University of Southern California Dissertations and Theses

Dropout - Stayin Personality Differentials And College Environments

(USC Thesis Other)

Dropout - Stayin Personality Differentials And College Environments

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content { 70-11,367
r
I
/
L HANNAH, William, 1919-
l DROPOUT-STAYIN PERSONALITY DIFFERENTIALS AND
COLLEGE ENVIRONMENTS. EAppendix D: "CUES
j , College and University Environment Scales",
I pages 125-131 and "Attitude Inventory, OPI-Form
Fx", pages 132-141, not microfilmed at request
I of author. Available for consultation at
University of Southern California Library].
I University of Southern California, Ed.D., 1969
i Education, administration
f •
( University Microfilms, Inc., Ann Arbor, Michigan
© Copyright by
WILLIAM HANNAH
| 1970 j
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED
DROPOUT-STAYIN PERSONALITY DIFFERENTIALS
AND COLLEGE ENVIRONMENTS
by
William Hannah
A Dissertation Presented to the
FACULTY OF THE SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION
August 1969
This dissertation, written under the direction
of the Chairman of the candidate's Guidance
Committee and approved by all members of the
Committee, has been presented to and accepted
by the Faculty of the School of Education in
partial fulfillment of the requirements for the
degree of Doctor of Education.
T)ntp August 1969________________________
.....
/ Dean
Guidance Committee
ACKNOWLEDGMENTS
This study has benefited from the cooperation of a
large number of institutions and individuals across the
nation. The Project on Student Development under the aus
pices of the Council for the Advancement of Small Colleges
played an indispensable part. The colleges* the institu
tional representatives* and the students who participated
were necessary to the completion of the study and much
appreciation goes to them.
Dr. Arthur Chickering* the Project Director* gener
ously provided time* advice* encouragement* and editorial
comments without which this effort could not have been com
pleted. Mrs. Beverly Burns* the Project Assistant* gave her
skills to the task of proofreading* editing* and typing
which were invaluable to this researcher. To both of these
individuals the writer is greatly indebted.
To Dr. Earl V. Pullias* my advisor in the Department:
of Higher Education at the University of Southern California,
for his guidance, patience, encouragement, suggestions, and
especially for his warmth and understanding go my heartfelt
thanks. Appreciation is also expressed to the other members
of the writer's Doctoral Committee, Dr. Wallace R. Muelder
and Dr. William Georgiades, for their interest and willing
ness to be part of this effort.
Finally, to my beloved wife Kathryn, the one whose
encouragement helped motivate this effort, a constant com
panion and inspiration, I dedicate this dissertation. My
deep wish was to share this small triumph with her, but her
passing did not allow that wish to be realized.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS.......................................... ii
LIST OF T A B L E S .......................................... vi
LIST OF G R A P H S ............................................ vii
Chapter
I. THE PROBLEM AND ORGANIZATION OF THE STUDY . . . 1
The Problem
Organization of the Study
II. REVIEW OF THE RELATED LITERATURE............. 21
Introduction
National Studies
General Studies on Attrition
Personality Studies of Dropouts
College Environment Studies of Attrition
III. THE PROCEDURE.................................. 48;
The Project :
The Colleges ;
The Sample and the Instruments j
The Hypotheses I
The Assumptions j
Treatment of the Data j
iv
Chapter
Page
IV. THE FINDINGS 58
Dropout-Stayin Differences
College Campus Characteristics
Summary
V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . . . 84
Summary
Conclusions
Implications and Recommendations
APPENDIXES
Appendix A: Tables 100
Appendix B: Graphs 113
Appendix C : Colleges 123
Appendix D : Instruments 125
SELECTED BIBLIOGRAPHY 143
v
LIST OF TABLES
Table Page
1. Dropout-Stayin OPI Mean Standard Scores .... 60
2. Male Dropout-Stayin OPI Mean Standard
Scores......................................... 61
3. Female Dropout-Stayin OPI Mean Standard
Scores......................................... 63
4. Dropout-Stayin SAT Scores at Entrance ......... 64
5. CUES Patterns vs. OPI Differences.............. 73
6. CUES Pattern I vs. OPI Differences............. 74
7. CUES Pattern II vs. OPI Differences........... 75
8. CUES Pattern III vs. OPI Differences........... 76
LIST OP GRAPHS
Graph Page
I. CUES Score Pattern for Four Colleges........... 68
II. CUES Score Pattern for Three Colleges ......... 70
III. CUES Score Pattern for Two Colleges........... 71
CHAPTER I
THE PROBLEM AND ORGANIZATION OF THE STUDY
The Problem
Introduction
The phenomenon of the college dropout has become a
problem of increasing interest and concern during recent
years. Each year more than one million students enter col
leges and universities across the nation, and yet, according
to Iffert, less than half successfully complete their course
of studies within four years. He also states that of those
who enter college, fully one-third of the most intelligent
entrants decide to withdraw during the four-year period. Of
all dropouts, including the most intelligent, approximately
28 per cent withdraw during, or at the end of, the freshman
year, while an additional 15 per cent withdraw during the
sophomore year, according to Iffert.'*'
-*-R. E. Iffert, "Study of College Student Retentions
and Withdrawal," College and University, XXXI (Summer,
1956), 435-446.
1
2
Although in most cases, clear-cut reasons are given
for these decisions, other more intangible factors seem to
play a part in the whole process of withdrawal. These ele
ments, it is suspected, may include significant emotional or
developmental components for a particular group of college
withdrawals. Farnsworth says, "Although there are no accu
rate statistics available, it has been estimated that per
haps half of the total dropout population falls into this
„2
group.
It is believed that study of college leavers can
illuminate and identify points of general weakness in the
educational program of a college and, therefore, can raise
questions of value for all students. Educational change and
improvement are clearly needed in the face of rapid cultural
change and the approach of universal higher education in the;
United States.
Statement of the problem
The small private college, because of its tenuous
position in the academic community, needs to find answers to
its more serious problems, one of which is the costly
^D. S. Farnsworth, "We Are Wasting Brain Power,"
Journal of National Education Association. XLVIII (1959),
42-44.
3
investment in those students who are admitted to their pro
grams and who subsequently leave. Every admitted student is
subsidized in some way by the institution; consequently,
every dropout can be measured in terms of the time and en
ergy of the faculty, and in terms of the practical financial
investment of the institution.
Because the greater percentage of leavers withdraw
during their first two years of college experience, identi
fication of personality characteristics during this time
sequence, it is suggested, will produce the most fruitful
results in terms of discovery of the emotional and develop
mental components which may move the potential leaver toward
his final decision to withdraw.
In analyzing the problem, many studies have been
undertaken both by the colleges and universities and by
various agencies of higher education. Most of these studies
have been limited in nature, dealing with superficial rea
sons given by students themselves, or with reasons relating
to academics obvious to the researcher. A few studies have
considered not only these factors, but personality variables
3
and institutional weaknesses. Both Knoell and
O
D. Knoell, "A Critical Review of Research on the
College Dropout," in The College Dropout and the Utilization:
Summerskill have reported the need for research which can
identify basic student motivation and personality character
istics in specified college environments in order to sepa
rate the obvious, or more superficial, reasons for with
drawal from the more subtle and less obvious forces causing
students to leave.
The basic difficulty, then, is that of identifying
clearly the personality characteristics of dropouts and the
differences in college climates which are significant for
modification of educational practice, and which suggest
appropriate kinds of institutional policy change. The study
of personality characteristics and their complex relation
ships to other more obvious college characteristics is a
more subtle approach, and may be a fruitful way toward
understanding the problem of college attrition. It is this
level of complexity to which this study is addressed.
This investigation seeks to answer the following
questions:
1. What personality differentials exist between
of Talent, ed. by L. Pervin (Princeton, N. J.: Princeton
University Press, 1966), pp. 71-72.
^J. Summerskill, "Dropouts from College," in The
American College, ed. by Nevitt Sanford (New York: John
Wiley and Sons, Inc., 1962), p. 648.
5
leavers and stayers, including differences between males and
females, in the first two years of the college experience?
2. What aptitude differentials exist between
leavers and stayers, including differences between males and
females, in the first two years of the college experience?
3. What college climate differentials exist among
the colleges participating in the study?
4. What personality characteristics can be identi
fied in leavers, as compared with stayers, from the differ
ent college climates?
The hypotheses
It was hypothesized, for purposes of analysis, that
no differences would appear on the Omnibus Personality In
ventory (OPI):
1. Between personality characteristics of those
students who leave college and the personality character
istics of those students who stay in college.
2. Between personality characteristics of males who
leave college and personality characteristics of males who
stay in college.
3. Between personality characteristics of females
who leave college and personality characteristics of females
who stay in college.
4. Between the level of aptitude of those who leave
college and the level of aptitude of those who stay in col
lege .
The assumptions
It was assumed that:
1. The personality characteristics of both with-
drawers and persisters could be identified through use of
the OPI.
2. The aptitude scores as recorded through the
Scholastic Aptitude Test (SAT) and American College Test
(ACT) could differentiate between leavers and stayers.
3. The characteristics of colleges could be identi
fied and colleges classified as to type by the institutional;
scores derived through use of the College and University
Environment Scales (CUES)., if any distinguishing character
istics existed.
Importance of the study
Student attrition or retention is one of the meas
ures of the success or failure of colleges. If attrition is
too large, it reflects upon the effectiveness of the aca
demic program and tends to require defensive action on the ;
part of the college in order to justify its position before
its public and before the higher educational community at
large. Identification of personality traits of the student
before admission to the college is one means by which small
colleges in particular may continue to attack the problem of
student attrition.
The structured academic programs of small private
colleges in America are often publicized as designed to
provide method and atmosphere which foster development of
the "individual" student. Enrollments are small, social and
religious activities are available, teachers personalize
their relationships with students, and good opportunities
for the development of human relationships are emphasized;
because of these characteristics a great number of potential
college students choose these schools believing that indeed
such advantages will be helpful during the college experi-
5
ence.
Yet, while many find this environment conducive to
learning and adjust to this new climate, a large number of
other students who enter small colleges elect to leave soon
^Research Proposal to NIMH for Project on Student
Development by Council for Advancement of Small Colleges,
1964, p. 1.
8
after they enroll. An example of this is seen in the en
rollment of 4,000 students in the 1957-58 school year of
thirty small colleges holding membership in the Council for
the Advancement of Small Colleges. Of these, 47 per cent
left the colleges before the completion of the sophomore
year, while 67 per cent had withdrawn before they were
6
seniors.
In view of this high rate of withdrawal, the impli
cation which can be drawn is that for many students entering
these institutions, in spite of more favorable circumstances
than those found in larger schools, adjustment continues to
be difficult and persistence remains a problem for them.
This weakness, whether it is found in the colleges or in the
students, is a very significant obstacle to the contribution
these colleges make to the nation. It raises doubt about
the claims of these institutions as to their superior en
vironmental advantage and questions the very points which
they propose as their strongest.
If this is a serious handicap to the development of :
the institutions, it is a more serious difficulty for those
students who cannot cope with problems of adjustment and who
6Ibid.
9
finally solve their problem, eliminating the stress and
strain, simply by removing themselves from an untenable
situation.
Donald Scott indicates that the small private'col
lege makes a significant contribution to higher education in
the United States. Many distinguished people are products
of these colleges and their personal contributions to soci
ety have been recognized by many and by the nation as a
whole. These colleges provide a unique diversity of commit
ment to the whole of higher education, and this diversity is
of special value in a society where sameness may become a
developing trend. The special kind of intellectual and
emotional development which these schools foster is needed
and desired by many of those students who are unwilling to
see themselves as merely one of many in a large cosmopolitan
7
academic community. Therefore the identification of the
sources of weakness is a necessary step to the support of
these colleges. This study is important because it ad
dresses itself to an understanding of a central issue which
involves the ultimate product of these colleges— the student
^Donald G. Scott, "Small Colleges in Transition"
(unpublished Ph.D. dissertation, Florida State University,
1966), pp. 2, 3.
10
himself. This understanding includes both leavers and
stayers and the significant differences between the two. As
has been stated, the findings from this study will have im
plications which go beyond the central problem of college
dropout— it will describe the stayer as well. The descrip
tion will involve the personality characteristics of the
student in both groups and will also attempt to describe and
relate environmental patterns in colleges participating in
the study to these different types of student withdrawals.
The Organization of the Study
The procedure
This study of college attrition was carried on in
the context of the Project on Student Development. This
Project began in September, 1965 with the testing of 3,536
freshmen entering thirteen participating colleges located
from Maine to California. It was designed basically as a
study of student development and change through four years
of college.
This study of attrition and personality developed
within the context of the total Project and proposed to
answer the questions stated in the previous section de
scribing the problem.
11
The thirteen colleges are well suited for such a
study. They reflect diverse philosophies, patterns of cur
riculum, regulation and supervision, student-faculty rela
tions, institutional objectives, and student characteris
tics. They range from the conservative religious institu
tion to the liberal and experimental. Therefore analysis of
attrition among these colleges reflects not only differences
in the personalities of students, but differences in college
climates as well.
In addition to a review of the related literature,
the study examined response to instruments presented to the
students as freshmen at entrance. Relevant data were ob
tained from the OPI. In this instrument, mean standard
scores are indicated and are reported at the .01 and the .05
level of significance. Thus the findings reflect signifi
cant differences between the leavers and stayers.
The CUES instrument was used in order to describe
the predominant characteristics of groups of colleges which
proved to be similar in nature. The purpose of this ap
proach was to determine what types of college climate seem
to militate against the persistence of certain personality
types. Selected samples of each student body were chosen,
and this instrument was administered to them.
12
The appropriate SAT and ACT scores were analyzed
after compilation of these from existing college records.
A threefold approach to statistical analysis was
made for OPI and for SAT-ACT scores:
1. All leavers were compared with all persisters.
2. Male leavers were compared with male persisters.
3. Female leavers were compared with female per
sisters .
Patterns of personality characteristics were exam
ined and transferred to tables which present group differ
ences for clearer understanding.
Although not measured by statistical analysis, the
CUES scores of the colleges were inspected for similarity
among the colleges and transferred to graphs for pattern
classification.
The findings, conclusions, and recommendations were
developed and summarized.
Definition of terms
The terminology normally used in a study of college
withdrawals is often criticized because of unclear connota
tions ascribed to certain words. The following terms are
defined here for this study (and include terms used in the
13
instruments and others used in the gathering of data).
College.— The college or colleges participating in
the study, except where it is evident that the general term
means any college.
0
College and University Environmental Scales (CUES)
a. Awareness— The degree to which there is a con
cern with self-understanding, reflectiveness,
and the search for personal meaning.
b. Community— The degree to which the campus is
friendly, cohesive, and group-oriented. There
is a feeling of group welfare and group loyalty
which encompasses the college as a whole.
c. Practicality— The degree to which personal sta
