Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Math and reading test anxiety among three populations: Mexicans, Hispanic immigrants, and Anglo-Saxon children
(USC Thesis Other)
Math and reading test anxiety among three populations: Mexicans, Hispanic immigrants, and Anglo-Saxon children
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
MATH AND READING TEST ANXIETY AMONG THREE POPULATIONS:
MEXICANS, HISPANIC IMMIGRANTS, AND
ANGLO-SAXON CHILDREN
by
Myra Debborah Saltoun
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
(Education)
May 1984
UMI Number: DP71335
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
in the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMI
Dissertation R jblishing
UMI DP71335
Published by ProQuest LLC (2015). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
uesf
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106- 1346
UNIVERSITY OF SOUTHERN CALIFORNIA
TH E G R A D U A TE S C H O O L
U N IV E R S IT Y PARK
LOS A N G E LE S. C A L I PORN IA 9 0 0 0 7
This dissertation, written by
....
under the direction of h^x.... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillment of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
^ f s U
h.]).
> u
6 n-
Date AgrU___2,__1984_
Dean
DISSERTATION COMMITTEE
\irman
il
ACKNOWLEDGMENTS
I am delighted to acknowledge the assistance,
cooperation, and encouragement of a number of individuals
whose support contributed to the completion of this
dissertation. Specifically I would like to thank the
following individuals:
my chairman. Dr. Dennis Hocevar, whose
assistance and encourgeraent in the planning of this
research was invaluable. His generosity with time
particularly during the analysis of the data was far
beyond the call of duty.
. . .my committee members Dr. Myron Dembo and Dr.
Franklin Manis, whose guidance and assistance were espec
ially helpful throughout the development of this disser
tation.
. . . my mother and step father Beverly and Burt
Bernstein, who have always encouraged my professional
growth and development. Their consistent support, willing
ness to help and editorial skills contributed greatly to
the completion of this research.
. . . my sisters Synthia and Julie, and my father
Andre', for their listening and continuous help and
encouragement.
111
. . . the faculty and administration of all the
schools that participated in this project. In Los
Angeles, Hoover Street School, Renter Canyon, Canterbury
Avenue, and Dixie Canyon are to be commended for their
enthusiasm and flexibility during the research. In Puerto
Vallarta, Mexico, the schools 20 de Noviembre, Club de
Leones, and Isidro Castillo, were very generous and accom
modating in letting me conduct the present research. A
special thanks needs to be given to the Lorenzana family
who aided me in coordinating this project in Mexico.
. finally all the boys and girls both in
Mexico and the United States who participated in this
research project deserve special recognition for their
hard work, enthusiasm, and cooperation throughout the
study.
iv
TABLE OF CONTENTS
ACKNOWLEDGMENTS i i
LIST OF TABLES................ vi
LIST OF FIGURES ...... vii
Chapter
I. INTRODUCTION. . . ....... 1
Statement of the Problem
Theoretical Rationale
Purpose of the Study
II. REVIEW OF THE LITERATURE AND RESEARCH
HYPOTHESES.. ......... 7
Measurement of Text Anxiety in Children
Etiology of Test Anxiety
Early Research on the Effects of Test
Anxiety
The Impact of Test Anxiety on the
Cognitive Process
Worry as a Component of Anxiety
Attention and Anxiety
Summary and Implications for the
Present Study
Math Text Anxiety
Cross Cultural Studies of Test
Anxiety
Conclusions from the Literature
Hypotheses and Research Questions
III. PROCEDURES............. 4?
Research Design
Description of Subjects
Description of Instruments
Description of Procedures
Statistical Analysis of Data
V
Chapter Page
IV. RESULTS ........ 59
New Immigrants and Hispanic Controls—
Anxiety Level
ANCOVA Results
Reading Test Anxiety vs. Math Test
Anxiety
Anxiety and Achievement Correlations
Between and Within Areas
Measurement Error in the TASC
V. DISCUSSION, CONCLUSIONS, AND
RECOMMENDATIONS ............. 73
Anxiety Level of New Immigrant
vs. Hispanic Controls
Reading Test Anxiety vs. Math
Test Anxiety
Anxiety and Achievement Correlations
Between and Within Areas
Measurement Error in the TASC and DSC
Conclusions and Implications for
Future Research
BIBLIOGRAPHY ...... 95
APPENDIXES 106
A. Approval for Study........ 107
B. Test Anxiety Scale for Children...... 109
C. Defensiveness Scale for Children........... 113
D. Directions for TASC and DSC...... 116
E. Student Answer Sheets for TASC and DSC,.... 119
VI
LIST OF TABLES
Table Page
1. i , test Comparison of School Anxiety
Between Recent Immigrants and Hispanic
Controls.................. 60
2. Race Differences— Comparison of Adjusted
Group Means..................... 62
3. Sex Differences— Comparison of Adjusted
Group Means............ 64
4. Grade Differences— Comparison of Adjusted
Group Means.................. 66
5. Interaction of Sex and Race on Reading
Test Anxiety.............. 67
6. Reliability of the TASC and DSC for
Each Ethnic Population and the Total
Population.......... 72
vii
LIST OF FIGURES
Figures Page
1• Correlations Examined in the Research
Hypotheses. ...... 37
2. Achievement and Anxiety Correlations
Within and Between Areas........... 69
3. Anxiety Correlations Between and
Within Areas. ............. 70
CHAPTER I
INTRODUCTION
Statement of the Problem
In recent years, a growing concern has evolved
among educators and psychologists over the subject of test
anxiety. Research indicates that such apprehension is
well founded. It should be observed at this point that
anxiety is cited in the literature as a major cause of
poor performance on school-related tasks. Specifically,
it has been demonstrated to lead to school drop outs, low
grade point averages, poor self-concepts as well as mini
mal achievement tendencies (Tobias, 1979). As early as
1940, researchers began finding relationships between
anxiety and underachievement. By I960, however, the focus
had turned to compensatory education (Tobias, 1979). It
was not until the late 1960s and early 1970s that interest
in anxiety as a variable in performance was revived.
Since that time theorists have attempted to explain the
specific context that anxiety affects both achievement and
learning in the school system.
2
Invariably, tests and/or test-like situations have
been cited as one of the highest anxiety-inducing situa
tions in the schools (Phillips, 1978), Specifically,
children have been found to have had increased levels of
test anxiety throughout the elementary school years.
There has been, however, a tendency for anxiety to sta
bilize by sixth grade (Hill, 1972). Among those found to
have high levels of test anxiety, minority children repre
sent the largest proportion of this population (Phillips,
1978). A number of studies comparing minority and Anglo
children's level of test anxiety indicates that minorities
consistently demonstrate a higher level of test anxiety
(Clawson 1981; Phillips, 1978; Schmidt & Gallessick,
1971). A few studies have even gone further and have
attempted to examine the factors that might account for
this finding. In particular, they have sought out the
specific factors in given schools which might contribute
to a greater level of text anxiety demonstrated by minor
ities when compared to Anglos,
Due to the fact that the previous decade has been
characterized by a large influx of immigrant children into
the public school system, the importance of this question
has been commensurately upgraded. Moreover, the fact that
some school districts, such as Los Angeles City Unified,
have become populated with predominately minority stu-
3
dents, the need to examine this situation is not only
relevant— it is vital so far as the improvement of the
educational system is concerned. It is almost self
evident that only by investigating such controlling
factors as text anxiety can the school performance and
achievement of the children be improved.
Theoretical Rationale
Recent research findings indicate that the roots
of test anxiety are often buried in a long history of poor
parent-child relationships. Among other things, parents
often tend to set unrealistically high standards for their
children. In efforts to achieve such unrealistic goals,
children tend to experience repeated failure. Further
contributing to this syndrome, parental evaluation of the
child's performance tends to be overly negative. In the
process of striving to gain more positive feedback from
their parents, the children tend to develop a great depen
dence upon both adult support and direction (Dusek, 1980).
It thus happens that when the child enters school, depen
dence upon adult support is duly transferred to the
teacher (Dusek, 1980; Hill, 1972). As a result of this
process, the teacher tends to play a critical role in
either reinforcing existing anxiety or in reducing it. In
either case, it has been found that there are only small
differences in the test anxiety level of younger children.
4
These differences between childr^en's anxiety level do not
appear until the upper elementary years. Evidently, some
thing happens during elementary school that# gradually
increases the test anxiety level of some children. This
relationship needs to be examined carefully.
It is now an accepted fact that with the great
influx of immigrants into the United States, the ratio of
minority students to Anglos has been rising constantly.
This is particularly true of Latinos from both Mexico and
Central America, where in recent years they have become
one of the largest minorities in the United States.
Despite the fact that the aforesaid research indicates
that minority groups have a higher level of test anxiety
than Anglos, little specialized research has been con
ducted involving the Latino population as a group. There
has been only one study made comparing Mexican children
living in Mexico with Anglos living in the United States.
The results confirmed a higher test anxiety level for
Mexicans in their country than for Anglos in the United
States (Diaz-Guerrero, 1976). Unfortunately, few if any
researchers have compared the level of test anxiety among
Mexican children in Mexico to that of Mexican immigrants
in the United States. A study utilizing this latter group
would enable researchers to identify which ethnic group
had a higher level of test anxiety. Moreover, it could
5
determine if the level of test anxiety of Mexican immi
grants decreases as the number of years they reside in the
United States increases. Perhaps researchers need to
examine the influence each educational and cultural system
has on the corresponding population to determine why dif
ferences in test anxiety level exist.
Another needed area of research is the test
anxiety construct. Distinguishing reading from math test
anxiety may suggest that test anxiety exists for partic
ular subject area rather than being an overall general
state. Specifically, two possibilities exist. Students
may be test anxious in one subject area and not another or
test anxiety may affect performance in all subject areas.
Once these differences are defined, educators could have a
better understanding of why many children may perform
poorly on tests in math and not in reading or vice versa.
Purpose of the Studv
The purpose of this study was to identify the
relationship of race, grade, sex, and years of residence
in the United States to school test anxiety. Text anxiety
specifically associated with math and reading achievement
was examined. Additionally, the relationships between
math test anxiety and reading test anxiety were inves
tigated.
6
In particular, the study involved Mexican children
residing in Mexico, Mexican children who had lived in the
United States for varying lengths of time, and native born
Americans, There were some children who had arrived in
the United States as recently as a month prior to the
study. Other immigrants had been in the United States for
most of their life. By comparing Mexican children in
Mexico with Mexican immigrants and Anglo children in the
United States, the researcher was able to determine
whether or not trends in anxiety level existed for each
cultural group. Specifically, were Mexican children more
anxious than the other two ethnic groups? JTf this in fact
was the case, then it would be expected that the level of
text anxiety of immigrant children would decrease in pro
portion to the number of years they lived in the United
States. Additionally, if test anxiety can be defined as
being related to specific subject areas, educators could
be better equipped to evaluate the etiology of high scores
of children on the Test Anxiety Scale. More importantly,
by determining that test anxiety is subject-specific,
educators could develop treatments aimed at reducing the
amount of test anxiety presently existing in the public
school system.
CHAPTER II
REVIEW OF THE LITERATURE AND RESEARCH HYPOTHESES
Test anxiety is an important variable that often
determines performance in school-related tasks. The term
is used to refer to an unpleasant feeling or emotional
state generated by concern about possible failure on a
given task. It is often accompanied by both physiological
and behavioral responses. It generally occurs during
formal testing or evaluative situations (Dusek, 1980). It
is the purpose of this study to review the literature on
test anxiety, including measurement techniques. An inves
tigation into the manner in which test anxiety initially
develops and the way in which it interferes with school
performance are investigated. Particular attention is
paid to studies differentiating between math and reading
test anxiety. The few studies that have compared cross-
cultural test anxiety are evaluated. Finally, the impli
cations for the study at hand are stated.
8
Measurement of Text Anxiety in Children
The most widely used questionnaire for assessing
children's text anxiety was developed by Sarason, David
son, Lighthall, Waite, and Ruebush (1960). Test Anxiety
Scale for Children (TASC) was created to measure anxiety
in evaluative, test-like conditions whereby the presence
or absence of support from adults was expected to be an
important factor (Sarason et al,, I960),
The TASC is a group-administered paper and pencil
test containing 30 items concerning reactions to inter
personal and evaluative situations. The examiner reads
each question aloud while the children respond by circling
"yes" or "no" on their answer sheet. The questions ask
about concerns of "tests" and "worry" over classroom per
formance. The child's anxiety score is the number of yes
answers circled on the test.
Sarason, Davidson, Lighthall, and Waite (1958)
demonstrated that the TASC had suitable reliability. In
their longitudinal study at Yale they reported the follow
ing results: (1) the reliability for test-retest (2 month
interval) procedures was .71; (2) for split-half reli
ability the corrected r was equal to .82-.88 depending on
grade level. Furthermore construct validity was demon
strated to be significant when comparing teacher ratings
of the child's anxiety with the student's self-report.
9
Studies in school settings have generally indi
cated a negative relationship between TASC Scores on test
performance. However, in many studies the negative corre
lations have been low (Hill & Sarason, 1966). In some
cases where intelligence was partialled out, the relation
ship between anxiety and test performance has been non
existent, More recent research also suggests that this
relationship between anxiety and test performance may vary
from situation to situation. More recent research sug
gests that this relationship may be weak because children
often lie about their anxious feelings or become defen
sive. This results in depressing their true score on the
TASC (Phillips, 1968). Thus, Sarason et al. added two new
scales, the Defensiveness Scale for Children (DSC) arid the
Lie Scale for Children (LSC) which are used simultaneously
with the TASC. These scales guard against test-taking
attitudes of defensiveness and malingering. O'Reilly and
Wightman (1971) demonstrated that adjusting the original
TASC score utilizing these additional scales significantly
improved construct validity.
Etiology of Test Anxiety
Researchers generally agree that test anxiety
originates in parent-child interactions during both pre
school and early school years. It is a result of chil
dren's reactions to evaluative situations (Hill, 1972;
10
Sarason et al., I960). In the 1950s, Sarason and his
colleagues developed a theory within a psychoanalytic
framework concerning the etiology of test anxiety. They
contend that reactions of children who are anxious in
test-like situations in school are a reflection of evalu
ative experiences in the home during preschool years but
which tend to stabilize during the ensuing school years.
Poor parent-child relationships begin when a child's per
formance does not live up to their parents' unrealist
ically high standards for achievement. As a result,
parental feedback relating to the child's performance is
often negative. Subsequently, children internalize these
negative parental feelings and tend to develop a hostile
view of their rejecting parents. These hostile feelings
breed guilt in the child, who then begins to engage in
self-derogation and repression of negative feelings toward
their parents. Additionally, the child fears parental
retaliation for their growing hostile feelings. The
threat of retaliation then causes the child to engage in
behaviors aimed at seeking both approval and support from
parents. In trying to perform adequately, the high test-
anxious child may develop a strong dependence upon adult
direction and support in evaluative situations (Hill,
1972; Dusek, 1980; Sarason et al., I960).
11
Hermans, ter Laak, and Maes (1972) provide some
impirical evidence in support of Sarason's theory. These
researchers examined parent-child relations pertaining to
the development of test anxiety. While children were in
problem-solving situations, the researchers found that
parents of high-anxious children were negative or non
caring toward their children. In particular they ignored
their children's longing for security and did not provide
constructive help in solving the problem. As a result
they often taught their children to engage in task-
irrelevant behaviors. As a further outcome, these chil
dren began to rely on external support from parents or
other adults to confront problem solving and evaluation
tasks as a way of avoiding criticism. In contrast,
parents of low-anxious children, offered suggestions to
their child for effectively completing the task, while
additionally encouraging them to rely on their own
resources for solving tasks.
