children's Ages and Total--I.Q. Scores . . . 202
29. Correlations between Boys and Girls
Perceptual Style Scores and Their Ages,
I.Q., and School Achievement Test Scores . . 203
30. Correlations between Parental Block
Stacking Task Scores and Their Children's
Ages and Total — I.Q. Scores.............. 204
31. Mean Perceptual Style Scores for
High and Low Achieving Children.......... 205
32. Multivariate Analysis of Variance and
Covariance of Perceptual Style Scores
of High and Low Achievers and
Boys and Girls............................ 206
xiii
Table Page
33. Multivariate Analysis of Variance and
Covariance of Perceptual Style Scores
of High and Low Achieving G i r l s .......... 208
34. Mean Perceptual Style Scores
for Boys and Girls......................... 209
35. Mean Block Stacking Task Scores
for High and Low Achievers................. 210
36. Mean Block Stacking Task Scores
for Boys and Girls......................... 211
37. Multivariate Analysis of Variance and
Covariance of Block Stacking Task
Scores of High and Low Achievers
and Boys and Girls......................... 212
38. Mean Ring Toss Task Scores
for High and Low Achievers................. 215
39. Mean Ring Toss Task Scores
for Boys and Girls......................... 216
40. Multivariate Analysis of Variance and
Covariance of Ring Toss Task Scores
of High and Low Achievers
and Boys and Girls......................... 217
41. Correlations between Children's
Ring Toss Task and Block
Stacking Task Scores ..... ............. 219
xiv
INTRODUCTION
The purpose of this research was to focus on the
impact of childrearing practices on a child's perceptual
style and school performance. Of particular interest was
the extent to which parents1 expectations for independence
behavior changed with the child's sex and the influence
those expectations may have had on the perceptual field
independence--field dependence dimension. Also of interest
were the relationships between parental expectations and
school performance and between school performance and field
dependence in elementary school-age children.
Perceptual Style
The Field Dependence— Field Independence Dimension
Definition and Measurement
The field dependent--field independent dimension of
perceptual style, as described by Witkin, Dyke, Faterson,
Goodenough, & Karp (1962) and Witkin, Lewis, Hertzman,
Machover, Meissner, & Wapner (1954), is of interest for
several reasons. First of all, it can be clearly and
objectively defined on the basis of reliable laboratory
tests. The three perceptual tasks most widely used by
Witkin are: the Body-Adjustment Test (BAT), the Rod-and-
1
Frame Test (RFT), and the Embedded-Figures Test (EFT).
Each of these three experimental situations requires the
subject to keep an item (his body, a luminous rod, a
geometric design) separate from the context (perceptual
field) of which it is a part. The extent to which a
person achieves this separation is a measure of his
independence from, or dependence on, the perceptual field,
and thus a measure of his perceptual style.
Individuals who can break up "the field" and over
come embedding contexts on these perceptual tasks are
designated field independent by Witkin and considered to
have an "analytical field approach." However, individuals
who are unable to keep an item separate from its context
and submit to the dominant organization of the field are
characterized as field dependent and considered to have a
"global field approach." Although only the two divergent
perceptual styles represented by the extremes of the
continuum have been discussed, scores for any large group
show a continuous distribution (Neville, Workman, &
Johnson, 1969; Stuart & Murgatroyd, 1971; Vaught, 1968,
1970; Witkin, 1965; Witkin, et al.. 1954, 1962). The
perceptual style of these intermediate scorers tends to be
more variable and has not been as clearly defined.
Furthermore, the particular mode of field approach
an individual displays on one task will tend to be used on
all others. According to Witkin (1965, pp. 318-319):
3
People tend to perform in a consistent fashion in
these three tests. The individual who cannot separate
the simple figure from the complex embedding design
also cannot keep his body apart from the surrounding
tilted room or the rod apart from the surrounding
frame.
This consistency in performance from one perceptual task
to the other is supported by the extensive longitudinal
and cross-sectional research of Witkin and his associates.
Witkin, et al. , (1962, p. 44) indicate the following
correlations (statistically significant at the .05 to .01
levels) among the three tests: .39 and .43 (men) and .39
and .55 (women) on the BAT and RFT; .54 (men) and .58
(women) on the BAT and EFT; and .63 and .64 (men) (the
correlation for women did not reach the 5% level of
statistical significance) on the EFT and RFT. Other
investigators (Gruen, 1951, 1955; Linton, 1952) have
reported similar findings (cited in Witkin, et al.. 1962,
p. 44).
The reliability of these tasks has also been
extensively studied. Witkin, et al.. (1962, p. 40) report
test-retest correlations for a three-year interval of:
.84 (men) and .66 (women) on the RFT; .77 (men) and .74
(women) on the BAT; and .89 (men and women) on the EFT.
Split-half or odd-even item reliabilities on the EFT or
RFT tend to be even higher.
However, methodological issues do exist. In their
1954 book, Witkin, et al., point out that the EFT differs
from the other tests in that the S knows immediately
whether he has succeeded or failed. In their 1967 paper,
Witkin, Goodenough, & Karp mention that a practice or
learning effect is clearly evident on the EFT, making it
less useful for longitudinal studies. Goldstein & Chance
(1965), in a study specifically designed to measure the
effects of practice on EFT scores, found that all Ss (13
college men and 13 college women) showed improvement with
practice. After reviewing the original correlation
between the RFT and EFT (.76 for men and .26 for women)
obtained by Witkin, et al., (1954), Thornton & Barrett
(1967) point out that the two tests have very little
common variance for women. Thus, p. 631: "If the RFT is
operationally defined as a measure of field independence,
then it is clear that for women the EFT is a completely
unacceptable substitute test. Even for males it is quite
possible that the two perceptual tests are measuring two
separate factors, ..." As a result of their work with a
portable RFT and short-form EFT in 203 college men, Gross
& Moore (1970) suggest using the combined scores of both
measures, since this index would provide a multi-method
assessment of the field independence dimension. However,
based on the score distributions they obtained for these
two tasks, Gross & Moore seem to prefer the use of the
short-form EFT to that of the portable RFT for the male
college population. This preference appears contrary to
the stated preference of both Witkin, et al., (1967) and
Thorton & Barrett for the RFT as the best single indicator
of perceptual style.
The reliability of RFT scores as an indicator of
the mode of perceptual functioning has also been called
into question by a number of investigators. The comments
of Lester (1968) on control of head position, starting
position (of the rod relative to the frame) effects,
control readings (of vertical for each S), and instruction
effects are especially relevant. Also, Schneider &
Bartley (1962) found that mechanically induced tension of
the neck muscles significantly affected perception of the
vertical. Similarly, Sandstrom (1954) found that even
small tilts of the head changed an S's perception of the
position of a luminous line. Cohen & Tepas (1958),
following a somewhat different line of research found that
prolonged (four minutes or more) exposure to a tilted rod
and frame significantly increased the errors in judgments
of verticality.
In addition to these methodological issues, two
other individual variables have been identified which can
make RFT scores an unreliable index of field dependence.
The first is anisometropia, a condition in which the two
eyes differ in visual acuity. Culver (1967) found that
anisometropic Ss were considerably poorer in RFT
performance than Ss with normal acuity. The second
variable which may make RFT scores erroneous is the extent
of perceptual scanning. Silverman & King (1970) have
pointed out that accurate RFT scores can be obtained
either by constricting the field of attention so as to not
register, or only minimally register, background cues or
by actively inhibiting attention to background cues which
have been allowed to register. Only the extensive scanning
and differentiation of the perceptual field reflected in
the second alternative would be considered indicative of
field independence behavior.
In spite of the methodological questions raised
and the inappropriateness of one or the other of these
tests for a particular individual or a particular study,
the bulk of the research tends to indicate that both the
RFT and the EFT can give reliable measures of the field
independence--field dependence dimension.
Pervasiveness
A second reason perceptual style is of interest is
that it appears to relate in a meaningful and coherent way
to broad and highly diverse psychological areas ranging
from the intellectual sphere to that of personality and
even that of pathological functioning. In the
intellectual area, work by Witkin and his associates
(1954, 1962) indicates that field independent people
(children and adults) score significantly higher on
intelligence tests requiring "analytical competence"
(particularly the Block Design, Picture Completion and
Object Assembly subtests of the Wechsler), but do not
score significantly better than field dependent people on
tests requiring verbal abilities (specifically the
Vocabulary, Information and Comprehension subtests). A
replication of this work by Karp, Silberman, & Winters
(1969) in children supports Witkinfs findings. Studies by
others have not used the Wechsler as a measure of
intellectual ability and, thus, are difficult to compare
directly with Witkin's reports.
Using cross-sectional and longitudinal studies,
Witkin and associates have also found perceptual style
related to personality measures. Body concept (field
independents tend to score lower on the BAT and to draw
more highly articulated figures), sense of separate
identity (field independents tend to use internal and
field dependents external sources for definition of their
attitudes, judgments, sentiments, and views of themselves),
and defensive structure (field independents tend to use
isolation and intellectualization, while field dependents
use repression and denial) were the variables studied.
Work by Adevai, Silverman, & McGough (1968) on the
correlates of the RFT, Minard & Mooney (1969) on
perceptual defense, Weinberg (1970) on responses to social
influences, and numerous others tend to confirm these
relationships.
In addition to the relationships between
perceptual style, intellectual abilities and personality
characteristics already described, Witkin (1965) has
related perceptual style to pathological functioning. He
feels the field independence dimension is not related to
the presence or absence of pathology or to the adequacy of
adjustment, but rather to the kinds of problems and
symptoms which will be found when a person becomes
disturbed. Field dependents tend to have problems of
identity, dependence, inadequately developed controls,
passivity and helplessness. Clinically they are most
likely to be: alcoholics, obese, asthmatic, have ulcers,
or in the functional sphere, to exhibit hysterical
character structures, character disorders, functional
cardiac disorders, and catatonic or hallucinatory symptoms.
Field independents, on the other hand, tend to show
delusions, expansive and euphoric ideas of grandeur, out
ward direction ofaggression, overideation, and a
continuing struggle for the maintenance of identity when
they break down. Clinically, disturbed field independents
tend to be paranoids, obsessive-compulsive characters,
neurotics with organized symptom pictures, or ambulatory
schizophrenics with well-developed defensive structures.
Most of the research cited so far has originated
in the Witkin group. However, the literature is filled
with attempts to extend and further clarify the
implications of the field independent-field dependent
dimension. Only a few representative studies are listed
to give an idea of the range of psychological functioning
covered. In the perceptual sphere, research has been
extended to the relation between field dependence and
autokinesis (Vaught & Hunter, 1967); form discrimination
(Vaught & Ellinger, 1966); figural aftereffects
(Immergluck, 1966a, 1966b, 1970); the Stroop Color-Word
Test (Huckabee & McGown, 1971); eye movement patterns
(Conklin, Muir, & Boersma, 1968); eye dominance (Oltman &
Capobianco, 1967); hand preference (Silverman, Adevai, &
McGough, 1966); and auditory functioning (De Fazio &
Moroney, 1969). Along a more cognitive line, Bloomberg,
1967) and Spotts & Mackler (1967) attempted to establish a
connection between creativity and the field independence
dimension. In the area of pathology, Scallon & Herron
(1969) found that enuretic boys tended to be field
dependent. In the physiological sphere, investigators
have attempted to relate field dependency to: response to
insulin (Silverman, McGough, & Bogdonoff, 1967); autonomic
activity (Courter, Wattenmaker, & Ax, 1965; Goldstein,
Pardes, Small, & Steinberg, 1970; Hustmyer & Karnes,
1964); and cardiac conditioning (Hein, Sanford, Cohen, &
Shmavonian, 1966).
Stability
The third reason that perceptual style is of
interest is that it appears to be an exceptionally stable
characteristic of personality throughout a lifetime.
Individual differences in field dependence on the RFT have
been found as early as four years six months by Gill,
Herdtner, & Lough (1968) and four years eleven months by
Busch & De Ridder (1971). Schwartz & Karp (1967) found
differences on the RFT as late as age eighty, and Karp
(1967) found them on the EFT at age ninety-two. Through
out childhood and the growth years perception tends to
progress from a global, undifferentiated state to a more
articulated, discrete level of functioning. The field
dependent-field independent curve also follows this
developmental pattern. "Young children tend to perceive
in relatively field-dependent fashion, and, as they grow
older, their perception assumes a generally more field-
independent form." (Witkin, et al., 1962, p. 7) Based on
their own research and that of others, Witkin, et al.,
(1967) see the developmental curve as progressing toward
and then away from field independence, depending on the
particular age group studied. Specifically, findings
indicate that:
1. The trend toward field independence is
progressive from age five to seventeen;
2. from age seventeen to at least age twenty-
11
four, the development of field independence
tends to approach a plateau; and,
3. at some point between age twenty-four and old
age, a return to field dependence begins.
In spite of these development changes in perceptual style,
the relative stability of this dimension within
individuals over extended periods of growth and changed
life circumstances is impressive. Witkin, et al., (1967),
p. 300) conclude their discussion on this stability and
individual continuity in style as follows:
The 14-year span from age 10 to age 24 covered by
our most extended longitudinal study is of course a
time of great personal change. This period covers the
turbulent events of adolescence, the breaking of close
family ties, psychosexual maturation, occupational
commitment, and sometimes marriage and the assumption
of family responsibility. The psychological impact of
all this upon the developing person is of course very
great and inevitably contributes to important changes
in personal functioning. There is no doubt that many
significant psychological changes did take place in
our subjects over the 14-year period we studied them.
Even with these changes, we still find continuity in
relative level of differentiation over time.1
Numerous attempts to challenge the stability of
the field dependence dimension and alter it experimentally
have been made. Witkin, et al., (1962) report that
neither drugs (sodium amytal, dexedrine, chlorpromazine,
• ' • I n this passage "differentiation" can best be
defined as "perceptual style." The concept of differentia
tion emerged gradually from Witkin!s work and its
evolution is briefly summarized in the following section.
12
imipramine or alcohol) nor electroconvulsive shock
affected performance on the RFT. In a later study Karp,
Witkin, & Goodenough (1965) found that RFT and BAT scores
did not change in alcoholics given alcohol, but that
scores grew worse on the EFT. Using non-alcoholic Ss,
Kristofferson (1968) found that RFT scores showed an
increase one hour after consumption of alcohol. However,
it is assumed this change lasted only as long as the
effects of the alcohol, since no further testing was
reported.
From the time Witkin first found the individual
differences in perceptual style, he assumed these differ
ences must reflect the underlying psychological organiza
tion of the person. In his earlier work he tended to
characterize the aspect of the underlying personality
relative to the FD-FI dimension as a ". . . tendency
toward active coping with or passive submission to the
environment, . . ." (Witkin, et al., 1954, p. 489). Later,
Witkin started using the term "differentiation" in dis
cussing this underlying personality structure. He defined
the differentiation process as, "progress from an initial
relatively unstructured state, which has only limited
segregation from the environment, to a more structured
state, which has relatively definite boundaries, and which
is capable of greater specificity of function." (Witkin,
et al.. 1962, p. 22). He also postulated several
"indicators" of differentiations an articulated way of
experiencing the world; a differentiated self as reflected
particularly in a clearly defined body concept and a
developed sense of separate identity; and specialized,
structured controls and defenses. In discussing these
"indicators" of the level of differentiation further,
Witkin (1965, p. 323), suggests that perceptual style can
be considered a "tracer element" in the assessment of the
underlying psychological differentiation. Thus, in
addition to this evolution in the conceptual basis of
Witkin's thinking, the emphasis of Witkin's research
13
Transient changes in perceptual test performance
have also been obtained by other means. Sensory
deprivation has been found to increase field independence
on the RFT (Astrup, 1968; Jacobson, 1966, 1968; Kurie &
Mordkoff, 1970), while either decreasing (Heron, 1961) or
not changing on the EFT (Culver, Cohen, Silverman, &
Shmavonian, 1964; Kurie & Mordkoff, 1970). Stress has
also been found effective in altering RFT scores, Hill &
Feigenbaum (1966) used a threat to self-esteem as the
stressor and produced an increase in RFT scores. On the
other hand, Wolf (1965), using body rotation, produced the
opposite effect, a reduction in RFT scores. Another
method, which has proven successful in temporarily
increasing RFT scores, has been creating a set for uncer
tainty in the test situation. Gross (1959) was able to do
this by suggesting that a plain piece of glass placed
between the experimental Ss and the rod-and-frame was a
lens which would distort their vision in unknown ways.
Efforts to alter the field dependence reflected in
perceptual tests on a more permanent basis by training
have also been reported. Witkin, et al., (1962, pp. 372-
373) cite three different studies involving special
training. First, in their own early work (Witkin, 1948),
appears to have shifted from a study of perception and its
relation to personality to that of personality as assessed
by perception.
14
discussions of problems of orientation and practical
demonstrations in which the S participated produced
improved scores on the test in which the training was
given, but not on different kinds of perceptual tests.
Second, a preliminary study (using only four Ss) by
Schneck & Wexler found that suggestions made under
hypnosis to change Rorschach responses were effective in
the post-hypnotic period while those dealing with
performance on the perceptual tests were not. The third
study cited (Gruen, 1955) compared the perceptual
performances of a group of professional dancers with that
of college men and that of a less selected group of Ss.
No significant differences were found on the EFT and RFT;
differences were found between the dancers and college
group on the BAT, but not between the dancers and the less
selected group. Thus, Witkin concludes that though
special training may affect performance on a particular
perceptual test, perceptual style itself is not altered
and there is little transfer of training between tests.
Other efforts to influence the field dependence
dimension through training have been made. Elliot &
McMichael (1963) were able to improve RFT performance in a
group of Ss who had made large errors on an initial RFT by
lecturing them on the use of kinesthetic cues in
discerning the vertical as well as giving them feedback on
subsequent trials. However, upon being retested four to
15
seven weeks later they had returned to their former level
of performance--obtaining a test-retest correlation of .99.
Olson, Diehm, & Elfner (1965) and Weiner (1955) were able
to improve Ss RFT performance through training, but
significantly, reported no retest scores except those
taken during the trials immediately following training.
After a three-month interval, Sondheimer (1970) reported
no change in RFT scores for two groups of physically
disabled patients (one on medication and the other
involved in physical therapy) in spite of changes in their
physical and psychological functioning. Thus, although
transient fluctuations in perceptual scores have been
obtained, current research tends to indicate that
perceptual style is a relatively stable personality
characteristic.
Sex Differences
Throughout the voluminous research on perceptual
functioning, women tend to be more field dependent than
men. "The difference between the sexes is small in
magnitude, compared to the range of individual differences
within each sex, but is clear-cut and pervasive. More
over these sex differences are evident over a large
segment of the life span, . . ." (Witkin, 1967, p. 243).
This difference in perceptual style is already quite clear
by age eight in children (Witkin, 1967; Witkin, et al.,
16
1954, 1962, 1967) and has been found in some children as
early as age four and one-half (Gill, Herdtner, & Lough,
1968). While progressing along the developmental curve
with its trend toward increasing field independence with
increasing age, this difference between the perceptual
scores of boys and girls tend to remain constant. Between
ages fifteen and seventeen the mean scores of the sexes
become more divergent. Boys tend to continue or
accelerate their progression toward field independence,
while girls reverse the developmental pattern and become
more field dependent. As mentioned earlier, these
developmental changes in field dependence tend to level
out in early adulthood (around age twenty-four) and remain
constant until old age. Thus, though the developmental
changes in field dependence have ceased with increasing
maturity, the discrepancy in the perceptual functioning of
the sexes remains throughout adulthood and into old age
(Schwartz & Karp, 1967).
Two basic explanations, the inherent biological
differences between the sexes and the culturally imposed
differences in sex roles, are cited for the greater field
dependence of women. Based on their own work and an
extensive review of sex differences in abilities in people
and lower animals as well as studies of drug and hormone
administration on behavior, Broverman, et al., (1968)
hypothesize that the differences in abilities are
17
reflections of sex-related differences in physiology.
Quoting from their abstract, p. 23: "It is hypothesized
that these sex differences are reflections of differences
in relationships between adrenergic activating and
cholinergic inhibitory neural processes, which, in turn,
are sensitive to the "sex" hormones, androgens and
estrogens." Shafer (1962; cited by Witkin, Faterson,
Goodenough, & Birnbaum, 1966) in a study of girls with
Turner's syndrome (these girls have only one X chromosome
instead of the usual two in normal women) found that
eighteen of the twenty girls showed a discrepancy between
their verbal and performance scores on the WISC in favor
of the verbal score. In discussing Shafer's results,
Witkin, et al.t (p. 315) suggest that:
. . . the possibility of a linkage between sex
chromosomal makeup and cognitive characteristics
(whatever the mechanism by which the chromosomes exert
their influence on cognitive development) is of
special interest in view of the repeated finding that
men are consistently superior to women in analytical
ability (Witkin, et al., 1962) and that women are
consistently superior to men in verbal ability (Tyler,
1956).
Olson, Diehm, & Elfner (1965), in discussing the
significantly greater number of errors on the RFT produced
by women in their study, suggest that this difference in
performance may be due to innate factors. However, they
offer no support or further explanation for this
suggestion. In addition to the differences observed in
abilities, Anastasi (1958) points out the extensive
18
differences between boys and girls in physical features,
activity level, vital capacity, and developmental
patterns.
However, Witkin and a number of other
investigators also hypothesize that the field dependence
of women as a group is probably a result of culturally
determined differential learning experiences. Witkin,
et al.. (1954) emphasize the differences in social roles
assigned to men and women along the dimension of activity--
passivity. They feel women are encouraged to be dependent
and conforming, while men are encouraged to be active and
responsible. Sherman (1967) tends to attribute the clear
sexual differences in spatial perception to differential
learning and practice opportunities. This view is
supported by the work of Goldstein and Chance (1965). In
their study the usual (and initially observed) differences
in EFT scores of college men and women were eliminated by
practice in a long series of trials. Vaught (1965), in
discussing the results of his work on the relationship
between role identification and ego strength to sex
differences in the RFT, says that the attributes of the
male role (in contrast to the female role) imply learning
in a prescribed direction which overlaps with the demands
of field independence performance on the RFT. Lynn (1969)
suggests a somewhat different type of learning as the
origin of the predominant relationship between mode of
19
perceptual functioning and sex. He hypothesizes that
children (in our culture usually boys) who must struggle
to define their sex-role (because a same-sex model is
frequently not available) become "problem solvers" and,
therefore, field independent. However, other children
(usually girls) do not develop these problem solving
skills, since they can acquire their sex-role identifica
tion simply by imitating the available same-sex model
(mother). Thus, these children (girls) tend to be more
field dependent.
All of these explanations involving differential
learning experiences, but particularly that advanced by
Lynn, help to explain another facet of the sex differences
in perceptual style— the tremendous overlapping in the
score distributions of men and women. Though in our
culture children may usually receive differential training
based on biological sex to better equip them for their
differing social roles in adult life, this is not always
or even necessarily the case. Thus, girls who, for
example, are brought up to be "problem solvers" may well
be considerably more field independent than boys who are
merely "imitators." However, the role these differential
learning experiences may play in contributing to sex
differences in perceptual style has yet to be established.
The hypotheses advanced are interesting, but generally
untested.
20
Origin
Why are some children more field dependent than
others, even at ages four and five? How do these
differences originate? To what extent do different
parental practices affect the perceptual styles of their
offspring? Witkin and his co-workers (Dyk & Witkin, 1965;
Witkin, 1967; Witkin, et al.. 1962) have attempted to
answer these questions. Most of their work has been with
mothers and their sons and has implicated two major
factors in the rearing of field independent children:
1 . the extent to which the mother has encouraged the
child to separate from her (trained the child for
independence) by giving physical care appropriate
for his age, not limiting the child's activities
or ventures into the community, not limiting
curiosity, not stressing conformity and generally
exercising control toward the child's achieving
mature goals and becoming responsible; and
2 . the extent to which the mother has helped the
child develop impulse control by exercising
control toward the child's asserting himself.
Other than the work done by the Witkin group and a study
(Seder, 1957) using both boys and girls, cited by Witkin,
et al., (1962, pp. 351-355) as confirming their findings,
little research on the origin of perceptual style is
reported in the literature. Berry (1966), in a study of
the Eskimos, found them remarkably field independent as a
group. He suggested that one reason for this might be the
way in which the Eskimos live and raise their children.
Their children (both boys and girls) are given strong
21
encouragement for personal self-reliance, individualism,
skill, and ingenuity, and discouraged from dependence and
incompetence. In addition, the Eskimo language includes a
complex system of geometric-spatial terms, which could
help children achieve an articulated concept of space.
Thus, the available research indicates that parental
childrearing practices do influence perceptual style and
that parents who train their children to function in an
independent manner tend to have field independent children.
Relation between Perceptual Style and
Achievement Motives and/or Behavior
Are field independent children also more highly
motivated to achieve? The type of training found relevant
to field independence in children appears to be remarkably
similar to that implicated in the development of a strong
need for achievement (n-Ach). McClelland, et al., (1953,
p. 276) said: ". . . individuals with high achievement
motivation will have been forced to master problems on
their own more often and earlier than individuals with low
achievement motivation." and (p. 328): "... achievement
motives develop . . . where there is an emphasis on the
independent development of the individual." In a study of
middle-class, eight-year-old boys and their mothers,
Winterbottom (1958) found achievement motivation strongest
among boys whose mothers expected relatively early signs
22
of self-reliance and mastery from them. She found that
mothers of high n-Ach boys:
1 . make more demands at an earlier age;
2 . evaluate their sons1 accomplishments higher and
are more rewarding;
3. make more early restrictions and fewer later on;
and
4. make more demands than restrictions, overall.
Rosen & D’Andrade (1959), in an extensive experimental
study of the parent-child interaction observed in problem
solving situations, came to similar conclusions. They
found that parents (both middle and lower class) of boys
(nine-, ten-and eleven-years-old) with high n-Ach tended
to:
1 . have higher aspirations for their son to do well
at any given task;
2 . have a higher regard for his competence at
problem solving;
3. set up standards of excellence; and
4. react to his performance with warmth and approval,
or with disapproval (mothers only) if he performs
poorly.
The work of Whiting (1970) comparing expectations for
independent behavior of the parents of achieving and non
achieving elementary-school boys also supports these
findings and extends them to school achievement. Whiting
found that parents of achieving boys expected mastery at
an earlier age than did parents of non-achievers, chose to
23
emphasize different developmental goals for their sons,
were in greater agreement about childrearing practices,
and were less involved in their sons1 homework.
On the basis of these similarities in childrearing
patterns, it seems reasonable to hypothesize that an
achieving child is also probably field independent. How
ever, studies in children have not clarified this
relationship. A study by Elkind (1969) in fifth-grade
boys found that field independence (as measured by the
Children's Embedded Figures Test) and sense of responsi
bility for achievement (measured by the Intellectual
Achievement Responsibility Scale) were essentially
unrelated. Using elementary school children (ages six to
twelve), Crandall & Sinkeldam (1964) found that children
who tended to be achievement oriented (as measured by
concern with mastery of fine motor skills, task
persistence, time spent alone on tasks, and independent
achievement efforts) in social situations displayed more
field independence (as measured by the EFT).
Studies in adults on the relationship between
field independence and n-Ach have not produced such clear
results, either. Greenwald (1968), using the EFT and the
Thematic Aperception Test (TAT) in a group of women, found
no support for a relationship between field independence
and high need for affiliation and none for field
independence and high n-Ach. Wertheim & Mednick (1958),
24
using the same tests in college men (11) and women (31),
found high n-Ach and field independence significantly
related (r = -.40, p <.0l). In a replication of Wertheim
& Mednick's work using fifteen male and thirteen female
college students, Honigfeld & Spigel (i960) found the two
variables related in women (.42, 11 df), but quite
questionable (-.12, 13 df) in men. Commenting specifically
on these last two studies and more generally on others in
the literature, Thornton & Barrett (1967) indicate that no
relationship between field independence and n-Ach has yet
been established. They point out that the EFT is not a
valid measure of field independence in women (see earlier
section on reliability of perceptual tests) and cannot be
used to study the relationship between field independence
and n-Ach. They also point out that though the EFT is "an
approximate index of field independence" in men, it has
not been found related to n-Ach. Thus, it seems
impossible to reliably assess the relationship between
these variables unless the RFT is used as the measure of
field independence. Two studies are reported which do use
the RFT; however, both studies used the n-Ach scale from
the Edwards Personal Preference Scale as an index of n-Ach
^The difference in the signs of these two
correlations is an artifact of the statistical procedure
used rather than a meaningful reversal in the direction of
the relationship.
rather than some form of the TAT. McFall & Schenkein
(1970) report that a stronger correlation was found
between the RFT and n-Ach scores (r = .27) than between
the Concealed Figures Test (similar to the EFT) and n-Ach
(r = .07) in college women. Using male undergraduates,
Gruenfeld & Arbuthnot (1968) found that field independents
did tend to be task- and achievement-oriented, while field
independents preferred "loyalty, friendship, kindness and
sympathy to task accomplishments requiring skill and
effort" (j: = .27, p < .05). Though these last two studies
do tend to indicate that achievement orientation and field
independence on the RFT are somewhat related in college
men and women, whether or not this relationship exists in
children is still an unanswered question. Certainly, if
early independence training can be considered an
antecedent to both high n-Ach and field independence, and
the literature indicates it can, field independent children
should also be high in n-Ach.
Sex Differences in Independence Training
Since independence training does seem to be
implicated in the development of field independence and
n-Ach, and both of these characteristics seem more
prevalent in males than in females, is there some evidence
that boys are brought up differently than girls? Do boys
receive different kinds or different amounts of
26
•'independence training" than girls? Are the behavioral
expectations, particularly those dealing with independence
behavior, different for boys than they are for girls?
Research findings tend to answer all three
questions positively. From birth on, boys and girls look
different, act differently (boys are more active,
aggressive, impulsive), progress at different rates (girls
learn to talk and read somewhat earlier), and are treated
differently (Baumrind, 1970; Baumrind & Black, 1967;
Kohlberg, 1966; Lewis, 1972; Maccoby, 1966; Sears, et al.,
1953). In our culture, as well as in most others (Barry,
Bacon, & Child, 1969; D'Andrade, 1966), expectations for
the sexes are decidedly different (Aberle & Naegele,
1952; Biller & Weiss, 1970; Dameron, 1955; Katkovsky,
Preston, & Crandall, 1964; Lucco, 1967; Maccoby, 1969;
Milner, 1949; Mussen, et al.. 1970; Torgoff, 1958). Boys
are brought up to be the leaders, household heads, and
family providers and protectors, while girls are brought
up to be wives, mothers, and to be generally proficient in
domestic, social, and caretaking activities. Thus, to the
extent these cultural roles prevail, behavioral
expectations and childrearing practices are different for
the two sexes. Boys are encouraged to be active,
independent, self-reliant, competitive, and achieving,
while girls are encouraged to be nurturant, responsible,
and (hopefully) attractive as well as passive, dependent,
27
and conforming. Therefore, it seems reasonable to
hypothesize both that there are differences in the expec
tations and socialization practices used for boys and
girls and that these differences would influence both
their achievement levels and their perceptual style.
However, though the implication is clear that boys
do receive more of some kinds of independence training
than girls, only one study (Gray, 1968) has actually tried
to measure these differences. Gray asked parents the ages
at which they expected the average boy or girl to show
3 ...
"independent mastery" of 100 common activities.
Statistically significant differences between the sexes
were found on only nine questions.
Statement of the Problem
To summarize and enlarge on the issues that have
been presented so far, perceptual style, once established,
is a remarkably stable, pervasive characteristic of the
personality--seemingly as resistant to change as
intelligence or sexual identity. How does one become
field independent or field dependent or somewhere in-
between these two extremes? Is this personality
•^The term "independent mastery" is used here and
throughout this research in the same sense as in the work
of Gray (1968) and Whiting (1970), i.e., mastery "without
instrumental help. It need not imply a lack of interest,
emotional support, or approval on the part of the adult."
28
characteristic inherent in the genetic makeup of an
individual like sex, body build, and hair color or is it,
at least partially, learned and, therefore subject to some
degree of change? The literature indicates that
perceptual style is influenced, at least to some extent,
by childrearing practices and that children who are
brought up to function in a competent, independent manner
tend to be field independent.
Even if perceptual style is initially subject to
change, what real difference does it make if someone is
field independent or field dependent? Is it better to be
field independent rather than field dependent, field
dependent rather than field independent, or are they
diverse, but equivalent ways of perceiving the world?
Neither of these questions can really be answered in any
absolute, general sense. But, because field independence
is viewed by many (particularly the Witkin group) as
further along the developmental ladder, rather than an
alternate but equivalent pathway, it is considered prefer
able. However, there is some research (Bloomberg, 1967;
Haronian & Sugerman, 1967; Rudin, 1968) which indicates
that some field independents are able to use either the
analytical mode of perception of the field independent
or the global mode of the field dependent, depending
on the circumstances. To the extent that field
independence does imply this kind of flexibility and
29
increased choice, it does seem more desirable.
Another interesting but relatively unexplored
facet of perceptual style is its relationship to achieve
ment motivation and behavior. The literature (most of the
studies have been done with boys only) indicates that both
field independent children and children who show high
n-Ach and/or achievement behavior tend to be brought up in
homes that make early demands for (and provide
opportunities for) independence, self-reliance, and
competence. Are field independent children, therefore,
also achievers? No clear answer has yet been provided to
this question.
A third point clearly evident from the literature
is that as a group and from an early age, males tend to be
both more field independent and more achievement-oriented
(in terms of motivation, behavior and/or actual
accomplishments) than females of equivalent ability
(Maccoby, 1966, 1969). Why? Can this differential
between the sexes in perceptual style and in achievement
also be attributed to the differential in parental demands
and expectations for independence behavior for boys and
girls suggested by the literature?
Thus, the available research suggests that parents
of: high and low achievers, boys and girls, and field
independent and field dependent children may differ
significantly in their expectancies for independence in
their children. Yet, the reports on parental
expectancies, though frequently conducted along parallel
lines, tend to be widely scattered in the literature,
since they are originated by different groups of
investigators seeking to explore seemingly diverse
subjects (i.e., n-Ach, perceptual style, and child
development). The purpose of this study is to examine the
relationships between parental expectancies and the n-Ach,
sex, and perceptual style of their children as well as
between perceptual style and n-Ach in the children, in an
effort to look at these questions simultaneously, rather
than as separate, non-interacting entities. The general
hypotheses advanced and the theoretical and/or
methodological issues they are addressed to are:
1. Parents of high and low achievers differ
significantly in the ages at which they
expect independent mastery of common
activities; specifically, parents of high
achievers expect more independent mastery
at an earlier age.
This hypothesis arises from several unresolved issues
suggested by the literature. First of all, several
studies (Rosen & D'Andrade, 1959; Whiting, 1970;
Winterbottom, 1958) have related early independence
training to increased achievement behavior or motivation
in elementary-school children. However, Chance (1961),
using essentially the same age children, questionnaire,
and procedure reported exactly the reverse. In her sample
31
she found that children whose mothers favored earlier
independence training made less adequate progress in
reading and arithmetic relative to their intellectual
ability than children whose mothers favored later
independence training.
In discussing her results, Chance explained the
discrepancy between her findings and those of Winterbottom
as due to the differences between actual achievement and
motivation to achieve. However, since the results of
Whiting’s study are based on actual achievement and yet
correspond to the earlier n-Ach literature, this
explanation appears inadequate. In an attempt to resolve
the conflict between Whiting's and Chance's results, as
well as that between n-Ach and actual achievement, this
study is in part a replication of Whiting's work (in that
essentially the same: questionnaire, subject selection
procedure, and age children will be used).
Another issue is that none of the previous studies
(other than the study by Chance) relating independence
training and achievement has included girls as subjects.
Thus, the question of whether independence training