tus and practical benefit are emphasized in the
college environment. Status is gained by know
ing the right people, being in the right groups,
and doing what is expected. Order and super
vision are characteristic of the administration
and of the classwork.
d. Propriety— The degree to which politeness,
^Robert C. Pace, College and University Environment
Scales (Princeton, N. J.: Educational Testing Service,
1963), pp. 24, 25.
14
protocol, and consideration are emphasized.
e. Scholarship— The degree to which high academic
achievement, serious interest in ideas, and
emphasis on intellectual speculation are evi
denced .
9
Omnibus Personality Inventory Scales (OPI).
a. Altruism (Am)— The degree to which a person is
trusting, has concern for, and interest in the
welfare of others.
b. Anxiety Level (AL)— The degree to which a person
is affected by circumstances in terms of tense
ness, nervousness, or worry; and to what measure
the results of this retard adjustment to social
environment.
c. Autonomy (Au)— The degree to which students are
measured in the characteristic composed of non
authoritarian attitudes, need for independence,
tolerance of other viewpoints, non-judgmental
and intellectually liberal tendencies.
d. Complexity (Co)— The measure of tolerance for
^Omnibus Personality Inventory— Research Manual
(Berkeley, Calif.: Center for Study of Higher Education,
1962), pp. 31-38.
ambiguity, uncertainty, diversity, and the en
joyment of unusual events, experiences, situa
tions, and ideas.
Estheticism (Es)— The degree to which a person
endorses statements indicating diverse interests
in, as well as appreciation of, artistic matters
and activities. The facets of this interest are
toward painting, sculpturing, music, and include
interests in literature and dramatics.
Impulse Expression (IE)— The assessment of a
general readiness to express impulses and to
seek gratification either in conscious thought
or in overt action^ the exhibiting of active
imagination, the valuing of sensual reactions,
and evidence of thinking and behavior containing
overtones of feelings and fantasies.
Masculinity-Femininity (MF)— The assessing of
differences in attitudes and interests between
men and women. The measurement of adjustment
problems, anxiety, personal adequacy, social in
clination, sensitivity, and emotionality.
Personal Integration (PI)— The degree to which
a person is anxious, disturbed, or alienated j
due to loneliness, rejection, and isolation re
sulting in avoidance of, hostility to, and ag
gression toward others.
Practical Outlook (PO)— The measure of one's
interest in, and valuing of, practical applied
activities related to immediate utility. Au
thoritarianism, conservatism, and non-intellec
tual interests are frequently components of the
practical person.
Religious Orientation (RO)— The assessment of a
person's attitudes about religious beliefs and
practices, how liberal one's views of religion
are or how fundamentalistic they tend to be.
Response Bias (RB)— This assesses a person's
test-taking attitude. The measure reflects the
person's attitude about particular kinds of
questions and the response given relates to
acceptable or unacceptable societal attitudes.
Social Extroversion (SE)— The degree to which a
person's preferred style of relating to people
is seen. Measured here is a person's interest
in being with people, seeking social activities,
and accepting contacts and responsibilities for
17
personal satisfaction,
m. Theoretical Orientation (TO)— This measures in
terest in science and in some scientific activi
ties * including a preference for using the sci
entific method of thinking; it also reflects the
approaching of problems from a logical* analyti
cal* and critical viewpoint,
n. Thinking Introversion (TI)— The degree to which
one reflects a liking for reflective thought and
academic activities. Interest is measured by
the valuing of a broad range of ideas such as
those in literature* art* and philosophy.
Persister (Stayin, Stayer).— A student who remains
in the college; one who has returned for the third year of
study.
Withdrawer (Leaver, Dropout).— A student who* at the
end of his second year* withdraws from his college for what
ever reason.
Delimitations
This study involves thirteen colleges* each of not
more than 2*000 enrollment. Eleven of the participating
colleges are members of the Council for the Advancement of
18 |
Small Colleges (CASC), which is sponsoring the Project under
a grant from the National Institute of Mental Health (NIMH).
The other colleges were selected to add additional dimen
sions to the Project: that of high scholastic achievement
on the one hand, and that of a very structured course of
study as contrasted with other colleges, on the other.
The study is based upon the response to testing of
the entering freshman class of 1965.
All colleges participating in the attrition study
were chosen because of their participation in the Project on
Student Development, which rested on certain agreements be
tween CASC and NIMH. These were:
1. The college agreed to limited financial support
on a specific basis.
2. The college was willing to provide faculty time ;
for testing students.
3. The college agreed to participate in an annual
workshop to discuss the findings.
Limitations
1. The study is not intended to identify person
ality characteristics which will be applicable to all stu
dent withdrawals found in the total academic community, but
19
is limited to students found in small colleges with similar
developmental aims and objectives.
2. The participating colleges range across a con
tinuum of diverse philosophy, from the religious conserva
tive college to the liberal experimental college of this
certain size.
3. The sample includes Protestant, Roman Catholic,
and non-affiliated private colleges. Church-related and
independent colleges are represented^ one Negro college was
also included in order to examine that culture as compared
to others.
Organization of the remainder
of the study
Chapter II is a review of some of the related re
search literature covering all phases of the attrition prob
lem in colleges and universities in the United States. Par
ticular emphasis is placed upon those segments of the liter
ature dealing with student personality characteristics and
college characteristics related to freshmen and sophomores
who do not persist.
Chapter III reports the context in which the study
is carried on, presents the procedural steps, describes the
colleges and their orientation, considers the instruments
20
used, sets forth the hypotheses, and states the assumptions.!
Chapter IV reports the findings. The differences
between the dropouts and stayins are emphasized and the
differences in the college climates are presented. Aptitude
scores between the two groups of students are considered and
generalizations are presented.
Chapter V contains a summary of the study, conclu
sions, and recommendations.
CHAPTER II
REVIEW OF THE RELATED LITERATURE
Introduction
This study of dropouts and stayins examined the
personality differences between these two groups and sought
to identify what student personality types tend to withdraw
from what kinds of college environments. An investigation
of leavers' personality characteristics and related college
climates required a review of the existing literature on
student attrition dealing with these variables in order to
clarify the state of the research in these particular areas
of student withdrawal and to determine, in some measure, the;
direction of this investigation.
The extensive literature on college dropouts dates
back some forty years, and thus the task of review becomes
almost insurmountable if one decides to examine all the
literature available. Consequently, a sampling of these
data was taken mainly from publications of recent years.
21
22
However, some reference is made here to older studies in
order to give perspective to those articles and books of
more recent origin. The attached reference list gives cre
dence to the extensive research completed on attrition which
is typical of the literature, listing more than thirty-five
studies covering a wide range of research designs and types.
The literature gives some insight into this almost
exclusively American phenomenon of concern. On occasion one
comes across references to attrition in other countries, and
is left with the idea that dropping out of college elsewhere
in the world is not of particular concern to societies where
it is an accepted pattern of life, or where, because of the
sense of responsibility of the educational institutions of
a nation, it is almost not allowed. In the case of Europe,
dropout is acceptable, probably because of the system of
education which relegates the average and poorer student
automatically to a vocational education and thus to an
occupational life.'*' In the case of Japan, attrition in
college is not even considered because dropping out by the
student reflects a failure on the part of the institution
■^L. S. Kubie, "The Ontogeny of the Dropout Problem,"
in The College Dropout and the Utilization of Talent, ed. by
L. Pervin (Princeton, N. J.: Princeton University Press,
1966), p. 25.
23
2
in its responsibility to educate all students enrolled.
National Studies
In the United States, our concern dates back to the
early part of this century, and culminated early in the
McNeeley study under the auspices of the Department of Edu
cation, which studied over 15,000 students who entered
twenty-five universities in 1931, and which noted that over
62 per cent of this number withdrew during the next four
years. Previous to this, many studies in the 1920's and
1930's reflected the growing concern and, no doubt, gave
impetus to the government decision to examine the issue.
In this large analysis of dropouts the author concluded
that, excluding transfers from the 62 per cent already
noted, over 45 per cent of those entering were lost to
3
higher education.
During the 1940's again many studies from colleges
and universities and educational agencies indicated the
deepening interest in the problem of college attrition,
2 *
Edgar Levenson and Martin Kohn, "A Demonstration
Clinic for College Dropouts," Journal of the American Health
Association, XII (1964), 388.
^J. H. McNeeley, "College Student Mortality," U.S.
Department of Interior Bulletin No. 11, 1937.
24
probably more because of practical administrative concern
than from concern over student or college weaknesses. Dur
ing the war years fewer studies appeared, interest was sus
tained, but with the influx of veterans, a somewhat smaller
amount of attrition research was published.
However, in 1950 the Office of Education, under
again increasing interest, instituted a monumental study
under Iffert's direction based upon a sampling of approxi
mately 13,000 men and women who had enrolled in colleges and
universities in the fall of that year. All types of insti
tutions (147) were represented: universities, technical
institutes, liberal arts colleges, teachers colleges, and
junior colleges. The study showed that slightly less than
40 per cent of the freshman class remained at the institu
tion of first enrollment to graduate four years later. An
additional 20 per cent graduated later from the first col
lege or graduated from another institution in four years.
Therefore, about 60 per cent of entering freshmen eventually
received degrees. This study, which has stood as a landmark
ever since, dealt mainly with reported reasons for leaving,
but also elicited impressions of students concerning other
aspects of the college organization, services, and climate.
One of the main findings dealt with the fact that the first
year of college is the most critical period for dropouts,
showing that fully 28 per cent of entering students leave
4
during this time. In many ways the findings of this study
supported the earlier government study by McNeeley.
General Studies on Attrition
These two large studies set the stage for, and gave
an understanding of, the problem and presented a fairly
clear picture of the size of college attrition for the na
tion as a whole. This approach was needed, for it is
against such a background that individual colleges, in some
measure, can assess their own performance and seek to iden
tify their own particular problems, relating their attrition
rates to a national norm. However, such averages do not
solve the difficulties of individual institutions, because
there are as many variations in problems as there are insti
tutions ; and local and institutional factors always affect
a particular rate of attrition. This is clearly pointed out
by Nelson in a study made recently in which it was shown
that as many as fifteen college variables can affect the
attrition rates of institutions without any consideration
4r. e. Iffert, "Retention and Withdrawal of College
Students," Office of Education Bulletin (Washington, D. C.:
Government Printing Office, 1958), pp. 99-109. ■
26
5
being given to student input or output variables.
In Suiranerskill1 s review of the literature the fol
lowing facts about attrition were discussed. He states that
"age of the student per se does not affect attrition al
though older graduates may encounter more obstacles to
graduation." He also indicates that there is little sex
difference in terms of the rates of attrition; however, the
reasons for withdrawal are necessarily different as between
men and women. He found that socioeconomic background does
affect the adjustment to the college environment and, con-
. . 7
sequently, is a factor m affecting attrition rates.
Farnsworth showed that success in college was re
lated to the educational level and the parental attitude
toward intellectual matters, by giving evidence that where
parents prize education, everything else being equal, the
8
students also will prize academic achievement. This point
C
G. A. Nelson, "College Characteristics Associated
with Freshman Attrition," Personnel and Guidance Journal,
XLIV (June, 1966), 1049-1050.
C.
Suiranerskill, "Dropouts from College," p. 631.
^Ibid., p. 632.
®D. S. Farnsworth, "Some Non-academic Causes of
Success and Failure in College Students," College Admis
sions , II (1955), 74.
27
is supported by a recent study of engineering students by
Greenfield where interest or disinterest of parents in the
9
students 1 study affected the leaving rate. It is well
known that Jewish families are particularly conscious of
educational advancement, as Summerskill points out in his
study, and thus there may be ethnic influences which also
encourage persistence in the academic endeavor.^
In many instances the cultural and educational
climate of the hometowns of students has been studied in an
attempt to identify reasons why certain students fail to
persist. Results of these studies seem to show that the
level of educational opportunity and the interest in cul
tural advancement in these communities in some fashion re
late to college success. It has been seen in one study by
Strang that students from rural homes show a higher attri
tion rate than those from more cosmopolitan areas.^ These
are difficult variables to isolate, however, and do not
g
Lois B. Greenfield, "Attrition among First Semes
ter Engineering Freshmen," Personnel and Guidance Journal.
LII (June, 1964), 1008.
■^Summerskill, "Dropouts from College," p. 632.
■^•'■Ruth Strang, Behavior and Background of Students
in College and Secondary School (New York: Harper Bros.,
1937) .
28
allow for any measure of assurance that indeed the condition
is directly responsible for attrition. But it may be a fac
tor .
Socioeconomic factors have been examined extensively
in terms of the search for reasons for non-persistence.
These studies have been second in number only to the exami
nation of academic achievement in all the studies examined.
On the local and national level many studies have delved
into the problems of student achievement in an effort to
arrive at some method of prediction of college success.
There is a consensus among the writers that high school
grades are definitely related to college attrition. Where
it has been shown that students were successful on the
secondary level, the pattern of success tends to be contin
ued on into the higher educational environment. Of course,
exceptions do occur, and later in this chapter these will be
discussed by examining other factors which seem to work
against the success of those who, previous to their college
experience, were adequate students. Summerskill, in citing
Iffert, states:
It is possible substantially to reduce high attrition
rates by simply raising college admissions require
ments with respect to secondary school grades. Iffert
found that the attrition rate for men attending twenty
29
colleges and universities would have been reduced
from 61.2 per cent to 43.9 per cent if admissions
had been confined to the top fifth of high school
graduating classes. He also concluded that: "The
percentages . . . seem to show that standing in high
school graduating class was a much better indicator
of the probability of graduation than standing in
the placement tests.
Abel, in discussing the college student who is cer
tain of his vocational and academic goals, implies that most
certainty comes from previous successes related to academic
achievement; therefore, when a student is certain of what he
wishes to do, and his grade-point average is above 2.0, he
is more likely to persist in college. On the other hand, a
student who is also certain but who is performing below a
2.0 average will withdraw at a rate twice that of the higher
achieving student.^ The findings of Gadzella and Bentall
reveal that high school grade point average was the only
single variable which indicated a significant mean differ
ence between the college graduate and the college dropout,
whether male of female. They state clearly that this is the
14
best single source of data which are predictive.
■^"Dropouts from College," p. 634.
13W. H. Abel, "Attrition and the Student Who Is
Certain," Personnel and Guidance Journal. XLIV (June, 1966),
1044.
■^B. M. Gadzella and G. Bentall, "Differences in ;
30
Little found that in terms of high school rank 78
per cent of dropouts ranked in the lowest 30 per cent of
their high school graduating class as compared with 37 per
cent of dropouts who were recorded as being in the highest
30 per cent. He also found a proportionately higher number
15
of dropouts ranking in the lower third of mental tests.
Therefore, the evidence mounts that at the present
time the only real predictor of college success rests upon
our knowledge of past performance. We can lament this
paucity of information, because we have the uneasy feeling
that many other factors are involved; one thing we can be
certain of is that other factors are indeed an influence.