Additional data for linking child-rearing prac
tices with the development of test anxiety are lacking.
This has posed a problem for those attempting to test,
evaluate, or improve Sarason's theory. Fortunately there
currently exists a significant amount of both theory and
research describing the effects of test anxiety on learn
ing and school performance.
12
Early Research on the Effects of
Test Anxiety
Beginning in the mid 1950s many studies of test
anxiety were conducted. Investigators attempted to
explain and treat academic performance deficiencies in
test-anxious individuals (Benjamin, McKeachie, Lin, &
Holinger, 1981). Most researchers found that anxiety was
an important factor in producing relationships and discrep
ancies between one's potential and performance (DeBlassie,
1972). Specifically, as the level of test anxiety in
students increased, their level of performance decreased
(Benjamin et al., 1981; Dusek, 1980; Hill, 1972).
Sarason, Hill, and Zimbardo (1964) conducted a
longitudinal study relating anxiety to IQ, achievement
test, and classroom performance. The sample of subjects
consisted of two consecutive grades of a suburban, pre
dominantly white middle and working class school system.
When the study was initiated 670 children in first and
second grade served as the subjects. They were followed
for five years. Measures of test anxiety were obtained
from a combination of scores on the Test Anxiety Scale for
Children (TASC), the Defensiveness Scale (DSC), and the
Lie Scale (LSC). Several significant findings were de
rived from this study. Foremost, when the three scales
were used together, the relationship of test anxiety and
13
school performance was equally strong for boys and girls;
a negative correlation between TASC and IQ scores
increased steadily through the elementary years. A
similar relationship between the TASC and grades existed.
Second, the researchers found a consistency in TASC scores
over two-year intervals with the condition stabilizing
toward the end of elementary school. There was, however,
no relationship found between TASC scores obtained in
first and fifth grades. This indicates that few chronic
cases of test anxiety exist across the entire elementary
school years. Furthermore, this suggests the possibility
that school factors may contribute to an increase in test
anxiety. Finally, children who had higher defensiveness
scores admitted to less anxiety on the TASC. But, it
should be observed that those who reported increasingly
more anxiety showed increasingly poorer performance scores
than those whose TASC scores decreased.
Early experimental studies concentrated on the
evaluative aspects of testing conditions. They also
focused on the effects of short-term social interactions
as well as success-failure experiences. One such study by
Sarason et al. (I960) involved manipulating test condi
tions of both high and low test anxious children. One
group of children took the Stroop Test. This test pre
sents the child with tasks that are increasingly more
14
difficult thereby producing increasingly greater pressure.
This test maximized the interfering effects of text
anxiety. The remaining children were given the Porteus
Maze Test which minimized interfering effects of test
anxiety. There were no time pressures and the experi
menter was supportive. Results indicated better perfor
mance of high test anxious students on the Maze Test and
better performance of low anxious students on the Stroop
Test.
Grimes and Allinsmith (1961) studied the influence
of children's anxiety on the effectiveness of different
methods for teaching reading. Their findings suggested
that the sight-vocabulary method which involved trial and
error guessing and hence more chances of failure was not
the most effective method of teaching reading to high
anxious children. The more structured phonics method was
more suitable.
A third study by Gampeau (1968) compared the
effects of feedback on the performance of a programmed
learning task. Low test anxious girls performed better
with no feedback. However, high test anxious girls were
more successful in conditions where feedback was given.
The researcher concluded that feedback reduced the threat
of failure for the high anxious girls and encouraged task
15
relevant responses because the subjects could progress at
their own speed.
As noted in the etiology section of this study,
high anxiety is partly a result of parental or adult nega
tive reactions to a child's failure experiences. Hill
(1972) postulated that a child's history of success and
failure experiences could explain why they become test
anxious. Basically Hill claimed that since low-test
anxious children had greater success in school they experi
enced more positive interactions with adults in evaluative
settings'. Therefore, such children developed a high
motive to succeed rather than to avoid failure. In con
trast, high test anxious children who had a poorer history
of success in school, had experienced more punitive inter
actions with adults. Thus, they developed a strong moti
vation for avoiding failure and criticism rather than one
of striving for success. In this respect, understanding
children's evaluation of success and failure based on
adult feedback is relevant to describing the effects of
anxiety on children's performance (Dusek, 1980).
Several studies support Hill's hypothesis. As
early as I960, Ruebush (I960) found direct evidence
relating high test anxiety and cautious behavior in
problem solving. When compared to low anxious boys, high
anxious boys demonstrated significantly more cautious
16
behavior on the Witkin Embedded Figures Test. The ten
dency to be overly cautious was interpreted as a sign of
attempting to avoid failure. Messer (1970) confirmed
these results by manipulating the success and failure
experiences of subjects on an intellectual task. Simi
larly, he found more cautious behavior indicated by longer
decision times of subjects who were conditioned to fail
ure. Bradshaw and Gaudry (1968) measured the relation of
success and failure experiences and scores on the TASC of
ninth grade students. Students were given a multiple
choice test. They were divided into three groups: a
control group, a group that experienced failure on the
test, and a group that experienced success. After the
test, they immediately scored their answers, compared
their scores to a standard, and took the TASC. Results
indicated that pupils who experienced failure on the test
scored higher on the TASC than those who experienced suc
cess. This study as well as several other studies (Dusek
& Hill, 1970; Handler & Sarason, 1951; Sieber, Kaneya, &
Paulson, 1970) suggest that histories of success and fail
ure are important influences on the development of text
anxiety and subsequent performance.
A final set of studies suggested that the anxious
child's strong success-failure motives are based on strong
praise-criticism motives. Kozma (1969) found that high
17
test anxious children learned more readily during brief
periods of social isolation than did low test anxious
children. The researcher concluded that the high test
anxious children performed better in conditions of social
isolation. The reason here is that under such circum
stances no fear is generated of possible negative evalu
ation coming from an adult. A study by Cox (1966) showed
that the presence of an observer with grade school boys
decreased the performance level of the high test anxious
subjects and increased the performance rate of the low
test anxious. Lastly, Ganzer (1968) observed the behavior
of college students during a test when they were told they
were being observed through a one-way mirror. High test
anxious students performed poorly in this condition
because they were distracted and displayed more task-
irrelevant responses than did the low-anxious subjects.
The Impact of Text Anxietv on the
Cognitive Process
The aforementioned early studies certainly suggest
a substantial performance deficit for high anxious chil
dren. Furthermore, explanations for the decline in per
formance were given. More recent studies have addition
ally suggested that high levels of anxiety inhibit per
formance because it has a negative impact on one *s
1 8
cognitive process. This is primarily due to the worry
component of anxiety.
Tobias (1979) contended that the anxiety level of
students indirectly affects their learning by obstructing
the normal cognitive processes that moderate learning.
Specifically, Tobias stated that there are three stages of
information-processing! input, processing, and output.
Input refers to the introduction of instructional concepts
to students. Processing is basically what the student
does with the material: store it, record it, or organize
it. Output is the performance of the student when re
quired to use this instructional material. Tobias sug
gested that anxiety can interfere with cognitive processes
at three particular points: preprocessing, during pro
cessing, and immediately following processing. During
preprocessing anxiety can impair learning by interfering
with one's internal representation or registration of the
input material. Primarily, it has been found that anxious
students divide their attention between task demands and
self-concerns. The minimal attention given to the input
is often not enough to make it the effective stimuli neces
sary for proper processing. Moreover, anxiety inhibits
processing in three ways: reduces performance on diffi
cult tasks, interferes with the use of short and inter
mediate term memory, and generates a need for highly
19
organized input to insure success. Lastly, Tobias implied
that during postprocessing, anxiety hinders retrieval of
information previously mastered.
Support for the aforementioned model was derived
from a study by Benjamin, McKeachie, Lin, & Holinger
(1981). They compared high, medium, and low test anxious
college students in an upper-level psychology course.
Results indicated that high test anxious students did
poorly on essay and short answer type questions but did
well on multiple choice questions. This latter type of
question involves less active retrieval. This finding was
consistent with Tobias' contention that anxiety hinders
retrieval during postprocessing. Further Benjamin et al.
reported that high test anxious students had problems
learning material throughout the course, picking the impor
tant points in the reading assignments, and encoding infor
mation at more than a superficial level. These outcomes
are additional support for Tobias' model. It suggests
that anxiety impairs learning during preprocessing by
interfering with registration of input materials.
Worrv as a Component of Anxietv
Eysenck (1979) distinguished between two cognitive
components of anxiety: emotionality and worry. Emotion
ality is associated with uneasiness, tension, and nervous
ness. Worry is primarily reflected in concern for one's
20
level of performance and poor self-concepts. While emo
tionality appears to be unrelated to test performance,
researchers have found a significant relationship between
worry and test performance (Tryon, 1980), Benjamin et al,
(1981) suggested that worry was not simply a personality
characteristic of test anxious students. It may however
be caused by inadequate knowledge of the subject matter
because of preprocessing interference.
Morris and Liebert (1970) closely examined the
relationship between worry and test performance. In one
study subjects were given five Wechsler Adult Intelligence
Scale (WAIS) subtests. Half of the subjects were given
only easy items to complete and half were given only the
difficult items. Half of the subjects knew they were
being timed, while the others did not. Emotionality and
worry were evaluated. A 2 x 2 factorial design was estab
lished. Results indicated that there were no significant
interactions between emotionality and task performance.
Worry was highly related to both task difficulty and
timing, suggesting that worry is the key factor influ
encing performance. In two other studies similar results
were reported. Morris and Liebert (1970) used a sample of
college graduates and high school students. The Liebert-
Morris scale of worry and emotionality was administered
immediately before a course final. In both studies final
21
examination grades negatively correlated with worry but
not with emotionality.
Eysenck (1979) claimed that this relationship
between worry and performance exists because worry gen
erates task-irrelevant activities that hinder the quality
of one's performance. Specifically, Eysenck stated: "The
major reason for this is that the task-irrelevant informa
tion involved in worry and cognitive self-concern competes
with task relevant information for space in the processing
system" (p. 364). In other words, subjects spend a great
deal of their task time worrying about performance, suc
cess of others, and the choices they have. Moreover, they
become repetitive while trying to solve a task. As a
result, high anxious people divide their attention over
two-tasks (worry and the task at hand), while low-anxious
individuals devote their attention solely to the cognitive
task demands. Thus, it seems reasonable to assume, since
it is the difficult tasks that high anxious subjects usual
ly perform poorly on, that their poor performance is an
outcome of inadequate attention to task cues. The follow
ing section examines this assumption.
Attention and Anxietv
In the previous sections of this study it was
established that high anxious students often engaged in
self-oriented behavior such as worry during task perfor-
22
mance. Thus, while the low anxious students directed
their attention and efforts toward the task, the high-
anxious attended to their personal needs. Although this
appears as a logical assumption, empirical evidence must
be included here for support. Ganzer (1968) found that
high anxious subjects, more so than any other group,
emitted many task-irrelevant comments while performing a
task in front of an audience. Furthermore, the comments
were mostly self-evaluative or apologetic in nature.
Eysenck (1979) cited a study whereby Nottleman and Hill
(1977) recorded off-task glancing of high and low anxious
subjects performing an anagram task. Inadvertently, high
anxious students had significantly inferior performance
and emitted more off-task glancing than did the low
anxious children. Lastly, Deffenbacher (1978) compared
performance of low and high anxious subjects on an anagram
task but under evaluative or nonevaluative conditions.
Results of a post-task questionnaire indicated that high
anxious students in the evaluative condition spent only 60
percent of their time on task activity, while the low
anxious spent about 80 percent of their time on task. The
highly anxious children claimed that they were distracted
from their tasks, that they worried a lot, and that they
performed in a poorer manner than did low anxiety sub
jects. The data from all three studies point to the fact
23
that high anxious students often exhibit task irrelevant
behavior which in turn impairs their performance.
Summarv and Implications for the
Present Studv
Test anxiety is a trait that can be measured
reliably in children. The most frequently used scale is
Sarason's et al. (1960) Test Anxiety Scale for Children
(TASC). The TASC provides a good index of test anxiety
related to the quality of a child's performance in evalu
ative situations. When used with the DSC and LSC its
validity is improved.
The little research available on the etiology of
text anxiety indicated that its development was associated
with poor child-rearing practices of parents. Specif
ically, parents of high test anxious children tend to
insist on unrealistically high standards of achievement
for their offspring. Additionally, these parents tend to
provide little emotional support for their children in
problem-solving situations during periods when they are
teaching them to rely on outside assistance rather than on
their own resources. Under such conditions a dependency
on outside support in evaluative circumstances tends to
evolve.
In studies that have examined the effects of
anxiety on performance, it has been found that a negative
24
correlation between performance and anxiety level exists
(Dusek, 1980). It has been further found that TASC scores
seem to increase throughout the elementary school years,
but stabilize by the sixth grade (Sarason et al,, 1960).
More recent research suggests that performance becomes
impaired due to the fact that anxiety tends to have a
negative impact on one's cognitive processes. In partic
ular, the worry component of anxiety interrupts the pro
cessing of information. The high test anxious child who
is worried during evaluation, tends to divide his atten
tion between task irrelevant behaviors and the performance
of the given task.
Before educators and schools can aid highly
anxious students, however, more research is needed to
identify precisely which cultural and school variables
tend to influence test anxiety. The question as to
whether or not test anxiety is subject-specific needs to
be examined. In particular, do children acquire test
anxiety in one school subject and then are free of test
anxiety in other areas? Contrarily, test anxiety may be a
feeling that prevails throughout all school subjects.
Moreover, these variables may differ across cultures and
educational systems. If answers to these questions are
established, more precise prescriptions for reducing
anxiety in academic situations will be possible. The
25
present study attempts to investigate these issues.
A certain amount of research along these lines has
already been conducted. The remainder of this review will
be concerned with those findings.
Math Text Anxietv
Math test anxiety is an emotion or state that is a
part of two much wider phenomena. Foremost, as Aiken
(1970) reported, the term "attitude" is often utilized in
studies that are referring to anxiety level. Much of the
existing research in this area is focused on math atti
tudes which are often determined by math anxiety level.
Moreover, math anxiety and math test anxiety are different
concepts but closely related (Richardson & Woolfolk,
1980). In other words math anxiety is a large part of
test anxiety experience in mathematic situations. There
fore, even though few studies concerning math test anxiety
have been reported, the research involving attitudes and
math anxiety pertains also to math test anxiety. The
findings from such research will be reported here.
At the outset, it should be noted that the tools
for assessing math anxiety have been limited. Self-report
is the most common method of measuring attitudes. It was
not until the 1970s, when research on math attitudes was
most popular, that a reliable measure was developed. The
Mathematics Anxiety Rating Scale (MARS) developed by
26
Richardson and Suinn (Richardson & Woolfolk, 1980) is a
98-item scale containing brief accounts of ordinary life
situations where the manipulation of numbers often arouse
anxiety. A math anxiety score is calculated after each
question is completed by assigning a value from 1 to 5
indicating the level of anxiety. Studies concerning the
reliability and validity of the measure have verified its
utility.