relates to achievement in girls is almost completely
unexplored.
2. Parents of boys and girls differ significantly
in the ages at which they expect independent
mastery of common activities; specifically,
parents of boys expect more independent
mastery at an earlier age.
Although the literature clearly indicates that different
ial expectations for independent behavior exist for boys
and girls, only one study (Gray, 1968) has actually tried
to measure these differential expectancies. In general,
the findings were negative. However, this may have been
due to several methodological problems involved in her
design. In collecting her data on expectancies, Gray used
a lengthy questionnaire and then asked each respondent to
answer each item for both boys and girls. In addition,
she used groups of mothers and groups of fathers unrelated
to each other and asked each of them to answer each item
for the average child rather than for a particular child
with whom they were familiar.
3. Parents of field independent and field
dependent children differ significantly in
the ages at which they expect independent
mastery of common activities; specifically,
parents of field independent children expect
more independent mastery at an earlier age.
This hypothesis deals with two related questions: is
independence training related to the development of field
independence or field dependence? And, more generally, is
a particular perceptual style an inherent characteristic
of a person or is it initially subject to environmental
influence? Only a very limited amount of work has been
done on the origins of perceptual style and none of these
studies (other than Seder, 1957) asked for specific,
measurable responses. In addition, most of the work done
33
on origin of perceptual style was done by Witkin using
boys, but not girls.
4. High achievers are more field independent
than low achievers.
This hypothesis also tests a relationship which appears
reasonable but remains unclear. Only two studies (Crandall
& Sinkeldam, 1964, Elkind, 1969) of achievement behavior
and field independence have been reported in children--one
giving inconclusive results and the other indicating a
positive relationship. However, neither of these studies
used the RFT (generally considered a more valid measure of
perceptual style than the EFT), and only one study used
both boys and girls.
There are also few studies of this relationship in
adults and only two (Gruenfeld & Arbuthnot, 1968; McFall &
Schenkein, 1970) using the RFT. Both of these report a
low order correlation (r = .27) between field independence
and n-Ach.
Several other relationships, normally found in
studies of perceptual style, are stated in hypothesis form,
although they are not central questions of the study.
5. Boys are more field independent than girls.
6 . Field independence increases with age.
7. Scores on the EFT and the RFT are related.
METHOD
Forty-three middle-class children selected from
the fourth, fifth, and sixth grades of a Los Angeles
elementary school and their parents (both mothers and
fathers) were used as subjects in this study. The
children were selected so that they were divided approxi
mately equally into high and low achievement groups, half
of which were boys and half, girls. Although all the
children used as subjects were of at least average
intellectual ability, assignment to the high or low group
was based on academic achievement test scores. All
children were given two measures of perceptual style and
one of achievement motivation individually. Their parents
were asked, in a joint interview, to complete a
questionnaire surveying the ages at which they expected
their child to attain independent mastery of a number of
common activities. In addition, a behavioral measure of
their expectancies and their child's performance on an
achievement task was obtained. The parents expectancy
measures and the childrens' perceptual style and achieve
ment scores for high and low achievement groups and for
boys and girls were examined by analysis of variance,
analysis of covariance, and correlational techniques.
34
35
Subject
Selection and Classification
On the basis of school records and parental
response, a pool of 78 potential subjects was selected
from the fourth, fifth, and sixth grades at Marengo
Elementary School in South Pasadena. Children chosen as
potential subjects met the following requirements: they
1 . came from intact homes (homes in which the
child had lived with both his parents from
birth to the present);
2 . were free from any known organic deficit
which could be thought to impede academic
performance; and
3. had a Total-I.Q. score of 100 of more on a
group and/or individual intelligence test.
(At this school the Lorge-Thorndike
Intelligence Test was routinely administered
to all children at the third and sixth grade
levels. Thus, third grade Lorge-Thorndike
Scores were available for most of the
children.)
Parents of the students who met these three criteria were
contacted by letter (see Appendix C) and asked to take
part in this study. Of the 108 families contacted, 78
returned the form indicating that they wanted to
participate. The 43 families actually used in the study
were selected from this subject pool.
Selection from the subject pool for inclusion in
the study was based on the child’s school achievement test
score obtained through the regular school testing
program. (For fourth graders the only scores available
36
were those from a Reading Test given at the end of third
grade; for most fifth graders Comprehensive Tests of Basic
Skills scores obtained in fourth grade were also available
and used; most sixth graders had taken the Iowa Tests of
Basic Skills in fourth grade so those scores were available
and used.) Those with scores falling in the upper and
lower 26% of the group were selected as subjects. Deter
mining a child's achievement test rank relative to others
in the pool was done by subtracting the child's expected
achievement level (i.e., fourth grade, first month, if that
was when the test had been administered) from his tested
level, converting this adjusted score into a z-score, and
then comparing these z-scores. Those children whose z-
scores were 0.4 or greater were classified as high achiev
ers; those with z-scores of -0.4 or less were labeled low
4
achievers.
4
It should be noted that the subject selection pro
cedure used was comparable to that used by Whiting (1970)
in that Ss were classified as high or low school achievers
relative to the achievement of others in the S pool. An
alternate selection procedure in which the Ss in the pool
were rank ordered from highest to lowest achiever by the
discrepancy between their Total-I.Q. and school Achievement
Test scores (the S selection procedure used by Chance,
1961) or, in other words achievement relative to their own
ability, was also applied to the S pool. Interestingly
enough, Ss who were classified as "high achievers" by the
first procedure (Fig. 1) were often classified as "low
achievers" by the second procedure (Fig. 2) and vice versa.
Since performance relative to the group seemed to be a more
meaningful standard of school achievement than performance
relative to ability, the first method was used here.
37
In addition to selecting children who were (from
their test scores) clearly high or low achievers, an
effort was made to have an equal number of boys and girls
in each achievement group. Thus, the study was designed
in such a way that the subjects fell into one of four
cells, with the level of school achievement and sex as the
classification variables. In the few cases where siblings
both met the selection criteria, the choice which minimized
the differences between cells in terms of numbers, Total-
I.Q. scores, and/or age was made.
Using this selection procedure, 43 children (20
boys and 23 girls) and their parents were selected as
subjects. Eleven boys and ten girls fell into the high
achievers group (labeled Hi-Boys and Hi-Girls,'
respectively), while nine boys and thirteen girls fit into
the low achievers group (labeled Lo-Boys and Lo-Girls,
respectively). All of the families were white and lived
in South Pasadena (a relatively small, old, stable,
homogeneous, predominately Protestant, middle-to upper-
class community) within a few blocks of the school. The
children in the study were between the ages of nine and
twelve, with an average age of eleven. Six were in
fourth, nineteen in fifth, and eighteen in sixth grade.
As a group the children (as well as those in the school as
a whole) tend to score well above the national averages
for their age and grade on school achievement and
38
intelligence tests. For the 43 children in this sample
the overall Total-I.Q. mean score was 119.72 (Lorge-
Thorndike scores obtained in Grade 3 were available on 39
Ss and those obtained in Grade 6 were used for 2 other
Ss; Wechsler Intelligence Scale for Children and Stanford-
Binet scores were used in the 2 other cases). The overall
mean Adjusted Achievement Test score was .98, one year
above the national average (either a reading or achieve
ment test score was available on all the children).
When the four groups of children were compared on
variables which have been found associated with achieve
ment and/or perceptual style such as: number with
handicaps, hand preference, number of schools attended,
number of children in their family, birthorder, birth
place, and age, no statistically significant (at the 5%
level of confidence) differences were noted (see Table 1
in Appendix B; most of this data was obtained from the
parents on the "Information Sheet" shown in Appendix C).
Birthorder varied among the four groups, but did not
appear to follow any particular pattern (for example, the
high achievers groups did not have, on the average, more
"oldest" children than the low groups). However, the Hi-
Girls tended to be older and score higher on I.Q. tests
while the Lo-Girls tended to be younger and score lower on
I.Q. tests than the other groups. Thus, the Hi and Lo
groups of girls used in the study differ more from each
39
other in average age (7 months versus 3.18-3.82 months,
respectively), and I.Q. score (17.03 points versus 2.34-
14.69 points) than they differ from either group of boys
and, as a group, are considerably more heterogeneous (at
least on these two variables) than are the boys.
Overall the boys and girls in the sample do not
differ significantly in age or in Total-, Verbal-, or Non-
Verbal-I.Q., but the girls do tend to be somewhat higher
on Verbal-I.Q. and significantly higher (F^^g = 7.96, p
< .007) in Adjusted Achievement Test scores. However, the
differences between high and low achievers on these
variables is much larger. The high achievers have signifi
cantly higher I.Q. scores (for Total-I.Q., F^^g = 16.09,
p < .001; for Verbal -1. Q. , F-^^g = 26.94, p < .001; and
for Non-Verbal-1 .Q. , F^^g = 4.06, p <.05) and
Achievement Test scores (F1 3g = 117.81, p <.00l) than
the low achievers. The high achieving group in the sample
is also four months, on the average, older than the low
achieving group, but this difference is not statistically
significant.
Each parent participating in the study was asked
to complete a data sheet on himself giving general infor
mation such as: age, birthplace, religious preference,
number of years and type of formal education, and
occupation, since each of these parental variables has
been found related to the achievement motivation or
behavior pattern of children [i.e., McClelland et al.,
(1955) has reported that parents of high achievers tend to
be more highly educated and are more frequently Protestant
or Jewish rather than Roman Catholic, while Hollingsheaid
(1949) reported that parents of high achievers tend to
have higher socioeconomic status and are more frequently
born in the United States than are the parents of low
achievers] . Mothers ranged in age from 30 to 50 with an
average age of 39.88. They tended to have been born in
the United States, be Protestant, have completed some
college work, and not to work outside the home (see Table
2). Mothers of boys and mothers of girls did not differ
significantly on any of these dimensions. However,
mothers of high and low achievers did show two significant
differences. Mothers of high achievers were more likely
to have been born inside the United States than outside it
2
(^ (l) = 4.69, p < .05) and had attended college longer on
the average (^1^41 = 7.33, p <.0l) than the mothers of
low achievers. Though mothers of high achievers also
tended to be somewhat older and more likely to be
Protestant than Roman Catholic (the second most frequent
religious preference in this sample), these differences
were not statistically significant. These two groups also
did not differ significantly in the number of mothers
employed on a full- or part-time basis.
The fathers included in the study ranged from 32
to 56 years of age with an average age of 43.60. As in
the case of the mothers, the fathers tended to have been
born in the United States, be Protestant, and to have
completed some college work (see Table 3). Fathers'
occupations ranged from those requiring little formal
education such as: postal clerk, painter, and pressman to
those requiring a great deal of training, i.e., doctor,
lawyer, and college professor, but tended to fall into
Class III overall, according to Hollingshead*s Two Factor
Index of Social Position (Myers & Bean, 1968, pp. 235-
237). [The Index consists of five classes, Class I
(highest) -- Class V (lowest), and is based upon the
occupation and the number of years of school completed by
the head of the household (father)]. No statistically
significant differences were apparent in the data sheets
of the fathers of boys compared with those from the
fathers of girls. However, two significant differences
were noted when fathers of high achievers were compared
with fathers of low achievers. Fathers of high achievers
had significantly more years of education (F^^i = 16.31,
p < .0 0 1) and were more likely to be classified in the
higher socioeconomic classes (^1^41 = 4.02, p = .02) than
were fathers of low achievers. In addition, several
interesting, but non-significant, trends were noted--
fathers of high achievers tended to be somewhat older, had
more frequently been born in the United States, and were
42
more likely to be Protestant.
Measuring Instruments
The Measures of Perceptual Style
The Embedded Figures Test (EFT)
An individual form of the EFT was administered to
each child as one measure of perceptual style. The EFT is
described by Witkin, et al., (1954, pp. 83-84) as follows:
The embedded-figures test is a paper-and-pencil test
which requires the subject to find a particular simple
figure within a larger complex figure. The simple
figure is "hidden,” to a greater or less extent, by
being incorporated into the pattern of the larger
figure; for example, its outlines may form the
boundaries of several prominent subpatterns in the
complex figure. In addition the complex figure is
colored in such a way as to reinforce a given pattern
and its subpatterns. Depending on the structure of
the complex figure, detection of the simple figure may
be very easy or very difficult.
The standard test makes use of a series of 24 complex
figures, in each of which a simple figure is to be
located. The subject is first shown the complex
colored figure, then a copy of the simple figure that
it contains, and finally the complex figure again,
with instructions to locate the simple figure in it.
Simple and complex figures are never shown together.
To familiarize the subject with the procedure, a
practice trial is given at the beginning of the series.
The subject's score is the mean amount of time taken
to find the simple figures within the complex ones,
and it provides a measure of the extent in which an
item occurs.
A shortened, 12-item form of the EFT was used here,
since it correlates in the mid-nineties with the full-scale
EFT (Witkin, et al., 1962). In administering the test, the
43
basic, standard procedures and instructions developed by
Witkin (1950) were used; however, the instructions were
revised slightly to make them simpler and easier for
children to understand. (The actual instructions and
procedural notes used are given in Appendix D). After
each child had been given the instructions and completed
the practice trial, the test items were presented in the
order (A-l, B-l, C-l, D-l, E-l, A-2, F-l, E-2, C-2, G-l,
A-3, and H-l) originally suggested by Witkin.
Since the administration of EFT items to children
(and even to adults for that matter) with a 5-minute or
even without a time limit (both procedures used by
Witkin), is often an extremely stressful situation
(especially for those subjects who find the task diffi
cult), a 2-minute time limit was imposed on each item. In
cases where a child failed to find the simple figure
within the allotted time on six consecutive EFT items, the
test was discontinued and those items not administered
were scored as failures.
Each child's score was defined as the total number
of seconds used in attempting to solve the twelve EFT
items. For each item in which the child failed to find
the simple figure within the complex one in two minutes, a
score of 120 seconds was recorded. (A copy of the record
sheet is included in Appendix D). The lower the total
score, the more field independent the subject.
The Rod-and-Frame Test (RFT)
As a second measure of perceptual style, the RFT
was given to each child using a portable Rod-and-Frame
apparatus. In the RFT the S is asked to adjust a "rod" to
true vertical while viewing it within a visual field
limited to a square "frame" which is tilted. The standard
testing procedure developed by Witkin, et al.. (1954,
1962) prescribes a series of eight trials administered
with frame and initial rod tilts of 28° in the following
sequence: frame - LLRRLLRR; rod - LRRLLRRL. The S's
score is the mean number of degrees he deviates from the
vertical over the eight trials. The lower this score, the
more field independent the S.
The particular portable Rod-and-Frame apparatus
used in this work was Model V 1260 AR (manufactured by
Polymetric Company, Hobroken, New Jersey) developed by
Oltman (1968). It consists of a single unit, a sturdy
metal support structure within which is contained a
rotatable square drum. A headrest is located at one end
of the drum and the "rod" (a straight black plastic strip
about 8 inches long) is displayed at the opposite end on a
disk which can be rotated independently of the drum by
either the S or E. The back of this disk is calibrated in
degrees so that deviations from the vertical can be read
by E. Surrounding the disk is a 2-feet deep square frame
(12 inches on a side) outlined in black and enclosed with
translucent plastic. Light transmitted by these sides
illuminates the stimulus display. When the S is seated
properly in front of the apparatus with his head in the
headrest, his field of view is restricted to the interior
of the enclosure by means of a curved shield attached to
the headrest. Between trials a curtain is raised in
front of S's face to block his view of the inside of the
enclosure. Oltman found that scores on this type of
portable apparatus correlated highly (.89) with those from
the original standard laboratory Rod-and-Frame developed
by Witkin.
After the concept of vertical and the nature of
the task and of the equipment was explained to the child,
he was seated (in the upright position) in front of the
apparatus. Care was taken to adjust his chair height
properly so that his chin rested comfortably on the chin
rest. He was asked to keep his hands in his lap (except
when adjusting the rod) and his head in the headrest
throughout the task. (The details of these instructions
and procedures are given in Appendix D). They follow the
standard procedure, developed by Oltman and specified in
Polymetric's Operating Instructions, for use with the
portable apparatus, except for minor revisions designed to
make the instructions more appropriate for children.)
Prior to each trial E set the rod and the frame to their
46
prescribed angles, opened the curtain, and asked the child
to move the rod until it looked "straight up and down" to
him. As soon as the child announced that he had finished
adjusting the rod, the curtain was lowered and the rod's
position (direction and degrees of deviation from the
vertical) was recorded. (See Appendix D for a copy of the
record sheet used; this form was adapted from one included
in Polymetric 1s Operating Instructions.)
The Measures of Parental Expectancies
The Age-Independence Scale (AIS)
Description: The revised AIS used in this study
is a questionnaire containing 62 completion items, each of
which asks the respondent the age at which he expects his
child to first attain independent mastery of a task and
which of these tasks he considers most important in his
child's development. The Scale was designed for parents
with children in the five- to twelve-year-old range and
the items refer to tasks and activities typical of the
everyday life of the elementary school child. Each item
on the AIS is assigned to one of seven subscaies the first
six of which were named (by Gray, 1968) to identify the
type of independence training included. The seventh sub
scale is a more general category added to the AIS for this
study and composed of items previously used by
Winterbottom (1958). These seven subscale categories are:
47
I. Self-Care - contains 4 questions pertaining to
caring for one's clothing and room; preparing food;
using the telephone.
II. Cognitive Facility - contains 6 items pertaining
to such activities as following written instructions;
mastering certain reading, writing, and speaking tasks.
III. Physical Skills - includes 3 items about
physical skills requiring maturation and practice,
some of which would be learned at home, others at
school.
IV. Social Responsibility - contains 14 questions
about trustworthiness and about being responsible for
timeliness, money, belongings, chores, and small jobs.
V. Autonomy - includes 15 questions about self-
direction, choice and self-selection: choosing one's
friends, clothes, personal possessions, and activities.
VI. Wide Experience - contains 10 items about activi
ties which would widen the experience of the child,
including risk-taking activities, venturing into the
community, going away from home, and having contacts
with strangers.
VII. General - includes 9 items about learning to
enjoy being independent, competent, and responsible in
activities at home and school.
The framework of categories is not apparent on the final
questionnaire, since items are merely listed in
chronological order by expected age and numbered consecu
tively. However, there is a rough correlation between the
number of the item and the expected age, with the lower
numbers having lower expected ages.
Construction and Revisions: The AIS was
originally constructed by Gray (1968, pp. 48-52), after it
became apparent to her that none of the existing scales
48
could adequately assess parental attitudes toward
independence training. Starting with an item pool of
approximately 300 items derived from various sources (the
child development literature, casework, other scales, i.e.,
Winterbottom, 1953, 1958; Torgoff, 1961, Doll, 1947;
Danziger, 1960; Blair, 1963), various versions of the
Scale were constructed and tested in pilot studies. Her
final form of the Scale consisted of 100 items divided
among the first six categories described earlier.
More recently, the AIS was revised by Whiting
(1970) after another pilot study. She clarified
additional items, altered the Scale (by adding ten items
and dropping the response column for girls) to make it
more appropriate for her sample (boys only), and added
space for parents to check the items they considered
especially important in their son's development. No
other bhanges were made and she retained the six subscale
categories originally suggested by Gray.
A number of revisions were made in the Whiting
version of the AIS before it was employed in the current
work. The most significant change was a reduction in the
total number of items--from Whiting's 110 to the current
62. This reduction in length was accomplished by dropping
all items with age expectancies below age 8 , except for
those included in Winterbottom's Scale. (The age
expectancies were based on fifty middle-class mothers
49
responses to these items for boys obtained by Gray in
1968, pp. 191-195.) In addition an effort was made to
exclude items which appeared to be sex-biased in the sense
that they called for greater strength or were normally
performed by only one sex in our society. For example the
item:
Hit a baseball as far as second base,
was dropped because, on the average, boys are bigger and
stronger than girls of the same age. Several other items:
Use simple power tools (father present).
Have and use own jacknife.
were dropped because they seemed more appropriate for boys
than girls.
Since the 20-item Winterbottom Scale has been
found useful in differentiating responses from parents of
high and low achievers in many previous studies, all of
those items not already included by either Whiting or Gray
were added to the AIS (with the exception of items #14 and
#16 which had not been found significant discriminators by
Winterbottom, Chance, 1961; or Shaw, 1964). The nine
additional Winterbottom items were labeled Subscale VII so
that they could be looked at as a group rather than
integrated into the other subscales. However, the
remaining nine Winterbottom items originally incorporated
into the AIS remained scattered on various subscales
(three on I; one on IV; seven on V; and one on VI). A
50
list of the items, by subscale, is included in Appendix E.
Once the 62 items to be included on the revised
AIS had been selected, they were arranged in chronological
order by expected age and numbered consecutively as in the
earlier versions. The format was also essentially the
same as those previously used. Two blanks followed each
item. In the first the respondent was asked to fill in
the "Age My Child Should" be able to perform that task or
activity. If the respondent considered that item
"Especially Important" in his child's development, he was
asked to put a check in the second blank; otherwise he was
instructed to skip it. Whiting (but not Gray) had also
used this second blank so that parents could indicate
which items they considered more important, but had
limited their choice to five items per page. This limita
tion seemed unnecessary and so was left out of the current
revision (shown in Appendix E).
An untitled page (see Appendix E) referred to as
the "Summary Page" was also added to the AIS following the
62 items. It was hoped that this page might help assess
how the parent saw this child in terms of competence*
leadership, skills, need for attention, and curiosity as
well as the parent's reaction to the Scale itself.
Administration; Whiting and Gray had developed a
direction sheet to accompany the AIS and this was used
(after minor changes) in the current work (see Appendix E
51
for version used). In addition to standardizing the
instructions, the direction sheet was developed to make
the Scale self-explanatory so that it could be administer
ed by an untrained person or even mailed to respondents,
if necessary. However, many of the parents in the study
seemed to have trouble with the questionnaire even after
reading these directions so further verbal instructions
(summarized in Appendix E) were also given.
Reliability? A test-retest reliability study was
conducted by Gray, who administered the Scale (in its
original form) to 50 middle-class parents of boys and
girls at one- to four-week intervals. For both sexes,
approximately two-fifths of the correlation coefficients
were above .70 and two-thirds above .60; only one-eighth
of the coefficients fell below .50. Overall reliability
for boys was .66 and, for girls, was .68 (Gray, 1968, pp.
53-55). No further reliability assessments have been made.
Scoring ? There are several different ways in
which the responses to the AIS can be treated; the
responses to each item can be compared; a mean of the
responses to the items in each subscale category can be
computed and compared; and/or the overall mean of all the
AIS items can be computed and used for comparison. In her
research to determine the differences held by selected
parental groups (lower- and middle-class parents, mothers
and fathers, and large-, medium-, and small-size
52
familities) about the age at which a boy or girl should be
able to perform various activities independently, Gray
used the first method and analyzed the responses of each
group to each question separately. Whiting, in comparing
the independence-age-expectancies of parents of achieving
and underachieving boys, used the second method and
computed the mean age for each of the six scale categories.
In this study all three methods were used to some
extent. The overall mean score for all AIS items (third
method) and the seven subscale scores (second method) were
calculated for each respondent and used in the comparison
of groups of parents (those of high and low achievers,
boys and girls, and field independent and field dependent
children). In addition a listing of the mean, standard
deviation, and variance for each of the four groups of
mothers (those of: Hi-Boys, Hi-Girls, Lo-Boys, and Lo-
Girls) and of fathers, for all mothers and all fathers,
and then for all respondents was made for each item on the
AIS (see Appendix F). However, only the overall mean
scores and subscale scores were used in the data analyses.
Block Stacking Task (BST)
The BST was used as a behavioral measure of
parental expectancies in an achievement situation. In
addition, it provided information on the child's evalua
tion of his own competence, his level of performance, and
53
an index of his independence. It was developed by Rosen
& D'Andrade who describe the task (1959, pp. 191-192; 203-
204) as follows:
The boys were asked to build towers out of very
irregularly shaped blocks. They were blindfolded and
told to use only one hand in order to create a
situation in which the boy was relatively dependent
upon his parents for help. His parents were told that
this was a test of their son's ability to build
things, and that they could say anything to their son
but could not touch the blocks. A performance norm
was set for the experiment by telling the parents that
the average boy could build a tower of eight blocks;
they were asked to write down privately their estimate
of how high they thought their son could build his
tower. The purposes of this experiment were (a) to
see how high were the parents1 aspirations for and
evaluations of their son, e.g., if they set their
estimates at, above, or below the norm; (b) to see how
self-reliant they expected or permitted their son to
be, e.g., how much help they would give him.
There were three trials for this task. The first
provided measures of parental evaluations and
aspirations not affected by the boy's performance; the
second and third trial estimates provided measures
affected by the boy's performance. The procedure for
the third trial differed from the first two in that the
boy was told that he would be given a nickel for each
block he stacked. Each member of the family was asked
to estimate privately how high the boy should build his
tower. No money would be given for blocks stacked
higher than the estimate nor would the subject receive
anything if the stack tumbled before he reached the
estimate. Conservative estimates, hence, provided
security but little opportunity for gain; high
estimates involved more opportunity for gain but
greater risk. The private estimates were then revealed
to all and the family was asked to reach a group
decision. . . . The observes scored three types of
behavior in this (the decision-making) process. (l)
The number of acts each subject contributed to the
decision making, (2 ) the number of times each subject
initiated the decision process by being the first to
present a choice for consideration, (3) the number of
54
times an individual made the decision for the group or
stated the final judgment. . .
. . . . In addition to securing objective measures of
parental aspiration-evaluation levels, the observers
scored the interaction between subjects, thus obtain
ing data as to the kind and amount of instructions the
parents gave their son, the amount of help the son
asked for or rejected, and the amount and kind of
affect generated during the experiment.
The BST was administered essentially as just
described. After the child, his parents, and E were all
seated at a table, it was explained that the child was to
build a tower with some blocks. But, because he would
undoubtedly find this so easy, E would make it harder by
blindfolding him and allowing him to use only one hand;
however, his parents could help him by saying whatever
they wished, but could not touch the blocks. After this
general introduction to the task, the child was blind
folded and it was explained that E would read the rest of
the instructions, so that each child in the study would
have the same chance. The instructions read are in
Appendix G. A set of twenty blocks of various geometric
shapes from the game "Blockhead!" (manufactured by
Saalfield Publishing Company, Akron, Ohio) was used for
stacking and paper towels served as disposable blindfolds.
One slight change in Rosen and D'Andrade's
procedure was that the performance norm was lowered from
eight to five--a number which appeared to be within the
capabilities of most children in this age range. Lowering
55
the norm to this score made it possible for each child
(and his parents) completing the task to feel that he was
at least as good at it as most other children, while still
providing an objective measure of the parents estimate of
his performance (above, below, or at the stated norm).
During the BST the E made an effort to record a
number of different kinds of information for each trial:
each parent's estimate, the child's performance (number of
blocks stacked), the number of comments made by each
parent, the number of requests for or rejections of
parents' aid made by the child, and the child's estimate
on Trial #3. All of these scores were used in the data
analyses. Each parent's first estimate, unaffected by any
previous performance in this task, was conceived to be
primarily a measure of his aspirations for the child.
Their estimates for the second and third trials of this
task were combined to form a single score and considered a
measure of that parent's aspiration-evaluation of the
child. The total number of comments made by a parent
during the trials was also combined into a single score
and used as an index of the extent to which they
encouraged or allowed independent functioning. Similarly
the number of requests for or rejections of parents' aid
made by the child were added for each of three trials and
used to estimate his level of independence in this
situation. The child's estimate of his performance on
56
Trial #3 was considered his own appraisal of his
competence. The sum of the blocks stacked in the three
trials was used as his performance score.
A number of other more subjective measures, i.e.:
the decision-making process on Trial #3, a rating of the
content of parents comments along four dimensions
(specific vs. non-specific; supportive vs. critical; clear
vs. unclear; and pushing vs. non-pushing), whether or not
each participant (mother, father, and child) appeared to
be actively involved in the task, and whether or not they
each appeared to find it a positive experience, overall,
were also recorded. However, these subjective measures,
though interesting, were not analyzed, both because so
many objective measures were available and because their
reliability (and that of the objective measures to some
extent) was questionable, since it was not confirmed by a
second observer or recorded on tape and then scored. The
record sheet used for this task is shown in Appendix G.
The Measure of Achievement Motivation
Ring Toss Task (RTT)
The RTT was employed as a measure of achievement
motivation. Although this study was not designed to
resolve the discrepancies in assessment of actual achieve
ment and achievement motivation, it was of interest to see
how well this theoretical measure correlated with school
57
achievement scores. The task itself is simply the
familiar game in which one or more players throw a number
of rings in the hope of getting as many as possible onto
the post in front of them, but administered in a more
standardized way to one S at a time. In this study a
wooden base and post and lightweight rope rings were used.
The child played seven games in succession in each of
which he threw all nine rings one at a time from some
particular distance. In games 1, 3, 4, and 5 the distance
the child stood from the post was fixed according to his
height--those between four and five feet in height stood
on line five (five feet from the post); those between five
and six feet stood on line six (six feet from the post);
etc. For the remaining three games, the child could
choose to stand from three to ten feet away from the post.
Distance was marked off on the floor with pieces of
masking tape, each one foot farther from the post, and
each marked with the appropriate distance. Before each
game the child was asked to estimate the number of rings
he thought he would get on the post (instructions are
included in Appendix G). This estimate, his actual
performance, and the distance from the post were recorded
for each game on a score sheet (see Appendix G).
Determining the child's "actual performance"
proved to be somewhat of a problem with the equipment
used, since the lightweight rings would often be thrown
58
onto the post, but then bounce off or cause a ring already
on the post to bounce off. Thus, the task frequently
involved more luck than skill and could be quite
frustrating (though most of the children enjoyed it).
However, the performance score was arbitrarily defined as
the number of rings remaining on the post at the end of
the game. Rings were not picked up or removed from the
post during a game.
Frank (1935) initially used the RTT to assess
level of aspiration, one important aspect of achievement
motivation. His assessment was based on a score
quantifying the extent to which a S expected his future
performance to differ from his past performance. Garcia
(1971) used a similar score called "goal discrepancy"
found by computing the algebraic difference between a S’s
estimated score for a single game and his actual
performance in the preceding game to measure level of
aspiration. She also used a second score called "attain
ment discrepancy"--the algebraic differences between a S's
estimate and his actual performance for a particular game.
Both of these scores were used in the current work. How
ever, since it seemed unreasonable to compare the goal
discrepancy scores of children when their experiences with
the task varied greatly':, (as they did in games 2, 6 , and 7
when distance from the post was not standardized), this
score was calculated only for games 3, 4, and 5 in which
59
distance from the pin was fixed. "Attainment discrepancy”
was computed for games 2 through 7; game 1 was not
included in the computation, since the first estimate was
not based on experience with the task as were the later
estimates.
Three other scores were also obtained on the RTT
so that their relationship to achievement motivation and/
or behavior could be investigated. First, the distance
from the post each S chose in games 2, 6 , and 7 was added
together as an index of the amount of risk he was willing
to take. Second, a S's estimate of the number of rings he
expected to get on the post in game 1 was used as another
index of level of aspiration. Finally, the number of
rings actually thrown onto the post in the games where
distance was fixed (#1, 3, 4, and 5) was added to obtain a
performance level or score for each S.
Procedure
Gnce the selection process had been completed, the
43 families participating in the study were contacted by
phone and an appointment with the child and his parents
scheduled. On the day of the appointment the family was
contacted again by phone to confirm the appointment and
reschedule it if necessary. All families were seen in
their homes between January 8th and March 2nd, 1973. Most
appointments were at 7:00 or 8:00 PM on a weeknight,
60
though some families were available in the late afternoon
(after school) while others preferred a Saturday or Sunday
afternoon.
All appointments followed the same general format.
After introductions and getting everyone settled at a
table, the E spent a few minutes outlining the study and
session in general terms and answering questions. Then
each parent was given a copy of the Information Sheet and
the AIS and asked to complete them independently.
Following this the child was asked to join the group, the
Block Stacking Task was administered, parents were asked
to sign a Permission Slip (see Appendix C) allowing the E
to test the child at school, any remaining questions were
answered, the family was thanked for their participation,
and the parents were asked not to discuss the AIS or BST
with other parents until the study had been completed.
The session usually lasted about an hour and most families
seemed to enjoy it--particularly the BST, which was
usually quite exciting.
All of the children were tested individually at
school during their normal school day in a single session
usually lasting about 40-45 minutes. The EFT was given
first, followed by the RFT, and then the RXT. This
testing sequence was chosen because the EFT was the
longest and most difficult test and required the most
concentration. Giving the RFT next provided a welcome and
61
intriguing change from the EFT for most children, both
because it involved an unusual looking piece of equipment
and because it was a new experience. Finally, the
administration of the RTT gave them a chance to move
around and become more physically active. All three tests
were administered by the same E to all of the children,
except for the fourth graders (in these cases after E
administered the EFT, the RFT and RTT were given by a
second E).
All of the children were tested between February
27th and March 14th, 1973. Within that time period, all
■fourth graders were tested, then fifth graders, and then
sixth graders. However, the time of day at which a
particular test was given could vary greatly from one S to
another. This was due to the fact that there were three
different classes at each grade level and Ss were drawn in
alphabetical order from one class at a time until all the
Ss within that class had been tested.
Several precautions were taken to minimize the
possibility of E bias. First, although the two phases of
the study--testing the children and interviewing the
parents— were not related and could have been carried out
concurrently, the appointment with the family was always
completed before the child was tested at school. This
procedure was followed because the same E was involved in
both phases of the investigation and because the data
62
obtained during the family appointment seemed much more
susceptible to bias than that obtained from the child at
school. This particular procedure was also beneficial in
another way. Not only had the E met the parents and
obtained their cooperation, but also that of the child
involved. Thus, the child knew in advance that he would
be tested at school, knew who would be testing him,
approximately when, and that his parents thought it was a
good idea. In addition all of the children had earned
some money during the BST and most were enthused about
more tests--even though they knew they would not earn any
more money.
Another precaution used to eliminate E bias was
the assignment of a code number to each family. The
numbers assigned did not indicate the child's achievement
status or sex, but did tend to increase with age. Thus,
when the E called a family to make an appointment or
visited them, the only information available to her was
the family’s name, address, code number, and the child's
name, age, and grade level. In addition these numbers
were used on all forms and data sheets as a means of
keeping the information confidential.
After all of the families had been seen, the
children tested, and the data analyzed, each family was
sent a letter thanking them for their participation and
highlighting the results related to school achievement. A
63