The confusion of students and their grasping at reasons for
leaving college can be recognized readily, and from this,
as many writers indicate, multicausality is the obvious
conclusion. Without recognizing the other significant com
ponents relating to college attrition, we cannot rightly
prescribe remedies. The outward manifestation of the
High School Academic Achievements and Mental Abilities of
College Graduates and College Dropouts," College and Uni
versity, XLII (Spring, 1967), 355.
K. Little, "The Persistence of Academically
Talented Youth in University Studies," Educational Record,
XL (June, 1959), 237-241.
31
seeming weakness in our system, as seen primarily in low
achievement of some students, cannot be used solely to ex
plain the problem of attrition.
Again, Summerskill discovered that fully 33 per cent
of college dropouts are due to poor grades and academic
failure. He cites twenty-three studies, of which the above
percentage is an average. ^ Again and again studies show
the college GPA as a factor in withdrawal: Barger and
17 18 19
Hall, Carlson and Wegner, Forrest, Weintraub and
20
Salley, suggest that not only poorer achievement is the
cause but that other factors, especially among high ability
students, are additional causes of attrition. Howevex1 size
able the academic problem is, Summerskill suggests that
"motivational and adjustment problems have been ascribed to
"Dropouts from College," p. 636.
-^B. Barger and E. Hall, "Personality Patterns and
Achievement in College," Educational and Psychological
Measurement, XXIV (Summer, 1964), 341-342.
^®S. J. Carlson and K. W. Wegner, "College Drop
outs," Phi Delta Kappan, XLVI (March, 1965), 326.
■^D. S. Forrest, "High School Underachievers in
College," Journal of Educational Research, DXI (December,
1967), 149.
G. Weintraub and R. E. Salley, "Graduation
Prospects of an Entering Freshman," Journal of Educational
Research, XXXIX (October, 1945), 126.
32
21
an even larger number," and quotes Farnsworth, who says:
"No reliable statistics are available as to how many
of those who leave college do so because of emotional
difficulties or conflicts. However, we have good
reason to believe that in some institutions the pro-
22
portion is considerably more than half."
However, Summerskill says that clinicians tend to ascribe
more weight to this than actually appears. From his analy
sis, 10 to 15 per cent of dropouts report that personal ad-
23
justment problems are involved in their leaving school.
Nevertheless, it is suspected that there may be a difference
between the self-report of students and the actual cause of
withdrawal; consequently, more subtle approaches must be
sought in order to ascertain the influence on attrition of
these more obscure forces in the student, and this investi
gation is one attempt to clarify these problems.
To this point an attempt has been made to show the
state of college attrition and to identify the academic
variables associated with non-persistence in college. Now
let us examine those areas which are at present eliciting
more interest than heretofore, i.e., the areas of student
personality and college characteristics.
^"Dropouts from College," p. 643.
22Ibid. 23Ibid.. p. 644.
Personality Studies of Dropouts
Some studies have, in this investigation, come to
light as bases for developing some conclusions about the
personality characteristics of college dropouts, and al
though not conclusive, they direct attention to certain
peculiarities which seem to militate against college per
sistence. One such study by Suczek and Alfert is of par
ticular interest. This study of 1621 entering freshmen was
carried on at Berkeley in cooperation with the Office of
Education (DHEW). This investigation compared all dropouts
with stayins and the authors found that male leavers when
compared with male persisters were seen as more independent
and rebellious, while stayins were more cautious and duti
ful .
Among females, dropouts were more impulsive, while
stayins were more conforming. Failing male dropouts were
compared with male dropouts in good academic standing, and
it was evident that failures were less controlled and rigid,
while female dropouts in good standing were more controlled
and flexible. The failing females in this category were
least mature of all groups sthdied, and had the least devel
oped personality, while female leavers in good standing had
a higher impulse level, and were the most complex
34
individuals of all groups tested.
In examining transfers, non-transfer dropouts, and
returnees, the authors found that transfers are much like
stayins, showing less maturity than either the non-transfer
dropouts or the returnees; they are more conservative, con
ventional, compliant to authority, task oriented, and ambi
tious. The non-transfer dropouts were seen as valuing sen
sations more, enjoying fantasy and imagination, and moti
vated by rebelliousness. The returnees were seen as the
most mature of all groups as measured by the tests. They
were complex, flexible, realistic, tolerant, adventurous,
imaginative, and valued intellectual and esthetic pur-
. . 24
suits.
This last observation brings to mind a statement by
Dalrymple, who after much study about leavers suggested that
leaving college possibly was not the horrible tragedy often
described. He said: "Education is a preparation for life,
25
yes; but life should also be a preparation for education,"
R. F. Suczek and E. Alfert, Personality Charac
teristics of College Dropouts, Research Report, Cooperative
Research, Office of Education and University of California,
Berkeley, Project No. 5-82 32, 1966, pp. 37-42.
2^W. Dalrymple, "The College Dropout Phenomenon,"
NEA Journal. LVI (April, 1967), 12.
35
26
agreeing with Kubie, who presents the same idea. The im
plication is that maybe the returnees have, outside the edu
cational institution, found some maturing influences which
make them more ready and mature, willing to accept what the
college or university has to offer.
Professional school transfers from Berkeley were
described by Suczek and Alfert as inhibited, dependent on
authority and convention, and intolerant of differences and
ambiguities, implying that narrowness, or the desire for
narrowness in education has a stultifying effect upon per
sonality as measured by the OPI. The general results of the
college experience are limited, according to this study, by
the frequent conflicts involving establishment of autonomy
in the student, and dropping out is a frequent result of
this stress. The authors further implied that for many
students, rebelliousness against externally imposed values
was a factor in withdrawal involving the need to provide a
27
time-space for reassessment of directions and values.
Other studies consider the same and other variables
with the resultant developing of patterns relating to
26
"Ontogeny of the Dropout Problem," p. 18.
^Suczek and Alfert, "Personality Characteristics of
College Dropouts," p. 37.
36
personality characteristics. Aiken discovered that aptitude
scores, when related to leavers and stayers, did not dis
criminate between them, while from biographical data he con
cluded that lack of motivation for academic matters was the
28
main reason. In a study involving the Minnesota Multi-
phasic Personality Inventory (MMPI), Barger and Hall found
that higher scores on the masculinity scale were identified
among both men and women who left college. These leavers
were seen as rebellious, non-conformist, and high-activity-
oriented persons, indicating that motivational elements were
related here. With these orientations the trait of quiet,
scholarly study was missing, implying lack of motivation and
29
inability to solve academic problems. In using the Minne
sota Counseling Inventory (MCI), Brown discovered that women
dropouts were withdrawn, introverted, depressed, and tended
toward social isolation, while men dropouts tended to be
30
irresponsible and non-conforming. Dalrymple further
2®L. r. Aiken, "The Prediction of Academic Success
and Early Attrition by Means of a Multiple Choice Biographi
cal Inventory,1 1 American Educational Research Journal. I
(1964), 134-135.
2®Barger and Hall, "Personality Patterns and
Achievement in College," pp. 339-346.
G. Brown, "Identifying College Dropouts with
the Minnesota Counseling Inventory," Personnel and Guidance
Journal. XXXIX (December, 1960), 280-282.
. 37
states that:
Psychological factors' which cause students to lose
motivation, quarrel with teachers or contemporaries,
or crack up, usually afflict students who have also
suffered personal dissatisfactions, academic diffi
culties, physical symptoms, disciplinary problems,
or some combination of these.31
Using the Gordon Personal Profile (GPP) and the
Gordon Personality Inventory (GPI), Daniel found that per-
sisters were more trustful and tolerant, perseverant, in
quiring, energetic, vigorous, calm and collected, and cau
tious . In this same study, leavers were seen as less per-
32
severing and inquiring. Grace discovered from his inves
tigation that withdrawers were dependent and irresponsible,
disagreeing with others who, as previously stated, saw in-
33
dependence as a prime trait in dropouts. Heilbrun, at the
University of Iowa, found that in high ability groups for
both men and women, dropping out was associated with more
assertive, less passive social behavior. He states that
31i'The College Dropout Phenomenon," pp. 11-13.
■^Kathryn b . Daniel, "A Study of College Dropouts
with Respect to Academic and Personality Variables," Journal
of Educational Research. LX (January, 1967), 230-235.
3%. Grace, "Personality Factors and College Attri
tion," Peabody Journal of Education. XXXV (June, 1957), 36-
40.
38
persistence relates to students in this ability group who
are more passive and task-oriented and who are influenced by
conformity to institutional values. The leavers, described
as more assertive and less task-oriented, encounter great
difficulty in value conformance and are more likely to leave
college prior to the second year. The author further ex
plains :
The greater academic and social regimentation imposed
upon incoming students of large universities is a
greater source of frustration to the bright student
than to those of lesser ability, and involves former
rewarding of intellectual pursuits [at lower levels].
Heilbrun, in a study of female freshmen using the
Adjective Check List Need Scales (ACL), suggests that it is
the relative absence or denial of certain needs that is cru
cial in implementing or deterring the female student's con
tinuation in school. He describes these students as higher
on heterosexuality and change and lower on achievement,
order, and endurance.
Another study by Hill involving high-aptitude
^A. B. Heilbrun Jr., "Personality Factors in Col
lege Dropout," Journal of Applied Psychology, XLIX (Febru
ary, 1965), 1-7.
■^A. B. Heilbrun Jr., "Prediction of First Year Col
lege Dropout, Using ACL Need Scales," Journal of Counseling
Psychology, IX (Spring, 1962), 58-63.
39
students* in part, deals with personality traits. His sam
ple was divided into three groups: graduates (G), voluntary
withdrawals (VW), and enforced withdrawals (EW), which were
then examined for differences in psychometric information.
He found that creative ability was lowest in the VW group,
highest among the G groupj strangely, the EW group rated
higher in creativity than did the VW's. In personality, the
only significant differences found were that G males were
greater socializers, and that females of the G group were
highest in achievement via independence. A significant
feature of this study was expressed by the author thus:
"Among the better predictors of withdrawal, simply asking
the student about his chances of graduation was highly sig
nificant. 1,36
Marks suggests the same idea, saying "Simply ask
37
him." The figures in Hill's study showed that the VW
group were much less certain of graduation than were either
the G's or the EW's.
•^A. H. Hill, "A Longitudinal Study of Attrition
among High Aptitude College Students," Journal of Educa
tional Research, LX (December, 1966), 166-173.
■^E. Marks, "Student Perceptions of College Per
sistence, and Their Intellective Personality and Performance
Correlates," Journal of Educational Psychology, LVIII
(August, 1967), 210-221.
40
A study by Jensen, again using the MMPI, indicated
that non-achieving students consistently differed from stu
dents of high ability on six of the eleven scales. It was
seen that generally low-scholastic-ability non-achievers
encountered more adjustment problems than other students.
Gifted students expressed themselves as having fewer ad
justment problems; this implies that greater scholastic
38
ability favors adjustment while low ability obstructs it.
In a study using the OPI made by Rose and Elton,
dropouts were clearly seen as distinct in personality traits
from other groups. The authors discovered that leavers were
the most maladjusted, had least interest in literature, art,
and philosophy, were illogical, irrational, and uncritical,
39
and disliked reflective and abstract thought.
One study involving freshmen dropouts revealed that
traits such as greater estheticism (interest in artistic
matters), less maturity (anxiety, hostility, and aggres
sion), greater independence, and irresponsibility militate
38
V. H. Jensen, "Influence of Personality Traits on
Academic Success," Personnel and Guidance Journal, XXXVI
(March, 1958), 497-500.
•^H. A. Rose and C. j. Elton, "Another Look at the
College Dropout," Journal of Counseling Education. XIII
(Summer, 1966), 242-245.
41 |
40
against persistence in the college environment.
Summary— Personality studies
of dropouts
These studies reflect the state of the research
efforts in the understanding of personality differentials of
college withdrawals at the present time. It is evident that
much disarray exists and much research must be done in order
to sort out the various forces working against college per
sistence. Realizing that no one study can adequately pro
vide that understanding, these many efforts are seen as a
start at bringing into focus the more subtle forces within
the student which tend to move him toward his decision to
leave the academic community for what appear to him to be
better and greener fields elsewhere. It is implied in these
studies that the students are confused and, in many cases,
unaware of the forces within themselves. They can see and,
in many cases, use external reasons for their decisions, but
how many indeed are aware of the internal pressures and un
conscious elements of their natures which drive them to seek
satisfaction away from the influence of academic striving?
^William Hannah, "Freshman Leavers and Stayers"
(paper presented at the 19th Annual Meeting of the Vermont
Psychological Association, April, 1967, Bolton, Vermont),
pp. 12-13.
42
It is hoped that this investigation may, in some measure,
clarify some of these forces and alleviate some of the con
fusion evident in the research.
College Environment Studies
of Attrition
One additional area of study with which this review
is concerned is that of college characteristics. Farnsworth
states:
The entire college atmosphere, both intellectual and
social, is different on a campus where almost the
entire student body will graduate, from that on a
campus where only a minority of students will gradu
ate .41
This has great implications, especially for those
colleges with large attrition rates. A transiency existing
and, more serious, a sense of transiency of a group, gener
ates instability and uncertainty affecting both those who
persist and those who do not. Sanford further emphasized
this when he said,
[Research] might obtain evidence that the phenomenon
[attrition] had less to do with factors in the student
than with a certain condition in the college itself,
4-*- d . S. Farnsworth, D. Funkenstein, and B. Wedge,
"Study of the Social and Emotional Adjustment of 'Early
Admission1 College Students," Report for the Fund for Ad
vancement of Education, 1955. (Mimeographed.)
43
and this condition might immediately assume greater
practical importance than withdrawal, because it was
now perceived as something that affected all the stu
dents .
Nelson's study, already cited, does not agree that
conditions in the college militate against persistence. His
study dealt with practical and tangible variables such as
library, faculty, size of college, size of local community,
selectivity, and affluence, but did not deal with other more
subtle variables measurable by use of validated instru
ments
Research people and administrators in colleges have
often felt that climate within the college could effect
changes in students, and have emphasized these feelings.
The question might be asked, "If there are positive effects,
is it not possible that unknowingly we might be fostering
negative effects as well?" It would seem that this is true,
as Dalrymple notes:
Colleges differ in their characteristics just as in
dividuals do. Some are authoritarian, others per
missive; some specific in their pedagogical approach,
others abstract; some overly paternalistic, others
42N. Sanford, "Personality Development during the
College Years," Journal of Social Issues. XII (1956).
42"College Characteristics Associated with Freshmen
Attrition," pp. 1046-1050.
44
excessively impersonal; some sectarian, others secu
lar. Under these circumstances, the student will
either fit or not fit the college or a particular
part of it. He may be stimulated, lulled, bored,
frustrated, angered, or nurtured by the environment
he finds and by its interaction with his personality
and talents.44
A few studies have attempted to assess these fac
tors, but very few. Already cited is Heilbrun's study in
which he speaks of regimentation and rigidity fostering
45
dropouts in the high-ability group of students. Iffert
has noted that counseling and guidance services for most
46
students are unproductive of value to the student. Holmes
states that criticism by students of the counseling services
47
was second only to criticism of food on campus. The
writer, in studying attitudes among members of the sophomore
class who were classified as potential dropouts, discovered
minimal participation between potential leavers and college
personnel, coupled with low value ratings of the college
counseling programs in terms of decisions to withdraw. In
^4"The College Dropout Phenomenon," p. 13.
^"Personality Factors in College Dropout," pp. 1-7.
4®"Study of College Student Retentions and With
drawals," pp. 106-107.
4?C. H. Holmes, "Why They Left College," College and
University, XXXIV (Spring, 1959), 295-300.
45
the same study it was seen that restrictive rules and regu
lations and unstimulating activities such as religious ser
vices were high on the list of criticism. "As students