A major concern among educators studying math test
anxiety is its relationship to achievement. Contrary to
popular belief, researchers have found a low positive
correlation between math attitudes and achievement (Kulm,
1980). However, most of the studies conducted used a
general measure of anxiety and achievement which does not
accurately reflect specific classroom situations. Thus,
key factors that may influence either variable such as
attitudes toward the teacher, attributions, etc., are
ignored.
More recently interest in sex differences in math
achievement has emerged. Researchers have agreed that in
elementary school girls and boys do not significantly
differ in math achievement. It is not until junior high
that notable differences favoring boys begin to appear.
This is true in other countries as well (Aiken, 1976). By
college, surveys indicate significantly fewer women than
27
men are in math classes. Some investigators claim that
biological differences between men and women attribute to
a low ability in math for females. However, as Aiken
(1976) suggested, there are so many inconsistencies
between countries as to the magnitude of attitude differ
ences among girls and boys that the biological explanation
is not sufficient. More importantly both Aiken (1976) and
Richarson and Woolfolk (1980) have suggested careful
examination of sociocultural expectations and rein
forcement schedules as predictors of attitude. Specif
ically, the effects of stereotyping cultural expectations
and socialization experiences, reinforcement for math
achievement, and career guidance need to be evaluated.
Additionally, studies comparing math achievement between
males and females often do not control for the number of
math courses taken. Due to female attitudes toward math,
men usually take more classes and thus appear on achieve
ment tests to have a superior ability in solving math
problems (Kaczala, 1980; Richardson & Woolfolk, 1980).
However, Richardson and Woolfolk reported that when males
and females had similar course preparation no significant
difference occurred in anxiety level or achievement. This
strongly suggests that previous amounts of interaction
with math may predict math anxiety rather than sex dif
ferences .
28
Aiken (1976) cited a number of other factors that
influence attitude towards math. Recently, personality
and social factors have been uncovered. Specifically,
positive math attitudes and high achievement have been
related to characteristics such as a high sense of per
sonal worth, a greater sense of responsibility, high
social standards, and high achievement motivation.
Children with positive attitudes toward math appear to
like more detailed work, are self-confident, and intui
tive. In high school, students with above average math
ability are often less sociable and are usually on the
extreme ends of anxiety scales. That is, they are either
very secure or severely anxious. Other variables affec
ting anxiety level involve the interactions between stu
dent and teachers, course content, mode of instruction,
and parental and peer support.
It is clear from this review that a large amount
of literature on math anxiety has evolved, but numerous
questions are yet remaining. The absence of literature on
test anxiety in other subject areas makes it difficult to
determine if test anxiety is a quality that can affect
just one subject area or whether it affects all curric
ulum. Existing studies have many flaws because they have
not controlled for the many significant variables that
appear in classroom situations and affect anxiety. Sex
29
and personality factors related to math test anxiety are
important but not at the expense of ignoring cultural
attitudes and stereotypes toward subject matter. These
defects in the literature need serious research attention.
Although a growing body of literature exists con
cerning attitudes and anxiety in mathematics, similar
research has not evolved in other subject areas such as
reading. This may be a result of popular belief that math
is a subject for which some people have no talent. Thus,
it has become commonly accepted for a person to have a
fear of math, especially if it is difficult for them.
However, in reading, the discrepancy between poor and good
readers is not as apparent and it is not as common to
become anxious about reading. Moreover, test anxiety in
math may be more widespread because the very nature of
trying to solve a computational problem can simulate a
test-like situation.
Cross Cultural Studies of Test Anxietv
Text anxiety is a near universal experience, espec
ially in countries where test giving is a regular pro
cedure (DeBlassie, 1972). Phillips (1968) asserted that
in the United States, which is a test-conscious culture,
minority children exhibit more school anxiety than do
Anglos. Specifically, he reported that fourth and fifth
grade Black and Mexican American children are generally
30
more anxious than Anglo children. Even though Phillips
admitted that some of this difference may be due to
defensiveness and coping style, he contended that there
were genuine differences in school anxiety levels between
these groups. Additionally, factor analytic studies
(Phillips, 1962, 1968, 1980) found the following four
major dimensions of school anxiety: negative evaluation
by others, taking tests, meeting expectations of others,
and low tolerance for stress, Non-Anglos tended to be
much higher than Anglos in all four areas with the great
est difference found to be in taking tests. Additionally,
females had higher anxiety scores than males and highly
anxious subjects had lower achievement and intelligence
scores (Phillips, 1962). Since a number of sex and socio
cultural status interactions occurred, Phillips suggested
that the role of the school experience in the development
of school anxiety was dependent upon sex and sociocultural
status.
In support of the above hypotheses, several
studies have compared specific ethnic groups with Anglos.
Clawson, Firment, and Trower (1981) compared the stan
dardized test scores of Blacks and Anglos who were seventh
and ninth grade students in Florida. In addition, the
students were administered the Test Anxiety Inventory. In
examining the results of both tests, the researchers found
31
a significant trend: Blacks did significantly worse on
the standardized test in both language and math. Most
important, it was shown that they had higher state anxiety
scores. McCormick and Karabinus (1976) compared differ
ences in measures of self-esteem and anxiety between
children from three ethnic groups (Blacks, Hispanics, and
Whites) in upper elementary grades. They were divided
into three groups, top 1/3, middle 1/3, and bottom 1/3
according to scores on achievement and attitude tests. In
general, high performance across cultures was associated
with low anxiety. Finally, in a study by Hawkes and Furst
(1971), Blacks in the inner city indicated more test
anxiety than did Anglos living in the suburbs.
Gotts (1968) described an example of culture
affecting anxiety levels. Contrary to popular belief his
study revealed that Japanese 9-year-olds reported lower
general anxiety scores than did French of American chil
dren of the same age. This finding was surprising because
previous studies showed that Japanese college students
were by far the most anxious. However, Gotts explained
these findings in the following cultural terms: In
Japanese society the young and old have greater freedom of
expression then do middle age people. This freedom helps
reduce the anxiety level of young children while increas
ing it during middle age. However, for Americans freedom
32
of expression increases with age which might explain why
anxiety decreases with age for Americans.
Two final studies are particularly relevant to the
present investigation. They compared anxiety level of
Mexican children to that of Anglos. Schmidt and Galles-
sick (1971) compared the adjustment of Anglo-American and
Mexican American pupils with respect to anxiety in two
conditions; a self-contained or a team teaching class
room. The subjects were first and sixth grade students
from experimental team teaching schools and self-contained
classroom schools. In each condition there was one pre
dominately Anglo class and one predominately Mexican-
American class. Anxiety was measured by the Picture
Anxiety Test for first graders as well as the Phillips
Anxiety Test for sixth graders. In general Mexican
American first and sixth grade subjects reported signif
icantly higher anxiety levels than did Anglo subjects.
Additionally, it was found that in first grade the two
class organization procedures did not affect the ethnic
groups differently, but by sixth grade the Mexican Amer
icans in the self-contained classrooms reported a signif
icantly higher level of anxiety than Mexicans who were
team taught. These differences did not exist for the
Anglo population. The findings of this study suggest that
team teaching classrooms may benefit Mexican children by
33
maintaining their anxiety level at a relatively low rate.
Finally, Diaz-Guerrero (1976) compared the test
anxiety levels of Mexican and American children during
school examinations. The subjects were 393 children, half
of them Mexicans living in Mexico and the other half were
American. The students were in either first, fourth, or
seventh grade. The children were given the TASC to
measure their test anxiety level. The data were evaluated
using analysis of variance. The variables under study
were: the two cultures, high and low social class, sex,
and age. The most relevant finding was that Mexican sub
jects scored significantly higher than American subjects
did on the TASC. It was further found that regardless of
culture, subjects from the lower socioeconomic class
scored higher on the test anxiety scale than those from
the upper socioeconomic level. Finally, it was found that
regardless of culture, girls scored higher on the TASC.
These results indicate a clear difference in the test
anxiety level of Mexican and Americans. It should be
observed, however, that variables contributing to this
difference need further investigation.
Conclusions from the Literature
A vast body of literature on test anxiety has been
reviewed in this study. Some conclusions need to be
formed so the reader does not lose sight of the relevance
34
that these findings have for the present investigation.
First there is the finding that test anxiety can be re
liably measured using the Test Anxiety Scale for Children
(TASC). Second, early studies indicate that test anxiety
impairs the performance of students (Sarason et al.,
1964). Additionally, some of the more recent research
suggests that test anxiety hinders performance due to the
fact that it interferes with cognitive processing. More
specifically, the worry component of anxiety tends to
interrupt the processing information. The high test
anxious child who worries about given performances tends
to divide his attention between the task at hand and task
irrelevant behavior. Thus, in evaluating scores on given
tests, instructors need to be aware of the anxiety level
of their students. It must be borne in mind that high
anxiety levels can present an inaccurate picture of a
child's ability, one which often masks real potential.
Theory concerning the etiology of test anxiety
suggests that it begins quite early in parent-child rela
tions (Sarason, I960). The theory further postulates
that, due to a combination of unreasonable parental stan
dards and harsh criticism, the child tends to develop a
strong dependency on the parents for evaluative support.
This need is often transferred to the teacher. This tends
to occur as soon as the child enters school. The teacher
35
thereafter plays an important role in determining whether
test anxiety is reinforced or reduced (Hill, 1972;
Phillips, 1980). Test anxious students usually enter
school with both a fear of failure in evaluative situa
tions as well as a strong desire to avoid criticism (Hill,
1972), The teacher is often in a position to alter this
attitude. As teachers are the ones who shape the class
room climate, researchers have begun to study which type
of atmosphere most positively affects high test anxious
students (Hill, 1972; Phillips, 1968). Additionally, it
has been suggested that test anxiety may be subject-
specific and the subject may vary across cultures. There
are many cultural and personality factors that may account
for such a discrepancy. Unfortunately, researchers are
only just beginning to uncover the many variables that
influence test anxiety levels. Particularly relevant to
this inquiry have been the cross cultural examinations of
test anxiety. Recent studies indicate that minority
children in the United States typically exhibit more test
anxiety than do Anglos (Phillips, 1962). Moreover it has
been found that both Mexican children in Mexico as well as
Mexican Americans residing in the United States report
greater levels of test anxiety than do native Anglos
(Diaz-Guerrero, 1976; Schmidt & Gallessick, 1971). Pre
cisely what factors contribute to this difference is not
36
as yet known. However, in light of the large influx of
Mexicans to this country, as well as the low performance
scores often reported for this population, it is a most
important question to ask.
Hvpotheses and Research Questions
In light of the preceding literature review, eight
hypotheses and/or research questions were tested for veri
fication in this study. The relationships that were
examined are best represented in Figure 1. Numbers (#)
refer to the specific hypothesis that examines that par
ticular correlation.
1. Immigrants who have resided in the United
States less than two years will have more test anxiety
than students who have lived in the United States longer
than two years.
Rationale : A few studies have been conducted to
compare the TASC scores of Mexican and American children.
The results suggest that Mexican children are much more
anxious about taking tests than American children (Diaz-
Guerrero, 1976; Holtzman, Diaz-Guerrero, & Swartz, 1975).
In comparing the mean scores between these two groups,
researchers attributed some amount of difference to the
problems of equivocally translating the TASC scale from
English to Spanish. However, the mean difference was to
great (Americans-9.8; Mexicans-16.8) to be explained sole-
+->
■ c
< D
E
^ O)
■M >
f O c u
o
I—
= # =
CO
CU >5
1 — -M
c u
^ • ! —
-M X
ftJ C
z: < = c
4 - J
C7> C
c cu
•r- E
~o cu
rO >
C U ( U
C d - 1 -
JC
o
/ j w
+ -> < = c
C O
oj=m/
> 5
CD +J
c c u
"5 X
CO C
cu < = C
q ;
37
A<
C d
I—
= # =
c u
4 — ) ■ I —
CO X
.Ï5
CO
CO
CO
O c u
O * 1 —
J d X
o c
( / ) * = c
Q
CO
O
ZD -a
fO X
CO
c u
CO
c u
o
O-
>>
o
fO
c u
CO
c u
Od
c u
-a
c u
c
fO
X
f O
c u
s-
o
C_)
c u
CD
oo CD C d
3 8
ly by this factor. Instead the researchers concluded that
there was a fundamental difference in the conscious way
school experiences were viewed by these two cultures.
Further insight into these cross cultural differences
evolved when Defensiveness (DSC) scores were compared for
both cultures. Again, Mexican children had a slightly but
significantly higher mean score than the Americans, This
indicated that the Mexican children may have even more
anxiety than they actually admit to on the TASC, which in
turn would broaden the mean score difference between these
two groups.
Holtzman, Diaz-Guerrero, and Swartz (1975) con
ducted a six-year longitudinal study on personality devel
opment in these two cultures. Based on their findings,
they concluded that a greater amount of test anxiety was
found in Mexicans, due largely to their view of school
experiences. In particular, Mexican children's coping
strategies in evaluative situations were completely dif
ferent than those of American children. Whereas Mexican
children generally had a passive-obedient coping style,
American children tended to have a more active-independent
style. While giving the children a series of cognitive
and personality tests, the researchers discovered certain
test-taking behaviors common to each group. Specifically,
Mexican children were cooperative, cautious, seldom took
39
the initiative, and tried to please the examiner. Ameri
can children more often saw these tests as a challenge to
be mastered and an opportunity to show how much they could
do. Hence, the researchers claimed that this former type
of passive coping style of the Mexican children leads to a
conforming behavior in evaluative situations which can
only serve to heighten anxiety.
If the greater amount of text anxiety occurring in
Mexican children is partly due to their distinct school
experiences and learned coping strategies, immigrating to
the United States should eventually decrease test anxiety.
Although at first the culture shock they experience may
increase their anxiety, their gradual assimilation to
American schools should eventually alter their school
experiences and hence their coping styles in evaluative
situations. As the number of years they reside in the
United States increases, their test anxiety should
decrease respectfully to a level more similar to that of
American students.
2. Mexican children in Mexico will have greater
test anxiety than Anglo children and/or immigrants who
have resided in the United States more than two years
Rationale ! In previous studies (Diaz-Guerrero, 1976;
Holtzman, Diaz-Guerrero, & Swartz, 1975) Mexican children
living either in Mexico or the United States scored higher
40
than did Americans on Test Anxiety Scales, Research has
suggested that both parent-child relationships and school
environment contribute to a child's test anxiety level
(Hill, 1972; Holtzman, Diaz-Guerrero, & Swartz, 1975;
Phillips, 1968; Sarason, I960). Specifically, test
anxiety often evolves as a result of strict parental
standards. Consequently a child develops a strong depen
dency on parental support in evaluative situations. These
conditions are typical of Mexican children's upbringing,
who are customarily reared in an environment surrounded by
their extended family and strong family traditions. Hence
their coping styles in stressful evaluative situations are
often based on passive obedience and a desire to please.
This appears to heighten anxiety (Holtzman, Diaz-Guerrero,
& Swartz, 1975). The Mexican school system tends to rein
force these patterns by encouraging a traditional, strict
school environment with high standards. Additionally, the
importance of individual differences is undermined as
reflected by the government's requirement of a standard
curriculum taught only with textbooks it publishes (Strom
& Johnson, 1981).
On the contrary, American children are encouraged
both at home and in the schools to acquire a more active
coping style that focuses both on initiating selfdirection
and developing independence in evaluative situations.