copy of the letter and this summary is given in Appendix
C.
Data Treatment and Analysis
The data were analyzed on the IBM 370/155 Computer
at the USC Computing Center. Prior to this analysis all
BST, EFT, RFT,5 and RTT scores were computed for each S,
transcribed onto IBM data sheets, punched on cards, and
verified. Each child's code number, achievement level
(high or low), sex, IQ scores, age, and adjusted achieve
ment scores were also transcribed and then punched on
these data cards. AIS responses were transcribed and
punched on a set of cards for each respondent. The over
all mean and the seven subscale scores were calculated by
the computer,,. In the cases where AIS items had been left
blank, those items were eliminated from the calculations.
Data from the Information Sheet and the AIS Summary Page
5In computing the RFT scores, it was noted that in
some Ss the mean deviation of the rod on trials in which
the frame was tilted to the left was quite different from
that obtained when the frame was tilted to the right.
Thus, a second RFT score (RFT-C) was developed to serve as
an index of the consistency of these two series of scores
in a particular S. The RFT-C score was calculated
according to the following equation:
Total deviation on Trials: (1+2+5+6) - (3+4+7+8)
RFT-C = Total deviation on Trials: 1 - 8
It was included in the data analyses so that its relation
ship to other measures could be investigated.
64
were tabulated by hand as were the number of "Especially
Important" and "Not Appropriate" responses on the AIS.
The frequency data from these three pages was analyzed
using the chi square statistic. The quantitative data
from the Information Sheet was analyzed by analysis of
variance. The Two Factor Index of Social Position (as
described by Myers & Bean, 1968, pp. 235-237), based upon
the fathers occupation and educational level, was used to
calculate the relative socioeconomic status of the family.
The basic questions of this study were whether
parental expectations, as expressed on the AIS and BST,
differed depending upon whether their child was a high or
a low achiever, a boy or a girl, and/or field independent
or field dependent. Two other questions of interest were
whether high achievers (on school achievement tests) would
tend to be more field independent (on the EFT and/or RFT)
than low achievers and whether their performance and
behavior would differ in other achievement situations (the
BST and the RTT). Analysis of variance was used to
investigate the differences between groups (Hi-Boys, Hi-
Girls, Lo-Boys, and Lo-Girls) on the dependent measures
(AIS, BST, EFT, RFT, And RTT) as well as the interactions
between achievement level and sex on these measures.
Analysis of covariance was used as a supplementary
technique to control for differences in age and I.Q.
scores when this was appropriate. All of these analyses
65
were done on the computer using a multivariate analysis of
variance Fortran program called, "MANOVA" developed at the
University of North Carolina Psychometric Laboratory and
stored in the USC University Computing Center library.
The program performs univariate and multivariate analyses
of variance, of covariance and of regression. It provides
an exact solution in either the orthogonal or non-
orthogonal case and includes a number of options in these
programs.
The correlation between EFT and RFT scores was
calculated to determine which measure seemed to be more
effective in measuring field independence in this sample,
or whether the two scores might be combined as an index of
perceptual style. The relationship between these two
measures was reasonably strong [ .45 (.02)] , for girls, but
essentially non-existent for boys. Thus, neither measure
by itself appeared to be the measure of choice, nor was
there any justification for combining them so both scores
were used in the calculations as measures of perceptual
style. In addition, since the children had not been pre
selected on the basis of their perceptual style, their
scores tended to fall along the continuum of field
independence— field dependence rather than clustering at
the extremes of the scale. Therefore, all of the children
could not be unambiguously classified as field
independent or field dependent, and correlational
66
measures were used (rather than analysis of variance) to
determine the extent of the relationship between their
parents expectations on the AIS and BST and their own RFT
and EFT scores. Correlational techniques were also used
to study the relationship between parental expectations on
the AIS and their behavior on the BST and between the
children's RTT and BST performances and their adjusted
achievement test scores. The Pearson Correlation sub
program of the Statistical Package for the Social Sciences
(SPSS written by Nie, Bent, & Hull, 1970) was used for all
of these computations.
Several special programs integrating various SPSS
subprograms were written in order to handle the AIS
responses. Starting with the raw AIS data from each
respondent the computer:
1 . calculated the overall and subscale means for
each respondent;
2. tabulated the AIS item, subscale, and overall
means by parent responding (mother or father)
and then by group; and
3. using the overall and subscale means already
obtained, produced two other series of AIS
scores (Father + Mother and Father - Mother)
for each set of parents
all of which were used in the data analyses.
RESULTS AND DISCUSSION
The general hypotheses of the study are:
1 . that parents of high and low achievers differ
significantly in the ages at which they expect
independent mastery of common activities;
specifically, parents of high achievers expect
more independent mastery at an earlier age;
2 . that parents of boys and girls differ
significantly in the ages at which they expect
independent mastery of common activities;
specifically, parents of boys expect more
independent mastery at an earlier age;
3. that parents of field independent and field
dependent children differ significantly in the
ages at which they expect independent mastery
of common activities; specifically, parents of
field independent children expect more
independent mastery at an earlier age; and,
4. that high achievers are more field independent
than low achievers
as stated in the Introduction.
These general hypotheses reduce to the following
tests:
(l) A. Parents of high achievers have lower
overall and subscale means on the AIS
than parents of low achievers.
B. Parents of high achievers give higher
estimates on all BST trials than parents
of low achievers.
C. Parents of high achievers make fewer com
ments during the BST than parents of low
achievers.
67
68
(2) A. Parents of boys have lower overall and
subscale means on the AIS than parents
of girls.
B. Parents of boys give higher estimates
on all BST trials than parents of
girls.
C. Parents of boys make fewer comments
during the BST than parents of girls.
(3) A. Parents of field independent children
have lower overall and subscale means
on the AIS than parents of field
dependent children.
B. Parents of field independent children
give higher estimates on all BST trials
than parents of field dependent
children.
C. Parents of field independent children
make fewer comments during the BST than
parents of field dependent children.
(4) High achievers score lower on both the EFT
and the RFT than low achievers.
Hypothesis la.
This hypothesis states that parents of high
achievers have lower overall and subscale means on the
AIS. Contrary to what was expected, parents of high and
low achievers did not differ in their AIS overall mean
scores, nor in most of the subscale means (as shown in
Tables 4, 5, and 6 ). A statistically significant differ
ence between groups in the direction expected occurred
on only one of the seven subscales, Scale 2--Cognitive
Facility, where F.^ 3g = 4.90, p < .03. Parents of high
69
achievers did tend to respond with lower ages on one other
subscale (#7--General), but gave higher ages on Scales 1
and 3 (Self-Care and Physical Skills, respectively),
though none of these tendencies reached levels normally
accepted as statistically significant.
These results are surprizing in view of those of
Whiting with an earlier version of the same Scale. She
found that high achievers' parents gave lower ages than
low achievers1 parents on all six subscales and that these
differences were statistically significant on five of the
six subscales. However, there were several methodological
differences in the two studies that might account for
these differences in outcome. First of all, the AIS used
in this study was considerably shorter than that used by
Whiting and some subscales (#1, 2, & 3) had been
abbreviated to the point that they contained fewer than
ten items. Furthermore, even if the majority of the sub
scales were long enough to be reliable, their validity was
questionable. A factor analysis of the items had not been
made by Gray, Whiting, or in the current revision. In
addition, some of the parents' reactions in completing the
AIS, as well as some of the data analyses, suggested this
might be the case. (This evidence will be discussed in
more detail at a later point in this section under the
heading "The AIS as a Measure of Parental Expectancies.")
A second area in which this investigation differed
from that of Whiting was in the sample used. Whiting used
only boys in her sample rather than both boys and girls.
Rather paradoxically, however, the responses of the
parents of girls in this study tend to correspond to
Whiting's AIS results better than those obtained from
parents of boys. As shown in Table 4, parents of high-
achieving girls tend to give lower AIS responses than
parents of low-achieving girls on five of the subscales,
while for boys, this is true on only three subscales.
The children in the sample differed from those
used by Whiting on two other important dimensions. They
probably tended to come from a somewhat higher socio
economic level (though differences in the information
reported as well as the passage of time make this diffi
cult to determine accurately; certainly the parents
involved in our study had spent considerably more years in
school). Second, the differences between the groups of
high and low achievers in school achievement, though clear
in this sample were much less pronounced than in Whiting's
sample. All of the children involved in the current
investigation tended, as a group, to be achievers, some of
whom were more accelerated than others. Whiting's high
and low achievers were much more discrepant in school
achievement, since her low achievers tended to perform
well below grade level on reading and/or achievement tests
while the high group scored well above grade level.
71
Finally, this study differed from Whiting’s in the
way parents' AIS responses were used in the data analysis.
In this work, a child's parents' overall and subscale
means were added together (Father + Mother) or subtracted
from each other (Father - Mother) as the case might be,
but always treated as a single score in the data analysis.
Whiting, however, did not combine parents' scores into a
single score, but used each parent's score individually.
Thus, in her analysis, she compared four groups of scores,
those of: mothers of high achievers, fathers of high
achievers, mothers of low achievers, and fathers of low
achievers rather than comparing scores of parents of high
and low achievers. How much each of these methodological
differences contributes to the difference in outcome is
difficult to ascertain. However, shortening the AIS and
particularly the subscales, using a relatively homogeneous
group of children, who as a group tended to achieve in
school, and decreasing the sample size by using a single
combined score rather than two individual scores to
represent the parents, certainly made it more difficult to
obtain clear support for this hypothesis.
Hypothesis lb.
This hypothesis states that parents of high
achievers give higher estimates on all BST trials than
parents of low achievers. For Trial #1, estimates from
72
parents of high achievers do tend to be slightly higher
than the estimates from parents of low achievers, as
shown in Table 7. However, this difference is not
statistically significant (see Table 8 ) even when analysis
of covariance is used to control for group differences in
the children's: ages, Total-I.Q. scores, ages plus I.Q.
scores, or block stacking performance (number of blocks
stacked on Trials #1 + 2 + 3).
On Trials #2 + 3, estimates from parents of high
achievers actually tend to be lower than those from
parents of low achievers. Again this tendency is not
statistically significant in most of the data analyses.
However, when the variance between groups due to the
children's performance is partitioned out, the F ratio
approaches the acceptable range (F = 3.18, p <.08). If
the estimates for high and low achievers are broken down
by sex of estimator (mother or father) as well as sex of
child, as shown in Table 9, it becomes clear that the
estimates from mothers of boys tend to be considerably
more discrepant than those from mothers of girls or
fathers of either boys or girls.
Using the same task, same data computation and
analysis procedures, but a sample composed only of boys
and somewhat more diverse in socioeconomic status [ half,
middle class (class II or III) and half, lower class
(class IV or V) on the Hollingshead Index of Social
Position] from that in this study, Rosen and D'Andrade
found that parents of high achievers gave higher estimates
on both Trials #1 and #2 + 3 than parents of low
achievers. This difference was significant at the .05
level of confidence on Trial #1 using a combined score for
each set of parents and at the .005 level on Trials #2 + 3
using mothers scores only (the difference in the fathers
scores was not statistically significant). Thus, their
results are similar to those reported here for Trial #1.
For Trials #2 + 3, they are similar also in that fathers
of high and low achievers differ little in their estimates
while mothers of these two groups, and particularly
mothers of boys, differ a great deal in their estimates.
The real contrast with Rosen and D1Andrade
results, with the relationship hypothesized here, and with
the Trial #1 estimates obtained is in the direction of the
estimates on Trials #2+3. It is difficult to understand
why parents of high achievers would give somewhat higher
estimates on Trial #1 than parents of low achievers and
clearly lower ones on Trials #2 + 3. One possible
explanation is that offered by Rosen and D1Andrade in
discussing the theoretical distinction they saw between
parents estimates on Trial #1 versus those on Trials #2 +
3. They conceptualized the parent's first estimate as
one primarily reflecting the parent's level of aspirations
for his child, while the later estimates (which were, of
course, influenced by the child’s previous performance)
were considered a mixture of the parent’s aspiration and
of his evaluation of his child's performance. If the
parent's aspiration for his child outweighted the child's
performance in the task (as the parent evaluated it), the
parent might well give higher estimates for the first
trial than for the later ones. Certainly this would
explain our results, though two other issues make this
explanation somewhat less convincing. First, from a
theoretical viewpoint, parents of high achievers are
usually characterized as being warmer, less critical, and
more positive and accepting than parents of low achievers.
Following this line of reasoning, one might expect their
evaluations of their child's performance to be more
positive and, therefore, higher. This is particularly
true in view of the second issue--the children's actual
performance in this task (see Table 35). For boys, the
number of blocks stacked during the three trials of the
task by high and low achievers varied little, though low
achievers actually stacked more blocks (25 versus 28; this
is contrary to the results reported by Rosen and
D'Andrade). High achieving girls, on the other hand,
stacked significantly more blocks (32 versus 24) than low
achieving girls. Thus, one might logically expect that
given the somewhat higher performance of high achievers as
a group on this task and the theoretically less critical
75
outlook of their parents, estimates given by parents of
high achievers for Trials #2 + 3 would have exceeded or at
least equalled those given by parents of low achievers.
In view of our results, parents of high achievers (or at
least those in this sample, involved in this task) though
higher in their level of aspiration for their child, are
apparently less positive in their evaluations of their
child's performance than parents of low achievers.
Hypothesis lc.
This hypothesis states that parents of high
achievers make fewer comments during the BST than parents
of low achievers. The data clearly refute this
prediction. Parents of high achievers make significantly
more comments to their children during this task than do
parents of low achievers even when the differences between
groups due to age, Total-I.Q. scores, age plus Total-I.Q.,
or their child's performance is statistically eliminated
by analysis of covariance (see Tables 7 and 8 ). In
addition, fathers, on the average, tend to make many more
comments in this task overall, than do mothers (102 versus
47) .
In predicting that parents of high achievers
would make fewer comments than those of low achievers
during this task, two assumptions were made:
1 . parents of high achievers encourage
independent functioning in their children to a
76
greater extent than parents or low achievers;
and,
2 . the number of comments made by parents in this
situation is a measure of the extent to which
they "encourage independent functioning'!;
specifically, the fewer comments (particularly
specific directions) made to the child, the
more the parents are "encouraging independent
functioning" and indicating their confidence
in their child's competence.
Though the results might be used to challenge both
assumptions, until the validity of "the number of comments
made by parents" has been clearly established as an index
of "encouraging independent functioning," it seems more
reasonable to question the accuracy of this assumption.
On this basis, if the number of comments made by parents
in this task is not considered an index of "encouraging
independent functioning," what, if anything, does it
measure? One possibility is that it might be considered
an indication of the extent to which the parent is
involved in his child's performance. This interpretation
seems reasonable in view of Rosen $.nd D'Andrade's
conclusion (p. 216):
In a way, it is this factor of involvement that
most clearly sets the mothers of high n Achievement
boys apart from the mothers of low n Achievement boys:
the former score higher on every variable. . . .
Another possible interpretation of the results on this
hypothesis is that assumption two is correct, but that
parents of high achievers value and encourage independent
functioning as a goal considerably less than they value
77
achievement. In the BST situation, allowing the child
more independence (by making fewer comments) made the
child's task much more difficult and increased the chances
of a lower score. Unfortunately Rosen and D'Andrade have
not reported their data in the same way as in this study
(rather than tabulating the total number of comments made
by a parent during the BST, they have categorized comments
on several different tasks as to specificity, affect, and
pressure applied and then combined these subtotals across
tasks) and so the results themselves are not directly
comparable.
Discussion of General Hypothesis 1
From our results, parents of high achievers can be
characterized as:
la. not differing significantly from parents of
low achievers on most of the AIS [however,
since
(1) parents of Hi-Girls did tend to give
lower ages than those of Lo-Girls;
and
(2 ) sampling and measurement problems
made confirmation of this hypothesis
particularly difficult the results
are not especially clear on this
point; certainly the data for girls
(but not boys) could be interpreted
as supporting the prediction that
parents of high achievers give lower
ages on the AIS than parents of low
achievers.]
lb. giving slightly higher BST estimates on Trial
#1 and significantly lower BST estimates on
Trials #2 + 3; and,
78
lc. making significantly more comments during the
BST than parents of low achievers.
In order to determine the actual relationship
between parents expectations on the AIS and their behavior
during the BST, correlational techniques were used. In
general the correlations obtained (see Table 10) are only
moderate (only three are greater than .5), but consistent
in their direction. The parents (as a set and
individually) who tend to put lower ages on the AIS also
tend to make higher estimates on both BST Trials #1 and #2
+ 3 and more comments during these trials. This relation
ship is similar to that found true for parents of high
achievers, except for BST Trials #2+3. The reason for
the inconsistency on these trials is not clear. However,
looking at Tables 9 and 10, the discrepancy on Trials #2 +
3 appears to be due more to the mothers' than the fathers'
estimates.
Parents of high and low achievers differ in a
number of other ways in their reactions to the AIS. When
they are compared on the number of AIS items marked as
"Especially Important to Me" (as shown in Table 11),
parents^ of high achievers check significantly more items
6In contrast to the way in which parents AIS means
were obtained, parents scores here were obtained by adding
the total number of such responses given by fathers to the
total number given by mothers from that group rather than
by combining a set of parents individual responses and
then totaling those responses across that group. This
79
(X(i)2 = 90.42, p < .001) than parents of low achievers.
However, when the two groups are compared on the number of
AIS items they consider "Not Appropriate" for their
particular child, though parents^ of low achievers,
especially fathers, tend to check more items (see Table
12), this difference is not statistically significant. (A
complete tabulation of the items checked, "Especially
Important to Me" or "Not Appropriate" arranged by parental
group and item number, is given in Appendix F.)
Parents^ of high and low achievers also differ in
their responses to a number of questions on the AIS
Summary Page (see Table 13) theoretically related to the
achievement dimension. The questions on which these
differences are statistically significant are:
2. More parents of high achievers said that
"their child had mastered the AIS items at an
earlier age" and fewer that "this had happen
ed at a later age" than was true of parents
of low achievers.
5. More parents of high achievers than low
achievers said that "their child had some
particular talent or skill", while fewer said
"No" to this question.
6 . Parents of high and low achievers did not
differ in the amount of attention they felt
their child demanded as a preschooler. How
ever, fewer parents of high achievers felt
their child demands a great deal of attention
now.
procedure allowed mothers and fathers scores to be compar
ed separately and simplified tabulating the data on the
Summary Page.
80
7. Parents of high and low achievers did not
differ in the amount of curiosity they felt
their child exhibited as a small child.
However, more parents of high achievers saw
their child as "very" curious now, while more
parents of low achievers saw their child as
only "somewhat" curious now.
Looking at all of the comparisons of parents of
high and low achievers that have been reported under this
hypothesis, some tentative conclusions can be drawn.
First of all, parents of high achievers tend to set higher
standards for their children by responding with somewhat
lower ages on the AIS and somewhat higher estimates on the
first trial of the BST. Second, parents of high
achievers seem to be more involved in or committed to
insisting that their child achieve. Data from both the
AIS on which parents of high achievers check many more
items as "Especially Important to Me" in their child’s
development, and the BST, during which these same parents
make significantly more comments, support this conclusion.
Closely related to both these conclusions is the third
one stating that parents of high achievers tend to stress
self-reliance and competent, independent functioning to a
greater extent than parents of low achievers. The
evidence for this conclusion is less clear, but
suggested by the tendency for parents of high achievers to
give slightly lower AIS ages.
A fourth conclusion, based on parents responses to
questions #2 and 5 on the AIS Summary Sheet, is that
81
parents of high achievers tend to evaluate their child's
accomplishments more positively. However, in conflict
with this conclusion is the fact that parents of high
achievers actually gave lower, and therefore apparently
less positive estimates of their child's performance on
the later BST trials. No reasonable explanation for this
conflicting evidence has been found (i.e., estimates from
parents of high achievers do not appear to be more
realistic than those of low achievers in their BST
estimates in view of their child's performance), except
for the obvious fact that the two measures are quite
different. The questions tend to cover a broad (and some
what ambiguous) range of activities many of which may
already have occurred, while the BST estimates are very
specific predictions subject to immediate (and public)
proof or contradiction.
Finally, parental response to question #6 suggests
a fifth, more speculative conclusion. The parents have
indicated that although high and low achievers did not
differ greatly in the amount of attention they demanded as
preschoolers, high achievers tend to need less attention
now (at ages 9-11). Since it seems logical to assume that
the more independent and self-reliant the child the less
attention he will demand, the response to this question
might be translated to read: although high and low
achievers tend to be about equal in dependency needs as
preschoolers, high achievers tend to become more
independent and self-reliant than low achievers as they
progress through elementary school. Based on this
assumption, it might be concluded that the differences
noted in high and low achievers demands for attention and
levels of competence, self-reliance, and independent
functioning are related to the differences noted in their
parents expectations, training, and involvement.
However, the proposed relationship between
parental expectations and their child's achievement does
not clarify an issue related to several of these
conclusions. Do parents of high achievers tend to set
higher standards, expect more independent functioning, and
evaluate their children more positively (when they do) due
to an attitude toward children inherent in the parents,
due to the fact that these children are in reality above
average in ability and/or accomplishments, or due to some
combination of the parents inherent attitude and the
child's performance? The evidence presented so far does
seem to indicate that parents of high and low achievers do
differ to some extent in their attitudes toward and
expectations of their children. The information discussed
later in this section under the heading, ''Relationship
between Measures of Children's Achievement,'' tends to
indicate that high achievers themselves do perform better
than low achievers on the measures used. None of this
83
data, however, specifies which came first--the child's
accomplishments or the parent's attitude. It seems most
probable that they interact.
Another dimension on which the parents of high and
low achievers have been compared by Whiting, Rosen and
D'Andrade, and many other investigators is the extent to
which they tend to agree or disagree on childrearing
practices. Rosen and D'Andrade compared the amount of
agreement obtained between pairs of parents of high and
low achievers on a number of different measures
theoretically related to achievement. They found that
parents of high achievers tended to: agree more on
setting high standards for their child (BST estimates #1,
2, and 3); give their child more autonomy in the decision
making process by themselves participating approximately
equally in it; and agree less on allowing their child to
function independently, than parents of low achievers.
Unfortunately, only the BST data in the current work is
comparable to their work and no clearcut trend is apparent
in it. For BST Trial #1 parents of high achievers were
less likely to give the same estimate than were the
parents of low achievers (for boys the correlation between
parents of high achievers was .11, and .35 for the parents
of low achievers; for girls the two correlations were -.01
and .2 2, respectively) though none of the correlations
obtained were significant at the 5% level of confidence.
For Trials #2 + 3 parents of high achievers did tend to
agree more frequently than did the parents of low
achievers (for boys, r = .69, p = .01, between parents of
high achievers and, r = .24, p > .05, for parents of low
achievers; for girls r = .85, p = .001 and r = .71, p =
.003, respectively). In addition, the relationship
between the number of comments made by mothers to the
number made by fathers, though about the same in high
(r = .78, p = .002) and low (r = .84, p = .002) achieving
boys groups was much stronger between the parents of high
achieving girls (r = .68, p = .0 2) than between the
parents of low achieving girls (jc = .39, p = .10). In
view of the various interpretations that could be made of
these three sets of correlations, it seems reasonable to
conclude that parents of high and low achievers do not
differ consistently in the amount of agreement shown on
setting a particular standard for their children in the
BST.
Using the AIS and an interview as measures of
parental attitudes toward independent functioning, Whiting
made three slightly different comparisons related to the
relative extent of agreement between parents of high and
low achievers. She found that:
1 . parents (as a set) of high achievers tended to
agree on the AIS items most important to
their child's development significantly more
often and disagree significantly less often
than the parents of low achievers;
85
2 . parents of low achievers argued or disagreed
significantly more frequently during the
interview with them than the parents of high
achievers. (However, these results may be
somewhat biased due to the fact that both
Whiting and her subjects were aware of the
achievement group they represented during the
interview.); and
3. the amount of agreement on the AIS ages
selected by fathers and mothers of high
achievers was not significantly different
from the amount of agreement shown by fathers
and mothers of low achievers.
Since information on the amount of agreement or disagree
ment between parents was not specifically noted in this
study, either during the interview or during the data
analysis, these results are directly comparable to
Whiting's only on the third point. Analysis of the
Father - Mother AIS means (the means are presented in
Table 14, and their analysis in Tables 15 & 16) indicates
that the amount of agreement between fathers and mothers
scores for high and low achievers does not differ on the
AIS, just as reported by Whiting.
To summarize the conclusions and issues presented
here in view of the results of this study and closely
related work, parents of high achievers tend to:
1 . set higher standards for their children;
2 . be more involved in their child's achievement;
3. stress, self-reliance and competent,
independent functioning more; (however,
results of the BST as well as Rosen and
D'Andrade's conclusions, strongly suggest
that independent functioning is not encouraged
86
or allowed when it might interfere with
achievementj)
4. evaluate their child's accomplishments
differently (sometimes more positively and
sometimes more negatively),
than parents of low achievers. In addition:
5. differences noted in parental expectations,
training, and involvement may be related to
differences in their childrens demands for
attention and levels of competence, self-
reliance, and independent functioning;
6 . parental attitudes and their child's
accomplishments probably interact; and
7. parents of high and low achievers do not
differ in the amount of agreement shown on
standards set for their child on a particular
activity, but may differ in the extent to
which they agree on the relative importance of
these activities.
Hypothesis 2a.
This hypothesis states that parents of boys have
lower overall and subscale means on the AIS than parents
of girls. Analysis of the AIS overall (Table 5) and sub
scale means (Table 6 ) does not support this hypothesis.
Though the overall means are not significantly different,
parents of boys tend to give slightly older ages on the
AIS than the parents of girls, as shown in Table 17.
Looking at the subscale means, parents of boys give some
what older ages than the parents of girls on four (#1, 3,
4, and 7) of the seven subscales, and the difference
between these groups is significant on Scale #?1
87
(Fj gg = 6.30, p < .02).
No explanation for these results is apparent other
than that the hypothesized relationship is incorrect for
these Ss. Parents of the boys and of the girls involved
in this study did not differ overall in the expectations
for their particular child as expressed on the AIS. The
few differences that occurred in subscale means all tended
to be opposite in direction to the prediction that had
been made. Since the boys and girls involved did not
differ significantly in age, I.Q. Scores, or socioeconomic
status (Table 1) and there were no significant differences
on relevant variables in the two groups of parents (Tables
2 and 3), none of these factors can account for the
results. It is possible either that girls mature faster
and thus are better able to handle the items listed on the
AIS at an earlier age and/or that parents set somewhat
higher (earlier) standards of performance for girls than
for boys. It is also possible that had the boys and
girls in the sample been more carefully matched on school
achievement test scores (the girls scored significantly
higher on these than did the boys), these results might
have been reversed. In view of Gray's findings that AIS
responses did not differ for boys and girls and our data,
the most reasonable conclusion is that, in general,
parents expectations for boys and girls do not differ when
boys and girls are equated on school achievement, age,
88
I.Q., and other relevant variables, or, at least, that
they do not differ on the AIS items.
Hypothesis 2b.
This hypothesis states that parents of boys give
higher estimates on all BST trials than parents of girls.
However, the mean BST estimates for Trials #1 and #2+3
(see Table 18) given by parents of boys do not differ at a
statistically significant level (see Table 8 ) from those
given by the parents of girls. Furthermore, although the
differences in the estimates are slight, parents of girls
tend to give higher estimates than parents of boys in
three of the four comparisons.
Again, no ready explanation of the results is
available. Apparently, the parents of boys and the
parents of girls involved in this study tend to have about
the same levels of aspiration for their child and evaluate
their child's performance and capabilities about the same
way on this task. This is somewhat surprizing in view of
the theoretical differences in parental expectations for
boys and girls discussed in the Introduction and also in
view of the fact that one might have expected parents to
consider block stacking ability more congruent with the
inclinations and abilities of boys than with those of
girls.
Hypothesis 2c.
This hypothesis states that parents of boys make
fewer comments during the BST than parents of girls. The
mean total number of comments made by parents' of boys and
parents of girls (Table 18) and its analysis (Table 8 )
indicate that these two groups of parents do not differ
overall in the number of comments made.
The data obtained clearly do not support this -
hypothesis. Overall these two groups of parents appear to
allow their child about the same amount of independent
functioning and/or are about equally involved in their
child's performance. Apparently they do not see boys as
needing more (or less) help than girls, as had been
expected.
Discussion of General Hypothesis 2
From the results of this study, parents of boys
can be characterized as not differing significantly from
the parents of girls in:
2a. overall AIS responses for their child, but
tending (significantly on Scale #l) to give
slightly older AIS responses on the subscales
2b. BST estimates given for their child, but
tending to give slightly lower estimates; and
2c. the number of comments made during the BST
trials.
Thus, the parents of boys and the parents of girls in this
sample tend to see their particular children as equally
90
competent, have about the same level of aspiration for
their children, allow (or expect from) them about the same
level of independent functioning, and/or are equally
involved in helping their children achieve on the AIS and
the BST.
When the responses of parents of boys and parents
of girls involved in the study are compared on the other
measures obtained, the two groups once again tend not to
differ overall. On the AIS parents of boys mark about the
same number of items as "Especially Important to Me" (see
column 1 in Table 19) as do the parents of girls and only
slightly (and non-significantly) more items "Not Appropri
ate" (see column 1 in Table 20) than the parents of girls.
Looking at the AIS Summary Page (see Table 21), the
tendency, again, is for parents of boys and parents of
girls to respond similarly on five of the eight questions.
There are no significant differences in the way parents of
boys and parents of girls evaluate their child on
Questions #2 and #3 (frequency with which child has
mastered the items on the AIS or the items considered
"Especially Important to Me") and #4 (whether child is
more of a leader or follower in group activities).
Questions on which parents of boys and parents of girls
evaluate their respective children differently (at the 5%
level of confidence) are:
91
5. More parents of girls say their child has "not
developed any particular talents or skills."
6 . More parents of boys said their child
"demanded a great deal of attention as a pre
schooler and now," while more parents of girls
said "no" to this question.
7. Parents of boys and parents of girls did not
differ in the amount of curiosity attributed
to their child when he was small, but more
parents of boys felt their child was "very
curious" now, while more parents of girls
labeled their child as only "somewhat curious"
now.
Thus, looking at this additional AIS data and questions #3
and 4 from the Summary Page, parents of boys and girls
tend to evaluate their children in the same general ways
and choose roughly the same number of developmental goals
(though not necessarily the same goals). However, there
is a slight (but non-significant) tendency on Question #2,
and a statistically significant tendency on Question #5,
for parents of boys to be more positive in their evalua
tions of their sons than are the parents' of girls in
evaluating their daughters. This tendency, in conjunction
with that noted on the AIS for parents of boys to give
higher ages than given by parents of girls and that noted
on the BST for parents of girls to give higher estimates,
suggests the possibility that parents of girls may expect
more of their children in terms of early independence and
achievement than the parents of boys. This higher
expectation of parents of girls for their daughters
92
suggested by the data would not be surprizing in view of
the well documented earlier maturation of girls (Anastasi,
1958) and may well account for the superior performance of
these girls on school achievement tests.
Another interesting comparison of parents of boys
with those of girls is the extent of agreement shown
between a set of parents in the ages used on the AIS items
to describe their expectations for their child. This
index of agreement was calculated by subtracting the
mother's AIS overall and subscale means from the father's.
For parents of boys, the average difference between
fathers and mothers in AIS overall means was .62, while
this difference was .05 for the parents of girls (see
Table 22). Thus, the extent of agreement is significantly
less (F]_}39 = 4.08, p < .05, as indicated in Table 15) for
parents of boys as a group than for parents of girls on
the AIS as a whole. Parents of boys also disagree more in
the ages selected for each of the seven AIS subscales than
parents of girls and this disagreement is significantly
(or nearly so) greater on three subscales (#1 , ^1^39 =
3.75, p < .06; #5, F1>39 = 3.95, p < .05; and #7, F;l>39 =
4.07, p < .05) as shown in Table 16. This greater
similarity in the type of age-specific expectations given
by parents of girls on the AIS may indicate that parents
of girls tend to be more consistent in their childrearing
practices than the parents of boys. If this
93
interpretation is correct, one would expect that parental
goals and expectations for their child would have greater
influence on the child and be more effective than those
from parents with more discrepant views. In addition, the
more similar the parents' expectations, the less freedom
the child would have to act as he wishes in these matters.
Thus, it is possible that parents of girls (in this
sample, at least) not only expect more of their children,
but also are more consistent and effective in their child-
rearing practices and allow their daughters less choice in
what and in how well they will do. However, though
parental childrearing practices and the consistency of
these practices may influence the level of their child's
achievement, the possibility that earlier maturation of
the girls and/or their inherent ability may have shaped
their parents expectations, cannot be discounted.
So far in the discussion of this hypothesis a
mother's and father's responses for their child have been
looked at collectively. However, if parents responses are
looked at individually, there are clear differences
between fathers and mothers responses depending on the sex
of the child on some of the measures. For example,
column two of Table 19 shows that fathers of boys as a
group tend to mark significantly more items as "Especially
Important" in the development of their sons than do the
fathers of girls for their daughters. Conversely, mothers
94
of girls tend to mark many more items than mothers of boys
(this difference is significant at the 2% level when only
the high achieving boys and girls are compared, but does
not reach the 5% level of confidence when all the boys and
girls in the sample are compared). These differences are
less pronounced (and less frequently significant) when the
number of items marked "Not Appropriate" (see Table 20) is
compared. In this case, fathers of boys mark about the
same number as the fathers of girls, while mothers of
boys check more items than the mothers of girls
(significantly so when low achieving boys and girls are
compared). In contrast, the patterns of fathers and those
of mothers responses on the AIS Summary Page are very
similar and so have not been tabulated separately. Per
haps the differences noted in the extent of parental
response (where differences occur) is a reflection of how
involved that parent is or is not in that child's develop
ment as well as the extent to which the parent finds this
involvement compatible with his role as a father or mother
of a boy or girl.
The following points summarize the data and
discussion presented under this hypothesis comparing the
responses of parents of boys with the responses from
parents of girls:
1. In general, the parents of boys and girls do
not differ significantly in either the level
of independent functioning and competence
95
expected from their children on the AIS and
BST, nor in the level of aspiration they have
for their children, nor in the extent of their
own involvement in these tasks.
In the few cases where differences in
expectations occur for parents of boys and
parents of girls, parents of girls tend to
expect somewhat more from their children and