leave, to whom do they go for advice?" The writer dis
covered that first, friends are contacted and discussions
ensue between both friends of the same and opposite sex;
second, students discuss with and listen to parents; fi
nally, almost after the decision has been solidified, the
college personnel are contacted— possibly only as a formal
ity— before the final exit is made. Harsh as this may
sound, little value seems to derive from these college
48
counseling systems. Johnson observes that proper coun
seling could remedy the attrition problem, and by implica
tion it could be said, if it were only instituted ade-
49
quately. Panos and Astm suggest that
Students are more likely to complete four years if
they attend colleges where student peer relationships
are characterized by friendliness, cooperativeness,
and independence; where the students frequently can
participate in college activities, where there is a
high level of involvement with, and concern for, the
individual student and where administrative policies
/O
W. Hannah, "The Withdrawal Process," unpublished
article.
^G. B. Johnson, "A Proposed Technique for Analysis
of Dropouts at a State College," Journal of Educational
Research. XLVII (January, 1954), 381-387.
46
concerning student aggression are relatively permis
sive. 50
Chickering has defined well the relationship between
institutional differences and characteristics of students
and implies that homogeneity of student bodies may foster
51
development in less stressful ways.
Summary— College environment
studies of attrition
The recorded research suggests that much needs to be
done in clarifying the variables relating to college attri
tion, and that more needs to be known about the students1
dispositions, personality characteristics, and influence of
college characteristics. These studies indicate an added
interest in the students' influence in the academic setting,
and the colleges1 responsibility in providing a proper cli
mate in which successful achievement can be made— this in
terms of both quantity and quality. The literature perused
also suggests that more needs to be known about fitting the
proper student to the proper environment in order to
j . Panos and A- W. Astin, "Attrition among
College Students," American Council of Education Research
Report, II (1967), 12.
■^A. W. Chickering, "Institutional Differences and
Student Characteristics," Journal of American College Health
Association, December, 1966, p. 181. j
47 i
maximize the benefit he is to receive from his experience in
college. However, the question of homogeneous college popu-;
lations and resultant benefit is still an open question. No
available data have measured whether "oneness of mind" in
certain groupings is better than "differences of opinion"
when the consideration is college attrition. This study
will seek to identify differences between those who persist
and those who do not, and relate these data to character
istics of colleges of different types. Certainly the fore
going studies show that these areas of concern need con
tinued study, and set the background in which this investi
gation has been carried through.
CHAPTER III
THE PROCEDURE
The Project
It was hypothesized for purposes of statistical
analysis that differences in personality between leavers and
stayers would be zero or near zero, and it was assumed that
college climates could be distinguished in thirteen small
private colleges in this study of student personality char
acteristics and college climate variables.
The study was conducted within the context of the
Project on Student Development under the auspices of the
1
Council for the Advancement of Small Colleges (CASC), funded,
by the National Institute of Mental Health.'*' Although the
staff of the Project is involved in the total endeavor, the
present writer has the responsibility of coordinating that
i
segment of the Project dealing exclusively with the dropout
1NIMH Project No. MH14780-04.
48
49
phenomenonj consequently, his responsibilities are exclu
sively with gathering, coding, and analyzing those data
dealing with college attrition involving the original sam
ple .
The Colleges
These data are compiled from thirteen different
colleges throughout the country; eleven are members of CASC;
two were chosen because of a desire to add additional dimen
sions to the Project, that of high scholastic achievement on
the part of one (College B.) and that of a very structured
course of study as contrasted with other colleges on the
part of the other (College K).
The participating colleges are located in thirteen
different states across the nation; they are identified in
this study by the letters A through M. All colleges par
ticipating in the Project and in the attrition study were
chosen because of their willingness to agree to certain
criteria based upon the concept of cooperative effort (see
Chapter I).
As described in Chapter I, the colleges range across
a continuum of diverse philosophy— from the religiously con
servative college with structured social mores to the
50
liberal experimental college with few social restrictions
and regulations.
The study is not intended to identify personality or
collegiate characteristics which will be applicable to all
student withdrawals found in the total academic community,
but is limited to describing students found in small private
colleges where similar developmental aims and objectives
exist.
The Sample and the Instruments
The Project began in September, 1965 with the ad
ministration of the instruments by the Institutional Repre
sentative of each participating college. The original sam
ple consisted of 3,536 students representing all the enter
ing freshmen of the thirteen colleges in the fall of that
year to whom the OPI was given. Because of various reasons,
such as insufficient preliminary information, college errors,
and incomplete responses, unusable instruments reduced the
original number to 2,874. Of these 2,874, there were 1,668
males and 1,206 females.
A second sample was selected randomly from the whole
student body of each college consisting of 100 students
across all four class levels from each college for the
51
purpose of identifying student attitudes about the colleges
themselves. The assumption was that students who had ex
perienced the college environment could more readily assess
its climate and determine its characteristics than could
entering students. To these the CUES was administered in
the spring of 1966.
The data were gathered from the following sources:
1. The Omnibus Personality Inventory provided a
measure across fourteen different scales identifying student
personality characteristics.
2. The College and University Environment Scales
provided the data concerning information relating to college
climate and characteristics.
3. Aptitude scores, either SAT or ACT, of all en
tering freshmen compiled from existing college records.
The OPI is an experimental instrument developed at
the Center for the Study of Higher Education at Berkeley.
The development of the OPI has been in progress since 1957
as a new personality inventory and builds on previous in
struments, but has been validated through extensive effort
over many years and with many students and colleges. Raw
scores (the number of items answered in the keyed direction)
have been converted to standard scores by using the table
provided in the OPI manual. The OPI has been developed
because of the authors1 belief that existing inventories
were inadequate at the time, permitting only the simplest
and most superficial distinctions among groups of students
in different institutions. Little was known, according to
the authors, about what seemed to be the more important
differentiations, for example, the varied motivational char
acteristics of students or the variations in academic at
mosphere within college cultures and subcultures.
CUES is a well-known commercial instrument developed
out of Stern's College Characteristics Index (CCI) . The
author, Robert Pace, a former colleague of Stern, decided
to reorient the original instrument from a psychological
structure to one which he felt would more accurately de
scribe a college. The structure of CUES is more educa
tionally oriented than is the CCI. The CUES are designed
to describe the institutional climate and consist of 150
statements about college life, college facilities, and
college rules and regulations, as well as faculty, curric
ula, instruction, examinations, and student life. By indi
cating which statements are generally characteristic of
their college, students act as reporters; their aggregate
judgment provides an opinion poll which reveals
53
characteristics of their college defining the intellectual-
social -cultural climate as they see it. Institutions, not
individuals, are scored along the five dimensions noted pre
viously. Scoring on a single item is represented by a sin
gle point for each item marked in the keyed direction by
more than two-thirds of the members of a particular college
sample.
In order to assess rightly the relative values of
SAT and ACT scores among the colleges, a conversion table
was used to convert one of the scorings into the equivalent
of the other to provide a common index. The decision
needed to be made as to which common index would be best.
It was decided, in view of the preponderance of SAT scores
received from the colleges, that the most reasonable con
version would be to convert ACT, received from a few insti
tutions, to SAT equivalents. This conversion is based upon
"A Table of Concordance between ACT and SAT," published by
2
C. I. Chase and L. S. Barrett. The measure resulting from
this table is given in terms of total verbal and quantita
tive scores combined. However, the fact that combined
scores are used in certain instances to measure aptitude was
^Journal of College Student Personnel, March, 1966,
pp. 105-108.
54
considered adequate for purposes of this study. It is
realized that this procedure imposes a limitation on any
conclusions to be derived, but the measure of limitation was
not deemed great enough to warrant exclusion of this par
ticular step in the procedure.
After the initial testing in 1965 it was necessary
to wait until the identification of the dropout could be
made. During the period between then and the fall of 1966
an Institutional Classification Sheet was formulated to be
sent to the Registrars of the thirteen colleges for purposes
of identifying those who withdrew during the first year.
All students who did not enroll in September, 1966 were re
corded on these reports by the college records offices.
Each student's name, age, SAT or ACT score, and classifica
tion as to reason for withdrawal were recorded along with
other pertinent data for later use in our study. These
sheets were then used to analyze for numbers and other data.
This process was again followed in the fall of 1967, with
the resulting information which indicated that in two years
1,212 students had decided not to return to study at the
college of original enrollment. Of these there were 715
men and 497 women.
Data from the instruments used in this study were
55
coded and recorded on IBM punch cards to facilitate the
statistical analysis. The resulting 2,874 sets of data were
then, in 1967, separated into two distinct groupings, one
dropouts, one stayins. The resulting n's were 1,212 leavers
and 1,662 stayers. This was further broken down into the
following:
Groups N1s
Dropouts
Males 715
Females 497
Stayers
Males 953
Females 709
The study had progressed to the point of handling
the data statistically.
The Hypotheses
It was hypothesized, for purposes of analysis, that
on the OPI no differences would appear:
1. Between personality characteristics of those
students who leave college and personality characteristics
of those students who stay in college.
2. Between the personality characteristics of males
who leave college and the personality characteristics of
56
males who stay in college.
3. Between the personality characteristics of
females who leave college and the personality characteris
tics of females who stay in college.
4. Between the level of aptitude of those who leave
college and the level of aptitude of those who stay in col
lege .
Treatment of the Data
The hypotheses were examined by the use of t tests
for independent groups (leavers and stayers) across all
colleges participating in the study. The .01 and .05 level
of confidence was used for a criterion of support. This
statistical test was used for those data deriving from the
OPI and the aptitude scores, and identifies significant
differences between the two groups. The student t was used
because of its adaptability to the small group samples used
throughout this study for individual colleges, and because
as sample size increases, the t values approach the values
of the z test which is normally used for large samples. The
writer felt that the test used was adequate and reasonable.
The final determination of use was made after consultation
with statisticians both at the Project office and at the
57
computer center of the University of Southern California.
The scores for colleges described previously in the
discussion of the CUES instrument were handled by use of
graphs which showed fairly similar patterns of college cli
mates, and allowed the division of colleges into types so
that individual analysis of these types could be made with
reference to the OPI data. The analysis of the relationship
between these data from OPI and CUES was done by inspection
only. The writer felt that the internal validity of the
CUES was sufficiently prescribed by its authors that little
value would be derived from statistical analysis between
instruments. The rationale followed the professional judg
ment of the writer that these were two different instru
ments, measuring two different sets of characteristics, and
using two different samples of students, and that there was
no real way of assessing any relationship between the two.
The writer also believes that unless the same sample is used
involving both instruments, little is to be gained from a
statistical treatment between the two. The CUES were merely
intended to provide a backdrop for the OPI data being ana
lyzed.
CHAPTER IV
THE FINDINGS
Dropout-Stayin Differences
It was predicted that no differences between leavers
and stayers would be found in any of the variables across
the scales of the OPI and that no differences would appear
in aptitude scores between the two groups being studied. It
was also assumed that* if differences did appear., college
climates would be different and that certain conclusions
could be formed from the resulting data. This last assump
tion was based upon the differences in philosophy and in
academic and social programs among the thirteen diverse
colleges involved in this investigation.
Hypothesis 1
Hypothesis 1 stated that no differences will appear
between personality characteristics of those students who
leave and personality characteristics of those students who
58
59
stay in college.
The hypothesis was tested through use of t tests of
differences between independent groups, and Table 1 shows
the mean standard scores of the two groups on fourteen
scales of the OPI.
This hypothesis is not confirmed for six of the
fourteen scales. Differences, beyond the .01 level, do
appear in the complexity, impulse expression, personal in
tegration, anxiety level, altruism, and response bias
scales. The evidence indicates that dropouts, compared with
persisters, are more complex, more impulsive, more anxious,
less integrated personally, less altruistic, and less will
ing to exert an effort to make a good appearance either
before their peers or their teachers.
Hypothesis 2
Hypothesis 2 predicted that no differences will
appear between personality characteristics of males who
leave college and personality characteristics of males who
stay in college.
Again, this hypothesis is not confirmed. The evi
dence in Table 2 shows that the same six of the fourteen
scales show significant differences as in the total group;
60
TABLE 1
DROPOUT-STAYIN OPI MEAN STANDARD SCORES
Scale Dropouts Stayins P Value t Value
(1212) (1662)
TI 46.41 47.08 NS -1.824
TO 45.48 46.08 NS -1.696
ES 48.59 48.99 NS -1.071
CO 50.01 48.70 .01 3.562
AU 47.72 47.26 NS 1.235
RO 48.47 47.98 NS 1.348
IE 53.19 50.45 .01 6.667
SE 47 .52 48.15 NS -1.752
PI 48.06 49.25 .01 -3.278
AL 47.15 48.34 .01 -3.333
AM 47.49 49.00 .01 -4.082
MF 49.66 49.69 NS -0.103
RB 44.88 46.43 .01 -4.525
PO 52.38 52.60 NS -0.637
Note : t value of 1.960 significant at the .05
level, t value of 2.576 significant at the .01 level.
Scores shown for each scale are mean standard scores
for groups as indicated. A score of 50 equals that achieved
by the norms group of 7,283 freshmen reported from the OPI
manual and derived from diverse colleges across the country.
61
MALE DROPOUT-
TABLE 2
STAYIN OPI MEAN STANDARD SCORES
Scale Dropouts Stayins P Value t Value
(715) (953)
TI 45.46 46.18 NS -1.562
TO 46.81 47.59 NS -1.762
ES 46.18 46.52 NS 0.502
CO 50.47 49.43 .05 2.291
AU 47.51 47.83 NS 0.664
RO 49.55 49.48 NS 0.147
IE 56.14 53.31 .01 5.731
SE 47.53 47.39 NS 0.350
PI 48.15 49.63 .01 -3.170
AL 47.51 48.79 .01 -2.755
AM 44.81 46.41 .01 -3.412
MF 54.39 54.44 NS 0.131
RB 45.15 46.79 .01 -3.668
PO 53.18 53.05 NS 0.296
Note: t value of 1.96 significant at the .05 level,
t value of 2.57 significant at the .01 level.
Scores shown for each scale are mean standard scores
for groups as indicated. A score of 50 equals that achieved1
by the norms group of 7,283 freshmen reported from the OPI
manual and derived from diverse colleges across the country.
62
complexity, impulse expression, personal integration,
anxiety level, altruism, and response bias are seen as sig
nificantly different between male dropouts and male stayins
on the OPI.
The same observations can be made for males as for
the whole group as described from the data in Table 1.
Hypothesis 3
Hypothesis 3 predicted that no differences will
appear between the personality characteristics of females
who leave college and the personality characteristics of
females who stay in college.
This hypothesis is not confirmed. As seen in Table
3, differences did appear in seven of the fourteen scales on
the OPI. These differences are in complexity, autonomy,
impulse expression, social extroversion, anxiety level,
altruism, and response bias. The sex differences from male
to female withdrawals are revealed in the scales for auton
omy, social extroversion, and personal integration.
The female dropouts are seen as more complex, in
dependent, impulsive, withdrawn socially, anxious\ less
altruistic and less willing to make good impressions than
are female stayins.
63
TABLE 3
FEMALE DROPOUT—STAYIN OPI MEAN STANDARD SCORES
Scale Dropouts Stayins P Value t Value