41
This generates a low test anxiety level. Thus, Mexican
children who immigrate to the United States will find a
very different set of family values among Americans and
will be forced to function in an unfamiliar school setting
with a new set of expectations. At first this task of
acculturation may heighten anxiety, but eventually this
assimilation to American standards may serve to reduce the
children's test anxiety level as it has similarly done
with American children. Thus, it seems reasonable to
hypothesize that Mexican children participating in Ameri
can schools and perhaps acquiring some American values
will be more likely to have lower test anxiety scores,
similar to Americans, than would Mexican children living
in a Mexican society.
3. Females in each country will score signif
icantly higher on the TASC than males in their respective
country.
4. Males in both countries will score signif
icantly higher on the Defensiveness (DSC) Scale.
5. Fifth grade students in each country will
score significantly higher on the TASC than second grade
students.
The above three hypotheses were tested in previous
studies (Diaz-Guerrero, 1976; Dusek, 1980; Phillips, 1962;
Sarason, I960; Schmidt & Gallessick, 1971). However,
42
replication in the present study will serve to confirm
these results.
These three hypotheses were derived from the
research that has established that anxiety is stable until
the later elementary years. Moreover, females in all
cultures tend to admit more to anxiety than males. This
becomes apparent because males score higher on the defen
siveness scale indicating that they have difficulty
admitting to anxiety.
These sex differences in TASC scores can best be
explained by examining sex role development. Girls in
both Mexican and American cultures have traditionally been
encouraged and allowed to admit to anxiety. It is re
ferred to as a feminine trait. Boys are taught not do
display feminine traits and thus may report less anxiety
than they actually feel. Hence, the reported sex differ
ences on the TASC may not represent real differences in
anxiety level but rather defensiveness among boys to admit
such anxiety. Due to these suspected tendencies, Sarason
and his colleagues (1980) developed the Defensiveness
Scale (DSC) to administer concurrently with the TASC.
Boys have consistently scored higher on these latter
scales indicating that while they may suffer equally from
test anxiety they are more defensive about admitting to
it.
.4 3-
There is substantial reason to believe that the
failure of boys to admit to anxiety is even stronger among
Mexican children. When comparing TASC scores of Mexican
children these sex differences are accentuated. Again
this can be explained by their development of sex role
identification. Unlike the United States where the
women’s liberation movement has successfully encouraged
the disappearance of many sex role traits, in Mexican
culture their still exists very defined sexual roles.
Holtzraan, Diaz-Guerrero, & Swartz (1975) administered a
Human Figure Drawing Scale to both Mexican and American
children. Mexican children had more defined differences
distinguishing each sex than did American children. This
reinforces the suspicion that Mexican culture tends to
emphasize sexual differences, while American culture is
attempting to do the opposite. Consequently, scores on
the Defensive Scale (DSC) between American and Mexican
subjects have been compared and Mexican children have
consistently had a slight but significantly higher score
than Americans (Diaz-Guerrero, 1976). This score differ
ence increases when only the male population are examined.
The present study will most likely find a similar differ
ence between Mexican children in Mexico and Mexican immi
grants due to the affects of assimilation to American
culture of the immigrant children. A larger difference
' 44
can be expected especially in light of the fact that sex
differences in America have decreased since the afore
mentioned studies were conducted.
6. Across the three populations there will be a
significantly greater amount of math test anxiety than
reading test anxiety.
Rationale ; Three authors (Aiken, 1976; Richardson
& Woolfolk, 1980) have suggested that a fear of math or
difficulty in this subject area is more commonly accepted
than it is for reading. Although there exists many poor
readers who thus have anxiety concerning reading it is not
common practice simply to eliminate reading from a stu
dent’s curriculum. However, students who find difficulty
in math often rectify the problem by taking fewer math
classes or only the minimum requirements. Thus, admitting
to math anxiety should be more frequent than admitting to
reading anxiety. This investigation will add to the liter
ature by providing support for this notion.
7A. There will be a significant relationship
between math achievement and math test anxiety and reading
achievement and reading test anxiety.
Rationale ; Most prior studies relating attitude
and achievement found a significant but low correlation
between these two variables. However, researchers failed
to control for many factors which could account for such a
45 '
low correlation. Foremost, amount of math preparation was
hardly if ever considered (Kaczala, 1980). In previous
studies high school and college students were often util
ized. These pupils had the opportunity to select most of
their classes. Thus, some had few math classes while
others had more. The problem in these studies was that
while investigating anxiety and achievement the research
ers failed to distinguish between students who had few
math classes and those who were math majors. The present
study is unique in that it includes an elementary popu
lation which has a predetermined curriculum and students
do not have the opportunity to select classes. Therefore,
the academic preparation they receive should be largely
equivalent. In addition, math and reading achievement
have not been differentiated in past studies.
7B. Are achievement and anxiety correlations
within areas (i.e. reading, math) greater than corre
lations between areas?
Rationale : It is predicted that the strongest
correlations will exist within each area. That is, the
highest correlations will be between reading test anxiety/-
reading achievement and math test anxiety/math achieve
ment. Although, general test anxiety may affect perfor
mance in all subject areas, if a student acquires anxiety
or fear in a specific subject area such as math, this
4 6
anxiety will affect math performance greater than it will
affect reading performance. Similarly, general school
anxiety will indeed influence general test anxiety and
specific subject anxiety. However, these correlations
will not be as strong as the previously designated
relationships.
8. Are there differences in the amount of measure
ment error within these populations?
Rationale ; This question is important in terms of
determining the appropriateness of the TASC and DSC among
the three populations. Although previous studies have
confirmed that there are no cross-cultural differences on
the TASC and DSC and that it is stable over time, nobody
has evaluated the test using the Mexican immigrant popula
tion (Holtzman, Diaz-Guerrero, & Swartz, 1975). A posi
tive response to this research question would not only
confirm the use of these measurement tools with the Mexi
can and American population but would validate its appro
priateness for use with Mexican immigrants.
47
CHAPTER III
PROCEDURES
Research Design
The research design of this study is unique. It
is one of the first attempts to compare levels of test
anxiety among Mexican children in Mexico, Mexican immi
grants in the United States, and Anglo children. The
purpose of this study was to determine which population
had a higher level of test anxiety and to investigate the
subject specificity (i.e. math vs. reading) of test
anxiety among these populations. Specifically, the inde
pendent variables examined were: sex, grade (second and
fifth), number of years immigrant children resided in the
United States and achievement in math and reading. The
dependent variable was test anxiety. The type of research
was correlational ; the relationship between each indepen
dent variable and the dependent variable was determined.
Moreover, correlational methods were used to distinguish
among reading test anxiety, math test anxiety, general
school anxiety, and general test anxiety.
48
Description of Subjects
A total of 755 students participated in this
research project. There were 250 students who lived in
Mexico, 265 were Mexican immigrants attending public
schools in Los Angeles, California, and 240 were Anglo
Saxons also residing in Los Angeles. Within each popu
lation approximately half of the students were in second
grade and half were in fifth grade.
Students in Mexico were from both rural and
urban areas in and surrounding Puerto Vallarta, Mexico.
In the city of Puerto Vallarta, two schools were utilized.
The first. Club de Leones, was in the heart of the city
and held classes from 2:00 p.m. until 7:00 p.m. Both
upper-lower class and lower middle class students attended
this school. The second school was located in a middle
class colony one mile north of the city. The school,
Isidro Castillo, was also in session in the afternoon.
Students in both schools came from families that owned
small businesses or whose parents worked for resort
hotels. The last school was located 20 miles north of the
city in a pueblo named Valle de Banderas. The school, 20
de Noviembre, served mostly children of farmers and agri
cultural workers. Children attended school from 8:00 a.m.
until 2:00 p.m., with no lunch break.
49
The second population was made up of students
who had immigrated to the United States from Mexico. They
had lived in this country for differing lengths of time.
They were from upper-lower class or lower-middle class
homes. All spoke Spanish as a first language, but dif
fered in their English proficiency. Students in this
population attended Hoover St. Elementary School. It was
located in the Wilshire Corridor area of Los Angeles. It
was a large overcrowded school with over 2,000 students.
Classes were in session all year round. Students attended
school for eight weeks, then had four weeks vacation.
School hours were approximately 9:00 a.m. to 3:15 p.m.,
varying according to grade level.
The final population was Anglo-Saxon children
born in the United States. Due to the large amount of
integration in Los Angeles among minority and white
9
students, and the tendency for many white children to
attend private schools, it was difficult to find one pre
dominately white school with similar SES background. Thus
students came from a variety of schools: Dixie Canyon
Avenue Elementary, Canterbury Avenue, and Renter Canyon.
Children were mainly from middle-class homes with a small
number from upper-lower class families. Approval to con
duct the study was obtained from the Research and Evalu-
50
ation Branch of the Los Angeles Unified School District
(Appendix A).
Description of Instruments
The Text Anxiety Scale for Children (TASC)
(Appendix B) was used to measure both children’s general
school anxiety and general test anxiety. Directions for
the TASC and answer sheets used appear in Appendixes D and
E. The Spanish edition, translated by R. Diaz-Guerrero
(1976) was administered to all students in Mexico and all
students in the United States whose dominant language was
Spanish. The English version was given to all students
who were most fluent in English. Sarason’s (I960) TASC
contains 30 items which are read orally by the examiner.
Children respond by circling yes or no on their answer
sheets. The number of yes responses is the child’s total
anxiety score. This measurement combined with the DSC
provides a reliable and valid measure of general school
anxiety and test anxiety.
The first 18 questions on the TASC measure
general school anxietv and do not use the word ’ ’test.”
Many items refer to things school children worry about,
i.e., passing to the next grade, falling behind when
missing school and inability to answer teacher-generated
questions. Two items specifically relate to reading and
two to math:
51
3. When the teacher asks you to get up in front of
the class and read aloud, are you afraid that you
are going to make some bad mistakes?
4. When the teacher says that she is going to call
upon some boys and girls in the class to do math
problems, do you hope that she will call upon
someone else and not on you?
7. When the teacher is teaching you about math, do
you feel that other children in the class under
stand her better than you?
10. When the teacher is teaching you about reading, do
you feel that other children in class understand
her better than you?
The above items measure general reading anxietv
and general math anxietv.
The second part of the scale measures general
test anxietv. After question 18 the examiner explains
what the word "test” means. It refers to "anytime the
teacher wants to find out how much you know" (Sarason,
I960). It can be oral, or written on paper, or on the
blackboard. Sarason had originally supplied 12 questions
that specifically spoke of test anxiety before, during,
and after tests.
The present researcher developed and integrated
eight questions into this latter section of the TASC.
52
These items would differentiate between math test anxietv
and reading test anxietv. Six of the questions were the
same questions Sarason had written for the general test
anxiety section. However, the word math and/or reading
was added to make it subject specific. For example,
Sarason asked: "Are you afraid of school tests?" The
present investigation additionally asked: "Are you afraid
of reading tests? Are you afraid of math tests? Two
others questions were reworded for this purpose:
24. Do you worry a lot before you take a reading test?
37. When you are taking a reading test, do you forget
some things you knew very well before you started
taking the test?
Two additional questions were created by the present
researcher to determine whether math or reading tests
generated more test anxiety:
26. Do math tests make you more afraid then reading
tests?
34. Do reading tests make you more afraid than math
tests?
The additional 8 questions brought the total
number of TASC questions to 38. The eight questions added
were randomly interspersed into the second section of the
scale. The entire test is given in Appendix B.
53
The defensiveness Scale for Children (DSC)
(Appendix C) contains 27 items concerned with inter
personal relationships in areas other than anxiety. This
scale was created to measure the tendency to deny negative
feelings, such as anxiety, hostility, and inadequacy.
Similarly to the TASC, questions are read orally by the
examiner and the children answer by circling yes or no on
their answer sheets. The number of no responses is the
child’s total defensiveness score. The items ask ques
tions that children universally will respond to similarly.
For example: " Do you like to go on trips with your
parents?" "If someone doesn’t like you, does it bother
you?" "Do you like to play in the snow?" Children gener
ally will answer yes to these questions unless they are
displaying some negative feelings by answering no. In
this case they would be considered defensive. Correla
tions between DSC and TASC scores range from -.34 to -.56
and are highly significant for both sexes and all ages
(Hill, 1966; Sarason, 1972).
Three parts of the Comprehensive Tests of Basic
Skills (CTBS) were given to all children in the language
in which they were most fluent. The reading vocabulary,
math computation, and math application sections were
administered. Directions in the teacher’s manual which
accompanied the test were followed. The text systemat
54
ically measures those skills prerequisite to studying and
learning in subject matter courses. The tests are in
tended for national use by students who have been taught
according to various approaches. They measure skills
common to all curricula.
In the second grade test each question is read
aloud by the examiner. In the fifth grade test the ini
tial directions for each section are read aloud and then
students work independently for a specified amount of
time.
In both the second and fifth grade Reading
Vocabularv test, each item contains a stem phrase (this is
orally read in second grade) and four discrete words for
alternatives are given in the test booklet. The student
needs to select the word that means the same or about the
same as the underlined word in the phrase. There are 33
items in the second grade test and 40 in the fifth grade
test. Fifth grade students are allotted 14 minutes to
complete the section.
The Math Computation section measures skills in
the fundamental mathematical operations. In the second
grade test there are 10 items in each of three sections;
multiplication, subtraction, and addition. Children have
12 minutes to complete each section. In the fifth grade
test there are 12 items in each of these three sections.
55
plus a fourth section with 12 division problems. A total
of 48 problems appear in this section with an allotted
time of 40 minutes to solve the problems.
The final section Math Concepts and Applications
was designed to measure the student’s ability to recognize
concepts, choose appropriate problem-solving operations,
and carry out such operations. There were 25 items on the
second grade test that dealt with money, time, and simple
word problems. There were 50 items on the fifth grade
test with 35 minutes to complete. Since the second and
fifth grade tests were different, they were standardized
by converting the results to standard scores.
Procedure
Second and fifth grade students both in Puerto
Vallarta, Mexico, and Los Angeles, California, partic
ipated in this study. In Puerto Vallarta three schools
were selected to participate, two located in the city and
one from a surrounding village, Valle de Banderas. This
was to insure that the sample of immigrant students
matched those in Los Angeles who originally came from both
small villages and larger cities. In each school there
was only one second grade and one fifth grade room.
Hence, both classes in all the schools were automatically
selected for use in this study. Classes ranged in size
from 40 to 80 students. In Los Angeles, four schools were
56
selected that were matched for SES and where the prin
cipals and teachers were willing to cooperate. Classrooms
were selected by drawing numbers from a hat. Class size
ranged from 25 to 35 pupils and thus twice the number of
classes were utilized in Los Angeles than in Mexico. One
of the four schools contained the Mexican immigrant popu
lation, while the Anglo-Saxons were taken from three
different schools.
Both in Mexico and the United States the admin
istration of the TASC and other scales was identical. The
examiner administered the scales in each classroom. The
TASC, DSC, and CTBS were given as a group test within each
room. The researcher read the directions suggested by
Sarason (I960) and the Teacher’s Manual of the CTBS in
either Spanish or English, depending on the primary lan
guage of the subjects. In classrooms where there were
some children who spoke Spanish and others who spoke
English, the class was divided into two groups by their
dominant language. Each group was given the test sepa
rately in the appropriate language.
Statistical Analvsis of Data
In the following paragraphs, the statistical
procedures used for evaluating each hypothesis are
described.
57
t Test Comparison
In order to evaluate the differences between new
immigrants and the Hispanic control group (immigrants who
had resided in the United States longer than two years) a
t test was calculated. The means for each group were
compared for all achievement variables and all anxiety
factors. The t values and probability levels were
evaluated to determine if the differences between groups
were significant.