evaluate their accomplishments somewhat less
positively.
Parents of girls are significantly more in
agreement in their expectations for their
child's level of competent, independent
functioning than are the parents of boys, and
thus may be more effective in influencing
their child's achievement. In addition, this
greater agreement between parents probably
allows girls less freedom of choice in matters
of achievement.
Fathers of boys are more involved in their
son's development than are fathers of girls in
their daughter's development. Conversely,
mothers of girls are more involved in their
daughter's development than are mothers of
boys in their son's development.
The first two points are essentially contradictions of the
relationship stated in this hypothesis and of the
literature cited in the "Introduction” referring to the
greater emphasis on competence and independence training
used in rearing boys as compared with girls in our
society. However, since the girls in this sample do score
higher on school achievement tests than the boys, it does
tend to confirm Hypothesis 1. The third point suggests a
possible explanation for the conflict between the data
and Hypothesis 2. Rosen and D'Andrade have suggested that
3.
96
n-Ach requires:
1 . achievement training--training which
emphasizes ’ 'doing well,”
2 . self-reliance--training which emphasizes
performance "by himself” and
3. autonomy--training which emphasizes freedom of
action in decision making
as well as parental sanctions to see that their demands
are fulfilled. Parents of the boys and girls involved in
this study may not differ to any significant degree (as
the data indicate) in the level of competence or the level
of self-reliance expected from their children in
activities such as listed on the AIS or involved in the
BST. But, they may differ to a much greater extent in the
amount of autonomy allowed--either in the autonomy granted
directly or that granted indirectly as a result of the
discrepancy in parental goals and expectations, and
ultimately, in sanctions. Thus, it seems more reasonable
to interpret these results as an indication that parents
of boys and girls do not differ in the extent to which
they emphasize independence training, but differ greatly
in the type of independence training they emphasize.
Relation between Hypothesis 1.
and Hypothesis 2.
Although no relationship between parental
expectations and their child's level of achievement and
97
sex was hypothesized, it was of interest to look at this
question. Do parents of high and low achieving boys
differ in their expectations for their children in the
same way that parents of high and low achieving girls do?
Data from a number of different AIS and BST measures are
available to answer this question.
Starting with the analysis of the AIS overall
means (Table 5), there is no significant difference
between the overall age-expectancies of parents of high
and low achieving boys and girls. However, these parents
do differ at a near-significant level on Scale 4--Social
Responsibility, where = 3.16, p < .08, as shown in
Table 6 . On this scale parents of boys tend to respond
with older ages for high than for low achievers; the
parents of girls give younger ages for high achievers (see
Table 4). The relationship between parental expectancies
for high and low achievers is about the same for boys as
for girls on the other six subscales.
The extent of agreement shown by a child's father
and mother in AIS Overall Means was also analyzed in
relation to this question. No difference was found in the
amount of agreement shown when differential expectancies
expressed on the whole AIS for high and low achieving
boys were compared with those for high and low achieving
girls (see Table 15). Again, this interaction approaches
the 5% level of statistical significance, however, when
the differences between father's and mother's AIS sub
scale means for the groups is analyzed. In this case, the
trend toward interaction occurs on four of the seven sub
scales (#1 where F-^^g = 3.14, p < .08; #2 where F^^g =
2.81, p < .10; #3 where F^^g = 3.06, p < .09; and to a
lesser extent on #6 ). On all four of the scales the
direction of the difference in the relationship is the
same. For boys, fathers of high achievers tend to give
older ages than their wives while fathers of low achievers
give closer to the same age as their wives (on two scales)
or a somewhat lower age than their wives (on two scales).
For girls, fathers of high achievers give younger ages
than their wives while fathers of low achievers either
give higher (on three scales) or about the same (on one
scale) ages as their wives do (see Table 14).
The interaction between a child's achievement
classification and sex is again evident when the number of
items considered "Especially Important" to his development
(Table 11) is surveyed. Fathers of boys mark signifi
cantly more (X(i)2 = 10.64, p < .0 1) such items for high
than for low achievers and then do the opposite [ mark many
more (but not significantly more) items for low than for
high achievers] for girls. This interaction is not
evident in mothers' responses, however.
There is less supporting evidence for this inter
action in the BST data. A non-significant interaction is
99
apparent in the estimates for Trials #2 + 3 (Table 7),
where parents of boys tend to give higher estimates for
low than for high achievers, while parents of girls tend
not to differ in their estimates for high and low
achievers. There is also another non-significant tendency
(Table 18) among the parents of high achievers for the
parents of boys to make fewer comments than the parents of
girls, while among parents of low achievers, parents of
boys make more comments than the parents of girls.
The following points summarize the data which
suggest a differential in parental expectancies for high
and low achieving boys compared with that for high and low
achieving girls:
1. When father + mother AIS Overall Means are
analyzed, there is no evidence for this inter
action; however, a non-significant interaction
does occur on one subscale (#4) on which
parents of boys respond with older ages for
high than for low achievers, while the parents
of girls give younger ages for high achievers.
2. When father - mother AIS Overall Means are
analyzed, there is no evidence for this inter
action; however, interaction tendencies occur
on four subscales (#1, 2 , 3, and to a lesser
extent, 6 ). On all four scales fathers give
higher AIS ages relative to mothers for high
achieving boys than for low achieving boys.
For girls, mothers give higher AIS ages
relative to fathers for high achieving girls
than for low achieving girls.
3. A significant interaction between these two
factors occurs in the number of AIS items
marked, "Especially Important" by fathers.
Fathers of boys mark many more items for high
100
than for low achievers while fathers of girls
reverse this trend. No interaction is found
in this type of response from mothers, however.
4. A tendency toward interaction occurs in the
estimates for Trials #2 + 3 of the BST.
Parents of boys tend to give higher estimates
for low than for high achievers while parents
of girls do not differentiate their respective
estimates.
5. A tendency toward another type of interaction
is apparent in the number of comments made by
parents during the BST. Parents of high
achievers tend to make more comments to girls
than to boys while parents of low achievers
make more comments to boys than to girls.
Thus, in answer to the question posed, parents of the boys
in this study appear to differ from the parents of the
girls in their expectations for high and low achievers,
particularly on the AIS.
Although the majority of these interactions are
non-significant, they are of interest due to their
consistency in direction. Parents of boys seem to expect
less and set lower standards for high achievers relative
to low achievers on the AIS (point l) and BST (point 4),
while parents of girls expect more on the AIS or about the
same on the BST from their high versus low achievers.
These differences in parental expectancies and standards
for high and low achieving boys and girls (points 2 and 3)
also raise the question of which (if either) parent's
expectations are more important in their child's develop
ment. Although the data cannot answer this question, they
do suggest the impact that the fathers expectations may
have on the child's behavior, either directly as he inter
acts with the child or indirectly as he influences the
mother's expectations of the child, and emphasize the
importance of studying the Father-Mother-Child rather than
just the Mother-Child relationship, when possible. In
addition, the data (points 3 and 5) reemphasize the
importance of parental involvement in their child's
development, and this is especially clear in the boys
sample. Here, though parental standards appear to be set
somewhat lower for high achievers than for low achievers,
parental involvement is significantly greater. Thus,
parental involvement may be a more significant factor in
achievement than the type or level of parental goals and
expectations.
Hypothesis 3
Although Hypothesis 3,, is parallel in form to
Hypotheses lj,,and 2., and the dependent measures are the
same, the data relevant to this hypothesis were treated
differently and, therefore, are reported somewhat
differently. This was necessary because the children in
the study were selected on the basis of their differences
on the criterion variables involved in Hypotheses 1. and
2 . (achievement and sex, respectively), but were not
evaluated as to their status on the criterion variable for
102
Hypothesis 3. (perceptual style). Thus,while the sample of
children is composed of an approximately equal number of
high and low achievers, half of whom are boys and half
girls, the children involved cannot be divided clearly and
unequivocally into field independent (Fl) and field
dependent (FD) groups. Therefore, the relationship between
parental expectations (as expressed on the AIS and BST) and
their child’s perceptual style was analyzed by correlation
(Pearson r) rather than analysis of variance or covariance.
In addition, it was found that the two measures of
perceptual style (EFT and RFT) were unrelated for the boys
in the sample (r = .01), but positively related (r = .45,
p = .02) for the girls. Since even the correlation
between these tests for girls does not account for most of
the variance involved in the scores, and, from a practical
viewpoint it is unclear which test measures perceptual
style more accurately, the relationship between parental
responses on the AIS and BST is reported for both
measures. The correlations for the parents of boys and
the parents of girls were performed separately so that any
differences in these relationships for boys and girls
could be noted.
Hypothesis 3a.
This hypothesis states that parents of FI children
have lower overall and subscale means on the AIS than
parents of FD children. The data obtained from the AIS
and EFT (presented in Table 23) for boys does not support
this hypothesis. None of the correlations obtained are
statistically significant, nor are they consistent in
direction. Thus, there is no clear relationship between
boys EFT scores and their parents' AIS overall and sub
scale means. For girls, most of the AIS and EFT data
tends to support the hypothesis. Though the correlation
between the girls EFT scores and their parents AIS
responses is low and non-significant, it is positive on
four of the seven subscales (near zero on two others and
negative on another) and nearly significant on Scale 6
(Wide Experience) where r = .30, p = .08. Thus, for girls
this relationship tends to be in the direction predicted
and suggests that the older the parental age-expectancies
on the AIS, the higher (more FD) their daughter's EFT
score.
Neither the data for boys nor that for girls
support the hypothesized relationship between parental AIS
responses and their children's RFT scores. For boys, a
clear negative relationship exists between their own RFT
scores and their parents AIS Overall Means (r = -.37, p =
.05) and Subscale Means (see Table 23). Thus, the data
for boys clearly contradict the prediction. Specifically,
boys who tend to score lower on the RFT and are therefore
considered FI, tend to have parents who respond with
104
somewhat older ages on the AIS and have higher overall and
subscale means. On the other hand, there tends to be no
relationship between parental expectations as expressed on
the AIS and their daughter's RFT, except for Scale 6 where
the relationship, though not statistically significant, is
once again positive (r = .29, p = .09) and thus, in
support of the hypothesis.
The results for boys not only do not support the
hypothesized relationship, but probably suggest more
questions than they answer. Looking only at the data for
boys, it is not surprizing that the relationship between
their EFT scores and the AIS and that between their RFT
scores and the AIS were quite different, since the EFT and
the RFT scores were not related for the boys in this
sample. (The question of why the EFT and RFT scores were
not related in this sample of boys can probably be given
a tentative answer--the homogeneity of the sample; but,
the question of which measure of perceptual style is more
appropriate for these boys is more difficult to answer and
will be discussed at greater length under Hypothesis 4.)
It is also not overly surprizing to find that parents of
FI and FD (on the basis of EFT scores) boys do not appear
to differ in their AIS responses, since this group of boys
and their parents is relatively homogeneous and since AIS
responses reported in this study for parents of high and
low achievers and of boys and girls have differed little.
105
However, it is surprizing to find that the parents of FI
(on the basis of RFT scores) boys tend to respond with
older ages than the parents of FD boys. Assuming that the
AIS is a reasonable index of the level of independent
functioning parents expect of their children over a wide
range of activities and that the RFT is a reasonable
measure of perceptual style, no explanation for the
results is apparent. They are clearly in conflict with
earlier research and the theoretical explanations of the
origin of perceptual style offered by Witkin and others.
Overall, the EFT-AIS data for girls tends to show
the expected relationship. It indicates that parents of
FI girls expect more independent functioning from their
daughters at an earlier age than do the parents of FD
girls. Except for the correlation with Scale 6 supporting
the hypothesis, however, the RFT-AIS data shows no
consistent or significant relationships. An interpreta
tion of this set of data would indicate that parents of FI
and FD girls do not differ in their expectations for
independent functioning in their daughters. Thus, though
the difference between the relationship of the AIS with
the EFT and with the RFT is less extreme for the girls
than for the boys in the sample, the relative validity of
these two measures of perceptual style is clearly an
important issue here in the interpretation of the results.
This issue is discussed in more detail under Hypothesis 4.
106
However, the current work was not designed to test the
validity of these measures, nor can it. In view of this,
both interpretations will be considered in the discussions
to follow.
Hypothesis 3b.
This hypothesis states that parents of FI children
give higher estimates on all BST trials than parents of FD
children. The relationship between the BST estimates and
the measures of perceptual style tends to be in the
direction predicted for both boys and girls (see Table 24).
However, the correlations between these measures are all
relatively low and only those correlations between the EFT
scores and the estimates for Trials #2+3 for boys (r =
-.37, p = .05) and estimates for Trial #1 for girls (r =
-.29, p = .09) fall within or close to the range normally
considered statistically significant.
If the parents' estimates are looked at
individually rather than combined into a single estimate
representing both parents, fathers estimates on Trial #1
appear to show the strongest relation to their child's
perceptual style score, while mothers estimates on Trials
#2 + 3 appear to be more closely related. For Trial #1,
the correlation between fathers estimates and their
daughter's EFT scores is -.47 (p = .01), while the
relationship with mother's estimates is zero. Fathers
107
estimates are also significantly related to their son's
RFT scores on this trial, r = -.45 (p = .02), while
mothers are not (r = 0). For Trials #2+3, the situation
is reversed. In this case, the relationship between
mothers estimates and their son's EFT scores is -.40 (p =
.04), while between fathers and their sons, r = -.25 (p =
.14). On the RFT, the correlation is -.27 (p = .11) for
mothers estimates and their daughters scores and -.16 (p =
.23) for fathers and their daughters.
Thus, though the correlations between parental
estimates during the BST and their child's perceptual
style measures are not particularly high, the correlations,
and particularly those which are statistically
significant, are in the expected direction. Parents of FI
children do tend to give higher estimates on the BST
trials than parents of FD children. The strongest
relationships between BST estimates and their child's
perceptual style scores were found for fathers on Trial #1
and for mothers on Trials #2 + 3. No consistent pattern
appears to be associated with the relation between sex of
parent, sex of child, and strength of correlation
obtained. Overall then, the data indicate that parents
(especially fathers) of FI children tend to have higher
aspirations for their children and (especially mothers)
consider their children more competent (or evaluate them
more positively) than do the parents of FD children.
108
Hypothesis 3c.
This hypothesis states that parents of FI children
make fewer comments during the BST than |>arents of FD
children. Three of the four correlations indicating the
relationship between the number of comments made by
parents during the BST and their child's perceptual style
scores are opposite in direction to the hypothesized
relationship (see Table 24). Thus, parents of FI children
(both boys and girls) tend to make more comments during
the BST than parents of FD children. However, all of the
correlations are low and only the one relating the number
of comments made by parents of girls to their daughter's
EFT score approaches the 5% level of confidence (r = -.31,
p = .08).
When mothers and fathers comments are looked at
separately, the number made by fathers tends to be more
strongly related to their son's or daughter's score on the
EFT (r = -.26, p = .14; r = -.35, p = .05; respectively)
than does the mother's (r = -.05; r =• -.15, p = .25). On
the RFT, the number of comments made by mothers is more
closely related to their son's scores (r = -.40, p = .04)
than is the fathers (r = -.16, p = .25). The relationship
for the number of comments made to girls RFT scores is
very low and about the same for both mothers (r = .14) and
fathers (r = .15). It is interesting to note that even
though fathers do make twice as many comments as mothers
109
during the BST (102 versus 47), the number of comments
made by each parent is significantly related to at least
one measure of their child's perceptual style. As in the
previous hypothesis (3b.), however, no consistent pattern
is evident in the relation between sex of parent, sex of
child, and strength of correlation obtained.
The results just presented are consistent with
those discussed under Hypothesis lc. They indicate that
the number of comments made during the BST either may not
be an accurate index of independent functioning or that
allowing the child to function independently in this
situation (by making fewer comments) is less important to
the parents than having the child perform as well as
possible (thus, necessitating many guiding comments from
the parents). The number of comments made during the BST
may also serve as a measure of another aspect of parental
functioning--the extent of parental involvement in their
child's achievement. On the basis of these assumptions
and the data reported, parents of FI children tend to
value and encourage compentence more than independence in
their children and also tend to be more involved in
helping their child achieve than the parents of FD
children.
Discussion of General Hypothesis 3.
The results presented to this point relating
110
parental responses on the AIS and BST to their child's EFT
and RFT scores, indicate:
3a. (1) no consistent relationship between boys
EFT scores and their parents AIS
responses;
(2) that on the EFT, FI girls tend to have
parents who give lower age-expectancies
on the AIS than do the parents of FD
girls;
(3) that on the RFT, FI boys tend to have
parents who give significantly higher
age-expectancies on the AIS than do the
parents of FD boys;
(4) no consistent relationship on the RFT
between girls scores and their parents
AIS responses except for Scale 6 on
which parents of FI girls tend to give
lower ages than parents of FD girls;
3b. that parents of FI children tend to give
higher estimates on the BST trials than
parents of FD children;
3c. that parents of FI children tend to make more
comments during the BST than parents of FD
children; and
3b. that although the correlation between BST
estimates or number of comments made with
& their child's EFT or RFT score may be much
higher for one parent than for the other, no
3c. consistent pattern is evident in the relation
between sex of parent, sex of child, and
strength of correlation obtained.
Overall then, the BST data and measures of perceptual
style indicate that the parents of FI children (both boys
and girls) tend to set higher goals for their children,
see them as more capable, and are more involved in helping
Ill
their children achieve than the parents of FD children.
The AIS data are more difficult to interpret, since they
are not consistent for boys and girls and relate
differently to the two different measures of perceptual
style. However, using the EFT as the measure of
perceptual style, the parental expectancies expressed on
the AIS for girls tend to correspond with those expressed
by those parents during the BST. The AIS responses from
the parents of boys in contrast to the BST responses of
these same parents, either do not relate to the perceptual
style scores of their sons or are contrary to the
relationships found with the BST data.
Thus, parental expectations, as measured on the
AIS and BST, for the parents of FI girls compared with
those of FD girls, generally (with the exception of the
RFT-AIS data) tend to relate in the same way as they do in
the parental group as a whole. The parents who tend to
put lower ages on the AIS also tend to give higher
estimates on the BST trials and make more comments during
these trials. When the parents of FI and FD boys are
compared, the relation between parental AIS and BST
responses appear either to be non-existant or opposite in
direction to that found true of the parental group as a
whole.
Considering the confusing and sometimes
contradictory correlations obtained between parental
112
(Father + Mother) responses on the AIS and their son's or
daughter's EFT and RFT scores, the consistency of the
relationships found between these two measures and the
Father - Mother AIS means for parents of both boys and
girls is surprizing. A statistically significant positive
relationship between the EFT scores and Father - Mother
AIS Overall Mean was obtained for boys (r = .46, p = .02)
and a nearly-significant one for girls (r = .31, p = .08)
as shown in Table 25. A significant positive relationship
between Father - Mother Subscale Means and the EFT is also
found on three of the seven subscales (#2, 6 , and 7) for
boys and on two of the subscales (#2 and 4) for girls.
The correlations between children's RFT scores and
the average difference in their parents AIS means are
lower and less frequently statistically significant, but
consistently positive for both boys and girls. The
correlation between RFT scores and parental differences on
the AIS Overall Mean are .21 (p = .19) for boys and .19
(p = .19) for girls. Only the relationship between RFT
scores and Scale 3 (r = .42, p = .02) is significant for
girls; none of the correlations for boys are significant.
Thus, though analysis of the relationship between
the collective parental AIS responses (Father + Mother
means) and their child's perceptual style scores have not
clarified the relationship between parental expectations
for independent functioning and their child's perceptual
113
style, analysis of the relationship between parental
differences in expectancies and the measures of perceptual
style has been more useful. The data indicate that there
is a definite positive relationship between the
discrepancy in parental AIS responses and their child's
perceptual style. Specifically, the higher the Father -
Mother AIS mean the higher the child's EFT and (to a
lesser extent) RFT score. Thus, in families where the
father tends to give older AIS responses than the mother,
the children tend to have more trouble with the EFT and
RFT and are more likely to score in the FD range.
The following conclusions tend to summarize the
results and discussion relevant to the parents of FI and
FD children. Parents of FI children tend to:
1 . set higher standards for their children;
2 . see their children as more capable;
3. be more involved in their child's achievement;
and
4. agree more frequently in their age-expectan
cies for their child's independent functioning
(and/or the fathers choose younger age-
expectancies than the mothers do more
frequently)
than the parents of FD children.
5. Parents of FI girls also tend to expect more
independent functioning from their daughters
at an earlier age than the parents of FD
girls.
6 . Parents of FI boys either do not differ from
the parents of FD boys in their expectancies
114
for independent functioning or tend to expect
their sons to function independently at an
older age than do the parents of FD boys.
The AIS as a Measure of
Parental Expectancies
Since the AIS is a relatively new and untried
instrument as well as being a self-report questionnaire
subject to various kinds of distortions in responding and,
thus, discrepancies between actual and reported behavior,
it seems desirable to use the data obtained in this study
to evaluate its usefulness. Assuming the attitudes of
parents of high and low achievers, boys and girls, and FI
and FD children differ in some systematic way toward
independent functioning in their children, how well does
the AIS measure this? Is there any evidence that parental
responses are affected by factors other than the level of
independent functioning expected from their children? If
so, what ’ are these factors?
The data already presented under Hypotheses 1.,
2., and 3. is relevant to the first question. Analysis of
the AIS Father + Mother Overall (Table 5) and Subscale
Means (Table 6 ) for parents of high and low achievers,
boys and girls, and FI and FD children (Table 23), in
general (with the exception of the correlation with the
RFT for boys), does not indicate clear, statistically
significant differences between these groups, though some
115
non-significant trends are apparent. However, analysis of
the Father - Mother Overall (Table 15) and Subscale Means
(Table 16), though not differentiating parents of high and
low achievers, did show clear, significant differences
between the responses of parents of boys and girls and FI
and FD (on the EFT) children (Table 25) as well as between
fathers and mothers in the group as a whole (see Tables
26 & 27). In addition, parental responses on the AIS
showed a low, but consistent relationship to their
behavior during the BST (Table 10), another measure of the
amount of independent functioning allowed. In answer to
the first question, then, the version of the AIS used in
this study appears able to measure differences in parental
expectancies, when they exist, reasonably well.
However, these same analyses also point out the
need to revise the subscales. In several cases, although
the multivariate test of a factor indicates that its
effect on the AIS as a whole is significant, the separate
univariate tests for that factor on the various subscales
are not significant. A particularly clear example of this
phenomena is shown in Table 6 and involves the tests for
the significance of the Achievement X Sex interaction in
Father + Mother AIS Subscale Means. The multivariate F
test is highly significant in this case, yet none of the
univariate tests of this interaction are significant. In
addition to this statistical evidence that overall AIS
116
variance is not necessarily accurately reflected in the
subscale means, some of the parents reactions to the AIS
emphasize the shortcomings of the subscales. Parents, and
particularly fathers of boys, objected to a number of AIS
items as "Not Appropriate" in their child's development.
Three of the items most frequently objected to as
inappropriate for boys (see tabulation in Appendix F)
were:
27. Wash dishes as a regular chore. (Scale 4)
35i Sew on a button. (Scale 3)
47. Iron own clothes. (Scale l)
Although the three activities are different and may
require somewhat different kinds of abilities and
responsibilities, all three of them deal basically with
domestic chores and it seems unlikely that they should
fall on three different scales. Thus, the evidence
indicates that the differences found between parents on
the AIS items as a whole are not accurately reflected in
the subscale scores. Decreasing the number of subscales
and assigning items to subscales on the basis of factor
analysis would increase both reliability and validity and
make the AIS a more sensitive and useful measurement
instrument.
Data relevant to the second question is presented
in Table 28. It indicates that parental responses are
affected by at least two factors other than the level of
117
independent functioning expected from their children,
namely their children's age and general intelligence (as
reflected in I.Q. scores). Thus, parents of the older
children (both boys and girls) in the sample tended to put
down older ages on the AIS than parents of younger
children. In addition, parents of children with higher
I.Q. scores tended to put down younger ages than parents
of children with lower scores.
Just what effect these two factors may have had on
the results obtained under Hypotheses 1. to 3. is
difficult to determine. It seems unlikely that it was
very great in the comparison of parents of boys with
parents of girls, since the boys and girls did not differ
in mean age or I.Q. scores. High and low achievers also
did not differ in age, but did differ in I.Q. scores, a
factor which should have worked to lower AIS means more
for parents in the high achievers group than in the lower
group. Looking at the relationships between perceptual
style and age and I.Q. for this sample shown in Table 29,
the more FD children would have been younger and therefore
had parents who tended to put down younger ages than the
parents of FIs. However, at least among the girls, those
who were younger .would also have tended to have lower I.Q.
scores and, thus, parents who put down older ages--so the
net effect of age and I.Q. score on parental response (for
girls) in relation to perceptual style would most likely
118
have been zero. For boys, since EFT scores are not
related to I.Q. (in this sample), the relationship of age
to AIS responses would have worked against the predicted
EFT-AIS relationship. Since the RFT is apparently not
related to either age or I.Q. in these boys, it is not
possible to assess the influence of these two factors on
the RFT-AIS relationship.
The analysis of Father - Mother AIS means
presented in Table 26 also indicates another factor which
influences the responses obtained. Fathers tend to expect
their children to be able to master the AIS items at a
later age than do mothers. For the group as a whole,
fathers expect their children to have mastered the 62
items seven months, on the average, later than mothers do
and this difference is significant at less than the .03
level of confidence (Fis43 = 5.06) when the Father -
Mother Overall Means are analyzed. Fathers give older age-
expectancies than mothers on five of the seven subscales
and this difference is statistically significant on three
of these scales (#2, 5, and 7) as shown in Table 27.
Thus, in addition to the data discussed under Hypothesis
2 . indicating that mothers and fathers differed in their
AIS responses depending on the sex of the child involved,
the results of this analysis indicate that mothers and
fathers differ anyway, on the average, in their responses
for the same child (whether the child is a boy or a girl).
119
In summary of the data relevant to the second
question, a number of factors other than parents
expectations for independent functioning influence their
responses on the AIS. These factors are:
1 . the child's age— parents of older children
tend to put down older ages than do parents
of younger children;
2 . the child's general intelligence as reflected
in his I.Q. score--parents of children with
higher I.Q. scores tend to put down younger
ages than do parents of children with lower
scores;
3. sex of parent--fathers tend to put down older
ages than do mothers;
4. sex of child combined with sex of parent--
fathers of boys tend to mark more items as
"Especially Important" in their sons develop
ment than do fathers of girls; conversely,
mothers of girls mark more items for their
daughters than do mothers of boys.
Although the data discussed gives clear evidence for the
existence of these factors in this study, it seems
unlikely that they have appreciably changed the results
reported for Hypotheses 1., 2., and for the girls on
Hypothesis 3., since the factors were often counter
balanced either in the groups themselves or in the data
analysis. However, their effect may have worked against
the predicted relation between EFT and AIS scores for
boys on Hypothesis 3., and had an unknown effect on the
RFT-AIS relationship.
Two other factors which appeared to be potential
120
problems of the AIS have apparently had no effect on
parental responses. The first of these possibilities was
a form of response set in which the respondent, in
completing the Scale, might tend to use the same age
throughout the 62 items. Brief scanning of the completed
forms gave no evidence of this type of response. The
second factor which appeared to present a problem in the
use of the AIS in this population was that the AIS includ
ed a number of items a nine- to eleven-year-old child has
already accomplished as well as those he has not. Thus,
the parent is forced, first of all, to rely on his memory
for many items. Secondly, in responding to the Scale, he
is forced to switch back and forth between sets, i.e.»
"My child should accomplish __________ at what age?" and,
"My child did accomplish _____________ at what age?"
However, significant differences between parents were
found on both Scale 5 and Scale 7 (as shown in Table 16)—
subscales on which the average age-expectancy for the
whole group of parents (see Appendix F) was at least six
months below that on the AIS as a whole. In addition, the
mean age-expectancy for the entire population of parents
on the AIS was 8.79 years, almost three years below the
average age of the children involved. Thus, including
items on the Scale which are below the age level of the
children involved does not seem to present any serious
measurement problems.
121
However, most parents seemed to find the AIS
difficult to complete. Many of the initial items were not
only difficult for parents to remember, but also vague and
ambiguous as well. In addition, each item required a
specific, definitive response. Thus, even if the parent
found that item clear, and relevant to his child and to
the current time span, he would often find it difficult to
answer. These comments are not meant to suggest that the
AIS should be discarded or that great effort is needed to
clarify it before any further use. Rather the intention
here is to point out that completion of this type of
questionnaire is a difficult task and certainly requires
the full cooperation of the respondent in order to obtain
a meaningful result. Considering the difficulty of the
task and the many questions asked about the AIS (and the
purpose of the study), it seems unlikely that mailing the
AIS to respondents would be a very practical or profitable
procedure. Decreasing the length of the AIS to fifty
items would make the respondent1s task somewhat easier and
decrease the time needed to complete the Scale. (Though
no effort was made to time respondents, it usually took a
parent about twenty minutes to complete the 62 items;
parents seemed to be able to complete the last forty items
at least as fast as the first twenty--probably because
their ideas of what they were to do were clearer and
because the initial items tended to be more ambiguous,)
122
without changing its reliability to any great extent.
Deleting the requirement that respondents check each of
the AIS items that they feel are "Especially Important" in
their child's development would also make the AIS easier
to complete. However, the results obtained here indicate
that this measure is a significant index of the parent's
interest in and involvement with his child for this
population at least and should not be discarded.
In summary, then the AIS appears to be a reasonable
measure of parental attitudes related to the level of
competent, independent functioning expected from their
children. However, parental responses are also
influenced by their child's age, level of intelligence,
sex, as well as the sex of the respondent, though it seems
unlikely that these factors have significantly changed the
results reported here (with the exception of the EFT-AIS
and RFT-AIS relationships for boys). No evidence was
found to indicate that parents tended to respond in any
"set" way throughout the Scale or that items the child had
already accomplished were less valid indicators than items
he had not.
The utility of the AIS could be increased by:
1 . using factor analysis to determine the number
of subscales necessary and the items
appropriate to each subscale; and,
2 . reducing the overall number of items to fifty;
this would probably not significantly reduce
123
reliability and would make the respondent’s
task somewhat easier.
No changes in format or administration procedure are
recommended. The "Especially Important" column (added by
Whiting and retained in the current revision) appears to
be a valuable addition to the Scale. As such, it appears
to provide a clear index of parental involvement with
their child which relates to school achievement, while
taking little additional respondent time. Though the AIS
was designed to be self-explanatory so that it could be
mailed to respondents or administered by an untrained
person, it seems unlikely that most parents would bother
to complete it under those circumstances.
In conclusion, though the AIS has been discussed
here primarily as a measure of independent functioning, an
argument might be made that it is also a set of standards
to be mastered and, thus, measures both independence and
achievement expectancies. In addition, the data presented
(on the influence of factors such as the child's age and
I.Q. and the apparent lack of effect of using items
referring to activities the child has accomplished mixed
with those he has not) could be equally well interpreted
as indicating that the Scale is a measure of the parent’s
evaluation of his child's level of competence and
independent functioning in these activities, rather than
just his expectation. In line with these comments and the
124
material presented earlier in this section, the AIS
appears to measure parental attitudes (expectations and/or
evaluations) toward their child*s;
1 . level of independent functioning;
2 . level of competence; as well as the:
3. kinds of childrearing goals set;
4. number of childrearing goals set (apparently
an index of parental involvement with the
child); and,
5. extent of parental agreement on these goals.
The BST as a Measure of
Parental Expectancies
The BST developed by Rosen and D1Andrade appeared
to show clear differences in behavior between their high
and low achieving boys and their respective groups of
parents. How clearly did it differentiate the behavior of
the parents of high and low achievers, boys and girls, and
FI and FD children in this study? To what extent were the
BST scores influenced by extraneous variables?
In answer to the first question, the data reported
under Hypotheses 1., 2., and 3. indicate that significant
differences were found in the BST scores of parents of
high and low achievers (Table 8 ) and in those of parents
of FI and FD children (Table 24); no significant differ
ences were found in the scores of parents of boys compared
with parents of girls. Thus, it appears that all three
125
BST measures are able to identify differences in parental
behavior when they exist. Furthermore, the pattern of
scores— parents who give higher estimates on Trial #1 tend
to give higher estimates on Trials #2 + 3 as well and to
make more comments during the trials--tends to be consis
tent from one comparison to another and to be consistent
with the response of those parents on the AIS (see Table
10). In addition, since the correlations between parents
first estimate (prior to watching their child perform) are
not strongly related to their estimates for the last two
trials (for mothers the correlation is .29; for fathers,
.42), it seems likely that these two measures may differ
somewhat, as suggested by Rosen and D'Andrade.
Data relevant to the second question is presented
in Table 30, and indicates several statistically
significant relationships between parental BST measures
and their children's I.Q. scores and age. Parents of both
boys and girls tend to make more comments to children
having higher I.Q. scores than to children having lower
scores. Parents of girls also tend to make more comments
to older girls than to younger girls, while parents of
boys tend to decrease their number of comments as their
son's age becomes greater. The BST estimates of parents
were, in general, not significantly related to either
their child's age or I.Q. score, except for the positive
relationship found between boys I.Q. scores and their
126
parents estimates on the last two trials.
In view of the relationships between these
extraneous variables and the BST measures, how reliable
are the results reported under the first three hypotheses?
As can be seen in Table 8 , controlling the variance due to
age and I.Q. by using these variables as covariants had no
effect on the significance of BST scores for parents of
boys compared with parents of girls. When the scores for
parents of high and low achievers were compared, covarying
age again had little effect, but controlling for I.Q.
decreased the significance of the total number of comments
below acceptable statistical levels. However, since
controlling the variance due to the child's performance
increased the F ratio for parents of high and low
achievers, it seems probable that covarying age, I.Q., and
the child's performance simultaneously would still make
this comparison statistically significant.
The assessment of the impact of these variables on
the results reported under Hypothesis 3. is more complex.
For boys, neither the relationships reported for estimate
#1 nor that for #2+3 would have been effected, since
neither perceptual style measure is related to I.Q.
scores. The number of comments made by FI as compared to
FD parents was probably reduced artificially by the
tendency for parents of older (and, therefore, more FI)
boys to make fewer comments. Thus, for boys the negative
127
correlation between number of comments and EFT scores
should probably be somewhat higher and more significant
than that reported. Since parents BST estimates were
apparently not influenced by their daughters age or I.Q.
score, neither factor is likely to have influenced these
results. However, since FI girls tend to be older and
have higher I.Q. scores and both factors are positively
related to the number of comments made by parents of
girls, it seems likely that the negative correlations