(497) (709)
*
TI 47.78 48.29 NS -0.865
TO 43.59 44.04 NS -0.834
ES 52.03 52.41 NS -0.722
CO 49.36 47.72 .01 2.679
AU 48.02 46.48 .05 2.537
RO 46.93 45.97 NS 1.712
IE 48.99 46.64 .01 3.714
SE 47.51 48.74 .05 -2.174
PI 47.93 48.74 NS -1.401
AL 46.64 47.74 .05 -1.976
AM 51.32 52.45 .05 -2.130
MF 42.91 43.37 NS -1.084
RB 44.50 45.95 .01 -2.722
PO 51.26 52.03 NS -1.397
Note: t value of 1.960 significant at the .05
level, t value of 2.576 significant at the .01 level.
Scores shown for each scale are mean standard scores:
for groups as indicated. A score of 50 equals that achieved
by the norms group of 7,283 freshmen reported from the OPI
manual and derived from diverse colleges across the country.
Hypothesis 4
Hypothesis 4 predicted that no differences will
appear between the level of aptitude of those who leave col
lege and the level of aptitude of those who stay in college
as measured by SAT scores or equivalents.
This hypothesis is not confirmed. Those who re
mained in college reflected higher aptitude levels than
those who withdrew. Table 4 shows the differences.
TABLE 4
DROPOUT-STAYIN SAT SCORES AT ENTRANCE
Groupings Dropouts Stayins P Value t Value
(1212) (1662)
Male 981.45 1022.28 .05 -2.162
Female 945.66 990.22 .05 -2.542
Total 963.74 1007.58 .01 -3.382
Note : t value required at .01 level is 2.586. t
value required at .05 level is 1.965.
The evidence suggests that differences in aptitude
do have some relationship to dropping out of college. There
appears to be at least a forty-point difference between
withdrawers and persisters in each statistic shown in the
table, indicating some small measure of lower aptitude in
65
dropouts and therefore a lower level of academic efficiency.
This difference is small and caution should be used in in
terpretation. An examination of the SAT scores or equiva
lents at all colleges individually revealed no significant
differences between the two groups. However, a small and
consistently lower dropout raw score difference was evident
in all the colleges studied. When all those scores were
combined., a statistically significant difference did appear,
as shown in Table 4.
College Campus Characteristics
The second section of this study classifies col
leges, describing their predominant characteristics as seen
by CUES scores, to determine what types of climates militate
against the persistence of certain personality types. The
major question is, "What personality types withdraw from
what types of colleges?"
Appendix Table 1 shows the scores for all colleges
on the CUES scales. These scores are arranged according to
similarity of pattern, and nine of the colleges are trans
ferred to graphs.
Four colleges were not shown in the graphs for the
following reasons:
66
1. Colleges I and L have CUES profiles similar to
those of Pattern I; OPI differences also parallel the dif
ferences found in colleges of this pattern. Therefore,
since these findings support the conclusions in this study,
they were excluded for reasons of clarity on the CUES Pat
tern I graph. Fewer graph lines make for clearer presenta
tion .
2. College G is similar on four of the CUES scales
of Pattern II colleges, but on Propriety its score is well
above the national norm, much higher than the other colleges
in this pattern. This college was atypical of Pattern II
due to the ethnic background of its students.
3. College B is similar to Pattern III colleges on
Practicality, Community, and Awareness, but its scores on
Propriety and Scholarship are far above the national average
and well above the other two schools in this pattern.
The main concern in this part of the study was to
determine, as clearly as possible, distinct college types in
the thirteen colleges. This does not suggest that other
college types do not exist, but simply sharpens through
adequate selection the types readily seen by the data from
the CUES scales.
The nine individual college profiles are also shown
67
on graphs in the Appendix (see Appendix Graphs 1-1 through
III-2). These graphs are then combined according to college
climate types on Graphs I through III shown below.
Determination of college climate type is based upon
the higher or lower institutional score on the particular
scale in comparison with the national norm. The score is
relative to that norm and, therefore, the college climate is
described as being more or less strong in that particular
variable as reported by the majority of students in the in
stitutional sample.
On Graph I are shown the CUES patterns of four col
leges relating their patterns to the national norm. All of
these colleges are religiously conservative institutions.
Their climates are characterized by ( . 1 ) a normal amount of
order and supervision, (2) a high sense of community— a
friendly, supportive, and sympathetic atmosphere, (3) a
lower than average concern for awareness of self, of soci
ety, and of esthetic stimuli, (4) a high sense of propriety,
reflecting an environment which is polite and considerate,
non-demonstrative, and non-assertive, and (5) a slightly
lower than average sense of scholarship. This pattern can
be described as CUES Pattern I, with the following predomi
nant characteristics: a high sense of community, a low
68
GRAPH I
CUES SCORE PATTERN FOR FOUR COLLEGES
(CUES Pattern I)
Practi- Commu- Aware- Propri- Scholar-
cality nity ness ety ship
College A — a— a— a— a— College E — i — i— / —
College C --------------- College F ----------
National average------
Note: See Appendix Graphs 1-1 through 1-4.
69
sense of awareness, and a high sense of propriety.
On Graph II are shown CUES profiles of three col
leges , two commuter-resident institutions and one almost
totally resident college which enrolls students more gener
ally recruited from rural or conservative backgrounds.
These institutions are characterized by (1) an average
amount of order and supervision, (2) an average amount of
community spirit and feeling of group welfare and loyalty,
(3) a very low awareness of self, society, and esthetic
interests, (4) a lower than average level of politeness and
consideration among the students, and (5) a low sense of
scholastic achievement. This pattern can be described as
CUES Pattern II with dominating characteristics as follows:
very low sense of awareness and very low sense of scholar
ship .
On Graph III, two colleges show a CUES profile
decidedly unlike the others. This pattern reflects a con
dition on campuses where there is a highly selective admis
sions policy, a more liberal attitude toward social norms,
and a greater sense of independence accepted and approved by
both students and faculty. This pattern reflects (1) very
low emphasis on order and supervision— doing what is ex
pected is of less importance than on the average campus,
70
GRAPH II
CUES SCORE PATTERN FOR THREE COLLEGES
(CUES Pattern II)
Practi- Commu- Aware- Propri- Scholar-
cality nity ness ety ship
« - * — ■ . « i
24_ _24
22_ _22
20_ _20
18_ _18
16_ _16
14_ _14
12_ _ _12
~ ^ - io
\\
8 - / " ' ' x ' \ - 8 i
/ \ \\ !
6- 7 v \ - 6i
X _4
V \ - 2
0 0
College J ------------ College H -------------
College M — i — i — i — i— National average------
Note: See Appendix Graphs II-l through II-3.
71
GRAPH III
CUES SCORE PATTERN FOR TWO COLLEGES
(CUES Pattern III)
Practi
cality
Commu
nity
Aware
ness
___ i ___
Propri- Scholar-
ety ship
24.
22.
20.
18.
16.
14.
12.
10.
8.
6.
4_
2_
0
/ /
//
'i
_24
22
_20
_18
.16
.14
.12
.10
_ 8
_ 6
_ 4
_ 2
0
College D -------
National average
College K — i— i — i —
Note: See Appendix Graphs III-l and III-2.
72
(2) a high sense of community, indicating a supportive and
sympathetic climate, (3) a very high sense of the awareness
of self, society, and esthetics, (4) low emphasis on pro
priety, suggesting a demonstrative, assertive, convention-
flouting atmosphere and behavior, and (5) a sense of campus
scholarship near the national average. This pattern can be
described as CUES Pattern III, as follows: low sense of
practicality, high sense of community, and high sense of
awareness•
Tables 5, 6, 7, and 8 combine the CUES patterns,
deriving from the graphs, and the OPI scale difference to
tals and pattern averages between the two groups found on
Appendix Table XI. These totals and averages are separated
into four classifications:
1. Those common to all institutions (Table 5).
2. Those which are the most significantly different
for Pattern I (Table 6).
3. Those which are the most significantly different
for Pattern II (Table 7).
4. Those which are the most significantly different
for Pattern III (Table 8).
(The scales in each classification are ranked
according to their importance.) j
TABLE 5
CUES PATTERNS VS. OPI DIFFERENCES— DROPOUTS RELATED TO STAYINS
OPI Scales CUES Patterns
Dropout Differences
Common to All
Patterns
High
High
Low
(4
I
Community
Propriety
Awareness
Colleges)
II
Low Awareness
Low Scholarship
(3 Colleges)
III
High Community
High Awareness
Low Practicality
(2 Colleges)
Total Average Total Average Total Average
Impulse Expression + 19.42 (4) +4.86 + 8.38 (2) + 2.79 +4.87
(0)
+ 2.44
Complexity + 9.66 (2) + 2.42 +4.61
(1)
+ 1.54 +4.60
(1)
+ 2.30
Response Bias -12.04
(1)
-3.01 -2.79
(1)
- .93 -4.51
(1)
-2.26
Autonomy + 5.08
(1)
+1.27 +4.43
(1)
+ 1.48 + 3.75 (0) + 1.88
Note: The figures shown on this table are generated by subtracting the mean
standard scores of stayins from the same scores of dropouts at each college. (See Appendix
Table XI.) A plus (+) sign means that the dropouts' scores are that much higher on the
particular scale than stayins. A minus (-) sign means that the dropouts' scores are that
much lower on the particular scale than stayins. Totals reflect the sum of the differences
for each scale in each pattern, while averages are simply the result of dividing the
totals by the number of colleges in the pattern. The number in the parentheses ( ) de
notes the number of institutions in the pattern where statistical significance on the OPI
scale reached at least the .05 level.
u>
74
TABLE 6
CUES PATTERN I VS. OPI DIFFERENCES
DROPOUTS RELATED TO STAYINS
OPI Scales CUES Pattern I
Distinguishing Personality
Characteristics among
Dropouts in Pattern I
Colleges
High Community
High Propriety
Low Awareness
(4 Colleges)
Total Average
Personal Integration -13.41 (2) -3.35
Altruism -10.69 (2) -2.67
Anxiety Level -10.43 (2) -2.61
Religious Orientation + 9.08 (3) +2.27
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts at each college. (See Appendix
Table XI.) A plus (+ ) sign means that the dropouts' scores
are that much higher on the particular scale than stayins.
A minus (-) sign means that the dropouts' scores are that
much lower on the particular scale than stayins. Totals
reflect the sum of the differences for each scale in the
pattern, while averages are simply the result of dividing
the totals by the number of colleges in the pattern. The
numbers in the parentheses ( ) denotes the number of insti
tutions in the pattern where statistical significance on the
OPI scale reached at least the .05 level.
75
TABLE 7
CUES PATTERN II VS. OPI DIFFERENCES
DROPOUTS RELATED TO STAYINS
OPI Scales CUES Pattern II
Distinguishing Personality
Characteristics among
Dropouts in Pattern II
Colleges
Low Awareness
Low Scholarship
(3 Colleges)
Total Average
Practical Outlook -3.36 (1) -1.12
Theoretical Orientation +3.14 (0) +1.05
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts at each college. (See Appendix
Table XI.) A plus (+) sign means that the dropouts' scores
are that much higher on the particular scale than stayins.
A minus (-) sign means that the dropouts' scores are that
much lower on the particular scale than stayins. Totals re
flect the sum of the differences for each scale in each pat
tern, while averages are simply the result of dividing the
totals by the number of colleges in the pattern. The number
in the parentheses ( ) denotes the number of institutions in
the pattern where statistical significance on the OPI scale
reached at least the .05 level.
76
TABLE 8
CUES PATTERN III VS. OPI DIFFERENCES
DROPOUTS RELATED TO STAYINS
OPI Scales CUES Pattern III
Distinguishing Personality
Characteristics among
Dropouts in Pattern III
High Community
High Awareness
Low Practicality
(2 Colleges)
Total Average
Social Extroversion -5.70 (1) -2.85
Masculinity-Femininity -5.53 (1) -2.76
Anxiety Level -5.36 (1) -2.68
Estheticism +5.28 (2) +2.64
Practical Outlook -4.38 (1) -2.19
Personal Integration -4.00 (0) -2.00
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts at each college. (See Appendix
Table XI.) A plus (+) sign means that the dropouts' scores
are that much higher on the particular scale than stayins.
A minus (-) sign means that the dropouts' scores are that
much lower on the particular scale than stayins. Totals re
flect the sum of the differences for each scale in each pat
tern, while averages are simply the result of dividing the
totals by the number of colleges in the pattern. The number
in the parentheses ( ) denotes the number of institutions in
the pattern where statistical significance on the OPI scale
reached at least the .05 level.
77
Table 5 presents total and average differences be
tween leavers and stayers on the four OPI scales which tend
to be common to the nine colleges in this part of the study.
These differences, taken from Appendix Table XI, are pre
sented for the colleges in each of the three CUES patterns
shown on Graphs I, II, and III. This table shows that de
spite the substantial differences in CUES profiles, the
typical dropout at these colleges is relatively more impul
sive, complex, independent, and less apt to present himself
in a good light. Because these four characteristics appear
strongly among the institutions in this study, and because
they are supported by the results of the hypotheses in this
chapter and by other studies, it seems clear that these are
the most important dimensions of personality to consider
between leavers and stayers regardless of institutional
climate. It is apparent that the two strongest dropout per
sonality traits are greater impulsivity and complexity.
Students withdrawing from colleges classified as
Pattern I— high sense of community and propriety, and low
student self-awareness— tend to exhibit less personal inte
gration, suggesting a less organized personal internal
structuring. This low personal integration, coupled with a
higher anxiety level, less altruistic interests, and a more
78
liberal religious orientation, suggests that withdrawing
persons do not fit well in the highly structured communities
of Pattern I (see Table 6).
At these colleges, most of the stayers subscribe to
the community spirit, conform to the student model which is
set forth implicitly by the institution, do what is expected
of them in terms of proper student behavior, and question
self and society less than students who withdraw. The cli
mate supports their personal organization, provokes less
anxiety, supports their high altruistic interests, and the
conservative religious atmosphere is compatible with their
own views. In effect, these religiously conservative in
stitutions are sympathetic to persisters1 conservative views
of self, society, and their religious concepts, and thus
these students tend to remain.
Students withdrawing from colleges classified as
Pattern II— low sense of self-awareness and low sense of
scholarship— tend to appear less practical and show a
greater theoretical orientation (see Table 7). The drop-
puts, in addition to being more impulsive, complex, and
independent, are less authoritarian, less anti-intellectual,
and more interested in science and in some science activi
ties than are their persisting peers. They also approach
79
problems from a logical* analytical* and critical viewpoint*
suggesting a personality which is seeking intellectual
stimulation. The persisters are less complex people* less
independent and less impulsive* and more willing to attempt
to project themselves in a favorable way than are leavers.
These persisting students are much more interested in prac
tical achievement and show less interest in science and less
willingness to be critical and logical.
The college climates in this pattern give little
emphasis to the developing self and to society and its
growth* and are below normal in the development of a sense
of academic pursuit on campus. This description indicates
a practically oriented campus where much social contact is
emphasized* which may work against persistence on the part
of the more impulsive* the less practical* the more with
drawn student. In effect* this describes two community
colleges and one college which is a traditional* rural in
stitution.
Students withdrawing from colleges classified as
Pattern III— high sense of community and self-awareness*
and low sense of practicality— project greater anxiety*
withdrawal* femininity* artistic interests* less personal
organization* and less desire for practical achievement
80
(see Table 8). These data suggest a highly sensitive per
sonality , esthetic in nature, not quite as mature as that
found among the persisters.
The persisting students at these colleges, although
above average in estheticism, are less artistically minded
than leavers; they are less complex, less independent, more
extroverted, less anxious, less feminine, and relatively
more practical.
The college climates in this pattern reflect strong
community spirit, high self-awareness and awareness of so
ciety and its problems, and relatively little emphasis on
personal status. At these campuses the individual is im
portant; the hierarchy of the community structure is de
emphasized and encouragement to know oneself predominates.
These are campuses where humanitarianism is stressed, where
much experimentation takes place, and where freedom and in
dividuality in both academic pursuits and social mores ex
ist .
It should be pointed out that across all CUES pat
terns, although statistical significance between dropouts
and stayins is not reached on the OPI scales at every col
lege, significance is evident at many institutions in many
variables; nevertheless, the direction of differences on
81
all distinguishing scales consistently moves in the same
direction at all institutions classified within the three
patterns.
Summary
This study explored the personality traits of drop
outs and stayins across fourteen different variables of the
OPI and found significant differences on several scales.
These differences were examined for groups of colleges as
determined by the score patterns on the CUES, and it was
discovered that different types of students withdrew from
the different types of colleges.
The major findings were as follows:
1.0 It was hypothesized that no differences would
exist between leavers and stayers across the sample when
both males and females were combined. Differences do in