Analvsis of Covariance
Analysis of covariance was utilized in deter
mining the effects of race, sex, and grade on the
achievement and anxiety variables. Defensiveness was
controlled for as recommended by previous researchers.
The means presented in this study were adjusted for
defensiveness. Main effects and interactions were
examined.
Partial Correlation
Partial correlation coefficients were evaluated
to determine the anxiety and achievement correlations
between and within areas, i.e., math, reading. The follow
ing variables were partialled: sex, defensiveness, grade,
and race. The correlations were calculated for the com
bined total papulation. Correlations were considered
significant if the probability was less than .05.
58
Paired t test
In order to determine whether students felt more
math test anxiety or more reading test anxiety a paired t
test was performed. Means were compared for the entire
population. The t value and probability were recorded to
determine the significance of the findings.
Reliabilitv
The coefficient alpha for each anxiety subscale
was calculated to determine the appropriateness of each
scale for the sample populations. Reliabilities for both
the entire population and each individual ethnic group
were derived. If alpha was greater than .60, the scale
was considered appropriate for use with the population
being examined.
59
CHAPTER IV
RESULTS
In this chapter the results of this study are
reported. They are interpreted primarily within the
framework of the measurement instruments and in the con
text of the research hypotheses. Discussion concerning
the relationship of these findings to previous research is
examined in Chapter V. Results are summarized by tables
and specific hypotheses.
New Immigrants and Hispanic
Controls— Anxietv Level
Table 1 compares the mean scores between Hispanic
controls (students who lived in the United States more
than two years) and new immigrants (students who lived in
the United States less than two years). The means for
each measurement utilized are reported. The results of
the i s , test indicate a significant difference (p<.05)
between the two groups for these variables: school
anxiety, math anxiety, test anxiety, math test anxiety,
and defensiveness. In each of these cases new immigrants
displayed a greater amount of anxiety, but less overall
60
Table 1
t test Comparison of School Anxiety Between
Recent Immigrants and Hispanic Controls
Variable
Mean
New**^
Immigrants
Hispanic
Controls t p level
Reading
Achievement 15.78 17.75 -1.44 .152
Math
Computation
Achievement
19.19
20.66 - .76 .445
Math
Application
Achievement 17.12
17.97 - .53 .599
School
Anxiety 7.52 5.85 3.01
.003*
Math
Anxiety 1.64 1.26
1.97
.050*
Reading
Anxiety 1.64 1.38 1.28
.203
Test
Anxiety 7.64 6.04
2.29 .023*
Math Test
Anxiety 2.08 1.49 1.96 .052*
Reading Test
Anxiety 1.52
1.37 .57 5.69
Defensiveness 9.68 11.36
-1 .93
0.54*
Note: For new immigrants N=2S-27 depending on the
variable. For Hispanic controls, N ranged from
211 to 218 depending on the variable.
The group "New immigrants" was comprised of students
who lived in the United States two years or less.
*E<.05
61
defensiveness. In the other anxiety variables, the dif
ferences between groups was not significant at the .05
level; however, there was a predominant trend suggesting
that new immigrants exhibited both greater school anxiety
and greater test anxiety. Similarly, their achievement
level lagged behind that of the Hispanic control group.
These results lend support to hypothesis one, suggesting a
negative correlation between the number of years a child
lives in the United States and his or her other score on
the stated subscales of the TASC,
ANCOVA Results
In the results reported in this chapter, analysis
of covariance was used to determine the effects of race,
sex, and grade on the achievement and anxiety variables.
Defensiveness was controlled for as recommended by pre
vious researchers. Interactions were examined.
Race Differences
Table 2 compares the adjusted group means among
the three ethnic groups in the present study. Clearly,
the results indicate a significant (£.01) difference in
achievement due to race. This held true for all achieve
ment measures: reading achievement, math computation, and
math application. Consistently. Mexican children per
formed poorest on all three measures, followed by the
immigrant students. Anglo children appeared to have the
Table 2
62
Race Differences— Comparison of Adjusted
Group Means
Mexican
Immi
grants Anglos F p level
Reading
Achievement 46.16 47.42 56.87 89.48 .001*
Math Computation
Achievement
44.53 48.76 52.36 36.34 .001*
Math Applications
Achievement 46.02 48.92
55.43 53.53
.001*
School
Anxiety 8.10 6.16 4.45 132.11 .001*
Math
Anxiety 1.64
1.31 .94 29.67 .001*
Reading
Anxiety 1 .83 1.44
.91
57.40 .001*
Test
Anxiety 7.70 6.55 , 4.55 82.79
.001*
Math Test
Anxiety
2.29 1.65 1.18 41 .27 .001»
Ready Test
Anxiety 1.60
1.47
.72 33.75 .001*
Defensiveness 12.00 11.18 7.85 87.44 .001*
Note: For Mexicans N= 238
able.
For Immigrants N =
variable.
For Anglos N = 224-
-261 depending on the vari-
199-212 depending on the
235 depending on the variable.
*£<.05
6 3
greatest achievement level. Similarly, a significant
difference in both school and test anxiety level due to
race was disclosed. On all subscales of the TASC, Mexican
children expressed the largest amount of anxiety, immi
grant students to the next greatest level, while Anglos
were least anxious. Moreover, Mexican children were the
most defensive, followed by the immigrant pupils. Anglos
showed the least amount of defensiveness.
These results clearly supported hypothesis two
which stated that Mexican children would score higher than
either of the other two ethnic groups on the TASC.
Sex Differences
Table 3 compares the adjusted group means for all
variables between males and females in this study. Males
scored higher on all three achievement measures but this
difference was not statistically significant. On all
subscales of the TASC, males reported less anxiety than
did females. These differences again were generally not
statistically significant except on the test anxiety scale
and the math text anxiety scale. Also, males demonstrated
a greater level of defensiveness than did females.
In support of hypothesis three, females did in
fact score higher on the TASC than did males. However,
these differences in scores were generally not statis
tically significant. Results confirmed hypothesis four.
64
Table 3
Sex Differences— Comparison
Group Means
of Adjusted
Males Females F p level
Reading
Achievement 50.39 49.77 .93 .33
Math Computation
Achievement 48.85 47.91 1.76 .19
Math Applications
Achievement 50.65 49.44 2.96 .08
School
Anxiety
6.17 6.47 3.09
.08
Math
Anxiety 1.25 1.38 3.35 .07
Reading
Anxiety 1.35 1.47
3.34
.07
Test
Anxiety 6.00
6.67 12.67
.001%
Math Test
Anxiety 1.58
1.91
12.12 .001*
Reading Test
Anxiety 1.24 1.30
.49
.48
Defensiveness 10.64
10.09 3.97 .05*
Note: For males N
For females
able.
= 349-379 depending on the variable.
N = 312-329 depending on the vari-
* £<.05
65
Males scored significantly higher on the Defensiveness
Scale than did females.
Grade Differences
Table 4 compares the adjusted group means for the
variables under study between second and fifth grade stu
dents. Differences in anxiety level due to grade effects
were evident. On all subscales of the TASC fifth grade
students indicated a higher level of anxiety than did
second grade students. These differences were statis
tically significant. However, the reverse was true for
defensiveness. Second grade pupils implied greater defen
siveness than did fifth grade pupils. These differences
were statistically significant. Evidence confirms hypoth
esis five that fifth grade students would score higher on
the TASC than second grade students.
Interaction Results
While the hypotheses in general did not center on
interactions in the present investigation, they were exam
ined. The only interaction that qualified a hypothesis
was the interaction of race and sex on reading test
anxiety. The results of this interaction are reported in
Table 5.
66
Table 4
Grade Differences— Comparison
Group Means
of Adjusted
2nd grade 5th grade
f P
level
School
Anxiety 6.06 6.54
6.03
.01*
Math
Anxiety 1.12 1.49 21 .40 .001*
Reading
Anxiety 1.19 1.62 35.45 .001*
Test
Anxiety 5.26 7.31 92.76 .001*
Math Test
Anxiety 1.45 2.00 27.06 .001*
Reading Test
Anxiety 1.16 1.38
4.83 .03*
Defensiveness 12.45 8.40 217.67
.001*
Note: For 2nd grade N = 319-346 depending on the
variable.
For 5th grade N = 342-362 depending on the
variable.
*£<.05
6 7
Table 5
Interaction of Sex and Race on Reading
Test Anxiety
Mexicans Immigrants Anglos
Males 1.63 (131) 120 (128) 0.83 (120)
Sex
Females 1.35 (130) 1.65 ( 84) 1.03 (115)
Note : Numbers in ( ) represent N.
Reading Test Anxiety vs. Math Test Anxiety
Hypothesis sex was confirmed. Across the three
populations there was a significantly greater amount of
math test anxiety than reading test anxiety. A paired t
test indicated that the mean for math test anxiety was
1.73 while for reading test anxiety the mean was 1.27,
t(713), =8.01, £<.001.
Anxiety and Achievement Correlations
Between and Within Areas
In the next section partial correlation coeffi
cients were evaluated to determine the anxiety and achieve
ment correlations between and within areas, i.e., reading,
math. These variables were partialled: race, sex, grade,
. 68
and defensiveness. The correlations were calculated for
the combined total population.
The partial correlation represented in Figure 2
confirms hypothesis 7A. A significant relationship
between math achievement and math test anxiety and reading
achievement and reading test anxiety did exist in the
present study. The relationship was negative suggesting
that as test anxiety increases achievement decreases.
Furthermore, significant positive correlations were indi
cated between reading test anxiety and math test anxiety.
Small and insignificant correlations were found between
reading test anxiety and math achievement and math test
anxiety and reading achievement.
Figure 3 represents anxiety correlations between
and within subject areas. The highest correlations gener
ally existed within subject areas. Both reading test
anxiety and math test anxiety had high correlations to
general test anxiety. General school anxiety was strongly
related to general math anxiety, general reading anxiety,
math test anxiety, and reading test anxiety. Furthermore,
general math anxiety and general reading anxiety were
highly correlated. More moderate relationships existed
between general math anxiety and math test anxiety, gen
eral reading anxiety and reading test anxiety, and math
test anxiety and reading test anxiety. The lowest corre-
.21*
Math Text
Anx i ety
Reading Test
Anxi ety
- . 08*
Math
Achievement
Readi ng
Achievement
,48*
Note, N=634. Correlations reported for Math Achievement
were derived solely from the Math Applica
tions Measure.
Achievement and Anxiety Correlations
Within and Between Areas
Figure
70
32
General
Math
Anxi ety
. 27
Math Test
-> Anxiety
^ General ^
School
. 30
Anxiety
. 32*
.33*
.36*
General
Reading
Anxiety
.2 2*
Reading
Test
Anxiety .21*
.46*
General
Test
Anxi ety
Figure 3. Anxiety Correlations Between and Within
Areas
71
lations existed between subject areas. General math
anxiety was slightly correlated to reading test anxiety,
while similar relationships existed between general read
ing anxiety and math test anxiety. In reply to research
question 7B anxiety and achievement correlations were
greatest within subject areas In the present study.
Measurement Error In the TASC
Research question eight asked If there were differ
ences In the amount of measurement error within the three
populations. Table 6 provides a summary of the reliabil
ity coefficients of each subscale for each ethnic group as
well as the total population. Using ,60 or greater as an
Index of good reliability, the following results are
reported. Among all three populations and the total popu
lation, the reading anxiety scale and the math anxiety
scale were not reliable scales (alpha <,60), This Is
probably because of the small number of Items on these
scales. With the exception of the reading test anxiety
scale for Mexicans (alpha=.50), all other scales across
all three ethnicities demonstrated significant reliabil
ity. They were thus appropriate for each group utilized
In the present study. Differences In the amount of
measurement error within these populations were small.
Table 6
Reliability of the TASC and DSC for
Ethnic Population and the Total
Population
72
Each
Mexicans Immigrants Anglos
Total
Population
School
Anxiety .67 .65 .62 .70
Reading
Anxiety .28 .31 .42 .36
Math
Anxiety .32 .23 .43
.36
Test
Anxiety .84 .80 .80 .82
Math Text
Anxiety .69 .73 .73
.72
Reading Test
Anxiety .50 .61 . 66 .59
Defensiveness
(DSC) .76 .73 .70 .76
73
CHAPTER V
DISCUSSION, CONCLUSIONS, AND RECOMMENDATIONS
The results of this study generally confirm the
hypotheses. In this final chapter these findings are
discussed both in terms of the present research hypotheses
and prior research studies. The significance of these
outcomes to test anxiety research and the field of educa
tional psychology are also considered. Finally and per
haps most importantly, the numerous possibilities for
future research that have evolved from the present results
are discussed.
Anxietv Level of New Immigrants vs.
Hispanic Controls
Hypothesis one predicted a negative correlation
between the number of years a child lived in the United
States and his or her TASC score. In analyzing the
results, immigrants were grouped together as follows: new
immigrants consisted of those students who had lived two
years or less in the United States and Hispanic controls
were those who had lived longer than two years in the
United States. A ^ test comparing these two groups
74
indicated a significant difference (&<.05) for all anxiety
variables except reading anxiety and reading test anxiety
(Table 1). The failure to obtain a significant difference
in these two categories was not surprising. Little re
search exists concerning reading anxiety and/or reading
test anxiety. It may be that these constructs do not
exist and that there is no such concept as reading
anxiety. The fact that there was a significant difference
between the groups on all other measures confirms hypoth
esis one. The new immigrants had a significantly greater
amount of text anxiety than the Hispanics who had lived in
the United States a longer period of time.
These results were not shocking in light of past
research. The few studies that have compared American and
Mexican children have indicated that Mexican children are
more anxious about taking tests (Diaz-Guerrero, 1976;
Holtzman, Diaz-Guerrero, & Swartz, 1975). Since the mean
difference between these populations was so great, the
researchers concluded that there was a fundamental differ
ence in the conscious way school experiences were viewed
by these two cultures. Further studies by these same
researchers found that Mexican children's coping strat
egies in evaluation situations were passive-obedient thus
increasing anxiety, while American children had a more
active independent style, thus reducing anxiety. There
75
fore, it stands to reason that new immigrants were still
viewing school and school tests as they did in Mexico and
reacted with the same test-taking behavior. Moreover,
culture shock may increase school anxiety. It is reason
able to assume that as these new immigrants assimilate
into the American school system they will develop values
and behaviors that are similar to American children. The
longer they reside in the United States the greater the
similarity will be.
Although this rationale seems plausible, it has
not been empirically evaluated. The present study is one
of the first to do so. Immigrants have not been used in
comparing anxiety levels between Mexican and American
students. Moreover, whether immigrant students' school
anxiety increases or decreases with length of time in the
United States has not been examined. The present study
adds a new dimension to the school anxiety literature.
Since it is one of the few original studies investigating
anxiety levels of Hispanic immigrants, definite conclu
sions cannot be drawn. In order to provide stronger sup
port, further studies need to be conducted which examine
these issues. In reflecting upon the increasing number of
immigrants attending schools in the United States, in
particular California, the need for such studies becomes
vital.