between total comments and EFT or RFT are higher than they
might otherwise be.
The BST itself was a task most parents entered
into enthusiastically and seemed to enjoy. Many of the
children were initially reluctant to participate, but
became more relaxed as they found they could perform well
and were generally delighted with having earned some
money. The administration time was not systematically
noted but probably averaged forty minutes per family and
varied greatly from family to family, ranging from ten to
fifteen minutes to well over an hour, depending upon the
care taken by the parents and the child and the strength
of their collective determination to do "well 1 1 In
addition, some children seemed much surer of how to
procede and/or were more willing to go ahead without
extensive (and time consuming) instructions from their
parents. A record of the time necessary for the task and
128
the child's interest in "doing well" and assurance in
tackling the blocks might be useful additional measures.
In conclusion, the BST appears to be a reasonable
measure of the parents':
1 . level of aspiration for their child;
2 . level of competence attributed to their child;
3. level of involvement with their child; and,
4. extent of agreement on these measures
as well as a task which most families enjoy. However,
parents responses also appear to be influenced by the age
and relative intelligence (as reflected in I.Q. scores) of
their child. In spite of the operation of these
extraneous factors in this study, no re-interpretation of
the results is needed for Hypotheses 1. and 2. On
Hypothesis 3., it seems unlikely that the correlations
between parents BST estimates and their child's perceptual
style were significantly altered by the operation of these
extraneous variables. However, the relation between total
comments and boys EFT scores is probably somewhat de
flated and that with girls EFT scores slightly inflated.
Hypothesis 4
This hypothesis states that high achieving
children will score lower on both the EFT and the RFT than
low achievers. Looking at the EFT data in Table 31, and
its analysis (Table 32), high achievers do score
129
significantly lower (^1^39 = 5.88; p < .02) than low
achievers. However, when the variance due to age and
verbal-I.Q. is eliminated by analysis of covariance, the
significance of this difference decreases (F^^s = 3.91,
p < .06; and Fl,38 “ 3 .75, p < .06; respectively) and
disappears (1 ^ 3 7 = 1.32; p < .26) when age and verbal-
I.Q. are covaried simultaneously.
Looking at the data presented so far under this
hypothesis, it would seem that high and low achievers do
differ in EFT scores in the direction predicted, but that
much of this difference can be attributed to the variance
of verbal-I.Q. and age rather than perceptual style.
However, when the relationships between EFT scores and
age, I.Q., and school achievement scores for boys and
girls (Table 29) are examined, this interpretation must
also be reexamined. The literature (for example, Kagan &
Kogan, 1970) suggests that, contrary to Witkin's data,
verbal-I.Q. is sometimes found related to perceptual style
measures (particularly, the EFT), and thus, should be
controlled in studies of this type. However, since this
study includes both boys and girls and EFT scores are
related to verbal-I.Q. only for girls, controlling the
variance due to this variable in boys does not seem mean
ingful for this sample. In addition, since high and low
achieving boys do not differ in age (see Table 1), though
this variable is significantly related to EFT scores
130
(r = -.38; p = .05), it seems unnecessary and probably un
desirable to eliminate this variance statistically. Thus,
comparing the EFT scores of high and low achieving boys,
tis = .93, p > .05. On the basis of this discussion and
additional data, then, it can be concluded that high
achieving boys do tend to score lower (more Fl) on the EFT
than do low achieving boys, but that this tendency is not
statistically significant. For girls EFT scores are signi
ficantly related (or nearly so) to both age and verbal-I.Q.
Re-analysis of the data for girls separately (Table 33)
indicates once again that high achievers do score signifi
cantly lower on the EFT (F]_ 21 = 4.82, p < .04) than low
achievers. As soon as variance due to age or verbal-I.Q.,
or their combined variance is eliminated with analysis of
covariance, however, the F ratios become non-significant.
Thus, for girls as well as boys, high achievers do tend to
score lower (more Fl) on the EFT than do low achievers,
but this tendency is not statistically significant.
There appear to be no significant differences
between the RFT scores of high and low achievers when boys
and girls scores are looked at collectively (Tables 31 &
32). However, since boys RFT scores are not related to
age and verbal-I.Q. and girls scores tend to be related
to both variables, analyzing the two sets of scores
separately again seems more meaningful. No difference is
found in the RFT scores of high and low achieving boys
131
when these are compared (t = .25, p > .05). For girls,
though high achievers do tend to score lower than the low
achievers (Table 31), neither the analysis of variance nor
the analyses of covariance indicate that these differences
are significant at the 5% level of confidence (see Table
33). Thus, high and low achievers do not differ
significantly in RFT scores and on this basis do not
differ in perceptual style.
Though analysis of both sets of perceptual style
data show that high and low achievers do not differ
significantly when related sources of variance are con
trolled, there is a relatively strong trend in the EFT
data, which is not evident in the RFT data, confirming the
tendency for high achievers to be more FI than low
achievers. To the extent that these two sets of data dis
agree, how can the discrepant results be interpreted?
Does one measure of perceptual style appear to be more
valid and/or reliable in this study? Starting with the
second question, on the basis of the literature and
particularly the extensive, long term work of Witkin and
his colleagues, the RFT is generally considered the single
most valid and reliable measure of perceptual style.
However, in our data the RFT scores present several
unanswered questions. For some Ss the mean deviation of
the rod on trials in which the frame was tilted to the
left was quite different from that obtained when the frame
132
was tilted to the right. Since no obvious cause for this
type of left-right inconsistency in scores was apparent
and the two studies (Lester, 1968; Silverman, Adevai &
McGough, 1966) in the literature that mentioned this
inconsistency also could not explain it, a second RFT
score (RFT-C) was developed to serve as an index of the
consistency of these two series of scores in a particular
S. The RFT-C score was calculated by subtracting the
total deviation observed when the frame was tilted to the
right (four trials) from that observed when the frame was
tilted to the left (four trials) and then dividing this
number by the total deviation for all eight trials (left +
right). The RFT-C score was then included in the various
data analyses so that its relationship to other measures
could be investigated.
In general the RFT-C score was not related to the
test scores obtained (including the RFT where the
correlation was -.09 for boys and .11 for girls) or other
measures (age, visual acuity, or hand preference) avail
able on the children, with the exception of the positive
relation with age (r = .36; p = .05) and with verbal-I.Q.
(r; = .27; p = .11) for girls (see Table 29). (Though
there is no obvious reason for the older girls to be less
consistent on the RFT than the younger girls, the
relationship with verbal-I.Q. is probably coincidental,
since the older girls also tended to be somewhat brighter
133
in this sample.) In addition, RFT-C scores did not differ
significantly for high and low achievers or for boys and
girls (see Tables 31, 32, and 34). Thus, though the
amount of variation found in individual RFT scores is
puzzling and introduces questions about the reliability
and validity of this measure for some Ss, it does not seem
to have affected one group of Ss more than another in this
sample. Therefore, though suggesting that the source of
this variance should be determined before the RFT can be
considered a reliable measure for all Ss, there is no
evidence to suggest that this variance has biased the
results reported.
Turning to the EFT data, there is no reason to
suspect the reliability of the boys or girls EFT scores.
In view of the comments in the literature, the question of
validity, particularly for the girls, must be scrutinized
more carefully. Thornton & Barrett (1967) have suggested
that the EFT is not a valid measure of perceptual style
for women because it is not highly correlated with the RFT
(r = .26; reported by Witkin et. al.. 1954), which is
generally considered the single most valid and reliable
measure. Applying the same logic to the relationship for
girls in our data, this contention seems overstated.
Though the relationship (r = .45) between the two measures
accounts for only twenty percent of their total variance,
they are clearly related and thus, at least to that
134
extent, the EFT score must be considered a valid index of
perceptual style for the girls in this study. In addition
the consistent and negative relationships found between
girls' EFT scores and their age and I.Q. scores is
consistent with the literature and tends to confirm the
validity of the EFT here.
The lack of any relationship between the EFT and
RFT and between the RFT and age for boys, for that matter,
is unexpected and contrary to the literature. Assuming
again that the RFT is the single best measure of percep
tual style and using Thornton and Barrett's line of
reasoning, the EFT scores for boys in this sample could be
considered invalid. However, it seems more likely that
the zero-order correlation between boys EFT and RFT scores
is a reflection of the homogeneity of the boys in the
sample and the resulting restriction in range of their
scores rather than a question of reliability or validity
of the scores themselves. (On the RFT, boys scores range
from 1.4 to 12.6 with M = 4.72; girls range from 1.4 to
25.1 with M = 6.49. On the EFT, the range for boys is
from 262 to 1033 with M = 610.15; for girls the range is
180 to 1424, M = 652.35.)
To-summarize the discussion so far and attempt to
answer the second question on the basis of the data and
the assumptions made, both measures of perceptual style
seem to be equally valid and reliable measures of
135
perceptual style for the children in the study. However,
since the EFT provides cues relevant to achievement
motivation and behavior not present in the RFT situation
and since the EFT scores are clearly related to school
achievement test scores for both boys and girls while RFT
scores are not, the relationship between EFT and
achievement test performance can as easily (and probably
more reasonably) be attributed to the similarities in the
testing situation as to the hypothesized relationship
between perceptual style and school achievement. Using
this explanation for the EFT data, the interpretation
based upon the RFT data can be accepted. Thus, in this
group of boys and girls it seems more likely that high and
low achievers do not differ significantly in perceptual
style.
However, as pointed out in the discussion of
Hypothesis la., the children used in this study were not
as discrepant in school achievement test scores as would
normally be expected of children labelled high and low
achievers. In addition, the relative homogeneity of the
sample (particularly the boys group) on other variables
(socioeconomic status, interests and activities) may have
made significant support for the hypothesis more difficult
to obtain. One indication of this possibility is the much
greater differences between the RFT and EFT scores of the
less homogeneous high and low achieving girls (Table 31)
136
compared with these differences in scores for the more
homogeneous boys groups (however, part of this difference
for girls is due to differences in age and I.Q. and
probably should be discounted). Another indication is
that though most studies involving perceptual style
measurements show clear differences between scores of boys
and girls, no significant sex differences were found here,
though boys tended to get lower scores on both measures
(Table 34). Thus, it seems probable that this study was
not as definitive a test of the relationship between
achievement and perceptual style as might have been
obtained with a larger, more diverse group of Ss.
Looking at the measures of perceptual style them
selves, both the results presented here and the literature
indicate that much of the time these two tests appear to
have little common variance. Thus, if the RFT is accepted
as the best single measure of perceptual style (and our
results support this choice) and if the FI-FD dimension is
considered a unitary rather than a multifaceted one, then
most of the variance on the EFT is apparently due to some
thing other than perceptual style. [The work of Crandall
& Sinkeldam (1964); Thornton & Barrett (1967); and
Wertheim & Mednick (1958), as well as this study indicate
that it does measure achievement motivation and behavior.]
Therefore, though the EFT may be a useful auxiliary task,
it appears to be a poor substitute for the RFT in
137
measuring perceptual style for either men or women and/or
boys or girls. In addition, unless both the EFT and RFT
are used and the scores from the EFT are found to be
highly correlated with those from the RFT in that sample,
no accurate conclusions about the relationship between
perceptual style and achievement can be drawn.
However, this study and others have clearly
indicated a need to improve the precision of the RFT by
determining what factors effect the consistency of the
scores obtained. Culver (1967) has suggested a number of
different visual problems (refractive disparities between
the two eyes, various types of oculomotor imbalance) which
might make a S's judgment of verticality erroneous.
Certainly if these extraneous factors have an effect, they
should be eliminated or at the very least, the direction
and magnitude of their effect determined. Lester (1968)
has suggested that a much more extensive series of rod and
frame settings is necessary to insure accurate and
unbiased scores. In addition to the usual controls
employed in administering the RFT (controlling head
position, distance from the stimulus display, level of
illumination, number and order of trials, instructions)
according to the standard procedure (as was done here), he
recommends:
1 . tilting the frame at various angles;
2 . using trials in which the frame is omitted;
138
3. using trials in which the frame is
symetrically-oriented (upright); and,
4. varying rod starting position considerably so
that it is not close to true vertical or
parallel with the tilted frame.
It seems unlikely that most investigators will find this
extensive experimental procedure practical or even that
all of these steps are necessary. However, if it were
possible to modify the RFT apparatus so that the S were
exposed only to the rod on some trials, a control reading
of each S's perception of vertical could be obtained with
little added effort. If most Ss show a consistent small
error in one direction when setting the rod alone to
vertical as Lester mentions, this control reading might be
useful in correcting for this deviation.
The following points summarize the data and
discussion presented under this hypothesis:
1. The EFT scores obtained on both boys and girls
support the hypothesized relationship between
achievement and perceptual style— namely that
high achievers are more FI than low achievers.
2. The RFT scores obtained on both boys and
girls do not support the hypothesis or the EFT
data and indicate that high and low achievers
do not differ in perceptual style.
3. The relationship between RFT and school
achievement scores appears to be the more
accurate assessment in this sample because:
a. both the literature and our data tend to
indicate that the relationship between EFT
and school achievement test scores can be
accounted for on the basis of the achieve
ment cues present in these tasks that are
139
not present in the RFT;
b. the EFT and RFT share a relatively small
amount of common variance in this sample
and in a number of those reported in the
literature; and,
c. the RFT has been defined as the single
most reliable measure of perceptual style.
4. It appears probably that the expected
relationship between RFT and school achieve
ment test scores would be found if a larger,
less homogeneous sample were used.
5. The EFT cannot reasonably be substituted for
the RFT as a measure of perceptual style
unless the two measures are known to be highly
related in that sample.
6 . Both measures of perceptual style were
influenced by extraneous factors such as age,
verbal-I.Q., and sex to some extent.
7. The effect of additional factors such as
various types of visual problems on RFT scores
should be studied. Trials in which the frame
is absent should be added to the standard
administration procedure to ascertain each
S's perception of true vertical and eliminate
variance due to this factor.
In conclusion it seems that the relation, between
perceptual style and achievement must be tested with the
RFT rather than the EFT and that a more precise form of
the RFT used in a larger, more diverse population will
indicate that high achievers are more FI. It seems
unlikely that this relation will be especially strong,
however. Even if the development of perceptual style and
achievement is related to exactly the same factors
(parental attitudes toward independent functioning and
140
competence and parental involvement in their child's
development), it appears doubtful that the relative
balance of these factors found associated with high
achievement is identical with that found with FI.
Relation between Measures of
Children1s Achievement
BST and School Achievement Tests
Although the BST was primarily used as a
behavioral measure of parental expectancies in an achieve
ment situation, it also provided information on the
child's evaluation of his own competence, his level of
performance, and an index of his independence. Four
different scores were obtained by observing the children's
behavior during this task:
1 . the number of requests for or rejections of
^ parents' aid made by the child (both
interpreted as indexes of dependence or
2 . independence in this situation);
3. the child's estimate of his performance on
Trial #3 (his evaluation of his competence);
and,
4. the total number of blocks stacked during the
three trials (his performance score).
The mean BST scores for high and low achievers are
presented in Table 35; and for boys and girls, in Table
36. The outcome of the analysis of this data is shown in
Table 37. Although a number of trends can be noted in the
BST scores in Tables 35 and 36 such as:
141
1 . high achieving girls give higher estimates
than low achieving girls;
2 . high achieving boys and girls reject their
parents help more often than low achieving
boys and girls; and,
3. boys in general tend to ask for less parental
aid and reject aid that is given more
frequently than girls do
none of these trends is statistically significant at the
5% level of confidence even when the variance due to the
child's age or total I.Q. score or the number of comments
made by the mother, father, or both parents together is
statistically controlled. However, the analysis of the
data does indicate a significant interaction between sex
and achievement in children's BST performance (^1^39 =
7.28, p < .01), the significance of which is reduced only
slightly by covarying other variables (see Table 37).
Thus, in this sample high achieving boys stack fewer
blocks than low achieving boys, (this difference is not
statistically significant, however, t^g = -1*18, p > .05),
while high achieving girls stack more than low achieving
girls (t2i = 2.51, p < .05).
In their study Rosen and D'Andrade found BST
performance related to n-Ach in that boys classified as
high achievers on the TAT stacked significantly more
blocks than those classified as low achievers. In
addition, they found BST performance relatively unrelated
to I.Q. scores. The results presented here for girls are
in agreement with those of Rosen and D'Andrade, while the
results for boys are not. Since the boys sample is
considerably more homogeneous even in school achievement
than the girls sample and since the high and low achieving
girls do differ significantly in BST performance in the
direction predicted, it seems likely that the lack of
significant results for boys can once again be attributed
to the restricted range of achievement scores rather than
to other differences (i.e., differences in achievement as
defined by school achievement tests versus the TAT). In
addition the relatively small number of requests for, or
rejections of, aid noted per child made it difficult to
obtain meaningful comparisons of these scores. Thus, in
spite of differences in the selection criteria and sex of
the Ss used, our results for girls (and to some extent for
boys) tend to support those of Rosen and D'Andrade on the
relationship between BST scores and achievement.
RTT and School Achievement Tests
The RTT was used to measure various theoretical
aspects of achievement motivation and explore their
relationship to school achievement in these children.
Five different scores were obtained from the task:
1 . the distance the child chose to stand from the
post (index of risk);
2 . the child's estimate of the number of rings he
expected to get on the post on his first trial
(index of level of aspiration);
143
3. the total number of rings actually thrown onto
the post (performance score);
4. the extent to which the child expected his
future performance to differ from his past
performance (mixture of level of aspiration
and evaluation of competence); and,
5. the difference between a child's estimated and
actual performance (index of how realistic
estimates were).
The mean RTT scores for high and low achievers are
presented in Table 38; for boys and girls, in Table 39;
and their analysis in Table 40. Looking at Tables 38 and
39 the following trends are evident in the RTT scores of
high and low achievers and boys and girls:
Table 38
1 .
2.
3.
4.
5.
Table 39
6.
7.
high achieving boys stand further from the
post than low achieving boys;
high achievers (especially boys) give lower
estimates for the first trial;
high achievers (especially boys) perform
better than low achievers;
high achievers give estimates of their future
performance that are closer to their past
performance than do low achievers;
high achieving girls' estimates of their
performance are closer to their actual
performances than are low achieving girls;
boys stand further from the post than girls
do;
boys perform better than girls do;
8 . boys give estimates of their future
performance that are closer to their past
performance than do girls; and
9. boys estimates of their performance are closer
to their actual performances than are girls.
Only trend #3 (with total-I.Q. covaried) is significant at
the 5% level of confidence (see Table 40). Although
trends #2 and 7 initially appeared significant at this
level, the F ratio became non-significant when total-I.Q.
scores were used as covariants. Thus, though a number of
differences were noted in the RTT scores of high and low
achievers and boys and girls, only the differences in
performance (high achievers, particularly boys, perform
better) were statistically significant and related to
school achievement.
Two factors may have been operating in this study
to minimize the differences between high and low achievers
in the theoretical scores used by Frank (1935) and Garcia
(1971; score #4, future - past performance and score #5,
estimated - actual performance). They are:
1 . the fact that the high and low achievers in
this sample are not as different in school
achievement as might normally be expected;
and,
2. the fact that theoretical measures of n-Ach
are often not as highly related to achievement
behavior as one might wish.
In addition, the analyses in Table 40 indicate that both
these scores (as well as the other RTT scores) may be
145
influenced by extraneous factors such as the age and
general intelligence of the S. Until the meaning of these
various theoretical scores is more clearly specified by
experimental work, it seems desirable to use only the
performance score in defining achievement.
BST and RTT
In spite of the fact that neither the child's
estimate of his performance for BST Trial #3 nor any of
his estimates of RTT performance (RTT scores #2, 4, and 5)
were related to school achievement in our population,
since all of these scores dealt with level of aspiration
and/or perceived competence in an achievement situation,
it seemed reasonable to look for relationships between
these two tasks. The correlations between these measures
are given in Table 41. Only the correlation between the
estimates for BST Trial #3 and RTT Trial #1 for girls is
significant (r = -.38, p = .04) and it is opposite in
direction to what would be expected. The extent of the
relationship between performance on these two measures was
also determined and found non-significant and consistently
negative for both boys and girls. Thus, the available
data indicate either no relationship between children's
behavior and performance in these two situations or that
the children act one way in the BST and then the opposite
way in the RTT.
146
The differences found in the childrens behavior
during these two tasks may be related in part to the
differences in circumstances. In the BST the child is
asked to perform a difficult task, but also has his
parents present to assist him (assuming they want to do
this) whenever and as much as he wishes. No such help is
available during the RTT. In addition the accuracy of the
RTT scores may have been influenced to a greater extent
than the BST scores by the child’s age and intelligence.
The RTT performance score (and, therefore, the other
scores calculated from it) was also somewhat less accurate
because many of the rings thrown onto the post did not
remain on it until the end of that game and, thus, were
not counted.
In summary:
1. High and low achievers and boys and girls do
not differ significantly in BST estimates of
their performance on Trial #3 or in the number
of requests for or rejections of their parents
aid during this task.
2. High achieving boys stack fewer blocks during
the BST than low achieving boys, while high
achieving girls stack more than low achieving
girls.
3. When the variance due to total-I.Q. score is
kept constant, high achievers (especially
boys) get more rings on the post during the
RTT than do low achievers.
4. There are no other significant differences
between high and low achievers or boys and
girls on the RTT.
147
5. BST scores appear to be influenced less than
RTT scores by the age and general intelligence
of the child.
6 . With the exception of the negative relation
ship found between girls estimates on BST
Trial #3 and RTT Trial #1, boys and girls BST
and RTT scores were not significantly related.
Thus, neither BST or RTT scores correlated with school
achievement test scores or with each other to the extent
that might have been expected--probably due to the
relatively narrow range of school achievement represented
by our sample. Our data, in conjunction with that
reported by Rosen and D'Andrade suggest that the BST is a
useful measure of performance in an achievement situation
and may be useful in measuring a S's evaluation of his
competence as well as his tendency toward independent or
dependent functioning. On the other hand, though a
number of interesting trends were found in the RTT data,
neither our results nor those reported by Garcia show any
significant relationships between the theoretical RTT
scores and n-Ach. Until this task has been more
thoroughly studied, it would seem more practical to use
some other measure of achievement.
OVERVIEW
Turning from the morass of results and their
interpretations to the questions posed in the Introduction,
what does this study say about parents of high and low
achievers and how do these findings fit with those
reported in the literature? Though few of the results are
statistically significant, the general trend of the data
tends to indicate that parents of high achievers do set
higher standards for their children and stress independent
functioning more than do the parents of low achievers.
Both these results are in accord with previous research
and theory about the origins of n-Ach (McClelland, Rosen &
Andrade, Whiting, & Winterbottom), but conflict with the
results reported by Chance (1961), who found that mothers
of high achievers (in school) favored later independence
training.
Chance has suggested that the discrepancy between
her findings and those of Winterbottom are due essentially
to the differences between theory and practice--the
differences between motivation to achieve and actual
achievement as well as to measure problems in their assess
ment. However, a closer look at the details of her
procedure strongly suggest a very different explanation.
148
Rosen and D 1Andrade, Whiting, Winterbottom, and the
current work all selected high and low achievers by
choosing Ss who scored either high or low relative to
others in the S pool on the achievement measure (whether
that happened to be school achievement tests or the TAT).
Chance, on the other hand, classified Ss as high or low
achievers on the basis of their progress in school reading
and arithmetic relative to their (own) intellectual
ability. As noted in the section of Method, during the S
selection phase of this study both procedures were applied
to the S pool in an effort to obtain the clearest possible
separation of achievement groups. Interes'tingiy enough,
Ss who were classified as "high achievers" by the first
procedure were often classified as "low achievers" by
Chance's procedure and vice versa. Thus, if Chances Ss
were reclassified by their achievement relative to the
group, it seems likely that her "high achievers" might
become "low achievers" and her "low achievers" become
"high achievers." In this case, her results would also be
reversed in direction and would then be in accord with
those reported by others.
Two other significant differences emerge from this
data when parents of high and low achievers are compared.
Parents of high achievers are much more involved in help
ing their child achieve--whether this is measured in terms
of their questionnaire responses or of their behavior in
150
an achievement task. Thus, as other investigators have
suggested, early independence training "per se" does not
necessarily insure high achievement. If the parents goal
in emphasizing competent, independent functioning at an
early age, is to free the parent of responsibility or
reduce his level of involvement with his child or if early
independence is fostered due to lack of parental
involvement, it is more likely to be detrimental to n-Ach
than to encourage it.
These results also tend to support Rosen and
D'Andrade's contention that achievement training is much
more important in producing high n-Ach than is indepen
dence training. Certainly during the BST the behavioral
measures indicate that parents of high achievers were much
more anxious to see that their child "did well" than that
he worked "by himself." In addition, though the AIS was
intended to measure the age of independent functioning, it
can as easily be interpreted as a measure of the age of
competent functioning.
Turning to the groups of children involved in the
study, high achievers differ significantly from low
achievers on a number of relevant variables--high
achievers score higher on I.Q. tests, are more likely to
have mothers born inside the U.S.A., have parents with
more years of formal education, and come from families of
higher socioeconomic status. In school they perform
151
significantly better on achievement tests and in the study
itself they score significantly better in finding the
simple figures on the EFT, stack more blocks during the
BST (girls only), and throw more rings onto the post
(especially boys) in the RTT. Yet with all of these
advantages, superior abilities, and higher performances
high achievers are no better at separating the rod from
the frame in the RFT than low achievers. In view of the
theory and research which suggests that field independence
originates in an environment which encourages competent,
independent functioning, it is surprizing to find that the
high achieving children do not differ significantly in
perceptual style from their lower achieving counterparts.
To some extent this outcome may be an artifact of the
particular Ss used, since even the 1 1 low achievers" in the
sample were, on the average, achieving at their grade
level. Had the high and low achievers selected been more
extreme on this dimension, their perceptual style scores
might also have been less similar.
In addition to this possible explanation of these
results it seems likely that performing well on achieve
ment tests in school may well present a very different
proposition from that required by successful performance
on the RFT. In contrast with the achievement tests, the
RFT provides a situation full of conflicting cues to be
resolved in choosing a solution and then provides no clear
152
indication to the S of how successfully he has
accomplished this. Thus, other factors, such as self-
assurance, not necessarily implied by high achievement,
may be involved here. In support of this line of
reasoning, a study by Vaught (1965) has shown that scores
on the RFT tend to become increasingly field independent
as ego strength increases. Several of the results noted
in this study can also be interpreted in this way.
For example, although there are few significant
differences between parents of field independent and field
dependent children, these differences and the non
significant trends evident in the data generally indicate
that parents of field independent boys and girls have
higher aspirations for their children and evaluate them
more positively. Both of these findings, though in accord
with a general hypothesis of the study--that parents of
field dependent children expect more competence from their
children than parents of field dependents--can also be
interpreted as promoting self-esteem. Two other findings—
that parents of field independents are more involved in
helping their children (in the BST) and that the parents
of field independent boys (but not girls) expect
independent functioning at a later age (on the AIS)--which
are in clear opposition to the literature emphasizing the
importance of early independent functioning in the
development of field independence, can also be reassessed
153
in terms of the effect the parents actions could have on
ego strength. In the BST parents comments, though
generally limiting the child's opportunity to function "on
his own," did greatly improve his chances of "doing well"
and may also have improved his self-esteem. On the AIS,
boys, due to differences in maturation, may be less able
to handle many of these activities successfully than girls
of the same age. If this is the case, parents who are
sensitive to their son's needs and abilities may set
somewhat lower (and/or more appropriate) standards, which
in turn may encourage a more positive self-image.
Another example of support for this view is
evident when the parents of boys and girls are compared in
their expectancies for competence and independent
functioning in their children. Once again there are few
significant differences between these groups of parents in
the data. However, parents of girls tend to expect
somewhat more of their daughters and evaluate their
accomplishments somewhat less positively than do the
parents of boys. Whether or not this observation is
typical of the parents behavior and the extent to which
this type of behavior has affected their daughters is not
known, since no attempt was made to assess this here.
However, the girls did show a non-significant tendency to
be more field dependent than boys even when the girls were
somewhat brighter. Thus, though the evidence presented is
154
not conclusive, it suggests that differences in ego
strength alone may account for some of the differences
between the sexes noted in perceptual style.
A second factor, also suggested by the differences
found between parents of boys and girls, may also be
relevant to the sex differences in perceptual style as
well as to the more general question of the origin of the
differences between field independence and field
dependence. Parents of girls are significantly more in
agreement in their expectations for their child's level of
competent, independent functioning than are the parents of
boys, and thus may be more effective in influencing their
child's development. It seems likely that this greater
agreement between parents of girls may allow the girls
themselves considerably less freedom of action and less
autonomy than is allowed to boys and that this reduction
in autonomy could be related to girls' poorer RFT
performance. Further evidence for this speculation is
seen within the girls sample itself. Parents of field
independent and field dependent girls tend not to differ
in their AIS Subscale responses, except for Scale #6— Wide
Experience (which includes: risk-taking activities,
venturing into the community, going away from home, and
having contacts with strangers) on which parents of field
independent girls put down lower ages. Since the items on
this subscale seem to emphasize freedom of action more
155
than those on other subscales, this finding suggests that
autonomy may be especially important in the development of
field independence in girls.
Assuming that autonomy is an important factor in
perceptual style, its relationship to field independence
in boys is not clear from this data. For example, the
results indicate that the parents of boys disagree more
than the parents of girls in their expectations for their
child and thus, indirectly may grant boys more autonomy.
Yet, on Subscale #6 of the AIS, presumably a more direct
measure of autonomy, parents of field independent boys
respond with older ages than parents of field dependent
boys. Thus, field independent and field dependent
children and boys and girls do seem to differ in at least
this kind of independence training, but in much more
subtle and complex ways than had been expected.
In conclusion the results of this study, as well
as others in the literature, clearly show that parental
emphasis on early competence and independent functioning,
when used in the interest of the child, is related to the
development of achievement in girls as well as in boys.
Further pursuit of this type of research does not seem
useful. The relation of parental expectancies to their
child's perceptual style and sex, however, appear to be
more complex. Though gross differences in parental
expectancies for competent, independent functioning would
undoubtedly have been reflected in their children's
perceptual style, these differences were not found between
the parents of field independent and field dependent
children nor between the parents of boys and girls in this
sample. Instead the significant differences found in
these comparisons suggest that more subtle factors,
related to the child's self-esteem and to his autonomy,
may be involved. Since the differences found between
parents of field independents and field dependents and
between parents of boys and girls were similar, further
study of the differences in childrearing practices for
boys and for girls may be helpful in understanding the
origin of perceptual style.
BIBLIOGRAPHY
REFERENCES
Aberle, D. G., and Naegele, K. D. Middle-class fathers’
occupational role and attitudes toward children.
American Journal of Orthopsychiatry. 1952, 22,
366-378.
Adevai, G., Silverman, A. J., and McGough, W. E.
Perceptual correlates of the rod-and-frame test.
Perceptual and Motor Skills, 1968, 26, 1055-1064.
Anastasi, A. Differential psychology; individual and
group differences in behavior. (3rd ed.) New
York: Macmillan, 1958.
Astrup, A. M. Repeated short-term sensory reduction in
mining. Perceptual and Motor Skills, 1968, 27,
863-869.
Barry, H., Bacon, M. K., and Child, I. L. A cross-
cultural survey of some sex differences in
socialization. In D. Rogers (Ed.), Issues in
child psychology. Belmont, California: Brooks/
Cole Publishing Company, 1969. Pp. 275-281.
Baumrind, D. Socialization and instrumental competence in
young children. Young Children, 1970, .26, 104-
119.
Baumrind, D., and Black, A. E. Socialization practices
associated with dimensions of competence in pre
school boys and girls. Child Development, 1967,
38, 291-327.
Berry, J. W. Temne and Eskimo perceptual skills.
International Journal of Psychology, 1966, _1, 207-
229.
Biller, H. B., and Weiss, S. D. The father-daughter
relationship and the personality development of
the female. Journal of Genetic Psychology. 1970,
116, 79-93.
158
159
Bloomberg, M. An inquiry into the relationship between
field independence-dependence and creativity.
Journal of Psychology. 1967, 67, 127-140.
Broverman, D. M., Klaiber, E. L., Kobayashi, Y., and
Vogel, W. Roles of activation and inhibition in
sex differences in cognitive abilities.
Psychological Review, 1968, 75, 23-50.
Busch, J. C., and DeRidder, L. M. Note on control for
intelligence in studies of field dependence with
young children. Perceptual and Motor Skills,
1971, 32, 337-338.
Chance, J. E. Independence training and first graders1
achievement. Journal of Consulting Psychology.
1961, 25, 149-154.
Cohen, W., and Tepas, D. Temporal factors in the
perception of verticality. American Journal of
Psychology, 1958, 7jL, 760-763.
Conklin, R. C., Muir, W., and Boersma, F. J. Field
dependency-independency and eye-movement patterns.
Perceptual and Motor Skills, 1968, 26, 59-65.
Courter, R. J., Wattenmaker, R. A., and Ax, A. F.
Physiological concomitants of psychological
differentiation. Psychophysiology, 165, J., 282-
290.
Crandall, V. J., and Sinkeldam, C. Children's dependent
and achievement behaviors in social situations and
their perceptual field dependence. Journal of
Personality, 1964, 32, 1-22.
Culver, C. M. Anisometropia and the rod-and-frame task.
Perceptual and Motor Skills. 1967, 25, 377-384.
Culver, C. M., Cohen, S. I., Silverman, A. S., and
Shmavonian, B. M. Cognitive structuring, field
dependence-independence, and the psycho-
physiological response to perceptual isolation.
In J. Wortis (Ed.), Recent advances in biological
psychiatry. New York: Plenum, 1964. Pp. 119-128.
160
D'Andrade, R. G. Sex differences and cultural
institutions. In E. E. Maccoby (Ed.), The
development of sex differences. Stanford:
Stanford University Press, 1966. Pp. 174-204.
Dameron, L. E. Mother-child interaction in the
development of self-restraint. Journal of Genetic
Psychology, 1955, 8 6, 289-308.
De Fazio, V. J., and Moroney, W. F. Performance
characteristics of field dependent and independent
individuals on an auditory signal detection task.
Journal of Psychology, 1969, 71, 77-82.
Dyk, R. B., and Witkin, H. A. Family experiences related
to the development of differentiation in children.
Child Development, 1965, 36, 21-55.
Elkind, S. N. The relationship of orientation toward
achievement to convergent and divergent problem
solving. Dissertation Abstracts International,
1969, 30(3-b), 1356-1357.
Elliott, R., and McMichael, R. E. Effects of specific
training on frame dependence. Perceptual and
Motor Skills. 1963, 17, 363-367.
Frank, J. D. Individual differences in certain aspects of
the level of aspiration. American Journal of
Psychology. 1935, 47, 119-128.
Garcia, S. J. The relationship of hypnotic susceptibility,
EEG alpha, achievement need, and parent's child
rearing practices. Unpublished doctoral
dissertation, University of Southern California,
1971.
Gill, N. T., Herdtner, T. J., and Lough, L. Perceptual
and socio-economic variables, instruction in body-
orientation, and predicted academic success in
young children. Perceptual and Motor Skills.
1968, 26, 1175-1184.
Goldstein, A. G., and Chance, J. E. Effects of practice
on sex-related differences in performance on
embedded figures. Psychonomic Science, 1965, 3,
361-362.
161