fact appear on six of the fourteen scales, showing dropouts
to be more complex, impulsive, anxious, and isolated but
less altruistic and less willing to attempt to make good
impressions. These traits suggest a relatively immature
personality, still uncertain and seeking.
1.1 Male leavers and stayers differ in the same way
as the whole group as described in 1.0.
82
1.2 Female leavers and stayers do not conform so
closely. The female leavers are more complex, independent,
impulsive, and withdrawn, less altruistic, and less willing
to present themselves in a good light. Thus female leavers
are more independent than either male leavers or male and
female persisters. Female leavers are more withdrawn than
either of the above groups and there is evidence to suggest
that male leavers are more organized and integrated than are
female leavers.
1.3 Both male and female dropouts show significant
differences in aptitude scores at entrance from scores of
stayins, indicating that academic efficiency is greater
among persisters than that evidenced by those who eventually
leave these colleges.
2.0 The CUES patterns of college characteristics
indicate that there are three identifiable climates among
the colleges in this study.
2.1 One pattern suggests that campuses where a high
sense of community and propriety plus a low sense of aware
ness exist militate against persistence on the part of stu
dents who exhibit personality characteristics such as lower
personal integrative tendencies, higher anxiety levels, less
altruistic interests, and more liberal religious
83
orientations than those which describe persisting students.
An examination of the colleges in this pattern reveals that
they are religiously conservative institutions.
2.2 Another pattern suggests that on campuses where
both a low sense of awareness and a low sense of scholarship
exist, the climate tends to encourage withdrawal among those
students who are less practically oriented and who have a
greater interest in science and science activities. Looking
at these institutions, it is discovered that two of them are
community colleges and one is a rural, traditional institu
tion. Apparently the relative openness of the community
college, the relative lack of a higher sense of community,
a tendency to de-emphasize awareness, and a general lack of
academic concern move the above-described students to leave
the institution.
2.3 The third CUES pattern consists of college
climates which reflect a high sense of community and aware
ness and a low sense of propriety. From these institutions,
the distinguishing personality traits of students withdraw
ing are higher anxiety, social withdrawal, more dominant
feminine and esthetic tastes, fewer tendencies toward per
sonal integration, and desire for practical achievement.
CHAPTER V
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
Summary
Problem
The problem of college attrition is of continuing
concern to all segments of higher education in the United
States. Although rates of withdrawal vary from institution
to institution, the national experience, as evidenced by
federally supported studies, is that more than half of those
students entering college in a given year fail to graduate
from the college of their first choice four years later.
Extensive research over forty years reveals the
multifaceted problem and presents the many reasons for with
drawal elicited from students themselves, and from colleges
and universities as well. These efforts have uncovered much
data, but have not moved institutions very far toward solu
tions .
84
85
Students differ, institutions also differ. The
characteristics of both can be factors which militate
against persistence. Few colleges, however, have empiri
cally investigated the personal adjustment and needs of
differing student types, and the relevant environmental
characteristics to determine the effect of these upon with
drawals . This study was undertaken to determine what per
sonality characteristics distinguished dropouts from stayins
and what institutional characteristics of thirteen diverse
small colleges across the country might tend to explain why
certain kinds of students leave certain types of colleges.
The research was carried on in the context of the Project
on Student Development (NIMH Grant No. MH 14780-04) under
the auspices of the Council for the Advancement of Small
Colleges.
Procedure of the study
The first part of this study included 2,874 students
who enrolled as freshmen in the fall of 1965. Of these,
1,212 had withdrawn from their institutions by the fall of
1967 and for purposes of this study were considered as drop
outs . All students were tested with the Omnibus Personality
Inventory at the time of enrollment at each college. By
86
the fall of 1967 it was possible to separate dropouts and
stayins. All data from the inventory were coded and re
corded on IBM cards and processed through use of the IBM 360
system at the University of Vermont. The student t for in
dependent groups was used to test for significant differen
ces between the two groups. In addition SAT scores or
equivalents between the two groups were tested in the same
manner.
The second part of the study used the College and
University Environmental Scales to determine differences in
college climates. At each college, a sample of 100 students
from all grade levels was chosen by the Institutional Repre
sentative of each institution at which this instrument was
administered. The 66 per cent plus scoring method described
in the CUES Manual was used. Three patterns emerged from
these data and these were used to classify and describe
institutional characteristics. Four colleges were omitted
because they deviated from these patterns.
The final step was to apply the significant OPI
differences between the leavers and stayers to the patterns
of college characteristics emerging from the institutional
scores derived from CUES (see Tables 5, 6, 1, and 8).
87
Findings
Underlying the entire study were the assumptions
that if real personality differences existed between drop
outs and stayins, these could be identified, and that dif
ferent college climates would be revealed by the institu
tional scores of CUES. To test these assumptions the dif
ferences of OPI means between dropouts and stayins were
compared and the college scores on CUES across its five
scales were compared with the normative sample from the CUES
manual.
Four major null hypotheses based on the OPI data
were tested; these hypotheses dealt exclusively with the
differences between the two student groups.
Hypothesis 1.— No differences will appear between
personality characteristics of those students who leave
college and personality characteristics of those students
who stay in college. The findings indicate that real dif
ferences do appear between the two groups. Significant
differences do exist on six of the fourteen scales of the
OPI: Complexity, Impulse Expression, Personal Integration,
Anxiety Level, Altruism, and Response Bias. Hypothesis 1
is not confirmed.
88
Hypothesis 2.— No differences will appear between
the personality characteristics of males who leave college
and the personality characteristics of males who stay in
college. The data reveal that male students who leave col
lege are different in personality traits from those males
who remain in college. Significant differences among male
dropouts and stayins were found on the same scales as those
of the entire sample as listed above. Hypothesis 2 is not
confirmed.
Hypothesis 3.— No differences will appear between
the personality characteristics of females who leave college
and the personality characteristics of females who stay in
college. Significant differences were found between these
two groups on seven of the fourteen scales of the OPI:
Complexity, Autonomy, Impulse Expression, Social Extrover
sion, Anxiety Level, Altruism, and Response Bias. Hypothe
sis 3 is not confirmed.
Sex differences deriving from the findings of
Hypotheses 2 and 3 show that female dropouts are similar in
personality characteristics to male dropouts, except that
females are more independent than either male dropouts or
male and female stayins, are more integrated personally, and
are more socially withdrawn. ;
89
Hypothesis 4.— No differences will appear between
the level of aptitude of those who leave college and the
level of aptitude of those who stay in college. A compari
son between the means for male and female leavers, as well
as the means of male and female leavers combined, with the
corresponding means for stayers, revealed significant dif
ferences . Leavers exhibited lower SAT scores or equivalents
than did stayers. Hypothesis 4 is not confirmed.
The second part of the total study involved an in
spection of the institutional scores on CUES. Three differ
ent patterns emerged, involving nine of the thirteen col
leges. The distinguishing CUES characteristics (i.e., those
not scoring at or near the averages) for each pattern are:
CUES Pattern I describes a college climate in which
there is a high sense of community and propriety, and a low
sense of awareness.
CUES Pattern II describes a college climate in which
there is a low sense of awareness and a low sense of schol
arship .
CUES Pattern III describes a college climate in
which there is a high sense of community and awareness, and
a low sense of practicality.
The final step in the study was to identify by
inspection the distinguishing personality characteristics of
withdrawals from the various college climates• While
leavers from all the colleges were more impulsive, more
complex, more independent, and less willing to impress
others, there were certain characteristics which distin
guished withdrawers from the three individual climates re
vealed in the CUES data.
From a Pattern I campus, leavers are less well in
tegrated personally, less altruistic, more anxious, and more
liberal in religious attitude than stayins.
From a Pattern II campus leavers are less practi
cally oriented and more interested in science than are
stayins.
From a Pattern III campus leavers are more anxious,
more withdrawn, more feminine and esthetic, less integrated
personally, and less practically oriented than are per-
sisters.
Conclusions
Within the very express limits of the instruments
and procedures of measurement in this study, the findings
suggest the following conclusions:
1. Individuals of both sexes who think at a less
91
simplistic level., who exhibit greater tolerance for ambi
guity and experimentation, who tend to express impulses in
terms of overt action, who are more hostile, aggressive, and
anxious, and who tend to create poor personal impressions
will more likely than not withdraw from college.
2. Females who are more independent, who tend to
express hostility and aggression, and who are more apt to
withdraw from social contact and responsibility are likely
to leave the college campus before graduation.
3. Persisting students are likely to have signifi
cantly higher scores as measured on standardized aptitude
tests than do leavers. Both male and female persisters
score higher on such tests.
4. Individuals who have less tolerance for diverse
thinking, who are more conforming, who have a tendency to
accept authority, who are less apt to express hostility and
aggression, whose anxiety levels are lower, and who are more
cautious and conforming are more likely to remain in college
until graduation.
5. The small religiously conservative college cli
mate is characterized by a very high sense of community
spirit reflected in a friendly, supportive, sympathetic en
vironment . These campuses also tend to produce an
92
atmosphere of caution, thoughtfulness, and conventionality,
as well as a condition in which concern for self-understand
ing and search for personal meaning is below the norm; this
suggests a restrictive climate in which meanings concerning
self and society have already been developed, and in which
expansion and exploration in life's meanings are limited.
6. Small liberal and experimental college climates
are characterized by a high sense of community spirit which
is supportive and sympathetic. But, unlike the conservative
institution, they create an atmosphere in which there is
emphasis on the expansion and enrichment of personality,
societal horizons, and expression. At the same time per
sonal status is de-emphasized; knowing the right people,
being in the right group, doing what is expected, and the
corresponding order and structure are held in low esteem.
7. A group of small colleges exist at which a sense
of the search for personal and societal meanings and self-
understanding is low and where intellectual speculation,
interest in ideas, and pursuit of knowledge for its own sake
are low. These tend to be located in rural communities and
are generally classified as traditional, community, or
vocational colleges.
8. Individuals who drop out of the conservative
93
religious colleges where sense of community is high are more
likely to be those who intentionally avoid others, who are
anxious, hostile, or alienated. These leavers are less
altruistic and more religiously liberal. All these traits
speak to the personality types who value community less,
have little tolerance for propriety, and who seek more in
terms of meaning than the conservative college can provide.
9. Individuals who leave the rural, traditional,
or vocational college are lower on practical achievement
concepts and place a higher value on a discovery of self
and interest in science. These traits suggest personalities
which are seeking meaning and scholarship where neither are
found in great abundance.
10. Individuals who withdraw from the liberal,
experimental college are sensitive, withdrawn, artistic, and
feminine; they tend toward hostility and are less practi
cally oriented. The high sense of community and awareness
on campus may militate against their introverted tendencies.
Implications and Recommendations
These findings have a number of implications for
small colleges. Certain personality characteristics appar
ent in certain student types do not readily fit
94
particular environments. Identification of specific per
sonality traits early in the admissions process could go far
toward "screening out" those students who are not likely to
be successful in a particular environment.
As shown in this study, it is a relatively simple
matter to acquire data relating to student tolerance, inde
pendence, impulsivity, and honesty in personal projection.
Apart from utilizing the instruments used here, other
sources are available to the admissions personnel, that is,
biographical and autobiographical material and information.
It is suggested here that small colleges should develop pro
cedures which will allow the explicit listing of both strong
and less strong personality traits for each new student
enrolled in order to help those who are weak in certain
personality characteristics which, this study has shown,
militate against persistence at particular kinds of col
leges . Early identification and support for those who ex
hibit withdrawal tendencies might work to "short-circuit"
non-productive withdrawal experiences.
A selective procedure could easily be instituted
classifying, on the basis of personality and aptitude
scores, each new student into one of three classifications:
most likely persisters, marginal, and most likely
95
withdrawals. Each class could then be managed in a differ
ent way. Those most likely to remain through four years
could be handled through more traditional processes. The
marginal students could be watched closely for withdrawal
behavior and tendencies, and counseled toward self-
understanding and institutional values. Those most likely
to withdraw could be counseled toward other institutions and
programs or directed to more productive educational exper
iences at the institution of first choice in order that they
not be lost to higher education.
Most important is the institutions' consideration of
program modification. Pattern I colleges, where the more
restrictive atmosphere causes anxiety and discomfort in some
students, could develop programs intended to alleviate the
sense of constriction. These modifications might include
off-campus experiences for those who need periodic respite
from the traditional or religious functions of these
schools, or might include more flexible scheduling like in
dependent study or tutorial courses. Pattern II colleges
might develop social programs aligned with local social
agencies in order to increase interest in self and society.
To these programs the possible withdrawals could be di
rected. In addition, interest in scholarship for its own
96
sake only could be encouraged by the change of attitude and
emphasis of the faculty and administration toward raising
the level of campus scholarship. Pattern III colleges might
institute a greater number of artistic experiences in order
to hold those students who have greater interest in artistic
matters. Social withdrawal is often a result of the impli
cit community pressure^ always prevalent in a grouping where
much community spirit is in evidence. In such a climate.,
opportunities for seclusion or segregation of persons or
small groups from the total population, or freedom from
community pressures through faculty and counselor support
may help to encourage persistence.
Certainly, program modification would vary with the
institutional type and with the different types of student
withdrawals and, although modification is no panacea,
flexibility which considers individual student personality
traits might go far toward lessening the problem of college
attrition.
Much can be said about the direction of future re
search in this field. This study has been limited by the
classification of all dropouts from these colleges into one
grouping. It is obvious that further study would need to be
carried on, dividing this group of leavers into transfers
97
and non-transfer (real dropouts). Additional research needs
to be done regarding the types of programs and philosophies
of higher education* both of institutions and students. The
instruments used here could be utilized further to develop*
in depth* understandings of the distinguishing character
istics of dropouts and stayins through different kinds of
statistical analysis and inspection.
A PP EN DI XES
i
98
A P P E N D I X A
TABLES
99
APPENDIX A
The following tables and the graphs included in
Appendix B represent, in essence, the working papers con
taining the data finally used in the eight tables and three
graphs used in the body of the study. The combining of
these data, shown in the Appendixes, on only a few tables
and graphs in the narrative facilitates the understanding of
the findings in a more clear and concise manner.
100
101
TABLE I
CUES SCORES FOR NINE COLLEGES IN PATTERNS