76
Race Differences on TASC Scores
As expected in hypothesis two, Mexican children in
Mexico scored higher on the TASC than did Anglo children
and immigrants who had resided in the United States longer
than two years. Previous research compared the test
anxiety level of both Mexican children in Mexico to Anglos
in the United States and Hispanic immigrants to Anglo
children in the United States. (Diaz-Guerrero, 1976;
McCormick & Karabinus, 1976; Phillips, 1962, 1968). In
all cases, the Anglo children exhibited less test anxiety
than did the two other groups. Many researchers have
explained these differences as a direct result of the
contrasting parent-child relationships and school environ
ment that is a part of each culture. Inquiries into the
etiology of test anxiety have suggested that this may be a
valid conclusion. Investigators have found that children
whose parents encourage them to rely on their own re
sources for solving tasks, or who offer suggestions for
effectively completing tasks, develop low anxious behav
ior. To the contrary, children who are anxious in test
like situations have usually been subjected to both nega
tive evaluative experiences and unreasonable parental
standards. Furthermore, these children have developed a
strong dependence upon adult support in similar situations
77
(Dusek, 1980; Hermans, ter Laak, & Maes, 1972; Hill, 1972;
Sarason et al., I960).
When contrasting child-rearing practices of both
American and Mexican parents, the results of the present
study were fitting. The fact that Mexican children were
the most test anxious of either population is a reflection
of their strict upbringing. Mexican children are usually
reared in an environment surrounded by their extended
family and strong traditional standards (Holtzman, Diaz-
Guerrero, & Swartz, 1975). Similarly, the school system
may perpetuate anxiety by upholding a strict, authori
tarian environment with high expectations for students.
On the other hand American children often do not live with
an extended family and are encouraged both at home and in
the schools to acquire a more active coping style. Child-
rearing practices focus on initiating both self-direction
and developing independence, rather than achieving paren
tal demands. This can account for their lower anxiety
level found in the present study.
As mentioned earlier, few if any studies have
included immigrant students as a third population when
comparing test anxiety of Mexican and American students.
Studies comparing minority students with Anglo students
have found that Hispanics have a higher anxiety level
(McCormick & Karabinus, 1976; Phillips, 1968). Thus, it
78
was reasonable to assume similar results in the present
study. However, this was one of the first studies that
additionally compared Mexican immigrant students to Mexi
can students in Mexico. Results indicated that immigrants
who had resided in the United States longer than two years
had less school and test anxiety than did Mexicans but
greater anxiety than did Anglo children. In view of the
literature reviewed concerning the etiology of test
anxiety, this finding was also not surprising. After
living in the United States for two years, many of the
values and traditions of immigrant families become lost.
Living with an extended family, which serves to uphold
Mexican traditions, is often not possible. Many family
members are left behind in Mexico. The school systems are
different. Teachers in the United States do not have the
same expectations as teachers in Mexico, nor do they
encourage the same behaviors. As the immigrant family
assimilates to American society, their values begin to
reflect those of the American people. More often their
reasoning becomes a cross between the two cultures. Thus,
their children are not raised in the same fashion as they
would have been had they stayed in Mexico. Their anxiety
level likewise is often smaller than Mexican students
because they are reared in an atmosphere with different
expectations. However, their anxiety level is not as low
79
as that of Anglo students because they are still Mexican
and parents, of course, maintain some of their cultural
practices•
Once more these findings have shed light on a new
facet of the school anxiety literature. In comparing
these three populations— Mexicans, Mexican immigrants, and
Anglo Saxon children— a significant trend has emerged.
Mexican children in Mexico are the most anxious on all
subscales of the Test Anxiety Scale and achieve the least
in reading and math. Mexicans who have immigrated to the
United States are less anxious and correspondingly have
achieved in greater amounts than Mexicans in Mexico.
Anglo children are the least anxious and have achieved the
most. Research as to the causative aspects of this trend
needs to begin. Investigation evaluating whether or not
the family and school environment perpetuates anxiety in
Mexico and in the United States is lacking. Evidence for
specific factors that affect school anxiety and test
anxiety in Mexican children needs to be obtained. Perhaps
it is school anxiety that is reflected in the low achieve
ment scores of immigrant students who attend public
schools in Los Angeles. Only by beginning to answer some
of these research questions will educators find ways to
reduce anxiety and increase performance of these students.
80
Sex Differences on the TASC and DSC
It was forecast in hypothesis three that females
across all populations would score significantly higher on
the TASC than would males. Results indicated a consistent
trend: females admitted to greater anxiety on all sub
scales of the TASC. However, the differences between
their scores were only significant for the Text Anxiety
and Math Test Anxiety subscales.
The fact that females admitted to more anxiety
than males probably is due to sex role development. In
the three populations utilized in this study, girls had
been traditionally encouraged and allowed to admit to
anxiety. It is considered a feminine trait. Boys are
taught not to display feminine traits and thus are more
reluctant in admitting to anxiety. Hence, the reported
sex differences on the TASC cannot be utilized alone to
establish true anxiety differences. In the present study,
the Defensiveness Scale (DSC) was incorporated into the
TASC and additionally administered to all subjects. As
prognosed in hypothesis four, males in both countries were
significantly more defensive in admitting to anxiety than
were females. This reluctance of males to admit to test
anxiety could affect TASC score differences between the
two sexes. Thus, defensiveness was controlled for in the
data analysis of this research project.
81
The failure to obtain significant sex differences
on all subscales of the TASC could be the result of many
confounding factors. First of all the TASC was designed
to evaluate general test anxiety in children. By breaking
it down into several subscales, it was not possible to
compare overall TASC scores but only scores on the indi
vidual subscales. There was a significant difference on
the subscale measuring test anxiety which of course was
the focus of the entire scale. Perhaps if the general
score of the combined subscales was compared between the
sexes, this difference would be greater. Additionally,
the fact that a significant sex difference was reported
for math test anxiety and not for reading test anxiety may
be consistent with the aforementioned notion that reading
anxiety does not exist at a significant level in either
sex. Research concerning reading anxiety as a construct
is lacking. In the present study students reported
greater math anxiety than reading anxiety which reempha
sizes the possibility that reading anxiety may not be a
significant variable.
In summarizing, the present research indicated
that males were significantly more defensive than females.
Females admitted to greater test anxiety on all subscales
of the TASC. Significant score differences between males
and females were found for only text anxiety and math test
82
anxiety. In analyzing this particular study, these vari
ables were the most important to the present researcher,
since the existence of reading test anxiety is not well
documented. In evaluating the effects of sex differences
on the TASC scores, incorporating immigrant students into
the study made it unique. The results were similar to
prior research but somewhat weaker. A similar study
should be repeated to validate these findings.
Grade Differences on the TASC
In comparing the differences between the TASC
scores of second and fifth grade students among the three
populations, the present study supported the findings of
previous research. On the TASC, fifth graders admitted to
a significantly greater amount of anxiety than did second
grade students was less than .001 on all scales except
reading test anxiety where q, was less than .03). The
literature reviewed in this study suggested that this
trend was common (Sarason, Hill, & Zimbardo, 1964). In
the primary grades students generally maintain similar
anxiety levels. However, differences in anxiety-level
begin to appear in the middle elementary years and to
stabilize by fifth or sixth grade. Since this trend was
the same for the combined population, it suggests that
school factors attribute to increased anxiety in the
latter elementary school years. Inconsistent with these
83
findings was the observation in the present study that
second grade students were significantly more defensive
than fifth graders. This may mean that the second grade
students may have a higher level of anxiety than they
admitted to because of their defensive nature. But more
likely than not the greater defensiveness of second grade
students may be due to a possible reluctance to admit
their feelings to teachers and adult figures who are
authoritarian. Fifth grade students are less fearful of
adults and school experiences, and are less threatened by
teacher figures, whereas the mere presence of a teacher's
unhappy expression may be enough to alter a second grade
child's behavior. Also, younger children may worry more
about seeking adult approval or having the right answer
and may choose their answers with these criteria on the
DSC. A final possibility is that younger children may not
understand or be able to identify their true feelings.
In any case the fact remains that younger children
have admitted to less anxiety than the older children in
this research study. The use of the three populations in
this investigation supported previous studies for Mexican
and Anglo children and suggested that similar findings may
be true when Hispanic immigrants are included in the
evaluation.
84
Reading Text Anxietv vs.
Math Text Anxietv
In the present study, students across the three
populations admitted to a significantly greater amount of
math test anxiety than reading test anxiety. Reasons why
this may be true are lacking in the present literature.
Also missing are studies examining reading test anxiety.
Therefore, it is difficult to determine why students in
the present study indicated that they felt more test
anxious in math than in reading. In view of the liter
ature on math attitudes a few possible explanations exist.
Foremost, popular belief suggests that math is a subject
for which all people do not have a talent. Thus, it is
common for a person to have a fear of math, especially if
they believe it is difficult for them. However, in read
ing the discrepancy between poor and good readers is less
defined. In contrast to math, there are not as many right
or wrong answers in reading. Unlike reading, the very
nature of math, which involves trying to solve computa
tional problems, can simulate a test-like situation and
generate anxiety in students.
More recently, interest in sex differences in math
achievement has emerged. Researchers have found that
there are attitude differences concerning math between
girls and boys. Specifically, by junior high boys appear
85
to have greater achievement in mathematics than do girls.
Additionally, boys often take more math classes than do
girls. Whether or not there is a biological difference
between men or women that contributes to the differing
math attitudes or whether it is a result of cultural expec
tations is not certain (Aiken, 1976). In any case there
exists a significant difference in math attitudes between
the sexes that has not been verified for reading atti
tudes. Clearly, the unfavorable attitudes of females
toward math can possibly attribute to the results in the
present study that indicate the three populations have
greater math test anxiety than reading test anxiety.
The present study has made great progress in
attributing to knowledge of math test anxiety. Its design
corrected a major flaw found in prior studies. Research
ers who had measured math anxiety and achievement did not
control for the number of math classes subjects took. In
this study, since subjects were in elementary school and
students do not select classes, math preparation of all
students was equivalent. Thus, the results attained can
be seen in a more valid light. Additionally, the combina
tion of all three populations denotes cross cultural agree
ment as to the findings. Perhaps future research should
analyze the sexes separately to determine if it is really
the female population that is contributing to the greater
86
math test anxiety level. Also, research needs to begin on
reading test anxiety to determine if it exists apart from
math test anxiety and to what extent.
Anxietv and Achievement Correlations
Between and Within Areas
Similarly to prior research studies, this study
found a significant but low negative correlation between
achievement and test anxiety (Benjamin et al., 1981;
Dusek, I98O; Hill, 1972). As indicated in Figure 2, read
ing test anxiety and reading achievement have a -.11 rela
tionship, while math test anxiety and math achievement
have a -.08 relationship. This suggests, as past studies
have, that as test anxiety increases achievement in each
area decreases. Although these relationships were signif
icant, the present researcher expected a stronger corre
lation. Perhaps if grade was controlled for in future
studies, this would occur. Since there is substantial
support suggesting that second graders have a lower test
anxiety level than fifth grade students, it is plausible
that their academic performance is hardly impaired by the
slight amount of anxiety they feel in evaluative situa
tions. Hence, in evaluating the relationship between test
anxiety and performance, future research projects should
only select upper-grade children for investigation because
they are the ones who most often feel test anxious and
87
thus have achievement hindered. By combining second grade
pupils with fifth grade pupils, as was done in this pro
ject, the true relationship of these two variables may be
lost.
Further correlations were derived from the results
that were not specifically hypothesized. Significant
positive correlations were found between reading achieve
ment and math achievement and reading test anxiety and
math test anxiety. The relationship between the achieve
ment variables could have been expected; children who
achieve well in one area often achieve well in the other
area unless a specific disability occurs. The fact that
reading test anxiety and math test anxiety are signifi
cantly correlated in a positive direction suggests that
children who are test anxious in one subject may be
similarly anxious in another area. However, the research
determining whether or not test anxiety exists for spec
ific subjects is scarce. Thus, results in the present
study are hardly conclusive. Future research should
direct itself toward investigating the subject specificity
of test anxiety.
Research question 7B asked if anxiety correlations
within areas (i.e., reading, math) would be greater than
correlations between areas. Figure 3 presented the find
ings; the highest correlations generally existed within
88
subject areas. Specifically, a stronger correlation
existed between general math anxiety and math test anxiety
than for the correlation of either of these variables with
reading test anxiety. Additionally, both general school
anxiety and general test anxiety were significantly corre
lated to anxiety in each subject area. Once more, prior
literature in this area is scarce. There has been little
research completed that can either support or negate the
present research findings. All that can be summized at
this time is that these relationships are very logical.
General school anxiety should be positively correlated to
both reading and math anxiety. Test anxiety in each area
ought to be related to general test anxiety. Moreover,
relationships between these subject areas should be weaker
if anxiety is different for designated subject areas.
Measurement Error in the TASC and DSC
Research question eight asked if there were differ
ences in the amount of measurement error within the three
populations for the TASC and the DSC. Table 6 listed the
reliability coefficients of each subscale for each ethnic
group as well as the total population. In analyzing the
results, ,60 or greater was considered an index of good
reliability. Most of the subscales were appropriate for
use in all three populations. As the TASC and the DSC had
been used before with Hispanic and Anglo children, this is
89
not surprising. Prior research conducted by Holtzman,
Diaz-Guerrero & Swartz (1975) validated its use with
Spanish-speaking and English-speaking children. The
weakest reliabilities for all three populations were found
for both the reading anxiety scale and the math anxiety
scale (alpha <.60). This is not surprising in view of the
fact that each scale only contained three items. Adding a
few more items to both scales would probably increase the
reliability. Additionally, since the original TASC was
designed to be evaluated by one total score, separating
the items into subscales that were analyzed individually
may hinder reliability. The present findings also indi
cated that the reading test anxiety scale may have been
inappropriate for use with the Mexican subjects (alpha
= .50). Even though alpha was greater than .60 for the
other two populations, a trend suggested lower reliability
in this scale than the other scales. This may be related
to a common theme repeated over and over throughout this
discussion. Since few research projects have examined
reading test anxiety scientifically, it is not really
known if it exists. Educational researchers in the future
need to begin evaluating the existence or nonexistence of
this structure.
In summary, the results appear favorable when
evaluating the appropriateness of the scales used in the
90
present research study. Differences in the amount of
measurement error within these populations were small.
This study has reconfirmed the suitability of the TASC and
DSC for the Mexican and Anglo population. Moreover, it is
one of the original studies validating the use of these
scales with the Hispanic immigrant population in Los
Angeles.
Conclusions and Implications for
Future Research
The present research study was unique. It not
only added a new dimension to the knowledge of test
anxiety but initiated research in several untouched areas
of school anxiety. It was original in that prior studies
concerning test anxiety either compared Anglo children to
minority students or Anglo children to Mexican children in
Mexico. This study not only included both the Mexican and
Anglo populations but additionally incorporated a very
significant population that presently dominates Los
Angeles public schools: Hispanic immigrant students. The
large sample size of 775 students contributed to the reli
ability and validity of the findings reported. The fact
that second grade pupils as well as fifth grade pupils
were selected for the study enabled the researcher to
trace the developmental changes in anxiety level. The
careful examination of prior research studies added to the
91
successful design of the present research project.
The hypotheses chosen for research were those
whose results would contribute significantly to the
literature on test anxiety. Predictions made were based
on a combination of theoretical rationale, prior empirical
research, and in some cases logical reasoning. Research
questions touched upon areas of concern that were impor
tant to an understanding of test anxiety. Yet previous
studies were not able to examine these issues. The hypoth
eses of the present research study were supported by these
findings ;
1 . New immigrants residing in the United States
less than two years were more test anxious than Hispanic
immigrants who lived in the United States more than two
years.