Goldstein, H. S., Pardes, H., Small, A. M., and Steinberg,
M. D. Psychological differentiation and
specificity of response. Journal of Nervous and
Mental Disease. 1970, 151, 97-103.
Gray, A. W. Independence concepts of parents concerning
elementary-school children. Unpublished doctoral
dissertation, Claremont Graduate School, 1968.
Greenwald, E. R. Perceptual style in relation to role
choices and motivational variables. Dissertations
Abstracts, 1968, 29(6-B), 2192.
Gross, F. The role of set in perception of the upright.
Journal of Personality. 1959, 27, 95-103.
Gross, S. J., and Moore, S. F. The portable rod-and-frame
test and the short-form embedded-figures test: A
combined index of the field-dependence dimension.
Psychonomic Science. 1970, 20, 213-214.
Gruenfeld, L., and Arbuthnot, J. Field independence,
achievement values and the evaluation of a
competency related dimension on the least
preferred co-worker measure. Perceptual and Motor
Skills. 1968, 27, 991-1002.
Haronian, F., and Sugerman, A. A. Fixed and mobile field
independence: Review of studies relevant to
Werner's dimension. Psychological Reports. 1967,
21, 41-57.
Hein, P. L., Cohen, S. I., and Shmavonian, B. M.
Perceptual mode and cardiac conditioning.
Psychophysiology. 1966, 3, 101-107.
Heron, W. Cognitive and physiological effects of
perceptual isolation. In P. Solomon, P.
Kubzansky, P. Leiderman, J. Mendelson, R. Trumbell,
and D. Wexler (Eds.), Sensory deprivation.
Cambridge: Harvard University Press, 1961. Pp.
7-33.
Hill, E., and Feigenbaum, K. D. Altering field dependence
through stress. Perceptual and Motor Skills.
1966, 23, 1200.
162
Hollingshead, A. B. Elmtowr^s youth; The impact of
social classes on adolescents. New York: Wiley,
1949. In C. N. Cofer and M. H. Appley,
Motivation: Theory and Research. New York:
Wiley, 1964. P. 570.
Honigfeld, G., and Spigel, I. M. Achievement motivation
and field independence. Journal of Consulting
Psychology. 1960, 24, 550-551.
Huckabee, M. W., and McGown, W. P. Differences between
field independent and field dependent persons on
the Stroop color-word test: A failure to
replicate. Journal of Clinical Psychology. 1971,
27, 226.
Hustmyer, F. E., and Karnes, E. Background autonomic
activity and "analytic perception.” Journal of
Abnormal and Social Psychology, 1964, 6 8, 467-468.
Immergluck, L. Visual figural after-effects and field
dependence. Psychonomic Science, 1966a, 4, 219-
220.
Immergluck, L. Figural after-effects, rate of "figure-
ground” reversal, and field dependence.
Psychonomic Science. 1966b, 6 , 45-46.
Immergluck, L. Figural after-effect potency: A function
of sex or field-dependence? Psychonomic Science.
1970, 18, 317-318.
Jacobson, G. R. Effect of brief sensory deprivation on
field dependence. Journal of Abnormal Psychology.
1966, 71, 115-118.
Jacobson, G. R. Reduction of field dependence in chronic
alcoholic patients. Journal of Abnormal
Psychology. 1968, 73, 547-549.
Kagan, J., and Kogan, N. Individual variation in
cognitive processess. In P. H. Mussen (Ed.),
Carmichael^ manual of child psychology. Vol. 1.
(3rd.) New York: Wiley, 1970. Pp. 1326-1328.
Karp, S. A. Field dependence and occupational activity in
the aged. Perceptual and Motor Skills, 1967, 24,
603-609.
163
Karp, S. A., Silberman, L., and Winters, S. Psychological
differentiation and socio-economic status.
Perceptual and Motor Skills, 1969, 28, 55-60.
Karp, S. A., Witkin, H. A., and Goodenough, D. R.
Alcoholism and psychological differentiations
Effect of alcohol on field dependence. Journal of
Abnormal and Social Psychology. 1965, 70, 262-265.
Katkovsky, W., Preston, A., and Crandall, V. J. Parents'
achievement attitudes and their behavior with
their children in achievement situations. Journal
of Genetic Psychology. 1964, 104, 105-121.
Kohlberg, L. A cognitive-developmental analysis of
children's sex-role concepts and attitudes. In
E. E. Maccoby (Ed.), The development of sex
differences. Stanford: Stanford University
Press, 1966. Pp. 82-173.
Kristofferson, M. W. Effect of alcohol on perceptual
field dependence. Journal of Abnormal Psychology,
1968, 73, 387-391.
Kurie, G. D., and Mordkoff, A. M. Effects of brief
sensory deprivation and somatic concentrations on
two measures of field dependence. Perceptual and
Motor Skills. 1970, 31, 683-687.
Lester, G. The rod-and-frame test: Some comments on
methodology. Perceptual and Motor Skills. 1968,
26, 1307-1314.
Lewis, M. Culture and gender roles: There's no unisex
in the nursery. Psychology Today, 1972, 5, 54-57.
Lucco, A. A. Curiosity behavior and independent
functioning in preschool children. Proceedings of
the 75th Annual Convention of the American
Psychological Association. 1967, 2, 173-174.
Lynn, D. B. Curvilinear relation between cognitive
functioning and distance of child from parent of
the same sex. Psychological Review. 1969, 76,
236-240.
164
Maccoby, E. E. Sex differences in intellectual
functioning. In E. E. Maccoby (Ed.), The
development of sex differences. Stanford:
Stanford University Press, 1966. Pp. 25-55.
Maccoby, E. E. Woman's intellect. In D. Rogers (Ed.),
Readings in child psychology. Belmont,
California: Brook/Cole Publishing Company, 1969.
Pp. 76-84.
McClelland, D. C., Atkinson, J. W., Clark, R. A., and
Lowell, E. L. The achievement motive. New York:
Appleton-Century-Crofts, 1953.
McClelland, D. C., Rindlisbacher, A., and DeCharms, R.
Religious and other sources of parental attitudes
toward independence training,1955. In C. N. Cofer
and M. H. Appley, Motivation: Theory and
Research. New York: Wiley, 1964. P. 764.
McFall, R. M., and Schenkein, D, Experimenter expectancy
effects, need for achievement, and field
dependence. Journal of Experimental Research in
Personality. 1970, 4, 122-128.
Milner, E. Effects of sex role and social status on the
early adolescent personality. Genetic Psychology
Monographs. 1949, 40, 231-325.
Minard, J. G., and Mooney, W. Psychological
differentiation and perceptual defense: Studies
of the separation of perception from emotion.
Journal of Abnormal Psychology. 1969, 74, 131-139.
Mussen, P., Harris, S., Rutherford, E., and Keasey, C. B.
Honesty and altruism among preadolescents.
Developmental Psychology, 1970, 3, 169-194.
Myers, J. K., and Bean, L. L. A decade later: a follow-
up of social class and mental illness. New York:
Wiley, 1968. Pp. 235-237.
Neville, C. W., Workman, S. N., and Johnson, D. T.
Expected scores in the rod-and-frame test: Field
dependence is where you find it. Psychonomic
Science. 1969, 15, 321-322.
165
Nie, N. H., Bent, D. H., and Hull, C. H. Statistical
package for the social sciences. New York:
McGraw-Hill, 1970.
Olson, R. S., Diehm, D. F., and Elfner, L. F. Some
factors affecting the perception of verticality.
Psychological Record. 1965, 15, 51-55.
Oltman, P. K. A portable rod-and-frame apparatus.
Perceptual and Motor Skills. 1968, 26, 503-506.
Oltman, P. K., and Gapobianco, F. Field dependence and
eye dominance. Perceptual and Motor Skills. 1967,
25, 645-646.
Polymetric Company. Operating instructions: Vertical rod
and frame apparatus, portable, subject-actuated:
Model V-1260-AR. Hoboken, New Jersey.
Rosen, B. C., and D*Andrade, R. The psychosocial origins
of achievement motivation. Sociometry. 1959, 22,
185-218.
Rudin, S. A. Figure-ground differentiation under differ
ent perceptual sets. Perceptual and Motor Skills.
1968, 27, 71-77.
Sandstrom, C. I. A note on the Aubert Phenomenon.
Journal of Experimental Psychology. 1954, 48, 209-
210.
Scallon, R. J., and Herron, W. G. Field articulation of
enuretic boys and their mothers. Perceptual and
Motor Skills. 1969, 28, 407-413.
Schneider, C. W., and Bartlfey, S. H. A study of the
effects of mechanically induced tension of the
neck muscles on the perception of verticality.
Journal of Psychology. 1962, 54, 245-248.
Schwartz, D. W., and Karp, S. A. Field dependence in a
geriatric population. Perceptual and Motor
Skills. 1967, 24, 495-504.
166
Sears, R. R., Whiting, J. W. M., Nowlis, V., and Sears,
P. S. Some childrearing antecedents of aggression
and dependency in young children. Genetic
Psychology Monographs. 1953, 47, 135-234.
Shaw, M. C. Note on parent attitudes toward independence
training and the academic achievement of their
children. Journal of Educational Psychology.
1964, 55, 371-374.
Sherman, J. A. Problem of sex differences in space
perception and aspects of intellectual
functioning. Psychological Review, 1967, 74, 290-
299.
Silverman, A. J., Adevai, G., and McGough, W. E. Some
relationships between handedness and perception.
Journal of Psychosomatic Research. 1966, 10, 151-
158.
Silverman, A. J., McGough, W. E., and Bogdonoff, M. D.
Perceptual correlates of the physiological
response to insulin. Psychosomatic Medicine,
1967, 29, 252-264.
Silverman, J., and King, C. Pseudo-perceptual
differentiation. Journal of Consulting and
Clinical Psychology. 1970, 34, 119-123.
Sondheimer, R. Field dependency and rehabilitation
therapy in physically disabled patients.
Dissertation Abstracts International, 1970, 30
(8-B), 3878-3879.
Spotts, J. V., and Mackler, B. Relationships of field-
dependent and: field-independent cognitive styles
to creative test performance. Perceptual and
Motor Skills. 1967, 24, 239-268.
Stuart, I. R., and Murgatroyd, D. Field research model of
the rod-and-frame test. Perceptual and Motor
Skills. 1971, 32, 671-674.
Thornton, C. L., and Barrett, G. V. Methodological note
on N-achievement and field independence
comparisons. Journal of Consulting Psychology.
1967, 31, 631-632.
167
Torgoff, I. Some correlates of parental developmental
expectancies. American Psychologist. 1958, 13,
319. (Abstract)
Vaught, G. M. The relationship of role identification and
ego strength to sex differences in the rod-and-
frame test. Journal of Personality, 1965, 33,
271-283.
Vaught, G. M. Expected scores in the rod-and-frame test:
Fuel for the Immergluck-Pressey fire.
Psychonomic Science, 1968, 13, 248.
Vaught, G. M. Expected scores in the rod-and-frame test
revisited. Psychonomic Science. 1970, 18, 111.
Vaught, G. M., and Ellinger, J. Field-dependence and form
discrimination. Psychonomic Science. 1966, 6 ,
357-358.
Vaught, G. M., and Hunter, W. Autokinetic word writing
and field-dependence. Psychonomic Science, 1967,
7, 335-336.
Weinberg, H. J. Changing perceptions on the RFT by
conditioning subjects to relieve dissonance and/or
excape from the anxiety in a new manner.
Dissertation Abstracts International. 1970, 31
(4-A), 1637.
Weiner, M. Effects of training in space orientation on
perception of the upright. Journal of
Experimental Psychology, 1955, 49, 367-373.
Wertheim, J., and Mednick, S. A. The achievement motive
and field independence. Journal of Consulting
Psychology. 1958, 22, 38.
Whiting, A. Independence concepts held by parents of
successful and unsuccessful elementary school
boys. Unpublished doctoral dissertation,
Claremont Graduate School, 1970.
Winterbottom, M. R. The relation of need for achievement
to learning experiences in independence and
mastery. In J. W. Atkinson (Ed.), Motives in
fantasy, action or society. Princeton: Van
Nostrand, 1958. Pp. 453-478.
168
Witkin,
Witkin,
Witkin,
Witkin,
Witkin,
Witkin,
Witkin,
Wolf,
H. A. Individual differences in ease of
perception of embedded figures. Journal of
Personality. 1950, JL9, 1-15.
H. A. Psychological differentiation and forms of
pathology. Journal of Abnormal Psychology, 1965,
70, 317-336.
H. A. A cognitive-style approach to cross-
cultural research. International Journal of
Psychology. 1967, 2, 233-250.
H. A., Dyk, R. B., Faterson, H. p., Goodenough,
D. R., and Karp, S. A. Psychological
differentiation. New York: Wiley, 1962.
H. A., Faterson, H. F., Goodenough, D. R., and
Birnbaum, J. Cognitive patterning in mildly
retarded boys. Child Development. 1966, 37, 301-
316.
H. A., Goodenough, D. R., and Karp, S. A.
Stability of cognitive style from childhood to
young adulthood. Journal of Personality and
Social Psychology. 1967, 7, 291-300.
H. A., Lewis, H. B., Hertzman, M., Machover, K.,
Meissner, P., and Wapner, S. Personality through
perception. New York: Harper, 1954.
Body rotation and the stability of field
dependence. Journal of Psychology, 1965, 59, 211-
217.
APPENDIX A
FIGURES
169
3.0
2.0
1.0
•VI
0 .
u
w
t
N|
01
0)
VI
0
u
i n
• u
0)
01
4 J
c
01
E
01
>
0)
-H
s.
o
<
o
o
•g
o n
•o
a
% - 2.0
3
" 3.0
- 1.0
100
• •
4 •
105* • ' ,110 ' *11S ' | 1201 ' • .1251 ' i ^130* " 135* 140' ' ' 145 ' ' 150
Total-I.Q, Test Scores
• . •
Figure 1. School Achievement Relative to the Group and I.Q. for Children in the Subject
Pool
170
3.0*-
i
H
r t
4 J
o
e*
w
3
C
•H
£
V
u
o
0
t f l
1 —*
N| f t )
— M
O
( U u
C / 1
O f
O Nl
C /1
2.0* -
1.0
. «
*
4 - > a )
( / ) u
o o
e* u
V I
■ p
d **
ft) to
H V
OJH - 1.0
0 )
•H
£
o
<
o
o
*6
c / i
•o
f t )
4 J
(0
3
•n
•O
- 2.0
- 3.0
• •
Total-I.Q. Test Scores
t #
Figure 2. School Achievement Relative to Individual Ability and I.Q. for Children in
the Subject Pool
171
APPENDIX B
TABLES
172
TABLE 1 -- BACKGROUND INFORMATION ON CHILDREN USED IN STUDY
Handicaps Hand Schs.
Prefer- Att1
ence ed
Birth- Birth-
order place
Mean
Age
(mon.)
Mean
I.Q. Scores
Total Verbal
Non-
Verbal
Group N
0)
a
o
s
rH
to
• r H
>
CO
G -G
•HUM
U 0) 0J
m oj £
oj a +>
5C c / ) O
+j
C O
•H
P
qj
X
•H
2
o3>» • >>
rH • Pr—I + ■ >
O G O H - H in
COO to 2 - H £ + ■ > 0) Q J •
G u XHH to H CO <
Q J • 0) G O f c Q J P G > \ •
M C x . t d PPGrHCO
c tS • -H QJM h G H -H O G •
2 J O S O - H O S > i O P
u
0)
r:
4 - >
o
p
0) + - > (0
+* C Q J
CO 0 ) > H
G S O
p qj o
< > C / )
Q J
G -H H - >
f l j X! (0
Q J O Q J
2 < H
Achievers
Boys &
Girls
High
Low
Boys
High
Low
Girls
High
Low
21
22
11
9
10
13
10
7 1
10
1 1
1 1
O 2
19 2 15 6 3.45 4 7 8 2 19 2 133.33 126.00a 125.71b 125.86C 1.92'
19 3 14 8 3.65 4 10 6 2 19 3 129.50 113.73 107.95 119.00 .08
10 1 9 2 3.5 2 5 3 1 10 1 131.82 122.55 122.18 122.64 1.69
7 5 4 3.9 3 4 2 0 8 1 131.67 115.11 106.67 123.00 -.20
9 1 6 4 3.4 2 2 5 1 9 1 135.00 129.80 129.60 129.40 2.18
12 3 9 4 3.4 1 6 4 2 11 2 128.00 112.77 108.85 116.23 .28
17 1 14 6 3.7 5 9 5 1 18 2 131.75 119.20 115.20 122.80 .84'
21 4 15 8 3.4 3 8 9 3 20 3 131.04 120.17 117.87 121.96 1.10
10 1 9 2 3.5 2 5 3 1 10 1 131.82 122.55 122.18 122.64 1.69
9 1 6 4 3.4 2 2 5 1 9 1 135.00 129.80 129.60 129.40 2.18
7 5 4 3.9 3 4 2 0 8 1 131.67 115.11 106.67 123.00 -.20
12 3 9 4 3.4 1 6 4 2 11 2 128.00 112.77 108.85 116.23 .28
Boys 20
Girls 23
Highs
Boys
Girls
Lows
Boys 9
Girls 13
11
10
17
22
10
10
7
12
1 1
1
1
1 1
TABLE 1 — BACKGROUND INFORMATION ON CHILDREN USED IN STUDY— (Cont.)
aFl 39 = 16.08, p <.001
bpl,39 = 26.94, p <.001
CF‘ l,39 = 4.06, p <.05
dpl,39 = 117.81, p <.001
Spl,39 = 7.96, p <.007
174
TABLE 2 — BACKGROUND INFORMATION ON MOTHERS USED IN STUDY
Group N
Mean
Age
(years)
Birth
place
USA Other
Reliqion
+>
c
o
Emploved
Yes
Full- Part-
No
Mean
Years of
Education
0)
0)
+J
0
M
fi
•n
xime Time
C 0
S -P
0 r t
KU
t f l
•rl
•n
u
< D
£
+>
o
Achievers
Bovs &
Girls
High
•
•
21 40.57 19a 2 15 4 1 1 6 3 12 15.02b
Low 22 39.22 14 8 15 7 0 0 6 1 15 13.04
Bovs
High 11 40.72 10 1 8 2 1 0 3 1 7 14.86
Low 9 39.44 6 3 5 4 0 0 1 1 7 13.22
Girls
High 10 40.40 9 1 7 2 0 1 3 2 5 15.20
Low 13 39.08 8 5 10 3 0 0 5 0 8 12.92
Sex
H. & L.
Boys
20 40.15 16 4 13 6 1 0 4 2 14 14.12
Girls 23 39.65 17 6 17 5 0 1 8 2 13 13.91
Hicjhs
Boys 11 40.72 10 1 8 2 1 0 3 1 7 14.86
Girls 10 40.40 9 1 7 2 0 1 3 2 5 15.20
Lows
Boys 9 39.44 6 3 5 4 0 0 1 1 7 13.22
Girls
a v 2
13 39.08 8 5 10 3 0 0 5 0 8 12.92
=4.69, p <.05 bpl,41 = 7-33» P <-01
TABLE 3 — BACKGROUND INFORMATION ON FATHERS USED IN STUDY
Group N
Mean
Age
(years)
Birthplace
USA Other
Protestant
Reliqion
0
•H
H JG
GO 0)
ns .c -h
e h- > &
0 r t S Q J
Other
Mean
Years of
Education
Socio
economic
Status
Achievers:
Bovs & Girls
High 21 44.09 20 1 15 3 1 2 16.90a 27.61b
Low 22 43.13 18 4 14 7 0 1 14.22 34.95
Boys
High 11 44.36 10 1 9 1 1 0 15.82 33.73
Low 9 44.22 7 2 5 4 0 0 14.67 30.11
Girls
High 10 43.80 10 0 6 2 0 2 18.10 20.90
Low 13 42.62 11 2 9 3 0 1 13.92 38.31
Sex
Hiahs & Lows
Boys 20 44.30 17 3 14 5 1 0 15.30 32.10
Girls 23 43.00 21 2 15 5 0 3 15.73 30.73
Hiqhs
Boys 11 44.36 10 1 9 1 1 0 15.82 33.73
Girls 10 43.30 10 0 6 2 0 2 18.10 20.90
Lows
Boys 9 44.22 7 2 5 4 0 0 14.67 30.11
Girls 13 42.62 11 2 9 3 0 1 13.92 38.31
a Pl,41 = 16
• 31, p < .001
bpl,41 “ 4
• 0 2, ] p ~ .05
176
TABLE 4 — AGE INDEPENDENCE SCALE MEANS FOR PARENTS OF HIGH AND LOW ACHIEVERS
No. of AIS Means
Achieve Sets of (Father + Mother)
ment Parents Overall Subscales
Group 1 2 3 4 5 6 7
Bovs &
Girls
Highs 21 17.38 18.76 18.09 18.13 19.21 16.16 19.29 12.99
Lows 22 17.76 17.92 19.09 17.08 19.34 16.80 19.52 14.24
Bovs
High 11 17.68 19.65 18.19 18.14 19.96 16.06 19.44 13.32
Low 9 17.95 18.53 19.14 17.22 19.18 17.06 19.11 15.24
Girls
High 10 17.06 17.78 17.98 18.12 18.39 16.27 19.13 12.62
Low 13 17.63 17.49 19.05 16.98 19.45 16.62 19.80 13.55
177
TABLE 5
ANALYSIS OF VARIANCE OF AGE INDEPENDENCE SCALE OVERALL MEANS
FOR PARENTS OF HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Source
df MS F
P
Less
Than
Total 42
Achievement (A) 1 1.554 .48 .49
Sex (S) 1 2.273 .70 .41
AS 1 .235 .07 .79
Within Cells 39 3.255
178
TABLE 6 -- MULTIVARIATE ANALYSIS OF VARIANCE OF AGE INDEPENDENCE SCALE
SUBSCALE MEANS FOR PARENTS OF HIGH AND LOW
ACHIEVERS AND BOYS AND GIRLS
Sources
of
Variance
Multivariate
F Tests
Variable
(Subscale
Number)
Univariate
F Tests
F(7,33) p Less Than F(l,39) p Less Than
1 2.17 .15
2 4.90 .03
3 1.74 .19
Achievement 8.45 .001 4 .06 .81
5 .81 .37
6 .16 .69
7 2.52 .12
1 6.30 .02
2 .10 .75
3 .03 .87
Sex 2.52 .03 4 1.57 .22
5 .03 .87
6 .11 .74
7 2.23 .14
1 .52 .48
2 .02 .89
Achi evement 3 .02 .89
X
3.44 .007 4 3.16 .08
Sex 5 .21 .65
6 .77 .39
7 .39 .54
H
- n)
VO
TABLE 7
MEAN BLOCK STACKING TASK SCORES FOR
PARENTS OF HIGH AND LOW ACHIEVERS
Achievement
Group
Number of
Sets of Parents
Mean BST Scores (Fathers
Estimates
Trial # 1 Trials # 2+3
+ Mothers)
Total
Comments
Trials
# 1+2+3
Bovs & Girls
High 21 14.47 32.24 202.95
Low 22 14.19 33.59 98.28
Bovs
High 11 14.27 31.36 175.27
Low 9 13.89 33.78 118.00
Girls
High 10 14.70 33.20 233.40
Low 13 14.39 33.46 84.62
180
TABLE 8 -- MULTIVARIATE ANALYSIS OF VARIANCE AND COVARIANCE OF BLOCK STACKING TASK
SCORES FOR PARENTS OF HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Sources Multivariate Univariate
of Covariates F Tests Variable F Tests
Variance F p Less Than F p Less Than
Estimate 1 .11^ .74
(None) 2,95a .05 Est. 2 + 3 .34 .56
Comments 7.17 .01
Estimate 1 • 28e .60
Age 3.27b .03 Est. 2 + 3 .40 .53
Comments 7.40 .01
Estimate 1 .03e .86
Achievement
I.Q.
1.38b .26 Est. 2 + 3 .71 .41
Comments 2.48 .12
Estimate 1 .21f .65
Age + I.Q. 1.65c .20 Est. 2 + 3 .91 .35
Comments 2.37 .13
Child's Estimate 1 .0 1e .94
Performance 2.41b .08 Es t • 2 + 3 3.18 .08
Comments 5.28 .03
Estimate 1 . 28d .60
(None) . 12a .95 Est. 2 + 3 .10 .75
Comments .09 .76
Estimate 1 . 26e .61
Age .llb .95 Est. 2 + 3 .10 .75
Comments .09 .77
Estimate 1 . 24e .63
Sex
I.Q.
.08b .97 Est. 2 + 3 .06 .82
Comments .0 2 . .89
Estimate 1 .26 .62
Age + I.Q. .09° .97 Est. 2 + 3 .05 .82
Comments .02 .88
Child•s Estimate 1 . 15e .70
Performance ,08b .97 Est. 2 + 3 .03 .87
Comments .01 .92
181
TABLE 8 -- (Cont.)
Sources Multivariate Univariate
of Covariates F Tests Variable F Tests
Variance F p Less Than F p Less Than
Estimate 1 0. a .97
(None) .45a .72 Est. 2 + 3 .21 .65
Comments 1.35 .25
Estimate 1 .0 2e .89
Age . 50b .69 Est. 2 + 3 .15 .70
Comments 1.56 .22
Achievement
u
Estimate 1 . 0 1e .94
X I • Q • . 23 .88 Est. 2 + 3 .08 .78
Sex Comments .68 .42
Estimate 1 .0 2f .88
Age + I.Q. to
0
.87 Est. 2 + 3 .02 .89
Comments .72 .40
Child's Estimate 1 .30e .59
Performance • 93b .44 Est. 2 + 3 2.32 .14
Comments .18 .68
a df = 3,37
b df = 3,36
c df = 3,35
d df = 1,39
e df = 1,38
f df = 1,37
182
TABLE 9
MEAN BLOCK STACKING TASK ESTIMATES FOR TRIALS #2+3
OF MOTHERS AND FATHERS OF HIGH AND LOW ACHIEVERS
Achi evement
Group
Number
of
Parents
Mean BST Estimates
# 2 + 3
Mothers
for Trials
Fathers
Bovs & Girls
High 21 15.62 16.52
Low 22 16.95 16.64
Bovs
High 11 14.36 16.82
Low 9 16.11 17.67
Girls
High 10 17.00 16.20
Low 13 17.54 15.92
183
TABLE 10 — CORRELATIONS BETWEEN AGE INDEPENDENCE SCALE AND BLOCK STACKING SCALE
SCORES FOR PARENTS OF BOYS AND GIRLS
BST
Variable
Sex
Overall
Correlation with AIS
(Father + Mother)
Subscale
Means
1 2 3 4 5 6 7
Fathers + Mothers
Estimate Boys -.28 -.19 -.12 -.09 -.33 -.29 -.11 -.25
Trial # 1 Girls -.14 -.17 -.11 -.24 -.06 -.14 -.13 -.18
Estimate Boys -.22 -.35 .14 -.04 -.32 -.27 .02 -.19
Trial #2+3 Girls .00 .10 -.13 -.41* .17 .02 .05 -.11
Total Comments Boys -.32 -.02 -.26 -.04 -.18 -.38* -.26 -.35
Girls -.36* -.12 -.40* -.01 -.38* -.38* -.16 -.46**
Fathers
Estimate Boys -.08 -.12 -.26 .06 -.05 -.06 -.14 .03
Trial # 1 Girls -.37* -.38* -.23 -.30 -.37* -.31 -.26 -.45*
Estimate Boys -.20 -.11 -.18 .04 -.17 -.25 -.12 -.15
Trial #2+3 Girls -.25 -.25 -.26 -.41* -.19 -.19 -.37* -.21
Total Comments Boys -.43* -.09 -.33 .03 -.33 -.44* -.37* -.45*
Girls -.25 -.09 -.32 .02 -.33 -.25 -.09
i
to
Mothers
Estimate Boys -.45* -.19 -.19 .26 -.53** -.53** -.02 -.43*
Trial # 1 Girls -.01 -.01 -.35* -.04 -.01 -.02 -.03 .11
Estimate Boys -.18 -.26 .26 .07 -.33 -.26 .17 -.17
Trial #2+3 Girls .22 .31 -.06 -.32 .38* .20 .28 -.02
Total Comments Boys -.10 .11 -.16 .03 .00 -.18 -.08 -.09
Girls -.36* -.12 -.19 .12 -.34 -.33 .06 -.60***
Note: for Boys, N = 20 and for Girls, N = 23
*
P
< .05
**
P <*
.01
***
P
<
.001
oo
. f c .
TABLE 11 -- NUMBER OF AGE INDEPENDENCE SCALE ITEMS MARKED
"ESPECIALLY IMPORTANT TO ME" BY PARENTS OF
HIGH AND LOW ACHIEVERS
Achievement
Group
Number of
Fathers
+ Mothers
Number of AIS Items
Marked
"Especially Important to
Fathers + Mothers Fathers
Me" by
Mothers
Bovs & Girls
High
Low
42
44
588a
322
246d
187
342f
135
Bovs
High
Low
22
18
320b
133
164e
85
1569
48
Girls
High
Low
20
26
268c
189
82
102
186h
87
Note:--In the columns labeled "Fathers" and the one labeled "Mothers," N = \
the number given in the column labeled "Number of Fathers + Mothers."
— The "Fathers + Mothers" column is the algebraic sum of the total number
of items marked by the fathers plus that marked by the mothers in that
particular group; it was not obtained by combining a set of parents
individual scores and then totaling those scores across that group.
— df = 1
2 = 90.42, p <.001 2 = 40.48, p <.001 = 58.18, p < .001
2 = 10.90, p <.001
9 2
e 2 = 10.64, p <.01
X
= 34.58, p < .001
X
2 = 99.70, p <.001
2 = 80.38, p <.001
185
TABLE 12 -- NUMBER OF AGE INDEPENDENCE SCALE ITEMS MARKED
"NOT APPROPRIATE" BY PARENTS OF HIGH AND
LOW ACHIEVERS
Achi evement
Group
Number of
Fathers
+ Mothers
Number of AIS Items
Marked
"Not Appropriate" by
Fathers + Mothers Fathers Mothers
Bovs & Girls
High 42 51 29 22
Low 44 61 39 22
Bovs
High 22 25 13 12
Low 18 32 18 14
Girls
High 20 26 16 10
Low 26 29 21 8
Note:--In the columns labeled "Fathers" and the one labeled "Mothers," N = \
the number given in the column labeled "Number of Fathers + Mothers."
— The "Fathers + Mothers" column is the algebraic sum of the total number
of items marked by the fathers plus that marked by the mothers in that
particular group; it was not obtained by combining a set of parents
individual scores and then totaling those scores across that group.
--All x2 values are nonsignificant.
186
TABLE 13 -- TABULATION OF RESPONSES TO AGE INDEPENDENCE SCALE
SUMMARY PAGE FOR PARENTS OF HIGH AND LOW ACHIEVERS
Achieve
ment
Group
No. of
Fathers
&
Mothers
AIS Summarv
Number of
Paqe
Responses to
Questions Numbered
2 3
Mastered? Did .as
Expected?
4 5 6a
Atten-
Talent? tion
6b
Now?
7a
Curious?
7b
Now?
Same
Age
Birlier
Later
Always
Usual
ly
Occas.
Not
Often
Never
Leader
Follow
er
Yes
No
0)
0)
>* No
Can* t
Say
Yes
No
Can11
Say
Very
Some
what
Not
very
Can' t
say
Very
Some
what
Not
very
Can't
say
Bovs & Girls
Highs
42 19 21 0a 6 29 1 2 0 28 13 39 3d 6 33 1 4 36 0d 27 14 0 0 32 9 0 0f
Lows 44 26 14 4 1 22 4 1 0 27 17 29 32 8 33 3 8 33 1 23 17 3 0 22 19 1 0
Boys
High 22 9 11 0 3 16 1 0 0 16 6 21 1 6 14 0 4 16 0e 14 8 0 0 20 2 0 0g
Low 18 8 8 2 1 8 3 0 0 10 8 14 2 3 13 2 5 11 1 10 7 10 12 5 0 0
Girls
High 20 10 10 0 3 13 0 2 0 12 7 18 2C 0 19 1 0 20 0 13 6 0 0 12 7 0 0
Low 26 18 6 2 0 14 1 1 0 17 9 15 X) 5 20 1 3 22 0 13 10 2 0 10 14 1 0
Note:— The "Number of Responses" given under each response to each question was
obtained by adding the total number of such responses given by fathers to the total
number given by mothers from that group; it was not obtained by combining a set of
parents individual responses and then totaling those responses across that group.
a = 6.19, p <.05
(2) ‘
X
‘ dX(l)2 = 81.46, p <.001
Ml)'
9X
(1)
X(l)
2 = 5.15, p <.05
= 7.64, p < . 01
2 = 54.64, p <.001
c = 4.41, p <.05
X(l)
X (2 )
= 6.25, p <.05
00
TABLE 14
DIFFERENCES BETWEEN FATHERS AND MOTHERS AGE INDEPENDENCE SCALE
MEANS FOR PARENTS OF HIGH AND LOW ACHIEVERS
Achievement
Group
Number
of
Sets of
Parents Overall
AIS Means
(Father - Mother)
Subscales
1 2 3 4 5 6 7
Bovs & Girls
Highs 21 .30 .18 .21 -.19 .13 .67 .13 .46
Lows 22 .32 .25 .62 -.41 .26 .36 .08 .85
Boys
High 11 .68 .73 .78 .35 .41 1.10 .34 .83
Low 9 .54 .29 .60 -.78 .46 .79 -.01 1.38
Girls
High 10 -.11 -.42 -.42 -.78 -.17 .21 -.10 .06
Low 13 .18 .23 .64 -.16 .13 .06 .14 .48
188
TABLE 15
ANALYSIS OF VARIANCE OF AGE INDEPENDENCE SCALE DIFFERENCES
BETWEEN FATHERS AND MOTHERS OVERALL MEANS FOR PARENTS OF
HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Source df MS F
P
Less Than
Total 42
Achievement (A) 1 .004 .01 .94
Sex (S) 1 3.48 4.08 .05
AS 1 .48 .57
VO
•
Within Cells 39 .85
189
TABLE 16 — MULTIVARIATE ANALYSIS OF VARIANCE OF DIFFERENCES BETWEEN
FATHERS AND MOTHERS AGE INDEPENDENCE SCALE SUBSCALE MEANS FOR
PARENTS OF HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Source
of
Variance
Multivariate
F Tests
Variable
(Subscale
Number)
Univariate
F Tests
F(7,33) p Less Than F(l,39) p Less Than
1 .06 .81
2 1.29 .26
3 .20 .66
Achievement .77 .61 4 .18 .68
5 .60 .44
6 .02 .89
7 .89 .35
1 3.75 .06
2 2.44 .13
3 .24 .63
Sex 1.11 .38 4 1.97 .17
5 3.95 .05
6 .15 .71
7 4.07 .05
1 3.14 .08
2 2.81 .10
Achievement 3 3.06 .09
X 1.01 .45 4 .15 .70
Sex 5 .04 .85
6 .62 .44
7 .03 .87
190
TABLE 17
AGE INDEPENDENCE SCALE MEANS FOR PARENTS OF BOYS AND GIRLS
Number
of
AIS Means (Father
Subs cales
+ Mother)
Group Sets of
Parents
Overall 1 2 3 4 5 6 7
Hiqh & Lows
Boys 20 17.80 19.15 18.62 17.73 19.61 16.51 19.30 14.18
Girls 23 17.38 17.62 18.59 17.47 18.99 16.47 19.51 13.15
Hiqhs
Boys 11 17.68 19.65 18.19 18.14 19.96 16.06 19.44 13.32
Girls 10 17.06 17.78 17.98 18.12 18.39 16.27 19.13 12.62
Lows
Boys 9 17.95 18.53 19.14 17.22 19.18 17.06 19.11 15.24
Girls 13 17.63 17.49 19.05 16.98 19.45 16.62 19.80 13.55
191
TABLE 18
MEAN BLOCK STACKING TASK SCORES FOR PARENTS
OF BOYS AND GIRLS
Group
Number
of
Sets of
Parents
Mean BST Scores (Fathers + Mothers)
Estimates
Trial #1 Trials # 2+3
Total Comments
Trials # 1+2+3
Hiqh & Lows
Boys 20 14.10 32.45 149.50
Girls 23 14.52 33.35 149.30
Hiqhs
Boys 11 14.27 31.36 175.27
Girls 10 14.70 33.20 233.40
Lows
Boys 9 13.89 33.78 118.00
Girls 13 14.39 33.46 84.62
192
TABLE 19 -- NUMBER OF AGE INDEPENDENCE SCALE ITEMS MARKED "ESPECIALLY
IMPORTANT TO ME" BY PARENTS OF BOYS AND GIRLS
Group
Number of
Fathers
+
Mothers
Number of AIS Items
Marked
"Especially Important to Me" by
Fathers + Mothers Fathers Mothers
Hiqhs & Lows
Boys 40 453 249a 204
Girls 46 457 184 273
Hiqhs
Boys 22 320 164b 156c
Girls 20 268 82 186
Lows
Boys 18 133 85 48
Girls 26 189 102 87
Note:— In the columns labeled "Fathers" and the one labeled "Mothers,» ' N = • § the
number given in the column labeled "Number of Fathers + Mothers."
— The "Fathers + Mothers" column is the algebraic sum of the total number of
items marked by the fathers plus that marked by the mothers in that
particular group; it was not obtained by combining a set of parents
individual scores and then totaling those scores across that group.
— df = 1
a^ 2 = 19.86, p <.001 b 2 = 20.26, p <.001 c 2 = 5.44, p <.02
193
TABLE 20 — NUMBER OF AGE INDEPENDENCE SCALE ITEMS MARKED "NOT
APPROPRIATE” BY PARENTS OF BOYS AND GIRLS
Number of Number of AIS Items
Fathers Marked
Group + "Not Appropriate" by
Mothers Fathers + Mothers Fathers Mothers
Hiqhs & Lows
Boys 40
Girls 46
Hiqhs
Boys 22
Girls 20
Lows
Boys 18
Girls_________________________ 26
57 31 26
55 37 18
25 13 12
26 16 10
32a 18 14b
29 21 8
Note:--In the columns labeled "Fathers" and the one labeled "Mothers, i ' N = \
the number given in the column labeled "Number of Fathers + Mothers."
— The "Fathers + Mothers" column is the algebraic sum of the total number
of items marked by the fathers plus that marked by the mothers in that
particular group; it was not obtained by combining a set of parents
individual scores and then totaling those scores across that group.
— df = 1
a 2 = 3.92, p <.05 b 2 = 5 -56 , P <-02.
^ X
194
TABLE 21 -- TABULATION OF RESPONSES TO AGE INDEPENDENCE SCALE
SUMMARY PAGE FOR PARENTS OF BOYS AND GIRLS
Number of
Fathers Number of Responses to
Group + AIS Summarv Page Questions Numbered
Mothers 2 3 4 5 6a 6b 7a 7b
Mastered? Did as Talent? Atten- Now? Curious? Now?
Expected? tion
I
i
I
Same
Age
Earlier
Later
Always
Usually
Occas.
Not
Often
Never
Leader
Follow
er
Yes
No
Yes
No
Can11
Say
Yes
No
Can't
Say
Very
Some
what
Not
Very
Can't
Say
Very
Some
what
Not
Very
c s s ^ t
Boys 40 17 19 2 4 24 4 0 0 26
Girls 46 28 16 2 3 27 1 3 0 29
Highs
Boys 22 9 11 0 3 16 1 0 0 16
Girls 20 10 10 0 3 13 0 2 0 12
Lows
Boys 18 8 8 2 1 8 3 0 0 10
Girls 26 18 6 2 0 14 1 1 0 17
t 35 3a
29 16 33 12
6 21 1
7 18 2
9 27 2 9 27 ld 24 15 1 0 32 7 O O?
5 39 2 3 42 O 26 16 2 O 22 21 1 O
6 14 0 4 16 CP 14
O 19 1 O 20 O 13
8 O 0 20
6 O O 12
2 0 09
7 0 0
8 14 2
9 15 10
3 13 2 5 11 1 10 7 1 0 12 5 0 0
5 20 1 3 22 O 13 10 2 0 10 14 1 0
Note:--The "Number of Responses” given under each response to each question was
obtained by adding the total number of such responses given by fathers to the total
number given by mothers from that group; it was not obtained by combining a set of
parents individual responses and then totaling those responses across that group.
* 2 = 5• 24> P<-05 b _ = 4.41, p <.05 c - 7.76, p<.05 d _ 6.43,
X (D X (l> X(2)2 X (2)
p < .02
6y /. v2 = 4.44, p < .05 o = 8.82, p <.02 9 \2 = 4.22, p <.05
Ml) ^(2 ) Ml)
'O
( J l
TABLE 22
DIFFERENCES BETWEEN FATHERS AND MOTHERS AGE INDEPENDENCE SCALE
MEANS FOR PARENTS OF BOYS AND GIRLS
Group
Number
Of
Sets of
Parents
AIS Means
(Father - Mother)
Overall 1 2
Subscales
3 4 5 6 7
Hiqhs & Lows
Boys 20 .62 .53 .70 -.16 .43 .96 .18 1.07
Girls 23 .05 -.05 .18 -.43 -.00 .12 .04 .30
Hiqhs
Boys 11 .68 .73 .78 .35 .41 1.10 .34 .83
Girls 10 -.11 -.42 -.42 -.78 -.17 .21 -.10 .06
Lows
Boys 9 .54 .29 .60 -.78 .46 .79 -.01 1.38
Girls 13 .18 .23 .64 -.16 .13 .06 .14 .48
196
TABLE 23
CORRELATIONS BETWEEN PARENTAL AGE INDEPENDENCE SCALE
MEANS AND THEIR CHILDRENS PERCEPTUAL STYLE SCORES
Measure of
Perceptual
Style
Group
Correlations with AIS Means
(Father + Mother)
Subscales
Overall 1 2 3 4 5 6 7
EFT Boys .03 .20 -.13 -.19 -.03 -.02 .01 .20
Girls .16 -.11 -.03 -.01 .19 .16 .30 .22
RFT Boys -.37* -.31 -.36 -.25 -.25 -.27 .34 -.31
Girls 0.00 -.15 -.02 .17 -.08 -.07 .29 -.03
Note:— For boys, N = 20, and for girls, N = 23.
TABLE 24
CORRELATIONS BETWEEN PARENTAL BLOCK STACKING SCALE SCORES
AND THEIR CHILDREN'S PERCEPTUAL STYLE SCORES
Correlations with BST Scores
Measure of (Father + Mother)
Perceptual Group Estimates 'fotal Comments
Style Trial # 1 Trials # 2+3 Trials # 1+2+3
EFT Boys
H
•
1
-.37 -. 20
Girls -.29 -.06 -.31
RFT Boys -.26 -.05 -.24
Girls .06 -.24 .16
Note:--For boys, N = 20, and for girls, N = 23.
TABLE 25
CORRELATIONS BETWEEN CHILDRENS PERCEPTUAL STYLE SCORES AND THE
DIFFERENCES BETWEEN THEIR FATHERS AND MOTHERS AIS MEANS
Measure of
Perceptual
Stvle
Croup
Correlations
(Father
with AIS
- Mother)
Means
Subscales
Overall 1 2 3 4 5 6 7
EFT Boys .46* .23 .46* .22 .33 .20 .54** .45*
Girls .31 .28 .43* .22 .39* .20 .03 .19
RFT Boys .21 .02 .04 .20 .11 .29 .03 .19
Girls .19 .20 .27 .42* .14 .12 -.01 .19
Note:--For boys, N = 2 0; for girls, N = 23.
* p <.05
** p <.01
199
TABLE 26
ANALYSIS OF VARIANCE OF DIFFERENCES, BETWEEN FATHERS AND MOTHERS
ON AGE INDEPENDENCE SCALE OVERALL GRAND MEAN
Source df MS F p Less Than
Grand Mean 1 4.25 4.79 .03
Within + Residual 42 .89
200
TABLE 27
MULTIVARIATE ANALYSIS OF VARIANCE OF DIFFERENCES BETWEEN
FATHERS AND MOTHERS ON AGE INDEPENDENCE SCALE
SUBSCALE GRAND MEAN
Sources Multivariate F Variable Univariate
of Test (Subscale F Tests
Variance F(7,36) p Less Than Number) F(l,42) p Less Than
1 1.86 .18
2 4.93 .03
3 1.47 .23
Grand Mean 2.51 .03 4 1.56 .22
5 6.18 .02
6 .33 .57
7 10.03 .003
201
TABLE 28
CORRELATIONS BETWEEN PARENTAL AGE INDEPENDENCE SCALE MEANS AND
THEIR CHILDREN'S AGES AND TOTAL--I.Q. SCORES
Variable Group
Correlations
(Father
with AIS
+ Mother)
Means
Subscales
Overall 1 2 3 4 5 6 7
Age Boys .27 .09 .11 .15 .24 .42* .19 .10
Girls .32 .38* .27 .33 .30 .37* .12 .19
Total— I,Q. Boys -.29 .01 -.07 -.09 -.11 -.43* -.11 -.42*
Girls -.20 .04 -.28 .16 -.25 -.19 -.21 -.18
Note:--For boys, N = 20; for girls, N = 23.
* p < .05
202
TABLE 29
CORRELATIONS BETWEEN BOYS AND GIRLS PERCEPTUAL STYLE SCORES AND
THEIR AGES, I.Q., AND SCHOOL ACHIEVEMENT TEST SCORES
Variable Group
EFT
Correlations with
Perceptual Style Scores
RFT RFT-C
Age Boys -.38* .04 -.03
Girls -.39* -.39* .36*
Verbal— I.Q. Boys -.10 -.15 .11
Girls -.31 -.24 .27
Non-Verbal— I.Q. Boys -.09 -.04 .07
Girls -.66*** -.26 .19
Adj. Ach. Test Scores Boys -.24 -.01 -.13
Girls -.34* -.11 .18
Note:— For boys, N = 20 and for girls, N = 23.
*
P
< .05
**
P
<.01
***
P
< .001
203
TABLE 30
CORRELATIONS BETWEEN PARENTAL BLOCK STACKING TASK SCORES AND
THEIR CHILDREN'S AGES AND TOTAL— I.Q. SCORES
Variable Group
Correlations with BST Scores
Estimates
(Father + Mother)
Total Comments
Trial # 1 Trials # 2+3 Trials # 1+2+3
Age Boys -.05 -.11 -.35
Girls -.15 .11 .35*
Total— I .Q. Boys .15 .33 .42*
Girls -.01 -.09 .38*
Note:— For boys, N = 20; for girls, N = 23.
* p <.05
204
TABLE 31
MEAN PERCEPTUAL STYLE SCORES FOR
HIGH AND LOW ACHIEVING CHILDREN
Achi evement
Group
Number
of
Children
EFT
Mean
Perceptual Stvle Scores
RFT RFT-C
Bovs & Girls
High 21 541.86 5.32 .37
Low 22 719.46 5.99 .36
Bovs
High 11 574.91 4.88 .33
Low 9 653.22 4.51 .40
Girls
High 10 505.50 5.80 .42
Low 13 765.31 7.02 .32
205
TABLE 32 -- MULTIVARIATE ANALYSIS OF VARIANCE AND COVARIANCE OF PERCEPTUAL
STYLE SCORES OF HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Sources Multivariate Univariate
of Covariates F_Tests________ Variable F Tests
Variance_______________ F _p Less Than______________ F_________p Less Than
EFT 5« 88d .02
(None) 1.89a .15 RFT .26 .62
__________________________________________RFT-C .06____________.81
EFT 3.91e .06
Age 1.31b .29 RFT .03 .87
Achievement RFT-C 0.00 _.99
EFT 3.75e .06
V.-I.Q. 1.97b .14 RFT .18 .67
________________________________________RFT-C .38 .54
EFT 1.321 726
Age & V.-I.Q. 1.89c .15 RFT 1.28 .27
RFT-C 1.23 .28
3BBBBIB3Q82BB3
(None) . 53a
BBSS338IBB8B
.67
EFT
RFT
RFT-C
.09d
1.64
0.00
.77
.21
1.00
EFT ,08e .78
Age . 52b .67 RFT 1.62 .21
RFT-C 0.00 .99
EFT .07e
.80
V.-I.Q. . 86b .47 RFT 2.37 .13
RFT-C .08 .78
EFT .13f .72
Age & V.-I.Q. 1.07c .38 RFT 2.86 .10
RFT-C .12 .73
to
O
a
TABLE 32--(Cont.)
Sources
of Covariates
Multivariate
F Tests Variable
Univariate
F Tests
Variance F p Less Than F p Less Than
(None) .80a .50
EFT
RFT
RFT-C
1.51d
.35
1.20
.23
.56
.28
Achievement
Age .43b .73
EFT
RFT
RFT-C
.70®
.08
.75
.41
.78
.39
X
Sex V.-I.Q. .68b . 57
EFT
RFT
RFT-C
1.49®
.19
.89
.23
.66
.35
Age & V.-I.Q. .26c .85
EFT
RFT
RFT-C
stsasiassssBBBB:
.53*
0.00
.31
Baaaaaa
.47
.95
.58
BBaBaaaBBBBaaBBB
a df = 3,37 b df = 3,36 c df = 3,35 d df = 1,39 e df = 1,38
f df = 1,37
to
O
-4
TABLE 33 -- MULTIVARIATE ANALYSIS OF VARIANCE AND COVARIANCE OF PERCEPTUAL
STYLE SCORES OF HIGH AND LOW ACHIEVING GIRLS
Sources Multivariate Univariate
of Covariates F Tests Variable F Tests
Variance F p Less Than F p Less Than
EFT 4.82a .04
(None) 1.60a .22 RFT .31 .58
RFT-C .83 .37
EFT 2.36e .14
Age .94b .44 RFT .04 .84
Achievement RFT-C .07 .80
EFT 2 .21® .15
Verbal-I.Q. 1.28° .31 RFT .18 .68
RFT-C .003 .96
EFT 1 .10* .31
Age & Verbal- 1.74e .20 RFT 1.07 .31
I.Q. RFT-C .29 .60
a df = 3,19 t> df = 3,18 c df = 3,17 d df = 1,21 e df = 1,20 f df = 1,19
CO
O
00
TABLE 34
MEAN PERCEPTUAL STYLE SCORES FOR BOYS AND GIRLS
Group
Number of
Children
Mean
Perceptual Style Scores
EFT RFT RFT-C
Hiqhs & Lows
Boys 20 610.15 4.72 .37
Girls 23 652.35 6.49 .36 5
Hiahs
Boys 11 574.91 4.88 .33
Girls 10 505.50 5.80 .42
Lows
Boys 9 653.22 4.51 .40
Girls 13 765.31 7.02 .32
209
TABLE 35
MEAN BLOCK STACKING TASK SCORES FOR HIGH AND LOW ACHIEVERS
Achieve
ment No. of
Group Children
Mean BST Scores
Child's Est.
Trial # 3 Performance
Requests
for Aid
Rejections
of Aid
Bovs & Girls
High 21 8.24 28.62 5.33 .62
Low 22 7.41 26.09 4.86 .27
Bovs
High 11 8.09 25.36 4.82 .82
Low 9 8.11 28.44 3.56 .22
Girls
High 10 8.40 32.20 5.90 .40
Low 13 6.92 24.46 5.77 .31
210
TABLE 36
MEAN BLOCK STACKING TASK SCORES FOR BOYS AND GIRLS
Achieve
ment
Group
No. of
Children
Mean BST Scores
Child's Est.
Trial # 3 Performance
Requests
for Aid
Rejections
of Aid
Hiahs & Lows
Boys 20 8.10 26.75 4.25 .55
Girls 23 7.57 27.83 5.83 .35
Hiqh
Boys 11 8.09 25.36 4.82 .82
Girls 10 8.40 32.20 5.90 .40
Lows
Boys 9 8.11 28.44 3.56 .22
Girls 13 6.92 24.46 5.77 .31
211
TABLE 37 — MULTIVARIATE ANALYSIS OF VARIANCE AND COVARIANCE OF BLOCK STACKING TASK
SCORES OF HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Sources
of
Variance
Covariates
Multivariate
F Tests Variable
Univariate
F Tests
F p Less Than F p Less Than
Est. # 3
1.48C
.23
(None) .89a .48 Perform. 1.62 .21
Req. Aid .04 .84
Rei. Aid 1.66 .21
Est. # 3 1.4ld .24
Age 1.03b .41 Perform. 1.68 .20
Req. Aid .06 .82
Rei. Aid 2.15 .15
Est. # 3 .14b .71
Tot.-I.Q. .59b . 67 Perform. .17 .68
Req. Aid .27 .61
Rei. Aid 1.95 .17
Achievement Est. # 3 .34° .56
Mothers
Comments .64b . 64 Perform. .17 .68
Req. Aid .20 . 66
Rei. Aid 2.12 .15
Est. # 3 .03d .87
Fathers
Comment .83b . 52 Perform. .13 .72
Req. Aid .35 .56
Rei. Aid 2.72 .11
Mothers Est. # 3 .0 0d
.96
+ .81b .53 Perform. .08 .78
Fathers Req. Aid .42 .52
Comments Rei. Aid 2.64
.11 w
212
TABLE 37--(Cont.)
Sources
of
Variance
Covariates
Multivariate
F Tests Variable
Univariate
F Tests
F p Less Than F p Less Than
Est. # 3 .42° .52
(None) .98a .43 Perform. .48 .49
Req. Aid .52 .48
Re i. Aid .37 .55
Est. # 3 .41d .53
Age .95b .45 Perform. .46 .50
Req. Aid .50 .48
Rei. Aid .38 .54
Est. # 3 .64d .43
Tot.-I.Q. 'O
o
O '
CO
Perform. .29 .59
Req. Aid .31 .58
Rei. Aid .27 .61
Est. # 3
.72d
.40
Sex Mothers .85b .50 Perform. .21 .65
Comments Req. Aid
Rei. Aid
.29
.27
.60
.61
Est. # 3 • 67d .42
Fathers 1 . 01b .42 Perform. .33 ..45
Comments Req. Aid .50 .48
Rei. Aid .34 .56
Mothers Est. # 3
.78d
.38
+ .95b .45 Perform. .39 .54
Fathers Req. Aid .42 .52
Comments Rei. Aid .31 .58
213
TABLE 37--(Cont.)
Sources Multivariate Univariate
of Covariates F Tests____ Variable ____________ F Tests
Variance_________________________ F p Less Than___________________ F_________p Less Than
Est. # 3 1.18C .28
(None) 2.30 ,08a Perform. 7.28 .01
Req. Aid .06 .81
________________________________________Rej. Aid .86____________.36____
Est. # 3 1.15d .29
Age 2.19 .09b Perform. 7.15 .01
Req. Aid .04 .84
________________________________________Rei . Aid . 50 ___________.48____
Est. # 3 .59d 745
Tot.I.Q. 1.92 .13b Perform. 5.47 .03
Achievement Req. Aid .31 .58
X ________________________________________ Rei . Aid .56 ___________.46
Sex Est. # 3 .72d 740
Mothers 2.03 .llb Perform. 6.22 .02
Comments Req. Aid .28 .60
________________________________________Rej. Aid .65____________.43
Est. # 3 .30d .59
Fathers 1.89 .13b Perform. 5.82 .02
Comments Req. Aid .39 .54
________________________________________Rej. Aid .49____________.49____
Mothers Est. # 3 .35d .56
+ 1.90 .13b Peform. 5.79 .02
Fathers Req. Aid .42 .52
Comments_______________________________Rej. Aid .51____________.48____
a df = 4,36 b df = 4,35 c df = 1,39 d df = 1,38
to
H
.L
TABLE 38
MEAN RING TOSS TASK SCORES .FOR HIGH AND
LOW ACHIEVERS
Mean RTT Scores
Future-
No. of Distance Child's Past
Achi evement Children from Est. Perform- Perform Est.-Actual
Group Post Trial #1 ance ance Performance
Bovs & Girls
High 21 15.86 3.62 6.00 2.95 6.38
Low 22 14.09 4.59 4.41 3.64 7.55
Bovs
High 11 17.55 3.36 7.46 2.36 6.18
Low 9 14.78 5.00 5.00 3.00 6.44
Girls
High 10 14.00 3.90 4.40 3.60 6.60
Low 13 13.62 4.31 4.00 4.08 8.31
w
Ul
TABLE 39
MEAN RING TOSS TASK SCORES FOR BOYS AND GIRLS
Achi evement
Group
No. of
Children
Distance
from
Post
Child's
Est.
Trial #1
Mean RTT Scores
Perform
ance
Future-
Past
Perform
ance
Est.-Actual
Performance
Hicjhs & Lows
Boys 20 16.30 4.10 6.35 2.65 6.30
Girls 23 13.78 4.13 4.17 3.87 7.57
Hicjhs
Boys 11 17.55 3.36 7.46 2.36 6.18
Girls 10 14.00 3.90 4.40 3.60 6.60
Lows
Boys 9 14.78 5.00 5.00 3.00 6.44
Girls 13 13.62 4.31 4.00 4.08 8.31
TABLE 40 — MULTIVARIATE ANALYSIS OF VARIANCE AND COVARIANCE OF RING TOSS TASK
SCORES OF HIGH AND LOW ACHIEVERS AND BOYS AND GIRLS
Sources
Multivariate Univariate
of
Covariates F Tests Variable F Tests
Variance F p Less Than F p Less Than
Distance
1.82C
.19
Est. # 1 5.64 .02
(None) 1.93a .11 Perform. 2.77 .10
F-P Perf. .77 .39
E-A Perf. .55 .46
Distance
1.29d
.26
Est. # 1 7.13 . .01
Achievement Age 2.12b .09 Perform. 2.66 .11
F-P Perf. .78 .38
E-A Perf. .23 .63
Distance 1.19a .28
Est. # 1 .57 .45
Tot.-I.Q. 1.12b .37 Perform. 4.00 .05
F-P Perf. .02 .89
E-A Perf. .07 .79
Distance 3.1^ .08
Est. # 1 .04 .84
(None) 1.75a .15 Perform. 4.38 .04
F-P Perf. 2.15 .15
E-A Perf. .53 .47
Distance
3.07d
.09
Est. # 1 .03 .86
Sex Age 1 1.71b .16 Perform. 4.28 .05
F-P Perf. 2.11 .16
E-A Perf. .50 .48
to
i —1
v t
TABLE 40--(Cont.)
Sources
Multivariate Univariate
of Covariates F Tests Variable F Tests
Variance F p Less Than F p Less Than
Distance 3.05d
.09
1.64b .18
Est. # 1 .01 .92
Tot.-I.Q. Perform. 3.79 .06
F-P Perf. 2.55 .12
E-A Perf. .63 .43
Distance . 82c .37
Est. # 1 2.21 .15
(None) .97a .45 Perform. 1.13 .29
F-P Perf. .01 .92
E-A Perf. .21 .65
Achievement Distance
1.05d
.31
X
.72b .61
Est. # 1 1.38 .25
Sex Age Perform. 1.00 .33
F-P Perf. .00 .95
E-A Perf. .05 .82
Distance . 77d
.37
Est. # 1 4.65 .04
Tot.-I.Q. 1.23b .32 Perform. .59 .45
F-P Perf. .13 .72
E-A Perf. .08 .78
a df = 5,35 b df = 5,34 c df = 1,39 d df = 1,3S
to
H
CD
TABLE 41
CORRELATIONS BETWEEN CHILDREN’S RING TOSS TASK AND
BLOCK STACKING TASK SCORES
Correlation with
BST Score____
RTT Score Group
Child’s Est.
Trial # 3 Performance
RTT-1
Distance from Post Boys -.16 -.32
Girls -.36* -.37*
RTT-2
Child's Estimate Trial # 1 Boys -.11 -.06
Girls -.38* -.32
RTT-3
Performance Boys -.17 -.20
Girls -.28 -.31
RTT-4
Future-Past Performance Boys .16 .10
Girls .03 .17
RTT-5
Estimate-Actual Performance Boys .13 .04
Girls .00 .07
Note:--For boys, N = 20; for girls, N = 33. * p <.05
APPENDIX C
FORMS
220
U N IV E R S ITY O F S O U T H E R N C A L IF O R N IA
P S Y C H O L O G I C A L R E S E A R C H A N D S E R V I C E C E N T E R
7 3 4 W E S T A D A M S B O U L E V A R D
L O S A N O E L C S . C A L I F O R N I A S 0 O O 7
D E P A R T M E N T O P P S Y C H O L O G Y
(SIS) 74S.11S7
Dear Parents:
The Psychological Research and Service Center o f the U n iv e rs ity o f
Southern C a lifo rn ia w i ll be conducting a study a t Marengo Elementary School.
The study is designed to fu rth e r our understanding o f such questions as:
What makes a c h ild in terested in achievement?
How does a c h ild learn to function independently?
At what age does your c h ild learn to perform various tasks?
This study has been approved by the Superintendent, D r. Fouts, and by
Mr. Eastwood, P rin c ip a l o f Marengo School. I t w i l l be conducted by Mrs.
Johnston, M .A ., and supervised by the U niversity fa c u lty and your school
adm inistration.
Each c h ild chosen fo r the study w i l l be given three tasks to perform ,
two dealing w ith perception and another involving achievement. Since a l l
three tests resemble games, most children fin d them fun to do. Each c h ild
p a rtic ip a tin g in the study w i l l be tested a t his school during regular
school hours. The testin g w i ll take about 30 minutes. As soon as the study
is completed, Mrs. Johnston w i l l be happy to discuss the o v e ra ll re s u lts
w ith you as w e ll as the resu lts obtained on your c h ild .
I f you are in terested in having your c h ild p a rtic ip a te in th is study
and are w illin g to spend a h a lf hour describing your c h ild 's development,
please f i l l out the bottom part of th is sheet and return i t in the enclosed
envelope by December 8, 1972. I f your c h ild is selected fo r the study,
Mrs. Johnston w i ll c a ll you to schedule your appointment a t your convenience.
S incerely yours,
A lb e rt R. Marston, Ph.D
Professor o f Psychology
1 am in terested in having m y c h ild p a rtic ip a te in th is USC study.