Colleges
CUES Scales
Practi- Commu- Aware- Propri- Scholar-
cality nity ness ety ship
g
g
Q)
-P H
•P
i d
ft
A
C
E
F
14 22 8 20 10
12 17 12 22 15
12 16 5 20 5
8 21 10 22 10
c
J 15 12 3 7 2
( U 1 *
-p M
M 10 8 3 8 7
m
ft
H 6 13 4 6 3
c
, g
( U H
•P H
D 0 15 17 5 8
■ P H
i d
ft
K 1 13 16 4 13
•P
0 c
a -H t o
B 4 19 25 16 22
c
t o td g
( U 0 ) ( U
G 14 12 3 14 5
tn ti - p
0 ) P 4J
H H I d
H O ft
0 c
O H
I
L
11
12
19
23
8
11
20
16
6
8
National
Norm 11.2 11.4 11.9 10 11.5
102
TABLE II
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE A)
OPI Scales Dropouts Stayins Differences t Value
TI 44.17 44.57 - .40 - .223
TO 41.40 42.41 -1.01 - .653
ES 45.87 43.76 + 2.11 1.284
CO 46.57 43.86 + 2.71 1.687
AU 43.55 41.14 +2.41 1.641
RO 39.62 36.63 + 2.99 2.994b
IE 48.72 41.92 +6.80 3.345b
SE 48.66 48.98 - .32 - .166
PI 48.72 54.14 -5.42 -3.022b
AL 48.21 52.08 -3.87 -2.010a
AM 47.98 51.67 -3.69 -2.046a
MF 48.85 50.45 -1.60 - .840
RB 44.66 47.78 -3.12 -1.721
PO 54.89 54.98 - .09 - .063
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scalej a plus sign indicates higher
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.986.
t value required at the .01 level of significance
2.629.
103
TABLE III
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE C)
OPI Scales Dropouts Stayins Differences t Value
TI 46.77 48.50 -1.73 -1.328
TO 43.17 42.42 + .75 .601
ES 49.53 49.34 + .19 .155
CO 46.58 43.63 +2.95 2.563a
AU 44.32 44.25 + .07 .074
RO 41.90 38.32 +3.58 4.353b
IE 47.50 43.27 +4.23 3.012b
SE 48.25 48.64 - .39 - .279
PI 48.02 52.40 -4.38 -3.17lb
AL 46.92 50.23 -3.31 -2.502a
AM 50.04 54.12 -4.08 -3.520b
MF 46.92 46.75 + .17 . 140
RB 44.99 47.44 -2.45 -1.974
PO 54.54 53.50 +1.04 .983
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scale; a plus sign indicates higher
dropout scores on the OPI scale.
t value required at the .05 level of significance
1.984.
t avlue required at the .01 level of significance
2.626.
104
TABLE IV
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE E)
OPI Scales Dropouts Stayins Differences t Value
TI 43.35 45.08 -1.73 -1.129
TO 43.74 43.00 + .74 .524
ES 45.85 47.87 -2.02 -1.455
CO 48.44 45.46 +2.98 2.202a
AU 46.61 43.77 +2.84 2.246a
RO 45.20 42.62 +2.58 2.388a
IE 49.21 44.61 +4.60 2.869b
SE 47.50 49.30 -1.80 -1.141
PI 49.42 50.08 - .66 - .413
AL 48.21 49.30 -1.09 - .760
AM 49.98 52.78 -2.80 -1.777
MF 48.71 46.05 +2.66 1.738
RB 44.33 46.02 -1.69 -1.167
PO 54.11 53.86 + .25 .201
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scale; a plus sign indicates higher
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.976.
bt value required at the .05 level of significance
2.609.
105 i
TABLE V
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE F)
OPI Scales Dropouts Stayins Differences t Value
TI 44.64 44.93 - .29 - .186
TO 41.38 44.58 -3.20 -2.117a
ES 48.84 47.47 +1.37 .879
CO 44.73 43.71 +1.02 .746
AU 40.84 41.08 - .24 - .222
RO 39.93 40.00 - .07 - .070
IE 45.67 41.88 + 3.79 2.371a
SE 47.58 45.23 + 2.35 1.405
PI 47.67 50.62 -2.95 -1.599
AL 46.73 48.89 -2.16 -1.349
AM 50.04 50.16 - .12 - .083
MF 46.90 49.66 -2.76 -1.691
RB 44.85 49.63 -4.78 -3.200b
PO 58.00 57.33 + .67 .607
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scales; a plus sign indicates lower
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.979.
t value required at the .01 level of significance
2.616.
106
TABLE VI
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE J)
OPI Scales Dropouts Stayins Differences t Value
TI 42.69 42.20 + .49 .913
TO 44.21 43.58 + .63 1.007
ES 45.07 45.08 - .01 - .009
CO 48.43 47.53 + .90 1.561
AU 43.87 42.79 +1.08 1.955
RO 49.52 48.25 +1.27 2.486a
IE 56.18 53.52 +2.66 3.840b
SE 48.57 48.89 - .32 - .480
PI 46.80 46.83 - .03 - .032
AL 46.44 46.29 + .15 .225
AM 43.24 44.68 -1.44 -2.126a
MF 53.13 51.83 +1.30 1.930
RB 43.77 43.57 + .20 .313
PO 56.16 56.95 - .79 -1.578
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scale; a plus sign indicates higher
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.965.
* i _
t value required at the .01 level of significance
2.586.
107
TABLE VII
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE M)
OPI Scales Dropouts Stayins Differences t Value
TI 40.92 40.46 + .46 .403
TO 48.70 47.45 +1.25 .943
ES 39.76 39.93 - .17 - .162
CO 49.76 46 .38 +3.38 3.003'
AU 47.76 44.86 +2.90 2.801
RO 50.78 48.86 +1.92 1.504
IE 58.97 53.11 +5.86 4 .522'
SE 48.08 46.53 +1.55 1.052
PI 49.74 50.69 - .95 - .671
AL 49.28 49.48 - .20 - .150
AM 41.61 41.82 - .21 - .160
MF 58.51 59.38 - .87 - .822
RB 44.00 48.03 -4.03 -2.720J
PO 54.28 56 .05 -1.77 -2.055'
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scalej a plus sign indicates higher
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.972.
kf value required at the .01 level of significance
2.601.
108
TABLE VIII
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE H)
OPI Scales Dropouts Stayins Differences t Value
TI 47.00 45.75 +1.25 - .890
TO 46 .24 44.98 +1.26 - .959
ES 50.38 49 .40 + .98 - .707
CO 50.70 50.37 + .33 - .245
AU 50.03 49.58 + .45 - .347
RO 53.62 54.29 - .67 .526
IE 54.65 54.79 - .14 .093
SE 46.08 46.97 - .89 .600
PI 47.41 46.98 - .43 - .297
AL 46.83 44.81 +2.02 -1.383
AM 47.27 48.03 - .76 .548
MF 49.11 48.47 + .64 - .440
RB 44.19 43.15 +1.04 - .756
PO 51.71 52.51 - .80 .671
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scalej a plus sign indicates higher
dropout scores on the OPI scale.
109
TABLE IX
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE D)
OPI Scales Dropouts Stayins Differences t Value
TI 60.33 59.07 + . 26 1.190
TO 53.80 52.83 + .97 .714
ES 62.58 60.14 +2.44 2.216a
CO 64.81 63.28 +1.53 1.081
AU 64.39 62.41 +1.98 1.942
RO 62.30 61.65 + .65 .650
IE 62.16 59.98 +2.18 1.418
SE 42.77 46.32 -3.55 -2.2513
PI 46.28 47.86 -1.58 -1.042
AL 44.58 46.76 -2.18 -1.408
AM 51.58 52.45 - .87 - .559
MF 42.08 44.80 -2.72 -2.522a
RB 43.25 45.78 -2.53 -2.031a
PO 38.56 40.17 -1.61 -1.635
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scale; a plus sign indicates higher
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.976.
110
TABLE X
OPI MEANS AND DROPOUT-STAYIN DIFFERENCES
(COLLEGE K)
OPI Scales Dropouts Stayins Differences t Value
TI 57 .10 56.55 + .55 .436
TO 52.11 52.56 - .45 - .355
ES 58.58 55.74 +2.84 2.118a
CO 61.60 58.53 +3.07 2.166a
AU 61.25 59.48 +1.77 1.648
RO 59.58 57.10 +2.48 2.149a
IE 59.84 57.15 +2.69 1.762
SE 44.02 46.17 -2.15 -1.483
PI 48.03 50.45 -2.42 -1.833
AL 45.75 48.94 -3.19 -2.234a
AM 49.44 49.58 - .14 - .111
MF 46.95 49.76 -2.81 -2.097a
RB 46.40 48.38 -1.98 -1.555
PO 40.38 43.15 -2.77 -2.498a
Note: The figures shown on this table are generated
by subtracting the mean standard scores of stayins from the
same scores of dropouts. A minus sign indicates lower drop
out scores on the OPI scale; a plus sign indicates higher
dropout scores on the OPI scale.
at value required at the .05 level of significance
1.972.
TABLE XI
MEAN STANDARD SCORE DIFFERENCES BETWEEN DROPOUTS AND STAYINS
(ALL FRESHMEN AND SOPHOMORES COMBINED)
N'S
Colleges TI TO ES CO AU RO
DO SI
Pattern I
A 47 51 - .40 -1.01 +2.11 +2.71 + 2.41 +2.99
C 96 111 -1.73X + .75 + .19 +2.95 + .07 +3.58
E 66 93 -1.73 + .74 -2.02 + 2^98 +2 J34_ +2^58
F 73 73 - .29
Z3j- 2°
+1.37 +1.02 - .24 - .07
Totals Pattern I -4.15 -2.72 +1.65 +9.66 +5.08 +9.08
Average Pattern I -1.04 - .68 + .41 +2.42 +1.27 +2.27
Pattern II
J 369 330 + ,49 + .63 - .01 + .90 +1.08 +1^27_
M 74 125 + .46 +1.25 - .17 +3.38 +2.90 +1.92
H 60 142 +1.25 +1.26 + .98 + .33 + .45 - .67
Totals Pattern II +2.20 +3.14 + .80 +4.61 +4.43 +2.52
Average Pattern II + .73 +1.05 + .27 +1.54 +1.48 + .84
Pattern III
D 64 92 + .26 + .97 +2^44 +1.53 +1.98 + .65
K 112 89 + .55 - .45 12J34 +3 J37 +1.77 +2^48
Totals Pattern III + .81 + .53 +5.28 +4.60 +3.75 +3.13
Average Pattern ITT + .41 + .26 +2.64 +2.30 +1.88 +1.57
IE SE PI AL AM MF RB PO
+ 6 .8 0 - .32 5 .4 2
2* §.?
- 3 .6 9 - 1 . 6 0 3 .1 2 - .09
+ 4 . 2 3 - .39 - 4 . 3 8 - 3 .3 1 - 4 . 0 8 + .17 - 2 .4 5 + 1 .0 4
+ 4 . 2 0 - 1 . 8 0 - .6 6 - 1 .0 9 - 2 .8 0 + 2 .6 6 - 1 .6 9 + .2 5
+ 3_.79 + 2 .3 5 - 2 .9 5 - 2 .1 6 - .1 2 - 2 . 7 6 - 4 . 7 8 + .67
+ 1 9 .4 2 - .16 - 1 3 . 4 1 - 1 0 . 4 3 - 1 0 . 6 9 - 1 . 5 3 - 1 2 .0 4 + 1 .8 7
+ 4 .8 6 - .0 4 — 3 .3 5 — 2 .6 1 - 2 .6 7 - .3 8 - 3 .0 1 + .47
+ 2 .6 6 - .3 2
_
.0 3 + .15 - 1_.44 + 1 .3 0
+
.20 - .79
+ 5 .8 6 + 1 .5 5 - .95 - .2 0 - .2 1 - .87 - 4 .0 3 - U 7 7
- .1 4 - .8 9 - .43 + 2 .0 2 - .7 6 + .6 4 + 1 .0 4 - .80
+ 8 . 3 8 + .3 4
-
1 .4 1 + 1 .9 7 - 2 .4 1 + 1 .0 7 - 2 .7 9 - 3 . 3 6
+ 2 .7 9 + .1 1 - .47 + .66 - .8 0 + .36 - .9 3 - 1 . 1 2
+ 2 .1 8 - 3 i.55
_
1 . 5 8
_
2 .1 8 - .8 7 -2^ 72
_
2_.53 - 1 . 6 1
+ 2 .6 9 - 2 . 1 5 - 2 .4 2 — 3_. 19_ - .1 4
Z2j-81
- 1 .9 8 -2^ 7 7
+ 4 .8 7 - 5 . 7 0
_
4 . 0 0
-
5 .3 7 - 1 .0 1 - 5 . 5 3
-
4 . 5 1 - 4 . 3 8
+ 2 .4 4 - 2 . 8 5 - 2 .0 0 - 2 . 6 8 - .5 1 - 2 . 7 6 - 2 .2 6 - 2 . 1 9
Note: ,01 level of significance. ,05 level of significance. The figures shown on this table are
generated by subtracting the mean standard scores of stayins from the same scores of dropouts at each college. A plus sign
means that dropouts' scores are that much higher on the particular scale than stayins. A minus sign means that the drop
outs' scores are that much lower on the particular scale than stayins. Totals reflect, the sum of the differences for each
scale in each pattern. Averages are derived from dividing the pattern totals by the number of colleges in each pattern.
See Appendix Tables II through X for individual college OPI means and differences. Our experience at the Project has been
that a difference of 2.0 usually reaches statistical significance.
in
A P P E N D I X B
GRAPHS
112
113
GRAPH 1-1
CUES SCORE PROFILE
(COLLEGE A)
Practi
cality
Commu
nity
___ i __
Aware
ness
___ i ___
Propri
ety
Scholar
ship
24_
22.
20_
IS
IS.
14.
12.
10_
8.
6_
4_
2.
0
_24
_22
20
_18
_is
_14
_12
.10
_ 8
_ 6
. 4
_ 2
0
114
GRAPH 1-2
CUES SCORE PROFILE
(COLLEGE C)
Practi
cality
Commu
nity
■
Aware
ness
Propri
ety
Scholar
ship
■
24.
22.
20.
18.
I6-
14_
12_
1°_
8_
6_
4_
2_
0
_24
22
2.0
_18
_16
_14
_12
_10
_ 8
1 6
_ 4
2
0
115
GRAPH 1-3
CUES SCORE PROFILE
(COLLEGE E)
Practi
cality
Commu-
-nity
Aware
ness
i
Propri
ety
Scholar
ship
24_
22_
20_
18_
14_
12_
10_
8_
6_
4_
2_
0
_24
_22
_20
_18
_16
_14
_12
_10
_ 8
_ 6
_ 4
_ 2
0
GRAPH 1-4
CUES SCORE PROFILE
(COLLEGE F)
116
Practi
cality
Commu
nity
Aware
ness
___ i
Propri
ety
Scholar
ship
24.
22.
20.
18.
16.
14.
12_
10.
8_
6_
4_
2_
0
_24
22
.20
_18
_16
_14
.12
_10
_ 8
. 6
_ 4
_ 2
0
117
GRAPH II-l
CUES SCORE PROFILE
(COLLEGE J)
Practi
cality
Commu
nity
___ i __
Aware-
ness
___ I ___
Propri
ety
Scholar
ship
24_
22_
20_
18.
14_
12_
1°.
8_
6_
4_
2_
0
_24
22
20
_18
16
J 4
.12
- 10
- 8 !
. 8
. 4
. 2 i
0 |
GRAPH II-2
Practi
cality
CUES SCORE PROFILE
(COLLEGE M)
Coiranu- Aware- Propri-
nity ness ety
Scholar
ship
119
GRAPH II-3
CUES SCORE PROFILE
(COLLEGE H)
Practi
cality
Commu
nity
Aware-
ness
___ i ___
Propri
ety
Scholar
ship
2 4 _
22_
20_
18.
16_
14_
12_
10_
8_
6_
4_
2_
0
24
22
_20
_18
_16
_14
_12
_10 ;
I
_ 8 !
_ 6
_ 4
_ 2
0
120
GRAPH UI-1
CUES SCORE PROFILE
(COLLEGE D)
Practi
cality
Commu
nity
___ i __
Aware
ness
Propri
ety
Scholar
ship
24_
22_
20_
18_
16 _
14 _
12.
10.
8_
6_
4_
2_
0
_24
_22
_20
_18
_16
_14
_12
_10
_ 8
_ 6
_ 4
_ 2
0
121
GRAPH III-2 I
CUES SCORE PROFILE
(COLLEGE K)
Practi- Conunu- Aware- Propri- Scholar-
cality nity ness ety ship
l t i l I
24_ _24
22_ _22
20_ _20
18_ _18
16 16
14
12
10.
8 .
6.
4
2
A P P E N D I X C
COLLEGES
122
APPENDIX C
The following are the colleges participating in this
study:
1. Bryan College, Dayton, Tennessee
2. Earlham College, Richmond, Indiana
3. Eastern Mennonite College, Harrisonburg, Virginia
4. Goddard College, Plainfield, Vermont
5. Malone College, Canton, Ohio
6. Messiah College, Grantham, Pennsylvania
7. Morris College, Sumter, South Carolina
8. Nasson College, Springvale, Maine
9. Sacred Heart College, Wichita, Kansas I
i
!
10. Salem College, Salem, West Virginia
11. Shimer College, Mount Carroll, Illinois
12. Western New England College, Springfield, Massachusetts'
13. Westmont College, Santa Barbara, California ;
123
A P P E N D I X D
INSTRUMENTS
124
PLEASE NOTE:
Appendix D: "CUES College and
University Environment Scales",
@ 1962 by C. Robert Pace, pages
125-131
and
"Attitude Inventory, OPI-Form Fx"
@ 1959 and 1963, pages 132-141,
not microfilmed at request of
author. Available for consult
ation at University of Southern
California Library.
UNIVERSITY MICROFILMS.
S E L E C T E D BI BL IO GR AP HY
j
i
I
142
SELECTED BIBLIOGRAPHY
Books
Knoell, D. "A Critical Review of Research on the College
Dropout." The College Dropout and the Utilization
of Talent. Edited by L. Pervin. Princeton, N. J.:
Princeton University Press, 1966.
Kubie, L. S. "The Ontogeny of the Drop-out Problem." The
College Dropout and the Utilization of Talent.
Edited by L. Pervin. Princeton, N. J.: Princeton
University Press, 1966.
Strang, Ruth. Behavior and Background of Students in Col
lege and Secondary School. New York: Harper Bros.,
1937.
Summerskill, J. "Dropouts from College." The American Col
lege . Edited by Nevitt Sanford. New York: John
Wiley and Sons, 1962.
Articles
Abel, W. H. "Attrition and the Student Who Is Certain,"
Personnel and Guidance Journal. XLIV (June, 1966),
1042-1045.
Aiken, L. R. "The Prediction of Academic Success and Early
Attrition by Means of a Multiple Choice Biographical
Inventory." American Educational Research Journal,
I (1964), 127-135.
143
144
Barger, B. and Hall, E. "Personality Patterns and Achieve
ment in College.1 1 Educational and Psychological
Measurement. XXIV (Summer, 1964), 339-346.
Brown, F. G. "Identifying College Dropouts with the Minne
sota Counseling Inventory." Personnel and Guidance
Journal, XXXIX (December, 1960), 280-282.
Carlson, S. J. and Wegner, K. W. "College Dropouts." Phi
Delta Kappan, XLVI (March, 1965), 325-327.
Chase, C. I. and Barritt, L. S. "A Table of Concordance
between ACT and SAT." Journal of College Student
Personnel, March, 1966, pp. 105-108.
Chickering, A. W. "Institutional Differences and Student
Characteristics." Journal of American College
Health Association, December, 1966, pp. 168-181.
Dalrymple, W. "The College Dropout Phenomenon." NEA Jour
nal, LVI (April, 1967), 11-13.
Daniel, Kathryn B. "A Study of College Dropouts with Re
spect to Academic and Personality Variables." Jour
nal of Educational Research, LX (January, 1967),
230-235.
Farnsworth, D. S. "Some Non-academic Causes of Success and ,
Failure in College Students." College Admissions,
II (1955), 72-78.
_________. "We Are Wasting Brain Power." Journal of Na
tional Education Association, XLVIII (1959), 42-44.
Forrest, D. S. "High School Underachievers in College."
Journal of Educational Research, LXI (December,
1967), 147-150.
Gadzella, B. M. and Bentall, G. "Differences in High School
Academic Achievements and Mental Abilities of Col
lege Graduates and College Dropouts." College and
University, XLII (Spring, 1967), 351-356.
145
Grace, H. "Personality Factors and College Attrition."
Peabody Journal of Education, XXXV (June, 1957), 36-
40.
Greenfield, Lois B. "Attrition among First Semester En
gineering Freshmen." Personnel and Guidance Jour
nal, LII (June, 1964), 1003-1010.
Heilbrun, A. B. Jr. "Personality Factors in College Drop
out." Journal of Applied Psychology, XLIX (Febru
ary, 1965), 1-7.
_________. "Prediction of First Year College Dropout, Using
ACL Need Scales." Journal of Counseling Psychology,
IX (Spring, 1962), 58-63.
Hill, A. H. "A Longitudinal Study of Attrition among High
Aptitude College Students." Journal of Educational
Research. LX (December, 1966), 166-173.
Holmes, C. H. "Why They Left College." College and Uni
versity, XXXIV (Spring, 1959), 295-300.
Iffert, R. E. "Study of College Student Retentions and
Withdrawals." College and University, XXXI (Summer,
1956), 435-446.
Jensen, V. H. "Influence of Personality Traits on Academic
Success." Personnel and Guidance Journal, XXXVI •
(March, 1958), 497-500.
Johnson, G. B. "A Proposed Technique for Analysis of Drop
outs at a State College." Journal of Educational
Research, XLVII (January, 1954), 381-387.
Levenson, Edgar and Kohn, Martin. "A Demonstration Clinic
for College Dropouts." Journal of the American
Health Association, XII (1964), 382-391.
Little, J. K. "The Persistence of Academically Talented
Youth in University Studies." Educational Record,
XL (June, 1959), 237-241.
146
Marks, E. "Student Perceptions of College Persistence, and
Their Intellective Personality and Performance Cor
relates ." Journal of Educational Psychology, LVIII
(August, 1967), 210-221.
Nelson, G. A. "College Characteristics Associated with
Freshmen Attrition." Personnel and Guidance Jour
nal , XLIV (June, 1966), 1046-1050.
Panos, R. J. and Astin, A. W. "Attrition among College Stu
dents ." American Council on Education Research
Report, II (1967) .
Rose, H. A. and Elton, C. J. "Another Look at the College
Dropout." Journal of Counseling Education, XIII
(Summer, 1966), 242-245.
Sanford, N. "Personality Development During the College
Years." Journal of Social Issues, XII (1956).
Summerskill, J. and Darling, C. D. "A Progress Report on
the Student Stability Studies." Studies in Medi
cine , III (1955), 85-91.
Weintraub, R. G. and Salley, R. E. "Graduation Prospects of
an Entering Freshman." Journal of Educational Re
search, XXXIX (October, 1945), 116-126.
Reports
Farnsworth, D. S.; Funkenstein, D.j and Wedge, B. A. "Study
of the Social and Emotional Adjustment of 'Early
Admission' College Students." Report for the Fund
for Advancement of Education, 1955. (Mimeographed.)
Iffert, R. E. "Retention and Withdrawal of College Stu
dents." Office of Education Bulletin, No. 1 (1958).
Washington, D. C.: U.S. Government Printing Office.
McNeeley, J. H. "College Student Mortality." U.S. Depart
ment of Interior Bulletin No. 11, 1937.
147
Suczek, R. F. and Alfert, E. "Personality Characteristics
of College Dropouts." Research Report, Cooperative
Research, Office of Education DHEW, and University
of California, Berkeley, Project No. 5-8232, 1966.
Unpublished Material
Hannah, W. "Freshman Leavers and Stayers." Paper presented
at the 19th Annual Meeting of the Vermont Psycho
logical Association, April 1967, Bolton, Vermont.
_________. "The Withdrawal Process." Unpublished article.
Research Proposal to NIMH for Project on Student Development
by Council for Advancement of Small Colleges, 1964.
Scott, Donald G. "Small Colleges in Transition." Unpub
lished Ph.D. dissertation, Florida State University,
1966. —
Tests
Omnibus Personality Inventory— Research Manual. Berkeley,
Calif.: Center for Study of Higher Education, 1962.
Pace, Robert C. College and University Environment Scales.
Princeton, N. J.: Educational Testing Service,
1963.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Criteria For Directing Junior College Instruction