2. Mexican children in Mexico scored signifi
cantly higher on the TASC than Anglo children and Hispanic
immigrants who had resided in the United States more than
two years.
3. Females across the populations scored higher
on the TASC than males.
4. Males in both countries scored significantly
higher on the Defensiveness (DSC) Scale.
5. Fifth grade students scored significantly
higher on the TASC than second grade students.
92
6. Across the three populations there was a
greater amount of math text anxiety than reading test
anxiety.
7A. There was a significant relationship between
math achievement and math text anxiety and reading achieve
ment and reading test anxiety.
7B. Achievement and anxiety correlations within
areas (i.e. math, reading) were greater than correlations
between areas for the majority of variables examined.
8. Differences in the amount of measurement error
within these populations were generally small and insig
nificant. The Test Anxiety Scale for Children (TASC) and
the Defensiveness Scale (DSC) appear to be appropriate
instruments for all populations.
These findings barely scratched the surface of all
the factors that could be examined in a study such as the
present one. The opportunities for future research are
unlimited. Foremost, similar studies should be replicated
to confirm the findings identified here. Interactions
between the demographic variables, race, sex, and grade,
should be studied to determine their relationship to test
anxiety. Experimental studies should be conducted to
determine the causes for the existing relationships
between the various factors studied and test anxiety.
Once this knowledge is revealed, educators should be able
93
to reduce anxiety level by appropriately manipulating the
environment of the students. Perhaps, then, the achieve
ment scores of all children, minority as well as majority,
students will rise.
Finally, the present research suggests that test
anxiety may not be an overall feeling that is applicable
to all areas, but is related to specific subjects depend
ing on the individual student. The fact that students in
the present study indicated they felt more test anxious in
math in comparison to reading suggests one of two possibil
ities: test anxiety is subject-specific or reading
anxiety does not exist at a significant level among stu
dents. Since the present literature on test anxiety has
not yet begun to answer questions regarding this issue, it
is safe to conclude that research is lacking in this area.
Research on test anxiety has become a growing
concern of educators in recent years. As the achievement
test scores of school age children decrease rather than
increase, psychologists and educators have begun to focus
their attention on causal variables. Text anxiety has
emerged as a major factor affecting performance. More
over, it appears to increase in level throughout the
elementary school years. Hence, this concern of educators
is well founded. The present research study has investi
gated some important relationships between test anxiety
94
and correlated variables, but it is only a small fraction
of the kind of work and investigations that need to be
conducted if educational psychologists truly yearn to
resolve the crisis of test anxiety in the classroom.
BIBLIOGRAPHY
96
BIBLIOGRAPHY
Aiken, L. R. Attitudes toward mathematics. Review of
Educational Research, 1970, Ü0, 551-596.
Aiken, L. Update on attitudes and other affective
variables in learning mathematics. Review of Educa
tional Research, 1976, Ü6(2), 293-311.
Barnard, J., Zimbardo, P., & Sarason, S. Anxiety and ver
bal behavior in children. Child Development, 1961,
32, 379-392.
Benjamin, M., Mckeachie, W., Lin, Y., & Holinger, D. Test
anxiety: Deficits in information processing. Journal
of Educational Psychology, 1981, %1, 816-824.
Bradshaw, G., & Gaudry, E. The effect of a single exper
ience of success or failure on test anxiety.
Australian Journal of Psychology, 1968, 20,
219-223.
Brody, E. The relationship between student teacher effec
tiveness and student characteristics : A moderator
variable analysis. Dissertation Abstracts, 1968,
2^(3-A), 757.
Bronzaft, A., Murgatroyd, D., & McNeilly, R. Test anxiety
among Black college students: A cross-cultural study.
Journal of Negro Education, 1974, iL3, 190-193.
Butterfield, E., & Zigler, E. The effects of success and
failure on the discrimination learning of normal and
retarded children. Journal of Abnormal Psychology,
1965, 10, 25-31.
Campeau, P. Text anxiety and feedback in programmed
instruction. Journal of Educational Psychology,
1968, 3^, 159-163.
Clawson, T., Firment, C., & Trower, T. Text anxiety:
Another origin for racial bias in standardized test
ing. Measurement and Evaluation Guide, 1981, JL3
210-215.
Cohen, L. Educational research in classrooms and
schools. New York: Harper & Row, 1976.
97
Cox, F. Educational streaming and general test anxiety.
Child Development, 1962, 33, 381-390.
Cox, F. Some effects of test anxiety and presence or
absence of other persons on boys' performance on a
repetitive motor task. Journal of Experimental Child
Ps¥chol_Q ^Y, 196 6, 3 , 1 0 0 - 1 1 2 .
Cronbach, L., & Snow, R. Aptitudes and instructional
methods. New York: Irvington Press, 1977.
Culler, R., & Holahan, C. Test anxiety and academic per
formance: The effects of study-related behaviors.
Journal of Educational Psvcholoev, 1980, 72,
16-20.
DeBlassie, R. Test anxiety: Education's hang-up.
Clearing House, 1972, 46, 526-530.
Deffenbacher, J. Worry, emotionality, and task
generated interferences in test anxiety: An empirical
test of attentional theory. Journal of Educational
Psychology, 1978, IÛ, 248-254.
Deutsch, T., & Tobias, S. Prior achievement, anxiety and
instructional method. Paper presented at the annual
convention at the American Psychological Association,
Montreal, September 1980.
Diaz-Guerrero, R. Psychology of the Mexican culture and
personality. Austin, Texas: University of Texas
Press, 1975.
Diaz-Guerrero, R. Test and general anxiety in Mexican and
American school children. In C. Spielberger & R.
Diaz-Guerrero (Eds.), Cross-cultural anxiety (Vol.
1), New York: John-Wiley & Sons, 1976.
Domino, G. Aptitude bv treatment interaction effects in
college instruction. Paper presented at the Ameri
can Psychological Association, New Orleans, September
1974.
Duffey, J., & Martin, R. Effects of direct and indirect
teacher influence and student trait anxiety on the
immediate recall of academic material. Psychology in
the School, 1973, 10, 233-237.
98
Dusek, J. The development of text anxiety in children.
In I. Sarason (Ed.), Test anxietv: Theorv. research
and application. Hillsdale, N.J.: Lawrence Erlbaum
Ass., 1980.
Dusek, J., & Hill, K. Probability learning as a function
of sex of subject, test anxiety, and percentage of
reinforcement. Developmental Psychology, 1970,
3, 195-207.
Eaton, W. Cross-lagged analysis of children's test
anxiety and defensiveness. Journal of Educational
Psychology, 1980, 12, 404-407.
Entwisle, D., & Greenberger, E. Questions about social
class, internality-externality and text anxiety.
Developmental Psychology, 1972, 1(2), 218.
Evans, R. Relationship of three patterns of public
school grade organization to student general anxiety
and test anxiety scores, grades 6-12. Dissertation
Abstracts International, 1969, 20(6-A), 2374-2375.
Eysenck, M, Anxiety, learning and memory: A reconcep
tualization. Journal of Research in Personality,
1979, 11, 363-385.
Friend, R., & Neale, J. Children's perceptions of success
and failure: An attributional analysis of the effects
of race and social class. Developmental Psychology,
1972, 1(2), 124-128.
Ganzer, V. Effects of audience presence and test anxiety
on learning and retention in a serial learning situa
tion. Journal of Personality and Social Psychology,
1968, 8, 194-199.
Gotts, E. Note on cross-cultural by age-group comparisons
of anxiety scores. Child Development, 1968, 32,
945-947.
Grimes, J., & Allinsmith, W. Corapulsivity, anxiety and
school achievement. Merrill-Palmer Quarterly, 1961,
I, 247-271.
Gubbay, J. The relationship between self-esteem, general
anxiety, and test anxiety among Fijian school children
in grades 6-10 and Mexican school children in grades
3-8 in schools in their own countries. Dissertation
99
Abstracts International, 1978, 32(1-A), 215-
216.
Hawkes, T., & Furst, N. Race, socio-economic situation,
achievement, IQ, and teacher ratings of students'
behavior as factors relating to anxiety in upper
elementary school children. Sociology of Education,
1971, M, 333-350.
Heetner, B., & Wallace, D. The emotional impact of tests.
The Journal of Emotional Education, 1967, 2(1),
33-35.
Hermans, H., ter Laak, J., & Maes, P. Achievement
motivators and less failure in family and school.
Development of Psychology, 1972, 6, 520-518.
Hernandez, N. Variables affecting achievement of middle
school Mexican-American students. Review of Educa
tional Research, Winter 1973, &3(1), 1-39.
Hill, K. Anxiety in the evaluative context. In W.
Hartup (Ed.), Young Child, (Vol. 2). Washington,
D.C.! National Association for Education of Young
Children, 1972, 225-263.
Hill, K., & Sarason, S. The relation of text anxiety and
defensiveness to test and school performance over the
elementry school years: A further longitudinal study.
Monographs of the Societyfor Research in Child
Development, University of Chicago Press, 1966,
31(2), 1-75.
Holmes, P. R. The influence of anxiety upon academic
performance under varying conditions of task orien
tation and evaluation (Doctoral Dissertation, George
Peabody College for Teachers, 1966). Disseration
Abstracts International, 1970, 2Q, 4833A.
Holtzman, W. Diaz-Guerrero, R., & Swartz, J. Personality
Hawal in I-t-to nii 1 ii < a AuStin, TeXaS :
)zman, w. uiaz-uuerrero, k., &
development in two cultures. A
University of Texas Press, 1975
Hutt, M. "Consecutive" and "adaptive" testing with
revised Stanford-Binet. Journal of Consulting
Psychology, 1947, 11, 93-104.
Johnson, S. Children's fears in the classroom setting.
School Psychology Digest. 1979, 2, 382-396.
100
Kaczala, C. A longitudinal study of attitudes toward
mathematics in 5th through 12th grades: Age, and
sex differences. Paper presented to American Educa
tional Research Association, Boston, April 1980.
Kirkland, M. The effects of tests on students and
schools. Review of Educational Research, 1971,
41(4), 303-350.
Kozma, A. The effects of anxiety, stimulation and
isolation on social reinforcer effectiveness.
Journal of Experimental Child Psychology, 1969,
2, 1- 8.
Krohne, H. Parental child-rearing behavior and the
development of anxiety and coping strategies in
children. In I. Sarason & C. Spielberger (Eds.),
Stress and anxiety, (Vol. 7). New York:
Hemisphere Publishing Co., 1980.
Klum, G. Research on mathematics attitudes. In R.
Shumway (Ed.), Resear cJa _lD__inatjiema t i cs education.
Virginia: National Council of Teachers of Mathe
matics, 1980.
Madden, S., & Borich, G. Evaluating classroom instruc
tion . California: Addison-Wesley Publishing Co.,
1977.
Mandler, G., & Sarason, S. A study of anxiety and
learning. Journal of Abnormal and Social Psychology,
1952, 41, 166-173.
McCormick, C., & Karabinus, R. Relationship of ethnic
groups, self-estemra and anxiety to school success.
Education and Psychology Measurement, 1976, 36,
1093-1100.
McCoy, N. Effects of text anxiety on children's perfor
mance as a function of instructions and type of
task. Journal of Personality and Social Psychology,
1965, 2, 634-641.
Many, M., & Many, W. Relationship between self-esteem and
anxiety in grades four through eight ; Sarason's
General Anxiety Scale for Children and Text Anxiety
Scale for Children. Education and Psychology
Measurement, 1975, 32, 1017-1021.
101
Marlett, N., & Watson, D. Text anxiety and immediate or
delayed feedback in a test like advoidance task.
Journal of Personality and Social Psychology,
1968, 8, 200-203.
Meece, J. A theoretical framework for studying stu
dents* course selection in mathematics. Paper pre
sented at annual meeting of the American Educational
Research Association, Boston, April 1980.
Messer, S. The effects of anxiety over intellectual
performance on reflection-impulsivity in children.
Child Development, 1970, M, 723-735.
Morris, L., & Fulmer, R. Text anxiety changes during
academic testing as a function of feedback and test
importance. Journal of Educational Psychology,
1976, 68, 817-824.
Morris, L., & Liebert, R. Relationship of cognitive and
emotional components of test anxiety to physiologi
cal arousal and academic performance. Journal of
Consulting and Clinical Psychology, 1970, 32,
332-337.
Nicholls, J. When a scale measures more than its name
denotes ; The case of the Test Anxiety Scale for
Children. Journal of Consulting and Clinical
Psychology, 1976, M(6), 976-984.
O'Reilly, R., & Wightman, L. Improving the identifica
tion of anxious elementary school children through
the use of an adjusted anxiety scale : Test Anxiety
Scale for Children. Journal of Educational Measure
ment, 1971, 8, 107-112.
Papay, J., Costello, R., Hedl, J., & Spielberger, C,
Effects of trait and state anxiety on the performance
of elementary school children in traditional and
individualized multi-age classrooms. Journal of
Educational Psychology, 1975, 2, 840-846.
Peterson, P. Interactive effects of student anxiety,
achievement, orientation, and teacher behavior on
student achievement and attitude. Journal of Educa
tional Psvchologv, 1977, 29, 779-792.
102
Phillips, B. Sex, social class and anxiety as sources
of variations in school achievement. Journal of
Educational Psvchologv, 1962, 21, 316-322.
Phillips, B. Anxiety in elementary school children.
Childhood Education, 1968, 44, 340-342.
Phillips, B. School stress and anxietv. New York :
Human Sciences Press, 1978.
Phillips, B., Pitcher, G., Worsham, M., & Miller, S.
Test anxiety and the school environment. In I. Sara
son (Ed.), Test anxietv; Theorv, research and
applications. Hillsdale, N.J.: Lawrence Erlbaum
Associates, 1980.
Richardson, C., & Woolfolk, R. Mathematics anxiety.
In I. Sarason (Ed.), : Theory,
research, and applications. Hillsdale, New Jersey:
Lawrence Erlbaum Associates, 1980.
Rosenbaum, M., & Hill, K. Effects of success-failure
and praise-criticism on retardates* persistence and
performance. Journal of Experimental Research and
Personality, 1969, 1, 253-263.
Ruebush, B. Interfering and facilitating effects of test
anxiety. Journal of Abnormal and Social Psychology.
1960, 60, 205-212.
Sarason, I. Empirical findings and theoretical problems
in the use of anxiety scales. Psychological Bulle-
jLla, 1960, 51, 403-415.
Sarason, I. Experimental approaches to text anxiety:
Attention and the uses of information. In C,
Spielberger (Ed.), Anxietv: Current trends in theory
and research, (Vol. 2). New York : Academic Press,
1972.
Sarson, I. Test anxiety and social influence. Journal
of Personality, 1973, 41, 261-271.
Sarason, I. Text anxiety, stress, and social support.
Journal of Personality, 1981, 29, 101-114.
Sarason, S., & Davidson, K. A study of teacher behavior
in relation to children differing in anxiety level.
103
Washington, D.C.: Cooperative Research Program,
Office of Education, U.S. Department of Health, 1966.
Sarason, S., Davidson, K,, Lighthall, F., & Waite, R.
A test anxiety scale for children. Child Develop
ment , 1958, 22, 105-113.
Sarason, S., Davidson, K., Lighthall, F., Waite, R., &
Ruebush, B. Anxietv in elementary school children.
New York: Wiley, I960.
Sarason, S., Hill, K., & Zimbardo, P. A longitudinal
study of the relation of test anxiety to performance
on intelligence and achievement tests. Monograph
Society for Research of Child Development, Serial
No. 98, 1964, 22, (Whole No. 7).
Sarason, S., Mandler, G., & Craighall, P. The effect of
differential instructions on anxiety and learning.
Journal of Abnormal Social Psychology, 1952,
41, 561-565.
Sarason, S., & Mandler, G. Some correlates of text
anxiety. Journal of Abnormal and Social Psychology,
1952, 42, 810-817.
Sarason, I., & Stoops, R. Text anxiety and the passage
of time. Journal of Consulting Clinical Psychology,
1978, 46, 102-109.
Schmidt. L., & Gallessick, J. Adjustment of AngJ.o-
American and Mexican American pupils in self-
contained and team teaching classrooms. Journal of
Educational Psychology, 1971, 62, 328-332.
Sellinger, S. An investigation of the effects of organi
zational climate and teacher anxiety on test anxiety
of elementary school students. Dissertation
Abstracts International, 1972, 12, 5515A
Sieber, J., Kaneya, L., & Paulson, F. Effect of memory
support on the problem-solving ability of test anxious
children. Journal of Educational Psychology, 1970,
61, 159-168.
Silverman, I., & Waite, S. Text anxiety and the effec
tiveness of social and nonsocial reinforcement in
children. Child Development, 1969, 40, 3070314.
Spielberger, C., & Diaz-Guerrero, R. (Eds.). Cross-
cultural anxietv. New York: John Wiley, 1976.
Spielberger, C., Gonzalez, D., Reigosa, F. Martinez-
ürrietia. A., Natalicio, L., & Natalicio, D.
Development of the Spanish edition of the Strait
Trait Anxiety Inventory, Interamerican Journal of
Psychology, 1971, 2, 145-157.
Stantaon, H. Relationship between teachers* anxiety
level and test anxiety level of their students.
Psychology in the Schools, 1971, 11, 360-363.
Strom, R., & Johnson, A. The elementary school exper
ience of children from Mexico. The Elementary
School Journal, 1981, 82(1), 37-40.
Tobias, S. Anxiety research in educational psychology.
Journal of Educational Psychology, 1979, 1(5),
573-582.
Trentham, L. Anxiety and instruction effects on sixth-
grade students in a testing situation. Psychology
in the Schools, 1979, 16, 439-443.
Tryon, G. The measurement and treatment of test anxiety.
Review of Educational Research, 1980, 52(2),
343-372.
Wade, B. Highly anxious pupils in formal and informal
primary classrooms; the relationship between inferred
coping strategies. British Journal of Educational
Psychology. 1981, 51, 39-57.
Weinstein, C. Effects of test anxiety on learning at
superficial and deep levels of processing.
Contemporary Educational Psychology, 1982, %,
107- 112.
Wilkins, W. Desensitization : Social and cognitive fac
tors underlying the effectiveness of Wolpe*s pro
cedure. Psychological Bulletin, 1971, 12(5), 311-
319.
Wilson, R. Don*t be nervous, it * s only a test. Early
Years, 1978, 8, 56-57.
Wine, J. Test anxiety and direction of attention.
Psychological Bulletin, 1971, 12(2), 92-104.
104
1 05
Zimmerman, B, The relationship between teacher class
behavior and student school anxiety levels.
Psychology in the Schools, 1970, %, 89-93.
APPENDIXES
107
APPENDIX A
APPROVAL FOR STUDY
1 08
Los Angeles Unified School District H A R R Y H A N D L E R
S u p m n t w m i^ t o f S cko o ls
Administrative Offices; 450 North G rand Avenue, Los A ngeles. California 90012 floraline i stevens
M ailing A doress; Box 3307, Los A ngeles, California 90051 ^^I^i,amiBvaiux.tionBmnch
Telephone: (213) 625-6207
February 29, 1984
Ms. Myra D. Saltoun
416 S. Holt Avenue
Los Angeles, C A 90048
Dear Ms. Saltoun:
The Committee on Research Studies approved your request to conduct a
study w ith the d e s c rip tiv e t i t l e , "Test Anxiety Among Three
Populations: Mexicans, Hispanics in Los Angeles, and Anglo-Saxon
Children," in the Los Angeles Unified School D is tric t. Your study
w ill be restricted to the following elementary schools: Canterbury
Avenue, Dixie Canyon Avenue, Hoover Street, and Kenter Canyon.
This approval by the Comnittee is in no way a requirement fo r d is tric t
personnel to particip ate. All participation by school admininstrators
and teachers must be completely voluntary. The anonymity of a ll
participants must be maintained.
You may use this le tte r as an authorization to approach school
administrators and teachers who may participate in your study. At the
conclusion of your study, please send an abstract of the findings to
the Coiwnittee.
Sincerely,
- /a -
Floraline I . Stevens
Director
Research and Evaluation Branch
FIS:rp
cc: Dr. Linscomb
Dr. Brickman
Dr. Daniels
Dr. Juhnke
Dr. Rivas
Dr. Hocevar
Principals of Selected Schools
109
APPENDIX B
TEST ANXIETY SCALE FOR CHILDREN
1 1 0
COPY
TASC QUESTIONS
1. Do you worry when the teacher says that she is going
to ask you questions to find out how much you know?
2. Do you worry about being promoted, that is passing
from the___ to the grade at the end of the year?
3. When the teacher asks you to get up in front of the
class and read aloud, are you afraid that you are going
to make some bad mistakes?
4. When the teacher says that she is going to call upon
some boys and girls in the class to do math problems, do
you hope that she will call upon someone else and not on
you?
5. Do you sometimes dream at night that you are in
school and cannot answer the teacher’s questions?
6. When the teacher says that she is going to find out
how much you have learned, does your heart begin to beat
faster?
7. When the teacher is teaching you about math, do you
feel that other children in the class understand her
better than you?
8. When you are in bed at night, do you sometimes worry
about how you are going to do in class the next day?
9. When the teacher asks you to write on the blackboard
in front of the class, does the hand you write with
sometimes shake a little?
10. When the teacher is teaching you about reading, do
you feel that other children in class understand her
better than you?
11. Do you think you worry more about school than other
children?
12. When you are at home and you are thinking about your
math lesson for the next day, do you become afraid that
111
you will get the answers wrong when the teacher calls upon
you?
13. If you are sick and miss school, do you worry that
you will do more poorly in your schoolwork than other
children when you return to school?
14. Do you sometimes dream at night that other boys and
girls in your class can do things you cannot do?
15. When you are home and you are thinking about your
reading lesson for the next day, do you worry that you
will do poorly on the lesson?
16. When the teacher says that she is going to find out
how much you have learned, do you get a funny feeling in
your stomach?
17. If you did very poorly when the teacher called on
you, would you probably feel like crying even though you
would try not to cry?
18. Do you sometimes dream at night that the teacher is
angry because you do not know your lessons?
* Read special directions for the following questions.
See instruction sheet for TASC.
19. Are you afraid of school tests?
20. Do you worry a lot before you take a test?
21. Do you worry a lot while you are taking a test?
22. After you have taken a test do you worry about how
well you did on the test?
23. Are you afraid of math tests?
24. Do you worry alot before you take a reading test?
25. Do you sometimes dream at night that you did poorly
on a test you had in school that day?
26. Do math tests make you more afraid than reading
tests?
27. When you are taking a test, does the hand you write
with shake a little?
112
28. When you are taking a math test, do you forget some
things you knew very well before you started taking the
test?
29. When the teacher says that she is going to give the
class a test, do you become afraid that you will do
poorly?
30. Are you afraid of reading tests?
31. Do you worry a lot before you take a math test?
32. When you are taking a hard test, do you forget some
things you knew very well before you started taking the
test?
33. Do you wish a lot of times that you didn’t worry so
much about tests?
34. Do reading tests make you more afraid than math
tests?
35. When the teacher says that she is going to give the
class a test, do you get a nervous or funny feeling?
36. While you are taking a test do you usually think you
are doing poorly?
37. When you are taking a reading test, do you forget
some things you knew very well before you started taking
the test?
38. While you are on your way to school, do you sometimes
worry that the teacher may give the class a test?
113
APPENDIX C
DEFENSIVENESS SCALE FOR CHILDREN
114
COPY
DSC QUESTIONS
1 , Do you love to play sports best of all?
2. Should girls be just as brave as boys?
3. Do you sometimes dream about things you don’t like to
talk about?
4. Are you sometimes afraid of getting into arguments?
5. When someone scolds you does it make you feel badly?
6. When you get mad do you ever tell anyone else about
it?
7. Do you sometimes feel like hurting someone?
8. Do you like to play in the snow?
9* Are you sorry for some of the things you have done?
10. When one of your friends won’t play with you, do you
feel badly?
11. When someone makes you mad, do you ever tell them
about it?
12. Do you feel cross and grouchy sometimes?
13. Are there some people that you don’t like?
14. Since you started school, have you ever felt like
crying?
15. Do you feel it’s important to think about how you
can get people to like you?
16. Do you like to go to the beach in the summertime?
17. Sometimes when you get mad, do you smash something?
18. When you hurt somebody’s feelings, does it make you
feel badly?
115
19. Do you wish your teacher paid more attention to you?
20. Do you sometimes have arguments with your mother and
father?
21. Are there some things you just don’t like to talk
about?
22. If you think someone doesn’t like you, does it bother
you?
23. Do you like to go on trips with your mother and
father?
24. Do you feel terrible if you break something which
belongs to somebody else?
25. Do you lose your temper sometimes?
26. When you are worried about something, do you like
to talk about it?
27. Does it bother you if the teacher chooses someone
else instead of you to do something for her (or
him) ?
28. When you’ve done something wrong, is it hard for
you to say you’re sorry?
29. Is it hard for you to tell someone you’re scared?
116
APPENDIX D
DIRECTIONS FOR TASC AND DSC
117
TASC DIRECTIONS
My name is __________ I^m going to be
asking some questions— questions different from the usual
school questions for these are about how you feel and so
have no right or wrong answers. First I'll hand out the
answer sheets and then I'll tell you more about the
questions. . .
Write your name at the top of the first page, both
your first and your last name. . . . Write your age now
and the grade you are in. Then circle the picture of the
boy if you're a boy or circle the picture of a girl if
you're a girl.
As 1 said before, 1 am going to ask you some
questions. No one but myself will see your answers to
these quêtions, not your teacher or your principal or your
parents. These questions are different from other
questions that you are asked in school. These questions
are different because there are no right or wrong answers.
People think and feel differently. The person sitting
next to you might put a circle around "yes" and you may
put a circle around "no." For example, if 1 asked you
this question: "Do you like to play ball?" some of you
would put a circle around "yes" and some of you would put
118
it around "no." Your answer depends on how you think and
feel. These questions are about how you think and feel
about school; and about a lot of other things. Remember,
listen carefully to each question and answer it "yes" or
"no" by deciding how you think and feel. If you don't
understand a question, ask me about it.
Now let's start by everybody putting their finger
on Number 1. Here is the first question. Number 1. "Do
you worry when ________ ?" (read number 1) (Repeat this
procedure of introducing the questions for several of them
and continue throughout to say the number of the question
before reading it.)
PROCEED UNTIL NUMBER 18. Then after #18 say:
In the following questions the word "test" is
used. What 1 mean by "test" is any time the teacher asks
you to do something to find out how much you know or how
much you have learned. It could be by your writing on
paper, or by your speaking aloud, or by your writing on
the blackboard. Do you understand what 1 mean by "test"—
it is any time the teacher asks you to do something to
find out how much you know. (Continue reading questions
until scale is completed)
119
APPENDIX E
STUDENT ANSWER SHEETS FOR TASC AND DSC
1 20
T A S C
N A M E A G E G R A D E S E X
27 14
Y E S N O Y E S N O Y E S N O
28 15
Y E S N O Y E S N O Y E S N O
29 16
Y E S Y E S N O Y E S N O N O
30 17
Y E S Y E S N O N O Y E S N O
31 18
Y E S Y E S N O N O Y E S N O
32 19
Y E S Y E S N O N O Y E S N O
33 20
Y E S Y E S Y E S N O N O N O
34 21
Y E S Y E S N O N O Y E S N O
35 22
Y E S N O Y E S Y E S N O N O
36 23 10
Y E S N O Y E S N O Y E S N O
24 37
N O Y E S Y E S Y E S N O N O
12 25 38
Y E S Y E S N O Y E S N O N O
26
Y E S N O Y E S N O
D S C
G R A D E S E X A G E N A M E
16
N o Y E S
N O Y E S
17
N O Y E S
N O Y E S
18
N O Y E S
N O Y E S
19
N O Y E S
N O Y E S
20
N O Y E S
N O Y E S
21
N O Y E S
Y E S N O
22
N O Y E S
N O Y E S
23
N O Y E S
N O Y E S
24
N O Y E S
N O Y E S
25
10
Y E S N O
N O Y E S
26
N O Y E S
N O Y E S
21
12
N O Y E S
N O Y E S
28
N O Y E S
N O Y E S
29
14
N O Y E S
N O Y E S
15
N O Y E S
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The positive aspects of classroom instruction in nutrition and subsequent eating habits
PDF
Intercollegiate athletic drug testing at the University of Southern California: Antecedents, rationale, and the law
PDF
A structural and psychoanalytic analysis of the symbolic ritual process in three sets of Russian folk tales
PDF
Reading achievement influenced by certain home factors in Mexican-American homes
PDF
The reputation of Cicero among the English Deists (1696 - 1776)
PDF
Alianza Hispano-Americana, 1894-1965: A Mexican American fraternal insurance society
PDF
Anglo-Jewish art collectors of the Victorian Period: Patterns in collecting
PDF
SEC's enforcement actions and its effectiveness in an efficient market
PDF
William Lawes: His life and works
PDF
Outside the Whale: Reading the American Political Novel in the Age of Reagan
PDF
Immigrant status, security, and family role conflict: A comparative study of resident aliens and undocumented aliens from Mexico
PDF
A three generational study: Traditional Korean value systems and psychosocial adjustment of Korean immigrants in Los Angeles
PDF
An investigation Piaget's theory of cognitive development as a basis for the assessment of reading disability of Afro-American junior high school students and for developing remedial curricula
PDF
Women's autobiography and national identity: Natalia Ginzburg, Anna Banti and Renata Vigano
PDF
Crisis preparation in technical organizations: A study using a multi-dimensional approach
PDF
David of Sassoun: The Armenian folk epic
PDF
A study of Mexican American perceptions of law enforcement policies and practices in East Los Angeles
PDF
The effect of cooperative and individualistic goal structures on the preferences and attribution of field-dependent and field-independent students
PDF
The critical technique of Sainte-Beuve considered in its relationship to the modern biography as exemplified by Lytton Strachey and Andre Maurois
PDF
Bulimia nervosa: A response to trauma associated with parental alcoholism and sexual abuse in college sorority women
Asset Metadata
Creator
Saltoun, Myra Debborah
(author)
Core Title
Math and reading test anxiety among three populations: Mexicans, Hispanic immigrants, and Anglo-Saxon children
Degree
Doctor of Philosophy
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Education,OAI-PMH Harvest,Social Sciences
Format
application/pdf
(imt)
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c37-48348
Unique identifier
UC11632831
Identifier
DP71335.pdf (filename),usctheses-c37-48348 (legacy record id)
Legacy Identifier
DP71335.pdf
Dmrecord
48348
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Saltoun, Myra Debborah
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