Signature _ _
Phone Number
AGE INDEPENDENCE SCALE — ELEMENTARY SCHOOL FORM
INFORMATION SHEET
D ire c tio n s ! Please ansver the follow ing questions as com pletely
as possible.
Male ■ B irthp lace _ _ _ _ _ _ _ _ _ _ _ *90 _____
Female _____
R eligious Preference!
P rotestant _ _ _ Roman C ath olic _____ Jewish _____ Other _ _ _ _ _ _ _
Education — C irc le highest grade completed!
Grade School! 1 2 3 4 5 6 7 8
High School! 1 2 3 4
Trade School or Business C ollege! 1 2 3 4
C ollege! 1 2 3 4 . Graduate School! 1 2 3 4
Occupation!
Ages o f C hildren (do not put names)!
Boys _____ _____ ____ _ _ G irls _____ . ___________
Other ad ults (in ad d itio n to your spouse) in home (r e la tio n s h ip )!
Your c h ild !
B irthplace _______________________ B irthd ate Age
Hand Preference!
Sight ______ L e ft E ith e r _
School Enrollment H istory — C irc le a l l grades completed a t
Marengo School! K 1 2 3 4 5 6
Present Grade Level _____
llln e s s e s s o r handicaps th a t might e ffe c t school work! None
V isu al _ _ _ Hearing _____ Speech _____ Other _ _ _ _ _ _ _ _ _ _
AUTHORIZATION FOR PARTICIPATION IN
USC STUDY
Date i _ _ _ _ _ _ _ _ _ _ _ _ _ _
I hereby au th orize the p a rtic ip a tio n o f my c h ild , _ _ _ _ _ _ _ _ _ _ _ _ _
in the study to be conducted a t Marengo School under the auspices o f
the Department o f Psychology, U n iv e rs ity o f Southern C a lifo r n ia ,
D r. A lb e rt R. Marston, d ire c tin g .
I understand th a t my c h ild v i l l be given three tasks to perform
— tvo d ealing v it h perception (The Embedded Figures Test and The
Rod & Frame Test) and another involving achievement (The Ring Toss
T e s t).
In a d d itio n , I authorize Mrs. Johnston to o btain my c h ild 's
- in te llig e n c e and achievement te s t scores from the school records.
I understand th a t a l l inform ation obtained from me, my c h ild ,
o r the school records is to be tre a te d as c o n fid e n tia l.
Mother _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Father
U N IV E R S ITY O F S O U T H E R N C A L IFO R N IA
P S Y C H O L O G I C A L R E S E A R C H A N D S E R V I C E C E N T E R
7 3 4 W E S T A D A M S B O U L E V A R D
L O S A N G E L E S . C A L I F O R N I A S 0 0 Q 7
D E P A R T M E N T O P P S Y C H O L O G Y
June 13, 1973
Dear Parentsi
Thank you and your ch ildren fo r your p a rtic ip a tio n e a r lie r
th is year in the USC study conducted a t Marengo Elementary School.
The data analyses have been completed and a b r ie f summary o f the
re s u lts is enclosed. . .
If.y o u have questions about the re s u lts or vould lik e a d d itio n a l
d e ta ils , Mrs. Johnston w i ll be a t Marengo School on Thursday and
F rid ay, June 2 ls t and 22nd, PleaRe c a ll Marengo School by Tuesday
(June 19th) i f you vish to make an appointment w ith Mrs. Johnston.
. Again, thank you fo r your in te re s t, cooperation and assistance
in helping to id e n tify the s k ills necessary fo r successful achievement
in school and clsevhare.
S incerely yours.
A lb e rt R. Marston, Ph.D,
Professor o f Psychology
& Psychiatry
AM.pj
RESEARCH SUMMARY
This study vas desiqned to fu rth er our understanding of
achievement and the relatio n sh ip between independent functioning
and achievement in elementary school boys and g ir ls . C entral ques
tio n s of the study were*
1. Do children who perform b etter on school achievement te s ts also
do b e tte r on th e o re tic a lly important tasks such as Block Stacking
and Ring Toss, which measure achievement in several precise ways?
2. Do high and lov achievers tend to score d iffe re n tly on perceptual
tests?
3. Do high achievers tend to function independently a t an e a r lie r
or la te r age than low achievers?
4 . Do boys d iffe r from g ir ls on Block Stacking, Ring toss, percep
tu a l tasks, and independent functioning?
Results
Question 1 - For g ir ls , the answer to th is question is , "Yes".
G irls who score higher on achievement te s ts stack more blocks than
g ir ls who score lower and are also able to put more rings on the
post in the Ring Toss Task.
For boys, higher scores on the achievement te s t grt w ith higher
Ring Toss scores, but w ith lower Block Stacking scorea-,.
Question 2 - To answer th is question, the achievement te s t scores
o f the children were compared w ith th e ir scores on two d iffe re n t
perceptual te s ts , the Embedded Figures Test and the Rod & Frame.
Scores from the Embedded Figures Test, a te s t in which the ch ildren
were asked to fin d a series of simple geometric designs hidden in
more complex designs, did re la te to those from school achievement
te s ts . Thus, the higher a ch ild scored on the achievement te s t,
the fa s te r he was able to find the hidden designs. This was tru e
fo r both boys and g ir ls .
No re latio n sh ip was found between achievement te s t scores
and the second perceptual task, the Rod & Frame (th is task measures
the a b ilit y to distinguish "true" v e r tic a l).
Questio n 3 - Two measures were used as n r> index o f independent
functioning, the Age Independence Scale (the questionnaire completed
by parents) and the Block Stacking task. According to parents responses
on the Age Independence Scale, higher achievers were expected to
function independently a t an e a r lie r age,them were lo ver achievers
on school re la te d a c tiv itie s such ast
Follow w ritte n d irections (model, recipe, e t c . ) .
226
2
W rite a one-page re p o rt,
and more generally on items such asi
Shov pride in h is own a b ilit y to do things v e il .
Try hard things fo r him self w ithout asking fo r help.
Thus, le v e l of school achievement does re la te to th is index of
independent functioning.
In the Block Stacking task the number o f times a c h ild asked
fo r or rejected his parents help and the to ta l number o f comments
made by the parents in helping the c h ild were used as measures of
independent functioning. The le v e l o f school achievement did not
re la te to the number o f requests fo r . or re je c tio n s or parents
help. However, parents o f higher achievers tended to make many more
comments to th e ir ch ildren than parents o f lower achievers. This
p a rtic u la r outcome was the reverse of what had been expected. The
question o f whether the number o f comments made by parents varies
w ith school achievement le v e l has been c le a rly answered; however,
vhether th is number of comments made can be considered an adequate
measure o f independent functioing is s t i l l a t issue.
Question 4 - D ifferences between scores fo r boys and those fo r g ir ls
were found on many o f the measures used in th is study. To begin
w ith , the group of boys and the group o f g ir ls used in the study
were about the same age and scored about the same on group i n t e l l i
gence tests. Boys’ Achievement Test scores tended to be somewhat
lo vsr than those fo r g ir ls . On the King Toss boys chose to stand
fa rth e r from the pin (increasing th e ir ris k o f fa ilu r e ) than g ir ls
and tended to get more rings on the post (when standing the same
distance from the p in ). On the Rod & Frame boys were able to fin d
the v e r tic a l axis more accurately than were g ir ls . On the Age
Independence Scale, parents o f g ir ls (on the average) ’ tended to
expect independent functioning a t an- e a r lie r age on items such asi
Use telephone to obtain inform ation.
Vacuum and dust own room.
than did parents o f boys. No s ig n ific a n t d ifferen ces between boys
and g ir ls were found on other types o f items on the questionnaire,
on the Embedded Figures Test, or in Block Stacking.
227
3
Conclusions
Though the re s u lts o f th is p a rtic u la r study must be considered
te n ta tiv e u n t il supported by other research, these general conclusions
seem reasonablei
1> C hildren vho score higher on school achievement te s ts tend to
excel in non-academic areas req u irin g motor coordination (Ring
Toss & Block Stacking) and in a t le a s t some perceptual s k ills
(th e Embedded Figures T e s t). •
2 . Parents o f higher achieving c h ild re n tend to set higher goals
fo r th e ir c h ild re n in some kinds o f a c tiv itie s in th at they
expect independent functioning a t an e a r lie r age than those o f
lower achieving c h ild re n .
3 . In contrast to Conclusion #2, parents of higher achievers also
tend to become more a c tiv e ly involved in helping th e ir c h ild
succeed (more comments on Block S tackin g ).
4 . Although a number o f d ifferen ces between the scores o f boys and
g ir ls were noted, these d ifferen ces do not appear to form any
c le a r p a tte rn .
H otel This study is only a small p a rt o f the much la rg e r in vestig a
tio n o f achievement and achievement m otivation. The study was designed
to increase our yndersieiuiiiig of achievement a t a th e o re tic a l, ra th er
than a t an applied or p ra c tic a l le v e l. Thus, the re s u lts cannot and
Should not be considered as recommendations or guidelines on how
parents or teachers should re la te to a p a rtic u la r c h ild or group of
C h ild ren . U n til a d d itio n a l th e o re tic a l and p ra c tic a l studies are
completed, these fin d in g s , though in te re s tin g , must be considered
te n ta tiv e and subject to change.
APPENDIX D
MEASURES OF PERCEPTUAL STYLE
228
229
THE EMBEDDED FIGURES TEST
Simplified Directions
"This is a test to see how fast you can find
figures that are hidden in a design. First, I am going to
show you some colored designs. Each time I show you one
of these designs, I want you to tell me what it looks like
to you. After you look at each design, I will show you a
simpler figure, which can be found in that larger design.
Then you will be given the larger design again, and your
job will be to try to find the smaller figure in it.
Let's go through one for practice."
Demonstration
1. S is shown the practice complex figure (P-l) for
1 C
"This is the first design; please tell me what it
looks like."
2. P-l is removed and S is shown the practice simple
figure (P) for 10 seconds.
"And, this is the simple figure. Now each time I show
you the simple figure, I want you to look at it very
carefully, so you can remember exactly what it looks
like— its size, shape, number of sides, the direction
its pointing, and all."
3. P is removed, (P-l) is re-presented, and the stopwatch
is started. S is asked to locate the simple figure
in it.
230
"Please try to find the simple figure in this design.
As soon as you have found it, tell me."
4. When S indicates he has the correct solution, the
time is noted (but the watch is left running) and he
is required to trace it.
"Okay, now I would like you to take this tracing
paper and pencil and trace the simple figure for me
so that I can be sure that you have found the right
one."
5a. Timing is stopped and noted only after S has traced
the correct solution.
"Very goodI You've got it! Now, why didn't you
choose the other triangle in this design?" (Follow
their comment with appropriate instructions)
b. If the solution is not correct, ask the S to continue
his search until the correct solution is found or
time runs out.
"No, I'm sorry but that isn't it. Do you remember
what the small figure looks like? (If not, repeat
step #2) Okay, then, look at the design again and try
to find the smaller figure in it."
After the practice trial, S is given the following
instructions: (leave book open to P-l).
"This is how we will do all of them. In every case
the smaller figure will be in the larger design, and it
will always be right-side-up. There may be several of the
smaller figures in the same large design, just like there
are in this design (demonstrate on P-l), but you are to
look only for the one that is right-side-up and the same
231
size, shape, direction, etc. as the simple figure I have
shown you. As soon as you have found the figure, tell me,
and then begin tracing it.
If you forget what the small figure looks like, you
may ask to see it again and you may see it as many times
as you wish. If you are ever not sure you remember the
simple figure, be sure and ask to see it again, because
you can't find the simple figure if you can't remember it.
Any questions?" (All of the S's questions are answered by
repeating relevant portions of the instructions.)
"Work as quickly as you can, since I will be timing
you, but be sure that the figure you find is right-side-up
and exactly the same size, shape, and direction as the
figure I have shown you."
Trial Procedure Summary
1. Complex Design (15 seconds)
"This is the next design. What does it look like to
you?"
2. Simple Figure (10 seconds)
"And, this is the simple figure."
3. Complex design again (Start stopwatch)
"Let's see if you can find the simple figure in this
one. Be sure and tell me as soon as you find it."
232
THE EMBEDDED FIGURES TEST
Procedural Notes
1. The same presentation procedure is used on the 12
test trials.
2. S's score on each trial is the time taken to find
the simple figure.
3. A maximum of 2 minutes is allowed; if failed to
locate within time limits, the score is recorded as
2' (F).
4. S is allowed to re-examine the simple figure as often
as he wishes. During this re-examination the complex
figure is always removed. Discourage S from taking
more than 10 seconds for each re-examination. Timing
is stopped during the re-examination of the simple
figure. The time at which S is shown the simple
figure again is recorded.
5. Note time at which S first reports he sees the simple
figure within the complex, but continue timing until
simple figure is traced correctly and note that time
also.
6 . If S' traces simple figure correctly, the time at
which the simple figure was first recognized is the
time used in scoring.
7. The time at which each unsuccessful attempt is made
is recorded.
8 . The total score is the total time taken to locate the
simple figure in all 12 designs. (Time consumed in
locating and tracing an incorrect figure prior to
finding the correct solution is also included.)
9. Discontinue test after 6 consecutive failures TO FIND
THE DESIGN.
233
EMBEDDED FIGURES TEST
Experimenter _ _ _ _ _ _ _ _ _ _ _ _ Data
Time
T r ia l
Number
Time Record (in seconds) For T o ta l
S olution
Time
Solution Re-exam m a t ion
of Simple Figure Reported Traced ; | Correct In correct
Practice
1
I
I
I |
||
I
'
J
1
I
' 1
' . ...
• • I I
I
jj
1
1
|
1 |
1
1
234
PORTABLE ROD AND FRAME
Directions for Children
Note: Apparatus must be on a sturdy table and be level.
Before S enters the room, be sure frame is not
tilted and that curtain is closed.
Before seating S in front of apparatus say:
"Now we are going to play a game called 'Rod and
Frame.,' This game is played using a very special machine.
The inside of this machine is made to look almost like a
minature room with walls, a floor and a ceiling. On the
back wall of the minature room is a black line that we
will call a rod. The rod is something like this (Hold up
a ruler on a pencil), and as you can see, this may be
tilted or slanted (demonstrate), or we can move it so that
it is straight up and down (demonstrate), like the walls
of a house, or like a telephone pole." The experimenter
moves the rod slowly, and says: "Is this rod straight up
and down now?"
"The idea of the game 'Rod and Frame' is to see how
well you can tell when the rod inside this machine is
straight up and down. The rod may be tilted and the
minature room may be tilted too, but what I want you to do
is to tell me whether the rod is straight up and down.
Then, if you think the rod is tilted, I'll show you how to
move it so that you can make it straight up and down. Do
you have any questions?"
235
"When you first sit down in front of the machine, you
will not be able to see inside it and it will seem very
dark. As soon as we are ready to start, I will open a
curtain in the machine and you will be able to look inside
it. Okay, now let’s go over to our special machine and
get you settled comfortably so that we can start."
Seat S in front of apparatus and adjust head rest.
S's hands must be in his lap, not touching table.
Tell S to keep his head in the rest at all times.
"The idea of this game is to see how well you can
tell when the rod is straight up and down. Do you
remember what straight up and down is? Okay, now we are
almost ready to start." (Allow S to take his head out of
the head rest while reviewing the instructions.) "But
first let me remind you what I want you to do."
"When I open the curtain, your job will be to tell me
whether the rod is already straight up and down or if the
top of the rod is tilted to the left or to the right. If
you say it is tilted, then I want you to Start moving the
rod a little at a time until it is straight up and down.
You move the rod with this knob, right here." (Show knob
to S) As soon as you finish getting it straight up and
down, tell me and put your hand back in your lap. Then I
will close the curtain, readjust the machine, and we will
do it again." (Ask S to put his head back in the head
rest and his hands back in his lap.)
236
Trial 1: Adjust the frame to 28L and the rod to 28L.
Open curtain. Say:
'•Please tell me if the rod is straight up and down
or tilted."
a. If S says the rod is tilted, say: "In which
direction?" and record the response. Then say:
"Okay. Please put your hand on the knob I showed
you and move the rod until it is straight up and
down. Let me know when you finish."
When S finishes, say: "Okay, that's fine. Now
put your hand back in your lap and rest a minute."
Then close the curtain, record the position of the
rod, and reset rod and frame for next trial.
b. If S says the rod is already straight up and down,
say: "You mean the rod is straight up and down
with the walls of the room we are in? Is it
straight up and down like a flagpole?"
(1) If he now says the rod is tilted, proceed as in
"a" above.
or
(2) If he repeats that the rod is vertical, say:
"Okay, that's fine. Rest a minute." Then close
the curtain, record the position of the rod, and
reset rod and frame for next trial.
Use the same procedure on all 8 trials.
Group Code Number.
ROD AND FRAM E TEST
E xperim en ter____________________ Date
Time
T r ia l
Number
Settings Degrees
D eviation
Comments
"Frame" Rod
TOTAL MEAN DEVIATION
APPENDIX E
THE AGE-INDEPENDENCE SCALE
238
239
THE AGE-INDEPENDENCE SCALE
Verbal Directions
"These questions have been used in other studies in
other times and places and so they are not necessarily all
appropriate for in South Pasadena, right now.
(Child)
For example, #53 (Read aloud) may not seem to apply.
However, I would like you to stretch the situation a bit
and assume:
(1) there is a local bus, and
(2 ) there is someplace could go--to visit a
(Child)
friend, to a movie, to the library, or whatever, and
then ask yourself at what age you would feel comfortable
having ____. use the bus by himself."
(Child)
(Explain items #59--assume transportation not an issue;
and, for only or youngest children, items #60 & 62--assume
they have younger brothers and sisters--similarly.)
"Also, if has already accomplished an item,
(Child)
put down the ages at which he did it, rather than the
expected age."
"In the second blank put an "X" if you feel that item
is something you especially want your child to learn
sometime in his development--disregarding age."
"If you come across any items you feel are totally
inappropriate for your child to learn, ever, at any age,
then put a "No” in the blank for that item and on the last
page of the Questionnaire (which is a summary page) you
will find a place (Question #1) to list those items."
241
AGE INDEPENDENCE SCALE— ELEMENTARY SCHOOL FO RK
D ire c tio n s
The a tta c h e d AGE INDEPENDENCE SCALE is designed to help d e te r
mine tn e ages a t which mothers and fa th e rs b elieve th e ir c h ild ren
should be a b le to handle v a rio u s tasks o r a c tiv itie s by themselves.
I t v i l l a ls o p ro vid e in fo rm a tio n as to which o f these tasks or
a c t i v i t i e s p aren ts fe e l are e s p e c ia lly im portant in th e ir c h ild 's
developm ent.
S ix ty -tw o separate ta s k s and a c t iv it ie s are lis te d on the
attached s h e e ts . There are two blanks beside each. Please in d ic a te
in th e f i r s t th e age a t which you b e lie v e your c h ild should be able
to handle t h is ta s k o r a c t iv it y by h im s e lf. Please assume th a t the
c h ild has a lre a d y received any a d u lt h e lp necessary fo r learnin g
th e ta s k i n i t i a l l y .
Check th e second blank on any item th a t you consider e s p e c ia lly
im po rtant fo r your c h ild to m aster by th e age you in d ic a te . You.may
check as many item s as " e s p e c ia lly im portant to me" as you w ish.
There are no r ig h t o r wrong answers. Please do not omit any
q u e s tio n s . Thank you.
ExampleI AGE MY ESPECIALLY
CHILD IMPORTANT
SHOULD TO ME
Choose own h a ir s ty le
_12_
Learn t o p la y a m usical instrum ent . . . . . . . . . 8
S in ce i t vas e s p e c ia lly im portant to th is fa th e r th a t h is c h ild
le a rn t o p la y a m usical instrum ent by th e age of e ig h t years, he
checked th e space provided in th e r ig h t hand column.
242
AGE MY ESPECIALLY
CHILD IMPORTANT
SHOULD TO M E
Stand up fo r rig h ts v ith other ch ildren . . . . . . . 1. _ _ _ ___
Work on ovn hobbies o r in te re s ts . . . . . . . . . . . 2 , _____ _____
Go outside to play vhen he vants to be noisy . . . . . 3, _____ .. ..
Try nev things on h is ovn vith o u t depending on h is
mother fo r help .................................................................. 4. ____ _ _ _ _
P a rtic ip a te e n e rg e tic a lly in gases and sports . . . . 5. _____ ___
Show p rid e in h is ovn a b ilit y to do things v e il . . . 6. ____ _____
TaKe p a rt in parents' conversations 7, ____ _____
T ry hard things fo r him self v ith o u t asking fo r help . 8, _ _ .
Be able to eat v ith o u t help in cu ttin g end handling
foods 9. _____ _____
Lead o th er ch ildren and assert him self in childrens
. groups 10._____ _____
Choose most o f ovn frie n d s ...................................... .... 11. ____ .
Do some reg u lar tasks around the house 12. ____ _____
Be in te re s te d in obtaining good grades in school . . . 13. _____ ______
Do v e i l in com petition v ith other ch ildren; tr y hard
to come out on top in games and sports • . . • • 14. ____ _____
Work out most quarrels v ith other children . . . . . . 15, ____ .
S elect presents fo r others .......................... 16. _ _ _ .
Choose to jo in (o r to resign from) Scouts or other
org anization . . . . 17. _ _ _ . .
Choose hov to spend allovance . . . . . . . . . . . . 18. ____ .
Dress s u ita b ly fo r occasion 19. _____ ______
S elect some clo th ing to be purchased (a d u lt present) . 20._____ _____
Be sent to shop fo r groceries in nearby store . . . . 21, ____ _____
Take f u l l re s p o n s ib ility fo r returning lib r a r y books . 22._____ _____
Ride a skateboard . . . . . . . . . . . . 23._____ _____
S elect minor decorative items fo r ovn room 24. ____ _____
243
AGE MY ESPECIALLY
CHILD IMPORTANT
SHOULD TO M E
Begin homework v ith o u t being reminded . . . . . . . , 25.
F in is h homework on t i m e ...................................... • .........................26.
Wash dishes as a re g u la r c h o r e .................................................... 27.
Go to an afternoon movie v it h a frie n d . . . . . . . 28.
Ride b ic y c le in s tre e t v ith lig h t t r a f f i c . . . . . . 29,
Answer telephone and v r it e down messages ..........................30.
Use telephone to o btain in f o r m a t io n ............................ , . 31,
Follow w r itte n d ire c tio n s (model, re c ip e , e tc .) . . , 32.
Use a c h ild 's encyclopedia . . . . . . . 33.
W rite a one-page re p o rt ................................ . . . . . . . 34,
Sew on a button 35.
P lay a game o f s o ftb a ll . . • • • • • • « , , , , , « 36,
Go away to summer camp fo r a week . . . . . . . , , , 3 7 .
Be l e f t to swim in a p u b lic pool (guard on duty) , . . 38,
Work fo r neighbors fo r money 39,
Replace lo s t a r t ic le w ith ovn money 40,
Ovn and operate a sim ple camera . . . . . . . . . . . 41.
Answer the door a t n ig h t (p aren t h o m e )........................... . 4 2 .
Get up fo r school w ithout being c a l l e d ................................43.
S e le c t ovn TV programs ................................................ 4 4.
Save money fo r a s p e c ific purpose . . . . . . . . . . 45.
Choose ovn h a ir s ty le ...................... 4 6.
Iro n ovn clo th es . . . . • • • • • • ..................... . . 47.
Vacuum and dust ovn room 48.
Prepare a simple meal using th e stove . . . . . . . . 4 9.
B icycle 10 to 12 blocks in re s id e n tia l area . . . . . 50.
244
AGE MY ESPECIALLY
CHILD IMPORTANT
SHOULD TO M E
Look in the newspaper fo r inform ation . . . . . . . 51» - _ _
Give a 5-minute o ra l report .................. . . . . . . . 52, ____ _____
{Tea lO C ll pUhllC ^ rancrv>rf af j < * > n f Ijne J . ' • • • • • •
* w — V M W M W M S W
Know how to use house key and where i t is hidden . . • 54. ____ _____
Stay home alone fo r 2-3 hours — daytime . . . . . .
? 55, ....... , - —
Plan own p a rty fo r frien ds and choose whom to in v ite • 56, ____ _____
Walk downtown alone — daytime . . . . . . . . . . .
57.
Make an o u tlin e o f a chapter . . . . . . . . . . . . 58. ____ _____
Go in to d e n tis t's o ffic e fo r check-up w ithout parent • 59. ____ _____
Supervise brothers and s is te rs 2-3 hours — daytime • 60. '____ _____
Have and use ovn savings account . . . . . . . . . • 61,
Supervise brothers and s is te rs 2-3 hours — n ig h t . • 62.
1 . As you th in k about the s ix ty -tv o items you have ju s t fin is h e d
reading, are there any items on the l i s t th a t you fe e l are not
appropriate goals fo r your c h ild 's development?
Yes _ _ _ _ No _ _ _
I f answer is “Yes," please in d ic a te item numbers! _ _ _ _ _ _ _ _
2 . Do you fe e l your c h ild has mastered the s ix ty -tv o item s lis te d !
about the same aae e a r lie r la te r
than other children?
3. Considering only the items you have checked as "E s p ecially
Im portant To Me," do you fe e l th a t your c h ild has mastered these
a c tiv itie s by the age you indicated?
Alvays Usually Occasionally Not very o ften Never___
4 . In group a c tiv itie s is your c h ild more o f si
leader or follow er_____ ?
5. Does your c h ild have any p a rtic u la r ta le n ts or s k ills (m usic,
sp orts, e tc .) th a t he has developed?
Yes _____ No . .
I f answer is "Yes," please l i s t ) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
6 . Did you fin d th a t th is c h ild demanded a g reat deal o f your
a tte n tio n as a preschooler?
Yes ■ No _____ C an 't say _ _ _
Does he s t i ll ?
Yes _____ No _____ C an 't say _____
7 . Hov curious vas he as a small child?
Very Somewhat Not very C an 't sav
Hov curious is he now?
Very Somewhat Not very C an 't say
8. Does he get help w ith h is homework?
Yes No C ertain subjects only
From whom? Mother Father Other
9 . Do you believe your c h ild could do h is homevork by h im self o r
do you th in k he re a lly needs help? Doesn't need h e lp
C a n 't do without help Could i f he vould C a n 't say
10. How are the other ch ildren in the fa m ily doing in th e ir school work?
Average Above_Average_____ Below Average
246
II.
III.
IV.
AGE-INDEPENDENCE SCALE ITEMS LISTED BY
SUBSCALE CATEGORY
Self-Care--4 items
31. Use telephone to obtain information.
47. Iron own clothes.
48. Vacuum and dust own room.
49. Prepare a simple meal using the stove.
Cognitive Facility— 6 items
32. Follow written directions (model, recipe, etc
33. Use a child's encyclopedia.
34. Write a one-page report.
51. Look in the newspaper for information.
52. Give a 5-minute oral report.
58. Make an outline of a chapter.
Physical Skills--3 items
23. Ride a skateboard.
35. Sew on a button.
36. Play a game of softball.
Social Responsibility--14 items
2 1 . Be sent to shop for groceries in nearby store
2 2. Take full responsibility for returning
library books.
25. Begin homework without being reminded.
26. Finish homework on time.
27. Wash dishes as a regular chore.
30. Answer telephone and write down messages.
39. Work for neighbors for money.
40. Replace lost article with own money.
41. Own and operate a simple camera.
43. Get up for school without being called.
45. Save money for a specific purpose.
55. Stay home alone for 2-3 hours— daytime.
60. Supervise brothers and sisters 2-3 hours--
daytime.
62. Supervise brothers and sisters 2-3 hours—
night.
247
VI.
VII.
Autonomy— 15 items
1. Stand up for rights with other children.
2. Work on own hobbies or interests.
7. Take part in parents1 conversations.
11. Choose most of own friends.
15. Work out most quarrels with other children.
16. Select presents for others.
17. Choose to join (or to resign from) Scouts or
other organization.
18. Choose how to spend allowance.
19. Dress suitably for occasion.
20. Select some clothing to be purchased (adult
present). .
24. Select minor decorative items for own room.
44. Select own TV programs.
46. Choose own hair style.
56. Plan own party for friends and choose whom
to invite.
61. Have and use own savings account.
Wide Experience--10 items
28. Go to an afternoon movie with a friend.
29. Ride bicycle in street with light traffic.
37. Go away to summer camp for a week.
38. Be left to swim in a public pool (guard on
duty).
42. Answer the door at night (parent home).
50. Bicycle 10 to 12 blocks in residential area.
53. Use local public transportation (bus).
54. Know how to use house key and where it is
hidden.
57. Walk downtown alone--daytime.
59. Go into dentist's office for check-up without
parent.
3. Go outside to play when he wants to be noisy.
4. Try new things on his own without depending on
his mother for help.
5. Participate energetically in games and sports.
6 . Show pride in his own ability to do things
well.
8 . Try hard things for himself without asking for
help.
10. Lead other children and assert himself in
childrens groups.
12. Do some regular tasks around the house.
248
13. Be interested in obtaining good grades in
school.
14. Do well in competition with other children;
try hard to come out on top in games and
sports.
APPENDIX F
AGE-INDEPENDENCE SCALE DATA
249
Parental AIS Item Means
Arranged by Parental Group and AIS Item Number
Mean Aqe-Respons
e of
AIS Fathers of All Mothers of All All
Item Hi Hi Lo Lo Fathers Hi Hi Lo Lo Mothers Parents
No. Boys Girls Boys Girls Boys Girls Boys Girls
1 6.27 6.70 8.11 6.07 6.68 6.00 5.60 7.11 6.36 6.25 6.47
2 6.90 5.50 8.11 6.79 6.80 5.91 6.50 7.78 6.71 6.68 6.74
3 5.55 5.80 7.33 4.92 5.79 5.00 4.20 5.56 5.07 4.95 5.37
4 6.91 5.60 7.89 6.21 6.59 5.36 6.50 6.89 6.57 6.32 6.45
5 7.27 7.00 8.67 7.77 7.65 6.36 6.90 7.33 6.21 6.64 7.14
6 5.64 4.80 7.56 6.38 6.07 4.50 4.90 6.11 5.21 5.16 5.62
7 7.36 6.80 10.00 9.29 8.39 6.60 6.10 6.44 6.92 6.55 7.49
8 7.91 6.70 8.78 7.43 7.66 6.82 7.00 7.22 7.50 7.16 7.41
9 6.45 5.90 6.78 6.21 6.32 6.36 5.70 6.56 5.79 6.07 6.19
10 7.91 7.60 8.78 8.71 8.27 7.55 7.56 6.43 7.86 7.46 7.88
11 6.91 8.40 8.13 7.71 7.74 5.82 7.10 6.00 7.08 6.52 7.14
12 7.36 5.90 8.11 7.21 7.14 6.09 5.10 6.63 5.79 5.86 6.51
13 6.91 6.10 8.44 6.93 7.05 6.82 6.44 7.89 6.93 7.00 7.02
14 8.18 7.60 9.22 8.54 8.37 7.88 8.10 7.88 8.33 8.08 8.23
15 8.45 8.60 8.89 8.07 8.45 6.27 7.90 8.67 8.00 7.68 8.07
16 8.18 7.70 9.00 7.93 8.16 7.09 6.60 8.88 7.64 7.49 7.83
17 9.91 8.80 8.56 8.43 8.91 7.91 9.00 8.56 8.86 8.59 8.75
18 8.90 8.78 8.56 8.29 8.60 7.90 8.00 8.00 8.50 8.15 8.38
19 9.18 9.10 9.11 8.36 8.89 7.36 8.30 7.63 8.71 8.07 8.48
20 8.73 8.90 8.11 9.57 8.91 7.18 9.10 8.13 8.93 8.37 8.64
21 9.27 8.90 9.67 10.36 9.61 8.55 8.20 8.89 9.07 8.70 9.16
22 9.45 8.80 9.78 9.14 9.27 9.09 8.40 9.33 9.21 9.02 9.15
23 7.73 9.67 7.14 7.71 8.05 8.22 9.38 7.75 8.08 8.32 8.18
24 8.45 8.00 9.11 8.79 8.59 7.45 8.00 9.22 8.14 8.16
8 ’38 to
9.02 Ln
O
25 9.64 8.60 9.67 8.93 9.18 9.09 8.00 9.44 8.93 8.86
250
Parental AIS Item Means— (cont.)
Mean Aqe-Response of
AIS
Item
No.
Fathers of All
Fathers
Mothers of All
Mothers
All
Parents Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
26 10.36 9.67 8.30 9.07 9.34 8.55 7.80 9.00 8.93 8.59 8.97
27 9.88 8.50 10.00 9.23 9.32 9.33 8.78 9.29 9.14 9.13 9.22
28 9.45 10.20 9.67 10.08 9.86 9.00 9.22 9.78 9.57 9.40 9.63
29 10.00 9.50 9.78 9.92 9.81 9.18 9.60 9.11 9.71 9.43 9,62
30 8.64 7.70 9.00 8.79 8.55 9.09 7.30 8.78 8.36 8.39 8.47
31 9.82 8.90 9.11 9.57 9.39 9.09 9.40 8.11 8.43 8.75 9.07
32 9.45 8.90 9.33 9.71 9.39 8.45 8.80 9.11 9.21 8.91 9.15
33 8.55 8.00 9.33 9.00 8.73 7.73 8.30 8.44 8.57 8.27 8.50
34 8.82 8.20 9.44 9.50 9.02 8.55 8.50 8.88 8.64 8.63 8.83
35 11.67 8.80 10.00 8.71 9.62 10.60 9.00 11.13 9.00 9.79 9.70
36 9.00 7.90 8.11 8.79 8.50 7.91 9.90 7.71 8.86 8.67 8.58
37 10.27 8.80 9.22 9.92 9.60 10.00 9.90 9.89 9.57 9.81 9.71
38 8.36 8.20 8.00 8.85 8.40 8.91 8.00 8.22 8.43 8.41 8.40
39 11.09 10.20 10.11 10.93 10.64 10.00 11.40 9.89 10.50 10.47 10.55
40 11.10 8.89 8.78 11.07 10.12 10.00 10.90 8.89 10.29 10.07 10.09
41 10.27 9.00 9.89 9.21 9.57 9.73 9.30 9.22 8.93 9.27 9.42
42 9.09 7.90 8.78 8.38 8.53 9.00 9.22 8.67 9.00 8.98 8.76
43 9.64 8.60 10.00 8.50 9.11 8.91 8.40 9.44 9.79 9.18 9.15
44 8.55 8.67 7.67 7.71 8.12 8.45 10.00 6.75 9.21 8.71 8.41
45 8.36 7.70 8.22 7.64 7.95 8.27 9.30 7.56 8.21 8.34 8.15
46 11.36 9.00 10.67 9.14 9.98 10.27 9.20 10.33 8.79 9.57 9.77
47 12.63 10.10 12.38 10.36 11.15 11.50 11.60 12.86 10.57 11.44 11.30
48 11.00 8.80 9.50 8.64 9.40 10.00 9.00 8.75 8.57 9.07 9.24
49 11.64 9.70 9.56 9.71 10.16 10.45 9.80 9.89 10.07 10.07 10.11
50 9.91 9.30 9.00 9.92 9.58 9.91 9.60 8.89 9.57 9.52 9.55
51 9.64 8.70 9.67 9.79 9.48 8.55 9.20 9.22 8.86 8.93 9.20
251
Parental AIS Item Means--cont.)
AIS
Item
No.
Mean Aqe-Response of
Fathers of All
Fathers
Mothers of All
Mothers
All
Parents Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
52 9.82 9.40 10.67 10.57 10.14 9.09 9.90 10.00 9.64 9.64 9.89
53 10.55 10.89 11.38 12.62 11.44 10.64 11.30 11.67 12.64 11.64 11.54
54 9.45 8.40 9.78 9.31 9.23 9.09 7.90 8.78 9.36 8.84 9.03
55 10.45 9.60 10.11 10.15 10.09 10.45 9.10 9.22 9.36 9.55 9.82
56 9.45 9.00 9.00 9.64 9.32 9.00 8.60 8.00 9.50 8.86 9.09
57 10.00 10.60 9.11 10.31 10.05 9.64 10.70 10.33 9.71 10.05 10.05
58 10.64 9.50 10.78 10.43 10.34 10.00 10.50 9.75 10.14 10.12 10.23
59 11.82 11.70 11.00 11.85 11.63 10.18 10.50 10.00 10.64 10.37 11.00
60 11.82 11.10 10.67 11.69 11.37 12.09 10.50 10.50 11.43 11.21 11.29
61 10.27 9.80 10.78 10.29 10.27 9.09 10.60 9.56 11.93 10.43 10.35
62 12.73 11.50 11.89 13.31 12.44 13.64 12.70 11.44 13.21 12.84 12.64
(N=)
(1 1)
(1 0) (9) (14) (44)
(1 1)
(1 0) (9) (14) (44) (8 8)
to
01
to
Parental AIS Subscale and Overall Means
Arranged by Parental Group and AIS Subscale Number
Mean Aae-Response of
AIS Fathers of All Mothers of All All
Sub-
scale
No.
Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
Fathers Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
Mothers Parents
I 10.19 8.68 9.41 8.86 9.27 9.46 9.10 9.12 8.63 9.06 9.16
II 9.48 8.78 . - 9 .87 9.85 9.51 8.70 9.20 9.27 9.21 9t09 9.30
III 9.24 8.67 8.22 8.41 8.64 8.89 9.45 9.13 8.56 8.97 8.80
IV 10.18 9.11 9.82 9.79 9.74 9.78 9.28 9.36 9.66 9.54 . 9.64
V 8.58 8.24 8.93 8.34 8.50 7.48 8.03 8.13 8.28 7.99 8.24
VI 9.89 9.51 9.55 10.00 9.76 9.55 9.61 9.56 9.83 9.65 , 9.71
VII 7.07 6.34 8.31 7.02 7.14 6.25 6.28 6.93 6.54 6.49 6.81
Over
all 9.18 8.48 9.24 8.90 8.95 8.50 8.59 8.71 8.73 8.63 8.79
(N=)
(11)
(1 0) (9) (13) (43)
(ID
(1 0) (9) (13) (43) (8 6)
253
Tabulation of Parental 'Especially Important to Me1 ' AIS Responses
Arranged by Parental Group and AIS Item Number
Number of Responses by
AIS Fathers of All Mothers of All All
Item Hi Hi Lo Lo Fathers Hi Hi Lo Lo Mothers Parents
No. Bovs Girls Boys Girls Boys Girls Boys Girls
1 4 3 4 5 16 5 7 1 4 17 33
2 5. 2 3 4 14 5 7 0 2 14 28
3 1 0 0 0 1 1 1 1 0 3 4
4 7 3 1 6 17 5 7 0 3 15 32
5 6 3 6 2 17 1 4 • 0 1 6 23
6 5 5 4 7 21 6 6 2 4 18 39
7 5 2 2 3 12 5 3 0 0 8 20
8 8 5 2 3 18 3 9 1 4 17 35
9 2 0 0 2 4 2 2 1 3 8 12
10 3 0 2 2 7 2 1 0 1 4 11
11 4 4 1 1 10 2 3 1 1 7 17
12 7 6 3 5 21 4 7 1 3 15 36
13 8 4 3 7 22 3 8 3 3 17 39
14 3 2 3 0 8 1 2 0 0 3 11
15 5 2 3 4 14 4 7 2 4 17 31
16 2 0 1 1 4 1 2 0 1 4 8
17 3 1 1 0 5 2 1 0 1 4 9
18 4 0 2 0 6 1 2 2 1 6 12
19 2 1 1 2 6 2 6 1 2 11 17
20 2 0 2 1 5 2 2 1 0 5 10
21 4 1 1 0 6 2 2 1 1 6 12
22 3 1 0 2 6 2 4 0 2 8 14
23 0 0 0 0 0 0 0 0 0 0 0
24 1 0 0 1 2 2 2 1 2 7 9
25 3 4 4 4 15 5 10 5 4 24 39
254
Tabulation of Parental"Especially Important to Me" AIS Responses--(cont.)
Number of Responses by
AIS Fathers of All Mothers of All All
Item
No.
Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
Fathers Hi
Boys
Hi
Girls
Lo
Boys
Lo
Girls
Mothers Parents
26 5 3 2 6 16 6 9 2 3 20 36
27 2 1 0 3 6 1 2 0 1 4 10
28 0 0 0 0 0 1 0 0 0 1 1
29 1 0 0 0 1 3 1 0 1 5 6
30 5 2 1 3 11 3 9 0 2 14 25
31 4 1 1 1 7 2 6 1 2 11 18
32 3 4 1 1 9 2 7 1 3 13 22
33 3 2 2 1 8 6 0 1 2 9 17
34 2 1 1 0 4 4 1 0 1 6 10
35 0 0 0 1 1 0 1 0 2 3 4
36 0 0 0 0 0 2 0 0 0 2 2
37 0 0 1 0 1 1 0 0 1 2 3
38 1 1 1 0 3 3 0 0 0 3 6
39 2 0 0 0 2 3 0 1 0 4 6
40 3 1 4 4 12 3 4 2 2 11 23
41 0 0 0 0 0 1 0 0 0 1 1
42 2 0 0 0 2 2 0 0 2 4 6
43 2 1 0 3 6 4 4 1 2 11 17
44 1 0 1 0 2 0 0 0 0 0 2
45 5 1 4 2 12 3 6 1 1 11 23
46 1 0 0 1 2 0 0 0 2 2 4
47 1 1 0 1 3 1 0 0 1 2 5
48 2 3 1 2 8 2 4 1 2 9 17
49 2 0 2 2 6 2 3 1 1 7
13
50 1 0 0 0 1 2 1 0 1 4
to
5
Ul
255
Tabulation of Parental "Especially Important to Me" AIS Responses--(cont.)
Number of Responses by
AIS Fathers of All Mothers of All All
Item Hi Hi Lo Lo Fathers Hi Hi Lo Lo Mothers Parents
No. Boys Girls Bovs Girls Bovs Girls Bovs Girls
51 2 1 1 2 6 6 2 2 1 11 17
52 2 1 3 0 6 4 2 1 1 8 14
53 1 0 1 0 2 2 0 0 0 2 4
54 2 2 1 2 7 1 5 2 1 9 16
55 2 0 2 1 5 2 3 1 0 6 11
56 0 1 0 1 2 1 2 1 2 6 8
57 2 1 0 0 3 1 1 0 0 2 5
58 1 1 1 0 3 2 1 0 1 4 7
59 2 0 1 0 3 2 2 1 0 5 8
60 1 1 1 1 4 2 2 2 0 6 10
61 3 2 1 2 8
5
2 1 1 9 17
62 1 1 2 0 4 3 1 1 1 6 10
Total 164 82 85 102 433 156 186 48 87 477 910
(n= )
(1 1)
(1 0)
(9) (13) (43)
(1 1)
(1 0) (9) (13) (43) (8 6)
w
Ul
Ol
Tabulation of Parental "Not Appropriate" AIS Responses
Arranged by Parental Group and AIS Item Number
Number of Responses bv
AIS Fathers of All Mothers of All All
Item Hi Hi Lo Lo Fathers Hi Hi Lo Lo Mothers Parents
No. Boys Girls Boys Girls Boys Girls Boys Girls
1 1 1 1 1 2
3 1 1 2 2
5 1 1 1
6 1 1 2 2
10 1 1 1
11 1 1 1
13 1 1 1
14 1 2 1 4 3 1 1 4 9 13
18 1 2 3 1 2 3 6
21 1 1 1
23 1 3 3 1 8 2 3 2 2 9 17
27 3 4 1 8 2 1 3 6 14
28 1 1 2 1 1 3
29 1 1 1
35 2 4 6 1 2 3 9
36 1 1 1
37 1 1 1 1 2
38 1 1 1
39 1 1 1
40 1 1 2 2
42 1 1 1 1 2
43 1 1 1
44 1 1 1 1 2
47 3 2 5 3 3
257
00
257
Tabulation of Parental "Not Appropriate" AIS Responses— (cont.)
Number of Responses by
AIS Fathers of All Mothers of All All
Item Hi Hi Lo Lo Fathers Hi Hi Lo Lo Mothers Parents
No. Boys Girls Boys Girls Boys Girls Boys Girls
48 1 2 3 1 1 4
50 1 1 1
53 1 1 2 4 4
54 1 1 1
55 1 1 1
57 1 1 1
59 1 1 2 1 1 3
60 1 1 1
62 1 1 2 2
Total 13 16 18 21 68 12 10 14 8 44 112
(n= )
(1 1)
(1 0) (13) (9) (43)
(1 1)
(1 0) (13) (9) (43) (8 6)
258
APPENDIX G
MEASURES OF ACHIEVEMENT
259
260
BLOCK STACKING
Directions
Alright, now I would like __________ to join us. I
want to see how well he can build a tower using these
blocks. He will be blindfolded and may use only one hand.
You may say anything you wish to __________ to help him in
his building, but I don't want you to touch the blocks at
all. Okay?
(Get child to table, blindfold him, and spread out
blocks)
, I have some different shapes of blocks
(Child)
all mixed together on the table in front of you. Your job
is to build a tower by stacking up as many blocks as you
can on top of each other. You can take as long as you
wish to build your tower, but you may use only one hand in
building it. Your parents may give you as many hints as
they wish, but they cannot touch the blocks.
(To the parents) The average child your
__________________ age can build a tower 5 blocks high.
son or daughter's
I would like each of you to take this piece of paper and
write down privately how high you think __________ can
build his tower.
Okay, __________. Go ahead and build your tower.
When you finish be sure and keep the blindfold on because
I want you to do it again.
261
Trial 1
(Count blocks, write down score, mix blocks again,
ask parents to make another private estimate for
Trial 2)
Okay, __________ . Go ahead and build your tower.
When you finish be sure and keep the blindfold on because
I aim going to ask you to build it one more time after
this.
Trial 2
(Count blocks, record score, mix blocks)
Directions for Trial 3
Okay, that's fine. Now leave the blindfold on and
we will do it one-more time. This time we will do it a
little differently.
this time I will give you a nickel for
(Child)
each block stacked up in the tower. But, you must tell
me before you build the tower how high you will make it.
If the tower ends up higher than your guess, you don't get
paid for those extra blocks, just for the number you told
me about (give a numerical exsimple). And, if the tower
falls before you get to the number of blocks you guessed,
you don't get any money (give a numerical exsimple). Okay?
Now I want each of you to write down, privately, of
course, how many blocks you think __________ should stack
in his tower. (Get their answers.)
262
Okay, now , how many blocks do you think you should
(Child)
stack in your tower?
(Write down and meditate a minute)
Well, the estimates are not all the same. I think it
would be best for the three of you to talk this over
together and decide just how high the tower should be.
Mrs. ________ says ________ should stack ___ blocks;
Mr. ________ says ________ should stack ___ blocks;
and ________ says he should stack ___ blocks.
Now, what number do the three of you want me to write down
for the estimate?
(Allow discussion and record comments)
That sounds like a good number (write it down). Now,
__________ , go ahead and build your tower.
Trial 3
(Count blocks, record number, make payments if
earned, put blocks away, thank everyone and leave)
263
BLOCK STACKING
T r ia l Estimates C h ild 's
A ctively Involved P o s itiv e Experience
Number Mother Father C hild Performance Yes ?
No Yes ? No
1
2
3
• •
Group Estimate C hild C ollected *
T r ia l
Number
Mother Father
Chi
Asks fo r Aid
Id
Rejects A id ,
1 •
i
2
3 -
T r ia l Three
Number o f Statements in Goal Discussion
MOTHER FATHER CHILD
F b f i r s t
L e la s t
Total
C irc le Charsc te r is tic Type o f Comment
..........
MOTHER
FATHER
S pecific ?
N on-Specific S pecific ? N on-specific
Supportive ?
C r it ic a l Supportive ? C r it ic a l
C lear ? Unclear C lear ? Unclear
Pushing
'
Pushing
Examples Examples
264
RING TOSS
Directions
Measure height of S to determine how far he should
stand from the post. If S is 4' or between 4-51 in
height, have him stand at 5'; if he is 5’ or between 5-6’
have him stand at line 6 ', etc.
Trials 1. 3. 4. & 5
"Now we are going to play a game called 'Ring Toss'.
Have you played it before? Come over here so I can
measure you and see which line you get to stand on.
(Measure child and record height) Okay. All you have to
do is toss these rings onto the post in front of you. We
will play the game seven times and you will have nine
chances to throw the rings during each game. For the
first game I want you to stand here (pointing to the
appropriate line, according to height), when you throw the
rings. The only rule is that you must throw the rings.
You may stand and throw anyway you want--I just don't want
you to try to lean over the line and drop the rings on.
Okay?
Before we start, how many rings do you think that
you will be able to get on the post from here? (Record)
Any questions?
Okay, go ahead."
265
Trials 2. 6. & 7
"That finished that game. Now, we'll play the same
game again, but this time I'll let you choose the distance
where you will throw the rings. Do you see the pieces of
tape on the floor? You can stand behind any one of these
lines. Which line would you like to try? (Record)
How many rings do you think you will get on the post
this time? (Record) Remember, you aren't allowed to lean
over the line when you throw the rings.
Okay, go ahead and throw the rings."
(And, of course, record the number of rings S
gets on the post on each trial.)
266
Code Number
RING TOSS
.C h ild 's Height (inches)
Fixed Distance (fe e t) _
Experimenter
Date ________
Time ________
’ Attainment
i Discrppanry
! c o l. 2 minus
Number of rings | tioal Discrepancy
gotten on peg. col. 2 minus
(actual score) 9 col. 3 for
I or»ceilimi t r ia l
distance from
the peg
(Inches)
Number o f rings
estimated. Mevel
of expectation)
fix e d
fixed