PDF

A Historical Analysis Of Vocational Education: Land-Grant Colleges To California Junior Colleges, 1862-1940

PDF

Factors In The Selection Of State School Board Members And The Chief State School Officer

PDF

The Relationship Of Size To Current Expense Of Education In California Single-College Public Junior College Districts

PDF

The Impact Of Selected Colleges On Students' Values

PDF

The Administrative Implications Of Federal Projects On School Personnel Practices

PDF

A Study Of Management Functions Common To The Administration Of Commercial Businesses And Institutions Of Higher Education

PDF

The Selection, Training And Evaluation Of School Bus Drivers In California

PDF

The Image Of Higher Education In American Novels, 1920-1966

PDF

Objectives Of Higher Education As Reflected In The Inaugural Addresses Ofselected College And University Presidents

PDF

Functions Of Trustees Of United Presbyterian Colleges

PDF

Faculty Views Of The California Public Junior College

PDF

Relationships Of Variables Identified With Success In College Clothing-Construction Courses: A View Toward Advanced Placement And Improved Learning

PDF

English Programs In Selected Two-Year Community Colleges, Private Liberalarts Colleges, And State Colleges In California

PDF

Programs For Professional Preparation Of Instructors For California Public Junior Colleges

PDF

The Potential Of The Junior College In The Developing Nations Of The World

PDF

Capital Outlay Approval Procedures For Public Institutions Of Higher Learning

PDF

Grievance Procedures And Disciplinary Actions In Classified Personnel Administration In California Public Schools

PDF

An Historical Analysis Of The Origin And Development Of The College Of Medical Evangelists

PDF

Selected Non-Intellectual Factors As Predictors Of Academic Success In Junior College Intellectually Capable Students

Asset Metadata

Creator Hannah, William (author)

Core Title Dropout - Stayin Personality Differentials And College Environments

Degree Doctor of Education

Degree Program Education

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

Tag education, administration,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Pullias, Earl Vivon (committee chair), Georgiades, William (committee member), Muelder, Wallace R. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-391926

Unique identifier UC11361162

Identifier 7011367.pdf (filename),usctheses-c18-391926 (legacy record id)

Legacy Identifier 7011367.pdf

Dmrecord 391926

Document Type Dissertation

Rights Hannah, William

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