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Relationship Of Teacher Empathy And Student Personality To Academic Achievement And Course Evaluation

PDF

Weight Reduction As A Function Of The Timing Of Reinforcement In A Covertaversive Conditioning Paradigm

PDF

The Effects Of Self Vs. Ideological Advocacy On The Self-Esteem And Endorsement Of Black Power Ideology Of Black College Students

PDF

A Multiple Investigation Of Child-Rearing Attitudes

PDF

The Enhancement Of Eeg - Alpha Production And Its Effects On Hypnotic Susceptibility

PDF

Attention, retention, and incentive processes in observational learning

PDF

Factors Of Adaptation And Rehabilitation In Home Hemodialysis

PDF

Mental Imagery As A Function Of Muscular Tension And Suggestion

PDF

The Influence Of The Achievement Motive, The Affiliation Motive, And Incentive Conditions On Roleplaying Ability In Children

PDF

Death Anxiety In Leukemic Children

PDF

The Effects Of Anxiety And Threat On Self-Disclosure

PDF

The Development And Evaluation Of Three Therapeutic Group Interventions For Widows

PDF

Relationship Of Personality And Task Demands To Self-Disclosing Behavior And Psychological Health

PDF

Factorial Stability As A Function Of Analytical Rotational Method, Type Of Simple Structure, And Size Of Sample

PDF

Individual Differences In Imagery Among Sensory Modalities

PDF

The Effects Of Einstellung On Visual Perception Of Random Shapes: A Multidimensional Scaling Analysis

PDF

Modification Of Low Self-Confidence In Elementary-School Children By Reinforcement And Modeling

PDF

Role Playing Ability And Social Adjustment In Children

PDF

Effects Of Encounter Group Methods Upon Selected Measures Of The Body Image

PDF

Looking Back After Coming Down: Conformity And Commitment In Campus Protest

Asset Metadata

Creator Johnston, Patricia Kirby (author)

Core Title Relationship Between Perceptual Style, Achievement, And Childrearing Practices In Elementary-School Boys And Girls

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, general

Format dissertations (aat)

Language English

Advisor Marston, Albert (committee chair), R. (committee chair), London, Perry (committee member), Rigler, David (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-820535

Unique identifier UC11363107

Identifier 7409070.pdf (filename),usctheses-c18-820535 (legacy record id)

Legacy Identifier 7409070

Dmrecord 820535

Document Type Dissertation

Format dissertations (aat)

Rights Johnston, Patricia Kirby

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA