Citation
Academic and nonacademic factors related to the attrition rate of specially admitted Black university freshman students

Material Information

Title:
Academic and nonacademic factors related to the attrition rate of specially admitted Black university freshman students
Creator:
Stewart-Dowdell, Betty, 1950-
Publication Date:
Language:
English
Physical Description:
ix, 94 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
College bound students ( jstor )
College freshmen ( jstor )
College students ( jstor )
Colleges ( jstor )
Grade point average ( jstor )
Graduate schools ( jstor )
High school students ( jstor )
School dropouts ( jstor )
Self concept ( jstor )
Universities ( jstor )
African American college students -- Florida -- Gainesville ( lcsh )
Counselor Education thesis Ph. D
Dissertations, Academic -- Counselor Education -- UF
Dropouts -- Florida -- Gainesville ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1982.
Bibliography:
Includes bibliographical references (leaves 84-93).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Betty Jean Stewart.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Betty Jean Stewart. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
08849817 ( OCLC )
0028528122 ( ALEPH )

Downloads

This item has the following downloads:


Full Text













ACADEMIC AND NONACADEMIC FACTORS RELATED TO THE
ATTRITION RATE OF SPECIALLY ADMITTED BLACK
UNIVERSITY FRESHMAN STUDENTS















BY

BETTY JEAN STEWART



















A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1982



















G(o











Copyright 1982

by

Betty Jean Stewart
























THIS PROJECT

IS

DEDICATED

TO

MY MOTHER, MRS. DOROTHY HICKMAN,

AND

THE PAST, PRESENT, AND FUTURE

BLACK STUDENTS

AT THE

UNIVERSITY OF FLORIDA
















ACKNOWLEDGMENTS



All glory goes to God for the great thing he has done

through the many persons in my life who made this endeavor

possible. I know that I will not be able to include all of

the "significant" others who offered words of encouragement

and support during the course of my educational experience,

but I will attempt to list a few of these individuals by name.

To Dr. Joseph Wittmer, my chairman, consultant, and

most of all, friend, I am forever grateful for your ever-

present support, guidance and faith in my ability to succeed.

Thank you from the depths of my heart for your concern and

sensitivity. Special words of thanks are also extended to

the other members of my supervisory committee, Dr. Roderick

McDavis and Dr. Robert Ziller. Thank you Dr. McDavis for

your firm and loving guidance in completing this project.

I am also especially appreciative of the inspiration you

gave me throughout my tenure as a graduate student. Dr.

Ziller, your continuous support, helpful suggestions and

friendship greatly aided me in achieving this goal.

There are many true friends to whom I am sincerely

grateful because of their patience and understanding during

all of the "crazy" moments of my life. I am indebted to

my typist, Vita Zamorano (who is faster than the speed of


iv










light), and my computer analyst, Alecia Schmitt (who made

it all so very plain and simple). To Gloria, Suzan, Liz,

Juanita, and Lainee, I appreciate your moral support, al-

though you were facing the same pressures. I know that

each of you will likewise succeed. To Pat, Brenda, Ruthell,

Elmira, and Charlene, who allowed me to "steal" time from

our friendships in order to pursue and complete this pro-

ject; I thank you.

To the staff of the College of Liberal Arts and Sciences,

Dr. Charles F. Sidman, Dr. Harry B. Shaw, G.W. Mingo, Deloris

Slater, John Watson, and the PACT and Special Services peer

counselors, without whose faith, support, and cooperation,

I would not have been able to succeed; I am greatly indebted.

In conclusion, a special word of thanks is extended

to my family for their moral support, prayers, love, and

patience. Thank you Mom, Dad, Rita, Damon, Graylyn, Thelma,

Julius and Sister for your sacrifices, understanding, and

assistance throughout my educational experience.




















v

















TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS . . . . . . . .... . iv

ABSTRACT . . . . . . . .. . . .viii

CHAPTER

ONE INTRODUCTION . . . . . .. . . 1

Statement of the Problem .. . . . . 1
Purpose of the Study . . . . . 7
Need for the Study . . .. . . . 8
Definition of Terms . . . . . . 12
Organization of the Study . . . . . 13

TWO REVIEW OF RELATED LITERATURE .... . ... .14

Introduction . . . . . . . . 14
Academic Factors Related to College Attrition 14
Nonacademic Factors Related to College Attrition 21
Self-concept and College Student Attrition 24
Personality Characteristics and College
Student Attrition . . . . . . 29
College Student Satisfaction and College
Student Attrition . . . . . .. 33

THREE METHODOLOGY . . . . . . . ... 41

Introduction . . . . . . . ... 41
Hypotheses . . . . . . . . . 41
Population and Sample . . . . . .. 42
Instruments . . . . . . . . 44
Tennessee Self Concept . . . . .. 44
Myers-Briggs Type Indicator . . . .. 49
College Student Satisfaction Questionnaire 52
Procedures . . . . . . .. .. . 57
Analysis of Data . . . . . . . 59
Limitations of the Study . . . . .. 60

FOUR RESULTS AND DISCUSSION . .. . . . 62

Results . . . . ..... . . 62
Discussion . . . . . . . . . 71


vi










CHAPTER Page

FIVE SUMMARY, CONCLUSIONS, IMPLICATIONS, AND
RECOMMENDATIONS FOR FUTURE RESEARCH . . 76

Summary . . . . . . . . . . 76
Conclusions . . . . . . . . 77
Implications . . . . . . ... .78
Recommendations for Further Research . . 80

APPENDICES

A INFORMED CONSENT FORM . . . . .. 82

B MEANS AND STANDARD DEVIATIONS ON ALL
VARIABLES FOR SPECIALLY ADMITTED BLACK
UNIVERSITY STUDENTS . . . . . .. 83

REFERENCES . . . . . . . . . . . 84

BIOGRAPHICAL SKETCH . . . . . . . ... .94



































vii










Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



ACADEMIC AND NONACADEMIC FACTORS RELATED TO THE
ATTRITION RATE OF SPECIALLY ADMITTED BLACK
UNIVERSITY FRESHMAN STUDENTS

By

Betty Jean Stewart

May 1982

Chairman: Dr. Paul Joseph Wittmer
Major Department: Counselor Education

The purpose of this study was to investigate the

relationship between five independent variables (high school

grade point averages, college entrance exam scores, per-

sonality, self-concept, and college student satisfaction)

and one dependent variable (college grade point average)

concerning specially admitted black students at the

University of Florida. Further, this study examined the

relationship between college student satisfaction scores

for these black students and two other variables,

personality and self-concept.

A total of 103 specially admitted University of Florida

black freshmen participated in this study. Each was present

and participated in the first day of the five-day orientation

program conducted by the two special admission programs during

Summer, 1981. The Tennessee Self Concept Scale (TSCS) and

the Myers-Briggs Type Indicator (MBTI) were administered to

each subject as part of the orientation program. During


viii










the third week of Spring, 1982, the College Student Satis-

faction Questionnaire (CSSQ) was also given to each

student. Vital information regarding high school grade

point averages, college entrance exam scores, and cumulative

college grade point averages was requested and obtained

from the University of Florida offices of Admissions and

the Registrar.

A Pearson Product Moment Correlation indicated no

significant relationship between college grade point averages

and any of the five independent variables. Correlation co-

efficients also indicated no significant relationship be-

tween college student satisfaction and the variables of

personality type and self-concept.

The results of a stepwise regression analysis yielded

a multiple coefficient of determination (R2) of only .25

(p<.001) when the four subscales of Personal Self, Scho-

lastic Aptitude Test-Total, Identity, and Working Conditions

were entered into the prediction model.


















ix
















CHAPTER ONE
INTRODUCTION



The gap between the percentage of black
and white students who make it to gradua-
tion at Florida universities is.widening
. Almost half of the black students
enrolled in the nine state schools during
the last four years dropped out before
completing a bachelor's program.
The Independent Florida Alligator,
September 22, 1981



Statement of the Problem


The above statement is only one of many in recent

months which has focused on the attrition/retention/gradua-

tion problem being experienced throughout the State Univer-

sity System of Florida. Although the main objective of

colleges and universities is to graduate students, there is

a high attrition rate for black students. A recent study by

Egerton (1969) reported that of every 100 white 18 year olds

in the United States, approximately 75 finish high school,

about 45 enter college, and about 15 earn a baccalaureate

degree. However, the persistence and graduation rates show

a marked decline for the black population of 18 year olds.

Of every 100 black 18 year olds, approximately 65 finish

high school, about 30 enter college, and only about 5 actually

earn a baccalaureate degree.


1







2


An extensive follow-up study of 20,000 1972 high school

graduates was conducted by Eckland and Wisenbaker (1979).

While the overall college dropout rate for both white men

and women was found to be 34 percent, the rates for black men

and women were 43 and 46 percent, respectively. An examina-

tion of the graduation trends for the two groups revealed the

following rates: white men, 36 percent; white women, 46 per-

cent; black men, 27 percent; and black women, 32 percent.

Goodrich (1976) conducted a similar in-depth study on

the academic status of minority students. The results of this

study indicated that of the first-time enrolled freshmen in

1974, a year later 14 percent of the black students had been

dismissed for academic reasons, as compared to 3 percent for

white students and 4 percent for the university as a whole.

The reported overall dropout rate (for both academic and

nonacademic reasons) was 27 percent for blacks, 18 percent

for whites and 21 percent for the overall university popula-

tion. However, another follow-up study of black and white

students (White and Suddick, 1981) found that the graduation

rates for both groups over a period of five years were very

similar. There was also very little difference in the aca-

demic standing (cumulative grade point averages) of those

students who withdrew by race. In a comprehensive ten-year

study Astin (1977) however concluded that although the

absolute persistence rate for blacks was lower than that

for whites, the persistence rate for black women was actually

higher than that for white women of comparable preparation







3

and ability. Therefore, not only were differences found

to exist between races, but also between sexes within the

same race.

While the majority of the above studies suggests high

attrition rates for some black students, those initially

admitted under special guidelines which were established in

compliance with federal requirements face additional pro-

blems (Garber and Schell, 1977). Although special admission

criteria may vary among similar institutions, they are

usually based on minimal standardized test scores and/or

high school grade point averages. A recent report by the

National Advisory Committee on Black Higher Education and

Black Colleges and Universities recommended that the specific

needs of black students at predominately white institutions

should be defined. This report also suggested identifica-

tion of ways in which these institutions can increase their

response to the access to and completion of programs by

blacks (N.A.C.B.II.E., 1979).

In a majority of cases, a great deal of emphasis is

placed on the recruitement and admission of specially admitted

students (Carr and Chittum, 1980). These students are over-

whelmed with letters of welcome and congratulations from

various offices prior to entering the university or college.

Their initial face-to-face contact with university personnel

usually comes in the form of some type of formal or informal

orientation program (McDavis, Mingo, Stewart, and Hough,

1980). Haettenschwiller (1971) suggests that the establishment







4


of early contact is fundamental to the counseling of black

college students in special programs. Although these con-

tacts are basically positive in nature, they are not often

maintained at the same level of intensity over an extended

period of time. This relationship is usually not re-

established until the student incurs academic difficulty.

Too often the student's dissatisfaction with the college

environment is made apparent through low academicachieve-

ment. It is at this point that concern is shown for these

students and assistance is offered. Many times this concern

and assistance is too late to be of any benefit to the

students (Friedlander, 1980). Other specially admitted

students, although performing well academically, also may be

experiencing dissatisfaction with the college environment

and choose not to return to the institution for one reason

or another. Satisfaction with the college environment,

including the helpfulness of faculty, staff, advisors, and

counselors, seems to be an important determinant in the

attrition rate of these specially admitted students.

In contrast, the same set of circumstances may result

in a very different outcome for regular students because of

theirtendency to generalize high school experience to the

college situation (Haettenswiller, 1971). During the course

of previous daily interactions, they have internalized the

rules of the game and know the roles that they are expected

to play. When the expectations of the regular student are

not realized, the resources to make the appropriate







5


adjustment are readily available. Merton (1957) uses the

term "anticipatory socialization" to describe this process.

Since the beginning of special admission policies,

researchers have attempted to identify factors related to

academic success. The most common factors that have been

associated with college attrition are academic and non-

academic. There is a wealth of research concerned with

academic factors such as high school grade point averages

and standardized college entrance test scores (Aleamoni,

1977; Dalton, Anastasiow, and Brigman, 1977; Price and Kim,

1976; Rovezzi-Carroll and Thompson, 1980; Pfeifer and

Sedlacek, 1971). There have also been attempts to identify

nonacademic factors such as personality characteristics,

self-concept, socioeconomic status, biographic background,

and satisfaction (Green and Farquhar, 1965; Hannah, 1971;

Starr, Betz, and Menne, 1972; Samuel and Laird, 1974).

The University of Florida is part of a State University

System whose policy permits 10 percent of the freshman class

to be admitted without having met minimum admission require-

ments. At present, these requirements are a high school

grade point average of 2.0 and a total of 800 on the SAT

(Scholastic Aptitude Test) or 17 on the ACT (American College

Testing Program). Except in very few cases, the students

are required to enroll for the term immediately following

high school graduation. The rationale behind this require-

ment is that during the summer a more intensive orientation

program can be implemented and individual attention can be







6


given to these students since there is generally a lower

enrollment during this term.

Presently there are two programs at the University of

Florida primarily responsible for the overall coordination

of the orientation program, and subsequent retention and

graduation of these students. These two programs are

Special Services and the Program for Academic Counseling

and Tutoring (PACT). The Special Services Program is a

federally funded program designed to assist students from

"culturally and/or educationally" deprived backgrounds.

This program began in 1971 and provides supportive services

and alternative grading and test-taking practices to insure

the participants' success in the academic as well as social

environment. Currently there are hundreds of such programs

being conducted throughout the United States with 17 based

on college campuses within the State of Florida.

The PACT, which began summer 1979, is state funded and

provides essentially the same type of assistance that the

Special Services Program provides. These programs work

closely in the planning and implementation of an orientation

program and other activities which involve specially

admitted students at the University of Florida. These other

activities include tutoring, counseling, general information,

financial assistance information, academic advisement and

career information.

Another important aspect of these programs is the use

of peer counselors who serve as social and academic role







7


models. These former "program" participants are a valuable

asset providing assistance to students throughout the

critical freshman year. Although the type of assistance

which these two programs offer varies according to individual

student needs and resources available, they both have as

their main goal graduation of program participants.

While most of these students seem to make the initial

adjustment to college life, problemsarise for them after

this introductory period. Even though the special selec-

tion criteria are essentially used to predict which "marginal"

students will most likely succeed if admitted, predictions

based solely on standardized test scores and/or high school

grade point averages have not proven to be accurate.



Purpose of the Study


The purpose of the present study was to investigate

the relationship between five independent variables (high

school grade point averages, college entrance exam scores,

personality type scores, self concept scores, and college

student satisfaction scores) and one dependent variable

(college grade point average). This study also attempted

to determine the relationship between college student

satisfaction scores, personality type scores, and self-

concept scores of specially admitted black university

freshman students. Although this relationship has been the

source of much recent research, the establishment of a







8


predictive model by the use of multiple factors has not been

extensively explored and determined. This was the focus of

this study.

Specifically, this study proposed to answer the following

questions concerning specially admitted black university

freshman students:

(1) Does a relationship exist between college grade

point averages and high school grade point

averages?

(2) Does a relationship exist between college grade

point averages and college entrance exam scores?

(3) Does a relationship exist between college grade

point averages and self-concept?

(4) Does a relationship exist between college grade

point averages and personality type preferences?

(5) Does a relationship exist between college grade

point averages and college student satisfaction?

(6) Does a relationship exist between personality type

preferences and college student satisfaction?

(7) Does a relationship exist between self-concept

and college student satisfaction?



Need for the Study


Institutions of higher education, along with other seg-

ments of society, are facing difficult times due to financial

cutbacks and decreasing enrollments. These impending







9


budgetary difficulties have caused institutions to raise

admission standards for regularly admitted students and to

scrutinize even more closely those students who are admitted

through special admission policies.

Numerous national surveys have been conducted in an

attempt to monitor admission and retention trends for minor-

ity students attending large, predominately white universities

in the United States (Sedlacek and Brooks, 1970; Sedlacek,

Brooks, and Horowitz, 1972; Sedlacek, Brooks, and Mindus,

1973; Sedlacek, Lewis, and Brooks, 1974; Sedlacek and Webster,

1978). The results of these studies indicated that there

was gradual progress towards admitting more black freshmen

and searching for better methods of selection. Also, the

number of special programs demonstrated a decrease as well

as the number of schools utilizing different admissions

criteria for minority students.

The attrition/graduation rate of all students, and

especially those admitted through special guidelines,is a

major concern to university administrators. Rovezzi-Carroll

and Thompson (1980)concluded that college grade point average

and college graduation may result from the interaction of

different sets of antecedent and intervening variables. It

also has been suggested that a variety of factors should be

considered in making decisions about prospective students

(Mayhew, 1965). However, the problem exists in attempting

to operationally define some of those factors.







10

In recent years there have been questions raised

concerning the effectiveness of special admission programs and/

or the academic success of former participants in terms of

grade point averages and graduations. Past research in this

area has produced ambiguous and mixed results (Cleary, 1968;

Temp, 1971; Boyd, 1977; Pfeifer and Sedlacek, 1971). How-

ever, the decision as to whether a student is admitted or

denied admission is still most often based on the academic

factors of standardized test scores, high school grade point

averages, and/or letters of recommendation regarding their

academic potentials. It has been shown that decisions

based on the above mentioned factors are not foolproof in

most cases, and admission committees cannot be absolutely

certain of the accuracy of their decisions.

One researcher (Gibbs, 1973) suggested that in order to

reduce the casualties of the high risk admissions programs,

admissions officers should be more attentive to alternative

indicators that might be more predictive of a black stu-

dent's success than traditional indicators. Examples of

these nontraditional criteria may include positive self-

concept, leadership potential, goal oriented activities,

and personal maturity. Tinto (1975) also cited the failure

of past research to define more clearly the possibility of

multiple characteristics of dropouts. This researcher hypo-

thesized that dropout occurs because of academic and social

malintegration and is a longitudinal process of. interactions

between the individual and the academic and social systems

of the college.







11


Most certainly administrators and others responsible

for admitting, retaining, and graduating specially admitted

students would desire to admit those individuals with a high

probability of success based on as many factors as are

available. However, many questions still go unanswered as

to which factors combine to produce high retention and gra-

duation rates. Do some students suceeed because entrance

examination scores accurately predict their scores? Why do

other students with those same test scores fail? Is there

something unique about individuals who score low on standard-

ized tests and still manage to achieve academic success?

Answers to these and other questions would provide invalu-

able information to admissions committees, counselors, and

students.

During spring 1979 at the University of Florida, a

retention fact-finding study was conducted (Stewart, 1981).

The purpose of this study was to establish retention rates

for specially admitted students. The findings of this study

indicated that of the 762 students admitted between the sum-

mers of 1974 and 1979, 498 or 64 percent were still enrolled,

while 36 percent were no longer enrolled. Of those students

still enrolled, 38 or 8 percent had 3.0-4.0 grade point

averages; 320 or 65 percent had grade point averages of 2.0-

2.9; 66 or 13 percent were on academic warning; and 65 or

13 percent were on academic probation. This fact-finding

study provided evidence that not all specially admitted

students achieve at the same level and that nonacademic as







12


well as academic type factors may be predictors for the

success of these students.

It is apparent that individuals differ on character-

istics other than high school grade point averages and

standardized test scores. Thus, other factors which may

affect the academic performance of college students should

be studied. What makes one student succeed and another

student with identical or very similar academic background

fail? This question haunts university admissions committees

and administrators and was addressed by the present study.



Definition of Terms


The following terms have been operationally defined

for the purpose of this study:

Attrition. The tendency of students to leave an insti-

tution of higher education prior to receiving a degree due

to the failure of maintaining cumulative grade point averages

of 2.0 or better and/or nonacademic factors.

College entrance exam scores. The college entrance

scores of specially admitted black university freshman

students as measured by ACT (American College Testing) or

SAT (Scholastic Aptitude Test).

Grade point average. The quotient achieved when total

grade points are divided by total hours carried.

Personality scores. The personality scores of specially

admitted black university freshman students as measured by

the Myers-Briggs Type Indicator.







13


Retention. The tendency of students to remain at an

institution of higher education and receive a baccalaureate

degree.

Satisfaction scores. The satisfaction scores of

specially admitted black university freshman students as

measured by the College Student Satisfaction Questionnaire.

Self concept scores. The self concept scores of

specially admitted black university freshman students as

measured by the Tennessee Self Concept Scale.

Specially admitted students. First-time-in-college

freshman students admitted to the University of Florida

through special admission criteria during summer quarter,

1981.



Organization of the Study


The remainder of this study is divided into four

chapters. A review of the literature on specially admitted

students and the academic and nonacademic factors relating

to college student attrition is presented in Chapter Two.

The hypotheses, population and sample, instruments, proce-

dures, analysis of data, and limitations of the study are

explained in Chapter Three. The results of the study are

reported and discussed in Chapter Four. Chapter Five includes

conclusions, implications for further research, a summary of

the study, and recommendations for further research.
















CHAPTER TWO
REVIEW OF RELATED LITERATURE



Introduction


There are many factors which affect the college grade

point averages and attrition of university specially

admitted students. Basically, these factors can be divided

into two categories: academic and nonacademic. For the

purpose of the present study, the review of literature is

discussed in terms of those two major sections. However,

due to the extensiveness of the research in the area of

nonacademic factors and their relationship to college

attrition, the following subheadings are used: self-concept,

personality characteristics, and college student satisfaction.



Academic Factors Related to College Attrition


The results of a recent national survey entitled, What

Works in Student Retention (Beal and Noel, 1979) suggest that

retention efforts for both four- and two-year institutions

need improvement. Various authors have attempted to iden-

tify potential sources of problems associated with low

retention rates (Carney and Geis, 1981; Cleary, 1968; Price

and Kim, 1976; Stanley, 1971). The focus of the majority


14







15


of these studies has been toward academic factors and

their relationship to college attrition. Most of the

criteria for selection into college are based on previous

academic performance in the form of college entrance exam

scores or high school grade point averages. One author

(Aleamoni, 1977) concluded that if grades are to be used

to predict college grade point averages, then they should

be used as the criterion measure in the selection process.

Beal (1979) suggests that a combination of factors,

including high school grade point average and class rank,

first semester college grades, study habits, motivational

level and commitment, student-faculty relationships, and

the fit between the college and student should be considered

when addressing the problem of student retention.

Recent research has revealed a significantly strong

relationship between students who receive below average

first-term grade point averages and attrition (Pantages and

Creedon, 1978). Therefore it would seem that identification

of the reasons for poor academic achievement is a necessity

in the attempt to reduce attrition rates. One such study

(Hart and Keller, 1980) was designed to obtain self-reported

reasons for poor academic achievement among first-term

freshmen. The findings indicated that these students placed

the greatest responsibility for their low grades on them-

selves (i.e., their own lack of motivation, improper study

habits, and inattention to school work).






16


The freshman college year experience has been identified

as being quite different from any other year's experience.

Lokitz and Sprandel (1976) reported that freshman students

demonstrate a movement from academic to social concerns

during the first year. Following the establishment of

academic status, emphasis is focused toward building and

maintaining interpersonal relationships. The results of

another study (Madrazo-Peterson and Rodriguez, 1978) seem

to add credence to the idea that freshman perceptions are

different from those expressed by older and more experienced

students. Freshman students reported fewer feelings of

isolation and significantly greater satisfaction than other

students. These authors suggested that initially freshmen

perceive the campus environment in an idealistic and opti-

mistic light but as matriculation continues, idealism is

replaced by realism and optimism by pessimism. The outcome

is stress and dissatisfaction. Scott (1978) also reported

a significant change between the initial expectations of

black freshman students and their later perceptions about

the campus environment.

The research in the area of academic factors and their

relationship to college attrition is quite extensive. Many

researchers (Kalingal, 1971; Pfeifer, 1976; Bean and Covert,

1973; Thomas and Stanley, 1969) have attempted to correlate

previous academic achievements (SAT scores, high school

rank, and high school grade point averages) and attrition.

The findings, however, have yielded inconclusive and often

conflicting results.







17


Ikenberry (1961) cited the following limitations of

previous studies seeking to identify factors related to

college persistence: 1) too few variables had been in-

cluded; 2) no differentiation was made between students

who withdrew (in terms of achievement level or sex); 3) no

multivariate statistical techniques had been used to

analyze data; and 4) there had been little indication of

the interrelationships among variables.

One of the first studies which attempted to compare

grade prediction equations of Negro and white students was

conducted by Cleary (1968). This study reported mixed

results; no significant difference was found in two inte-

grated colleges in the East, but in a southwest college,

the Negro students' scores were overpredicted by the use

of the white regression equation. Temp (1971) replicated

Cleary's (1968) research and obtained similar results.

The author further suggested that since a predictive system

based on majority students serves to penalize black appli-

cants by underpredicting their potential college performance,

institutions using SAT and other admissions tests and pre-

dictors should conduct a validity analysis at their own

institution. Another researcher (Bowers, 1970) also compared

GPA regression equations for regular and specially admitted

freshmen. High school percentile rank and verbal quantita-

tive scores on the Cooperative School and College Ability

Tests (SCAT) were found to be useful predictors of college

GPA for both groups in this study.







18


One group of researchers conducted a multivariate

study of personality and academic factors in college attri-

tion (Maudal, Butcher, and Mauger, 1974). They concluded

that academic variables such as SAT, high school grade

point average and rank, etc., are useful predictors of

academic achievement with transfers, persisters, and drop-

outs. Lanning (1977) also investigated the relationship

between numerous variables (age, high school rank, SAT

scores, previous semester's grade point averages and number

of hours completed) and college persistence and voluntary

withdrawal. A consistently significant relationship was

found to exist between prior grade point average and prior

semester course load and withdrawal and persistence, that

is, prior to withdrawing, withdrawals had lower grade point

averages and took fewer class hours. Other research con-

ducted in this area has produced inconclusive results.

Price and Kim (1976) in a study designed to 1) identify

the specific factors most likely to be associated with suc-

cess of college education and 2) test the significance of

those factors on college performance, found college entrance

examination scores (ACT) to be more significant and important

as predictors than high school grades. However, just as

some studies have found a significant relationship between

prior academic achievement as evidenced by college entrance

exam scores or high school grade point averages and college

grade point averages, many others have found little or no

significant relationship. Pfeifer and Sedlacek (1971)







19


obtained mixed results suggesting that while high school

grades correlated highly with college grades for black

females, white males and females, the Scholastic Aptitude

Test-Verbal was the best predictor for black males.

In an attempt to determine the relationship between

underachievement and college attrition Dalton, Anastasiow,

and Brigman (1977) studied three variables (predicted

grade point average, Scholastic Aptitude Tests, and relative

high school rank). No significant relationship was found

between persisters, nonacademic dropouts and academic drop-

outs on these three variables. Boyd (1977) in yet another

study found SAT scores to be ineffective as predictors of

college graduation for minority students at selected

institutions.

The relative effectiveness of four variables, SAT-

Verbal, SAT-Math, high school GPA, and family income as

predictors for college graduation among low-income students

was investigated by Rovezzi-Carroll and Thompson (1980).

None of the four variables were found to be effective

either singly or in any combination as predictors of college

graduation, withdrawal, or academic dismissal. These authors

concluded that perhaps college grade point average and college

graduation result from the interaction of different sets

of antecedent variables (self-conCept, study habits) and

intervening variables (college environment, perceived iso-

lationism, provision of tutorial and support services).








20


Houston (1980)evaluated the relationship between

college grade point average (as the dependent variable)

and selection predictor variables (high school rank, and

the Verbal and Mathematics scores on the College Board

Scholastic Aptitude Test) in a sample of specially ad-

mitted black female undergraduates. Results indicated

that after 8 semesters, 19 percent of the students had

withdrawn or were transferred to another institution, 52

percent had graduated, and 27 percent had been dismissed

for academic reasons. Stepwise regression analysis using

the predictor variables revealed a coefficient of multiple

determination (R ) of .22, that is, only 22 percent of

the variance in college grade point average could be

attributed to the three variables. High school rank con-

tributed to the greatest proportion of the variance to R2

(.14). SAT-V and SAT-M accounted for approximately 7 and

1 percent of the variance respectively. Houston stressed

the need to identify predictors other than high school rank

and SAT-V or SAT-M scores because of possible underpredic-

tion in some cases.

The above section of this literature review cited

various studies concerned with academic factors such as

previous high school grade point average and/or rank and

college entrance exam score and their relationship to

college attrition. Although agreement has been reached on

the relevance of adequate academic factors and their

relationship to student attrition, agreement on specific

selection criteria has not proven conclusive. Therefore,







21


previous academic achievement is still a critical factor

in the overall issue of college student attrition.



Nonacademic Factors Related to College Attrition


Research studies concerning nonacademic factors related

to college student attrition are profuse. Although this

section of the literature review cites research investiga-

tions regarding numerous nonacademic factors related to

college student attrition, emphasis is placed on studies

relating to self-concept, personality, and college student

satisfaction.

Mornell (1973) in a study designed to access the suc-

cess of a special admissions program reported a 60 percent

retention rate. Mornell suggested that the traditional

cognitive measures used to predict academic success for all

students are valid only for the student with traditional

credentials, that is, the success of the program depends

on the selection process. Another researcher (Mayhew, 1965)

also suggested that a variety of factors should be taken

into consideration when making decisions about prospective

students. Factors mentioned by Mayhew include past academic

performance, academic aptitude, motivation, personality,

special interests or abilities, extra-curricular skills,

economic status, and physical characteristics. Two other

authors (Dawkins and Dawkins, 1980) similarly suggested

that because of the inconclusive results in the past,







22


additional noncognitive criteria such as racial perceptions

and experiences, goals and values, social participation in

campus activities and social background should be examined

in the context of academic performance.

Extensive research in the area of nonacademic factors

and their relationship to college student attrition has

been conducted (Menning, Bradley, and Cochran, 1975; Morri-

sey, 1971; Muskat, 1979; Wessell, Engle, and Smidchens,

1978). The various factors that have been researched cover

a wide spectrum of variables. One group of researchers

(Ellison, Murray, Fox, and Taylor, 1973.) explored the

effectiveness of biographical inventory data as possible

predictors of college performance. The results indicated

that biographical data were, in general, as effective or

in some cases more superior than previous high school per-

fonnance in predicting college grade point average.

Carney and Geis (1981) used data from a standardized

reading test and student background information to correlate

college academic performance and retention. Both college

entrance test scores and reading test scores were found to

be related to academic performance and attrition. These

authors suggested that the use of reading scores could be

helpful to colleges who seek to identify those students

who are in need of special assistance.

Other research (DiCesare, Sedlacek, and Brooks, 1972)

has explored the ways, if any, in which black students

returning to college are different from those not returning







23


on demographic and attitudinal variables. They reported

that blacks who returned 1) have more self-confidence

and higher expectations; 2) feel more strongly that the

university should influence social conditions; 3) see more

racism at the university; and 4) are more likely to have

lived on campus and made use of its facilities than did

nonreturning blacks. Students were also asked to identify

possible causes that would lead to their dropping out prior

to receiving a degree. The authors concluded that persis-

ters have strong self-concepts, take a more realistic look

at the university, and adapt to it to achieve their own

goals.

In a four-year longitudinal study of 6,660 high aptitude

students, Astin (1964) contended that students who drop out

of college demonstrated the following characteristics:

1) come from lower socioeconomic backgrounds; 2) have lower

ranks in high school; 3) plan initially to get lower college

degrees; and apply for relatively fewer scholarships. When

asked to identify and rate possible sources of academic

problems, black students in both predominately black and

white schools rated their high school preparation as less

adequate (Jones, Harris, and Hauck, 1975).

In an attempt to identify and describe the extent to

which certain influence sources affect attendance and reten-

tion in college, Brown (1976) conducted a survey at the end

of the college student's freshman and sophomore years.

The top three influences were 1) students' own career







24


plans; 2) the mother; and 3) the academic advisor. Webster,

Sedlacek, and Miyores (1979) in a comparison of the problems

perceived by minority students (Hispanic, American Indian,

Asian American, and black American) and white American stu-

dents found that black students expressed more serious pro-

blems in feeling like victims of racism and discrimination

because of race. Other problems identified by all students

included problems regarding vocational decisions, managing

time, studying, and negotiating the university system.

One author suggested specific program strategies beneficial

for both students and the university (Maynard, 1980).

Problems arise because of the inability of many universities

to respond to the special academic, sociocultural, and

economic needs of minority students (Maynard, 1980). The

conclusion reached by Cole and Hanson (1973) following a

study into the extent of racial ethnic bias in selective

college admissions was that college admissions personnel

should give consideration to the relation of selection pro-

cedures to the values and goals of their college.



Self-concept and College Student Attrition

One of the most researched factors associated with

college student attrition is self-concept. Mixed results

have been reported in the literature regarding the self-

concept and its relationship to college student attrition.

Astin (1977) reported positive changes in self-esteem and

suggested that college attendance may increase the indivi-

dual's sense of competence and self-worth. The longitudinal







25


analyses performed in this study demonstrated that students

undergo a variety of changes in self-concept, attitudes,

and values after they enter college. These changes are

evidenced by a more positive self-image and demonstrated

in a greater sense of interpersonal and intellectual

competence. Carey (1976) investigated the nature of the

relationship between the self-concept and academic perform-

ance of black students on white campuses who participated

in black studies and those who did not. No significant

correlation between self-concept and academic performance

was reported. Dowdle (1977) also reported no significant

relationship between self-concept (as measured by the

Tennessee Self-Concept Scale) and academic achievement of

freshman students, that is, students who performed low

academically did not have a lower self-concept than those

who performed higher academically. Similarly, Samuel and

Laird (1974) found no significant differences in the self-

concepts of black females on a predominately black campus

as measured by the Tennessee Self-Concept Scale when com-

pared to black females on a predominately white campus.

In a study designed to determine the effectiveness of

the Special Services Program in a southern university, Cole

(1974) compared the grade point averages and self-concepts

of program participants with nonparticipants. Even though

the Special Services Program appeared to have had a posi-

tive effect on academic achievement, the two groups did

not differ significantly with respect to self-concept.







26


During pretesting, 82 percent of the students had initial

total positive self-concept scores within plus or minus

one standard deviation of the normative population. Cole

hypothesized that the initial level of self-concept pro-

bably accounted for the small change score and the fact

that there was no significant difference between the two

groups as a result of the program treatment. Still another

researcher (Copeland, 1974) studied peer counselor effects

on the self-concept and academic adjustment of a group of

Special Services students who participated in group counseling.

The researcher used the Tennessee Self-Concept Scale and

the College Inventory of Academic Adjustment as measures

of self-concept and academic adjustment. Three groups of

students were used in .this study. Group one consisted of

students with peers as group counselors. Group two consisted

of students with Educational Opportunity Program staff

counselors as leaders. Group three consisted of students

who did not participate in the Educational Opportunity Program.

Groups one and two both changed positively, but not sig-

nificantly on all subscores on the Tennessee Self-Concept

Scale. Peer counselors therefore were found to be as effec-

tive as group counselors when Special Services students were

compared on the two variables.

In contrast to the above mentioned studies, other re-

searchers have reported significant relationships between

self-concept and college attrition. Gadzella and Fournet

(1976) found a significant difference between high and low








27


achievers, with high achievers rating themselves signifi-

cantly higher than low achievers. Green and Farquhar (1965)

investigated the relationship of personality and cognitive

factors with academic achievement of eleventh grade Negro

and white students. These researchers used the Michigan

State M Scales (as a measure of academic motivation), the

verbal score of the School and College Ability Test and the

Verbal Reasoning score of the Differential Aptitude Test

(as measures of aptitude), and current grade point average

(as a measure of school achievement). No correlation was

found between verbal aptitude and achievement for Negro males.

However, a significant correlation between verbal aptitude

and achievement was found for Negro females. Among the

four subsets of the M scales, the single best predictor of

achievement for the Negro sample was the self-concept.

Therefore, a strong relationship between the students' self-

perception and school achievement was reported.

Jones (1975) researched the effect of personal growth

group counseling on the self-concept and academic achieve-

vement of Special Services Program participants. This re-

searcher defined self-concept as a multi-dimensional phenom-

enom and therefore hypothesized that change may take place

in one or more facets without that change being evident in

the total self-picture. The effect of personal growth group

counseling on five scales (the physical, personal, social,

family and the moral-ethical) of the Tennessee Self Concept

Scale was compared to grades at the end of the summer







28


session in two classes (sociology and English composition).

Students were randomly assigned to one of three groups:

1) those who received group counseling; 2) those who

received tutoring; and 3) those who received no treatment.

The findings indicated that while the personal growth

group counseling appeared to improve the total self-concept

of males, the same did not hold true for females. However,

overall, the students who received group counseling had

more positive self-concepts and greater general academic

achievement than those not receiving group counseling.

One study (Morrison, Thomas, and Weaver, 1973) sought

to determine the relationship between self-esteem and self-

estimates of academic performance. These researchers hypo-

thesized that students with low self-esteem would predict

receiving low grades and students with high self-esteem

would predict receiving high grades on an examination.

Mixed results were obtained for two separate measures of

self-esteem. The hypothesis was confirmed for the Cooper-

smith Self-Esteem Inventory, but not for the Ziller Social

Self-Esteem scale or for the subscale of the Coopersmith

inventory specifically relevant to school self-esteem.

Another researcher (Pazandak, 1975) also investigated the

role of self-estimation as a moderator variable in the

prediction of college academic achievement. Valine (1976)

studied the effects of dropping out of school on the self-

concept of students initially identified as underachievers.

The findings indicated that those who had attained senior







29


status generally had a statistically more positive self-

concept (as measured by the Tennessee Self-Concept Scale)

than those, who for various reasons, dropped out of college.

The studies cited above have produced inconclusive

results regarding the relationship between various nonacademic

factors, including self-concept and college student attri-

tion. Therefore, it is apparent that there is needed addi-

tional research in this area.



Personality Characteristics and College Student Attrition

Various attempts have been made to relate both academic

and nonacademic factors to college student attrition. One

nonacademic factor which has been researched is the person-

ality of the students involved. In a historical three-year

follow-up study on the fate of minority group freshmen who

were admitted to the University of Florida under a 5 percent

exemption quota, Cranney and Larsen (1972) concluded that

"early identification of college potential and appropriate

programs to develop it may further enhance success rates of

marginally admitted freshmen" (p. 40). Therefore, it would

seem that any information which could lead to improving

this early identification process would be welcomed, in-

cluding information regarding personality characteristics.

Pfeifer and Sedlacek (1974) utilized two personality

measures (the California Psychological Inventory and the

Holland Vocational Preference Inventory) in an attempt to

determine the relationship between these two variables and







30


college grades. The results indicated some relation between

the measured college grades. The researchers therefore

suggested that nonintellectual measures should be used in

combination with intellectual measures to predict which

ones uniquely and appropriately reflect the experiences of

blacks in the society.

Hannah (1971) investigated the relationship between

personality traits and college entrance test scores of drop-

outs and "stayins." The personality traits of dropouts were

found to differ significantly from "stayins" on several

scales of the Omnibus Personality Inventory. Persisters

also had obtained higher college entrance test scores.

Another study (Tillman, Millott, and Larsen, 1974) found no

significant differences in mean reading test scores (as

measured by the McGraw-Hill Reading Test) and personality

type (as measured by the Myers-Briggs Type Indicator) when

student preferences were compared on the four indices of

the Myers-Briggs Type Indicator. However, higher end-of-the-

quarter grade point averages were obtained by those students

who preferred Feeling than those who preferred Thinking.

Pandey (1972) utilized the Minnesota Multiphasic Per-

sonality Inventory in an attempt to differentiate personality

characteristics of successful, dropout, and probationary

students. This researcher reported few significant differences

between the three groups of students. Johnson (1970) also

compared-personality characteristics and grade point averages

of persisters and nonpersisters and found distinct differences







31


in regards to personality characteristics (as measured by

the Minnesota Counseling Inventory). However, both persis-

ters and nonpersisters had above 2.0 cumulative grade point

averages and thus, no significant difference was reported in

regards to grade point averages.

Rossmann and Kirk (1970) examined the differences in

ability (as measured by the Omnibus Personality Inventory),

and attitude (as measured by questionnaire data) and found

that voluntary withdrawals had higher verbal ability and

were more intellectually oriented than those required to

withdraw for academic violations. Leon (1974) hypothesized

that personality and college grade point averages were sig-

nificantly related. However, no significant differences were

found and Leon suggested that the reason may lie in the role

that academic success may play in the lives of specially

admitted students.

Millott (1974) reported significant correlations ranging

from .065 to .186 between the MBTI proference scores and the

McGraw-Hill Reading Test scores of 2,514 University of Florida

freshmen.. However, although the correlations with the MBTI

were in the direction predicted by type theory, this cor-

relation was not large. Introverted-Intuitive (IN) types

tended to be more successful in academic endeavors and scored

higher on achievement tests than types in the other three

quadrants of the MBTI. These types (IN) were also found to

have higher scores than the other types on scholastic

aptitude tests. This study reported no statistically sig-

nificant differences by sex between type and reading.






32


McCaulley (1972) researched Expanded Educational

Opportunities Program freshmen at the University of Florida

in an attempt to identify personality types. The findings

indicated that out of the total sample of 135, the majority

(105 or 77 percent) of the students were identified as

Sensing types. A total of only 30 or 22 percent were on

the Intuitive side of the continuum. The three type pre-

ferences which yielded the highest number of students were

ESTJ (N = 29 or 21.5 percent); ESFJ (N = 20 or 14.8 percent),

and ISFJ (N = 17 or 12.6 percent). The three type pre-

ferences which yielded the lowest number of students were:

ENFJ and ENTP (both with Ns of 1 and percentages of 7 per-

cent each) and ESTP and INFP (both with Ns of 2 and per-

centages of 1.5 percent each).

McCaulley (1974) conducted another research study with

Special Services Program participants and obtained results

very similar to those reported above. The majority (N = 129

or 84 percent) of the 154 students in this study were iden-

tified as Sensing types. A total of only 25 or 17 percent

were on the Intuitive side of the continuum. The three type

preferences which yielded the highest number of students

were ESTJ (N = 32 or 20.8 percent) and ISFJ and ESFJ

(both with Ns of 24 and percentages of 15.6 percent). The

lowest percentages were recorded by the following type pre-

ferences: ENTP (N = 0), ESTP (N = 1 or .6 percent), and

INFJ, INTJ, and INTP, all with Ns of 2 each and percentages

of 1.3 percent each.







33


In a recent follow-up study Schroeder, Warner, and

Malone (1980) utilized the Myers-Briggs Type Indicator in

conjunction with retention-attrition data and reported the

following retention rates by personality type: Intuitive,

71 percent; Feeling, 71 percent; Sensing, 78 percent; and

Thinking, 57 percent.

The results of these studies indicate that although

attempts have been made to correlate specific personality

characteristics with certain levels of college academic

achievement, attrition, and subsequent graduation, research

in this area is still needed. While the evidence generally

indicates that some type of relationship does exist between

personality characteristics and college student attrition,

the extent and preciseness of this relationship is still

a topic of much debate in higher education.



College Student Satisfaction and College Student Attrition

Until recently, the least researched phenomenon relating

to college student attrition has been college student satis-

faction. However, college student unrest and the turbulence

of the Vietnam era ushered in the need for research in this

almost forgotten area. Thus, there has been a tremendous

increase in the number of investigations concerning college

student satisfaction during the last ten years. Numerous

attempts have been made to correlate several factors with

college student satisfaction. These include age (Sturtz,

1971), tenure in college (Starr, Betz, and Menne, 1972),






34


institutional assessment (Hallenbeck, 1978), and student-

college congruence (Pervin, 1967, Morstain, 1977).

Prior to the disturbances of the 1960's, Berdie's

(1944) was the only recorded report on the relationship

between college academic achievement and student satisfac-

tion. Berdie investigated engineering students' satisfac-

tion with the curriculum as it related to academic achieve-

ment. This study utilized first-year honor points, high

school grades, and scores on a series of ability tests as

measures of performance. Even though satisfaction was found

to be significantly related to academic achievement, no

single factor was found to relate to a student's satisfac-

tion with his/her curriculum. Berdie suggested a need for

a more complete measure of student satisfaction.

Martin (1968) attempted to evaluate freshman students'

perception and their degree of satisfaction by using a

modified Q-sort which consisted of college-oriented items.

Students were assumed to be satisfied with college if their

real and ideal Q-sorts were similar. If Q-sorts were dis-

similar, students were assumed to be dissatisfied with col-

lege. The results indicated that initially freshmen were

moderately satisfied with college but that this satisfac-

tion decreased by the end of the year. Results also indi-

cated that there was no statistically significant relation-

ship between initial freshman satisfaction with college and

academic achievement at the end of the year. King and

Walsh (1972) utilized the College and University Environmental







35


Scales (CUES) to investigate change in environmental

expectations and perceptions of college students. The

findings from this study indicated that freshman year

experiences have a direct impact on student perceptions

of the environment.

Astin (1977) reported that students, in general,

maintained relatively high satisfaction with most aspects

of their college experience as measured by a follow-up

questionnaire. Women students were found to be slightly

more satisfied than men with their collegiate experience

and whites were found to be relatively more satisfied than

blacks. The most important institutional characteristics

affecting student satisfaction were academic selectivity,

prestige, and institutional size.

Scott (1978) attempted to determine if black and white

freshman students had different initial expectations and

later perceptions of their college environment. Two of the

many variables that this researcher hypothesized would in-

fluence student expectations and actual perceptions of the

campus atmosphere were high school scholastic achieve-

ment and freshman year scholastic achievement. Students

were asked to complete the College and University Environ-

ment Scales, Second Edition (CUES 2), on two occasions with

a seven-month interval between administrations. During

the first administration, students were asked to complete

the CUES 2 in terms of what they expected to be true or

untrue of their college environment. On the second occasion,

students were requested to complete the CUES 2 in terms of







36


their actual perceptions of the campus atmosphere. Scott

hypothesized that there would be a significant difference

between high school grade point averages and student ex-

pectations and freshman year accumulative grade point

averages and student expectations and/or actual percep-

tions. However, the findings indicated that neither of

the two variables seemed to influence or contribute to

student expectations and perceptions.

In spite of the tremendous gains in knowledge that

have come from these recent studies, the question still

remains: What is the relationship between college student

satisfaction and attrition? It seems clear that more data

are needed both as a needs .assessment tool for planning

and implementing institutional policy as well as for a

general indication of college students' attitudes and

satisfaction with regard to their college experiences.

Betz, Menne, Starr, and Klingensmith (1971) attempted

to systematically investigate college student satisfaction

based on a factor analytic study. Their study produced

consistent results across two groups of college under-

graduates who were administered the questionnaire indepen-

dently during the fall and winter terms. In yet another

study, Betz, Starr, and Menne (1972) proposed to investigate

the degree of student satisfaction with colleges and univer-

sities. In a comparison of large public university students

with small private colleges, public university students

appeared to be more satisfied with social life and working






37


conditions in their institutions than did private college

students. On the other hand, private college students

appeared to be more satisfied with recognition, quality of

education, and compensation for effort. However, both

groups were found to be satisfied with various aspects of

their college experience as measured by the CSSQ.

Hallenbeck (1978) views satisfaction indicators as

primary sources of information regarding institutional

assessment. Thus, the information obtained from satisfac-

tion studies can be used as baseline data to guide in pro-

gram planning, implementation, and change. Schmidt and

Sedlacek (1972) used an activities and attitude inventory

to survey variables related to university student satisfac-

tion. Their findings reflected the optimism and idealism

expressed by freshman students. In a study designed to

investigate the relationship between age and college student

satisfaction, Sturtz (1971) reported that older women (25

or above) were found to be generally more satisfied than

younger women (18 to 24).

In a comparison of black and white student satisfaction,

Robertson (1980) utilized the College Student Satisfaction

Questionnaire. Black student satisfaction was found to be

significantly lower than that of white students only with

respect to the university's system of rewards (compensation).

On the scales of social life, working conditions, and quality

of education, the satisfaction level of black students was

found to be slightly higher than that of their white







38


counterparts. Although black student satisfaction on

the recognition scale was slightly lower than that of white

students, the difference was not statistically significant.

Also, even though the overall level of black student satis-

faction was lower than that of white students, no significant

differences were found.

In an effort to identify differences in college student

satisfaction among academic dropouts and nonacademic drop-

outs, Starr, Betz, and Menne (1972) utilized the College

Student Satisfaction Questionnaire. These researchers hypo-

thesized that 1) students who remained in college would

be more satisfied than those who dropped out; and 2) drop-

outs who left for nonacademic reasons would be more satisfied

than academic dropouts. There were three scales of the

College Student Satisfaction Questionnaire which were found

to discriminate between students who remained and those who

dropped out. These were compensation, recognition, and

quality of education, that is, dropouts were consistently

less satisfied on these three scales.

In a recent study Lindsey (1981) utilized the College

Student Satisfaction Questionnaire (CSSQ) along with the

University Involvement Questionnaire (UIQ) in an attempt

to investigate black and white undergraduate university

students' degree of satisfaction with various aspects of

their university environments and their levels of involvement

in those environments. A total of 800 undergraduate students

from two predominately white (Florida State and the University







39


of Florida) and one predominately black (Florida Agri-

cultural and Mechanical University) institution were in-

cluded in the study. Lindsey hypothesized that

1) there would be no differences between black and white

undergraduate university students' degree of satisfaction

with their university environment; 2) there would be no

differences between the degree of satisfaction of black

undergraduate students attending predominately white

universities and those attending a predominately black

university; and 3) there would be no differences between

the degree of satisfaction of white undergraduate students

attending a predominately black university and those

attending predominately white universities.

The means for the 373 blacks in the study on the five

scales of the CSSQ were reported as follows: Working

Conditions, 40.87; Compensation, 42.65; Quality of Education,

37.72; Social Life, 40.32; and Recognition, 42.64. The

total score mean was 204.22. These means compared to 42.17,

45.08, 39.44, 39.83, 41.77, and 208.29, respectively, for

the 427 whites included in the study. The total score mean

for black undergraduate students attending the University

of Florida was reported as 200.95. Although mean total

satisfaction scores for black undergraduate students at the

two predominately white institutions were slightly lower

than that of their white counterparts, the differences were

not statistically significant. Lindsey therefore concluded

that black undergraduate students attending predominately





40


white universities were as satisfied with their university

environments as black and white students attending pre-

dominately black and white universities. However, black

students were found to be less satisfied than white students

on the scales of working conditions, compensation, and

quality of education in their university environments as

measured by the CSSQ.

Recent studies in the area of college student satis-

faction have proven to be beneficial in the attempt to

identify factors related to college student attrition. The

ability to identify exit-prone students prior to their

leaving seems a necessity for those persons delegated the

responsibility for increasing graduation and retention rates.















CHAPTER THREE
METHODOLOGY



Introduction


The purpose of this study was to investigate the

relationship between five independent variables (high

school grade point averages, college entrance exam scores,

personality type scores, self-concept scores, and college

student satisfaction scores) and one dependent variable

(college grade point average) in a sample of black specially

admitted university freshmen. The hypotheses, population

and sample, instruments, procedures, analysis of data, and

limitations of the study are discussed in this chapter.



Hypotheses


The following hypotheses were investigated:

HO1: There is no significant relationship between

college grade point averages and high school

grade point averages of specially admitted black

university freshman students.

HO2: There is no significant relationship between

college grade point averages and college entrance

exam scores of specially admitted black univer-

sity freshman students.

41






42


HO3: There is no significant relationship between

college grade point averages and self-concepts

of specially admitted black university freshman

students.

HO4: There is no significant relationship between

college grade point averages and personality

type preferences of specially admitted black

university freshman students.

HO5: There is no significant relationship between

college grade point averages and college student

satisfaction of specially admitted black univer-

sity freshman students.

HO6: There is no significant relationship between

personality type preference and college student

satisfaction of specially admitted black univer-

sity freshman students.

HO7: There is no significant relationship between

self-concept and college student satisfaction of

specially admitted black university freshman

students.



Population and Sample


The target population to which the results of this

study is generalizable to includes black freshman students

who have been admitted to four-year public universities

under special admission guidelines. These students enter

the university as first-time-in-college freshmen directly







43


from high school and participate in special admittance

programs such as those described in the statement of the

problem section of this study. These students have not

met the minimum academic admission standards set by univer-

sities which would allow them to be admitted as regular

students and are thus considered "high risk." However,

they are granted admission through special procedures which

may include a special subcommittee which reviews their

records along with recommendations from counselors, teachers,

and administrators who feel they have the potential for suc-

cess in college. The majority of these students receive

some type of financial aid from either federal and/or state

assistance programs.

From a total population of 142 specially admitted

students who were admitted to the University of Florida

during summer 1981, a sample of 103 black freshmen (69 females,

34 males) were chosen for this study. All of the subjects

were present and participated in the first day of the five-

day orientation conducted by the two special admission pro-

grams (PACT and Special Services). The SAT (Scholastic

Aptitude Test) or ACT (American College Testing) scores of

these students were either below the minimum admission

requirement of 800 and 17, respectively, and/or their HSGPA

(high school grade point averages) were below 2.0.






44


Instruments


The Tennessee Self Concept Scale, the Myers-Briggs

Type Indicator, and the College Student Satisfaction

Questionnaire were the instruments used in this study.

The Tennessee Self Concept Scale is a 100-item instrument

which is also self-descriptive and was used to identify

"self-perceived" attributes of specially admitted black

students. The Myers-Briggs Type Indicator is a self-

descriptive instrument and was used to "type" and identify

personality characteristics of specially admitted black

students. The College Student Satisfaction Questionnaire

consists of 70 items relating to various aspects of college

and university life and was used to assess college student

satisfaction of specially admitted black students.



Tennessee Self Concept Scale

The Tennessee Self Concept Scale consists of 100 self-

descriptive, self-administering items by which responses

are recorded in a Likert-type scale of five choices ranging

from completely false to completely true. The manner in

which individuals perceive themselves influences their

behavior and their relationships and is related to their

overall personality. Therefore, being aware of how people

view themselves is both useful and necessary in order to

assist them.

Two forms of the TSCS are available for use, a Counseling

Form and a Clinical and Research Form. The Counseling Form






45


was used for the purpose of this study since it is less

time consuming and the only differences between the two

forms is the method of scoring and the profiling system.

Both forms can be scored either by hand or machine. The

time required to complete the scale varies from 10 to 20

minutes.

The Counseling Form generates five scores which

provide information regarding an individual's self-concept.

The following is a brief description of each of these scores:

1. The Self-Criticism Score (SC) is composed of 10

items taken from the L-Scale of the Minnesota

Multiphasic Personality Inventory (Hathaway and

McKinley, 1951). These are statements (even though

negative to some extent) which most persons agree to

as being true for them. In general, low scores

(denial) indicate defensiveness and high scores

indicate a normal capacity for self-criticism.

2. The Positive Scores (P) consist of three horizontal

categories (Row 1, Row 2, and Row 3) which re-

present an internal frame of reference and five

vertical categories (Column A, Column B, Column C,

Column D, and Column E) which represent an external

frame of reference. In addition, a Total P score

is also calculated.

a. Total P Score is the most important single

score on the Counseling Form because it reflects

the overall level of self-esteem.







46


b. Row 1 P Score (Identity) indicates "what I am."

The individual describes what he perceives to

be his basic identity.

c. Row 2 P Score (Self-Satisfaction) indicates

"how I feel about myself." This score reflects

how satisfied an individual is with himself.

d. Row 3 P Score (Behavior) indicates "what I do,

or how I act." This score reflects the

individual's perception behavior.

e. Column A (Physical Self) indicates how an

individual views his body and overall physical

well-being.

f. Column B (Moral-Ethical Self) describes how

an individual views himself in terms of moral

worth, religion, relationship to God, etc.

g. Column C (Personal Self) measures the indivi-

dual's feelings of adequacy without regard to

his body or relationships to others.

h. Column D (Family Self) measures the individual's

feelings of value and usefulness as a family

member.

i. Column E (Social Self) describes the individual's

feelings of adequacy and usefulness in relation

to other people in general.

3. The Variability Scores (V) reflect the inconsistency

which may exist among different areas of the indi-

vidual's self-perception. Three (V) scores are

measured: Total V, Column Total V, and Row Total V.






47


a. Total V measures the sum of all variability.

b. Column Total V measures the sum of the varia-

tions within the columns.

c. Row Total V measures the sum of the variations

across the rows.

4. The Distribution Score (D) summarizes how an indi-

vidual varies answers across the five available

response categories. This score may also indicate

how certain an individual is about the choices made.

5. The Time Score measures the amount of time it

requires the individual to complete the scale.

Norms for the scale were developed from a group of 626

people from various parts of the United States and varying

in age from 12 to 68. Subjects included members from high

school and college classes, and employers at state institu-

tions in addition to other sources. The sample included

both black and white subjects, representatives from all

social, economic, and intellectual levels and educational

levels from sixth grade through the Ph.D. degree, and an

approximately equal number of males and females. The test-

retest reliability coefficients based on a group of 60

college students over a two-week period (Fitts, 1965) are

as follows: Self-Criticism, .75; Total Positive, .92;

Total Variability, .67; Distribution, .89; and Time .89.

The four procedures used by Fitts (1965) to establish

validity were content validity, discrimination between






48


groups, correlation with other measures, and personality

changes under particular conditions. Content validity was

established by having clinical psychologists analyze each

item. Only unanimous agreement by the judges that it was

classified correctly allowed an item to remain a part of

the scale. Various groups were used to demonstrate that

the scale is a reliable discriminator between psychiatric

patients and nonpatients, delinquents and nondelinquents,

and the average person and a psychologically integrated

person.

Further assessment of validitywas established by com-

paring the scale with other personality measures. In com-

paring the scale with the Minnesota Multiphasic Personality

Inventory (MMPI), it was found that most of the scores of

the scale correlated with MMPI scores. Correlations with

the Edwards Personal Preference Schedule indicate a nonlinear

relationship between scores on the two tests (Sundby, 1962).

In order to validate personality changes under parti-

cular conditions, the effects of positive and negative life

experiences were studied. Gividen (1959) reported a sig-

nificant decrease in scores following a stressful situation.

Ashcraft and Fitts (1964) studied changes in patients' scale

scores due to psychotherapy. They reported that the therapy

group changed significantly and in the expected direction on

18 of the 22 variables studied, while the control group

changed in only two variables.






49


The TSCS was used to identify self-concept because

the development of this instrument is based on the premise

that the way a person perceives himself influences his

behavior. Therefore, the ability to obtain and identify

self-reported self-concept scores would be of invaluable

use for the purpose of this research.



Myers-Briggs Type Indicator

The Myers-Briggs Type Indicator is a 166-item, forced-

choice, self-administering, self-validating instrument,

first published in 1962 by Educational Testing Service and

later in 1975 by Consulting Psychologists Press. It was

designed to facilitate use of that part of Jung's theory

which describes psychological types. In essence the theory

proposes that variations in human behavior which appear to

be random at first glance are in actuality orderly and

planned. Perceived differences only exist in the manner by

which people choose to make use of their perception and

judgment. The purpose of the MBTI is to identify and "type"

individuals in terms of four preferences: Extraversion-

Introversion (EI); Sensing-Intuition (SN); Thinking-Feeling

(TF); and Judging-Perception (JP). The kinds of perception

are sensing and intuition; thinking and feeling are the two

kinds of judgment. Myers (1962) postulates that if people

differ systematically in what they perceive and the con-

clusions they come to, they may show corresponding differ-

ences in their reactions, interests, values, needs,






50


motivations, in what they do best, and in what they like

best to do. The indicator aims to obtain people's basic

preferences as they relate to perception and judgment so

that the outcome of the preferences and their combinations

may be established by additional research and put to prac-

tical use.

Each of the four independent preferences is "scored"

on a continuum and yields atotal of 16 possible "type"

combinations. The following is a brief description of the

four basic preferences:


DIRECTION OF INTEREST

Extraversion E Introversion I
Prefers the outer world of Prefers the inner world
actions, objects, and people, of ideas and concepts.

PERCEPTION

Sensing S Intuition N
Present-oriented; relies on Future-oriented; looks
immediate reality, known into inferred meanings
facts, and direct experience, and relationships.

JUDGMENT

Thinking T Feeling F
Views events objectively, Views events subjectively;
scientifically; uses the uses personal values and
scientific method to make importance to make decisions.
decisions; logical order--
cause and effect.

LIFE STYLE

Judging J Perceptive P
Prefers planned, decided Prefers a flexible, spon-
orderly way of life; looks for taneous way of life; open-
closure in decision making. ended, tentative schedule.


Figure 1
Myers-Briggs Type Indicator Preferences






51

Studies with academic populations have yielded fairly

low intercorrelations for EI and TF, with the median

absolute interrelation ranging from .03 for males and .06

for females. However, the JP index has been shown to

correlate consistently with SN. The range of this correla-

tion is from .20 to .47. Split-half reliability correla-

tions were obtained by computing tetrachoric r's and

applying the Spearman-Brown Prophecy Formula. The correla-

tions for types are EI, .74 to .88; SN, .77 to .88; TF, .66

to .90; and JP, .76 to .93.

The results of the extensive validity studies that have

been conducted are difficult to summarize. Even though

theoretically the most important correlates for the MBTI

are the "types" themselves, evidence is available to support

the validity of the theory and the indicator. This evidence

is found in the ability of the "type" preferences to cor-

relate positively with interests, values, and needs identified

by other tests, or to correlate with any other external in-

dication of internal differences. The Gray-Wheelwright

(which also proposes to identify the Jungian types) and the

MBTI yield correlations of .79, .58, and .60 for EI, SN, and

TF, respectively. The Gray-Wheelwright has no scale for JP.

The MBTI correlates with 103 of the 180 correlations of the

Strong Vocational Interest Blank. These correlations are

significant at the .01 level. When compared to the Allport-

Vernon-Lindzey study of values, two-thirds of the 24 correla-

tions with the MBTI are significant at the .01 level (Myers,1962).







52


All instructions are contained on the cover page of

the test booklet, and there is no time limit. Two forms

of the MBTI are available, Form F and Form G, both with

standard IBM answer sheets that may be either hand or

machine scored. For the purposes of identifying the "type"

preferences of each subject in the present study, Form G

was used. Continuous scores transformed from preference

scores were computed. A brief description of this trans-

formation process is explained by Myers (1962) in the MBTI

manual (p. 9). For an I, N, F, or P score, the continuous

score is the preference score plus 100. For an E, S, T, or

J score, the continuous score is 100 minus the preference

score. A table is provided to speed the transformation

process (p. 10).

The MBTI was used to identify personality types because

it seeks to identify how individuals use perception and

judgment in making decisions. Myers (1962) suggests that

identifying whether an individual is influenced by intellec-

tual or economic values could contribute to a more accurate

matching of applicants to the college of their choice.



College Student Satisfaction Questionnaire

The College Student Satisfaction Questionnaire (CSSQ)

was designed to measure five selected dimensions of college

student satisfaction. The theory behind its development

stems from research based on the satisfaction of employees

in business and industry (Herzberg, Mausner, Peterson, and






53


Capwell, 1957). Job satisfaction research has shown a

consistently negative relationship between job satisfac-

tion and job turnover, that is, the greater the satisfac-

tion, the less the turnover. Therefore, if the college

environment can be viewed as a place of "employment" and

the student as an "employee," then student satisfaction

with college should be negatively related to turnover

(dropping out of college). It has been the purpose of much

of the research with the CSSQ to test the validity of this

analogy.

The initial form of the CSSQ (Form A) consisted of a

139-item instrument. The model for the format of the instru-

ment is the Minnesota Satisfaction Questionnaire which was

developed by Weiss, Dawis, England, and Lofquist (1967) to

measure job satisfaction. The 139-item instrument proposed

to measure six selected satisfaction dimensions: Policies

and Procedures, Working Conditions, Compensation, Quality

of Education, Social Life and Recognition.

An analysis of data following the administration of the

measure to 643 Iowa State University students yielded in-

ternal consistency reliability coefficients for each of the

six scales. These reliabilities ranged from .85 to .92 with

a median of .88. Relatively normal scale score distributions

were also found. Between scale correlations ranged from .39

(Social Life and Compensation) to .77 (Policies and Proce-

dures and Quality of Education), with a median correlation

of .54.







54


The present form of the CSSQ (Form C) was developed

on the basis of the above analyses and consists of a 70-

item, five-choice, Likert-type scale. There are five

response alternatives offered which range from "Very Dis-

satisfied," through "Satisfied," to "Very Satisfied," and

are scored one to five points, respectively. Scale scores

are derived from the sum of each of the 14 selected items.

A total satisfaction score is obtained by summing all 70

item responses.

The five dimensions of college student satisfaction

which are outlined by Starr, Betz, and Menne (1971), can

be summarized as follows:

1. Working Conditions assesses the student's physical

environment such as the cleanliness and comfort

of his living space, adequacy of study areas on

campus., quality of meals, facilities for relaxing

between classes.

2. Compensation judges the amount of input (e.g.,

study) required relative to academic outcomes (e.g.,

grades) and the effect of study requirements on

the student's satisfaction of his other needs and

goals.

3. Quality of Education surveys the various academic

conditions relative to the student's intellectual

and career development such as the competence and

'helpfulness of faculty and staff, including ad-

visors and counselors, and the adequacy of curricu-

lum requirements, teaching, methods, assignments, etc.







55


4. Social Life refers to the opportunities to satisfy

socially relevant goals such as dating, meeting

compatible or interesting people, making friends,

participating in campus events and informal social

activities.

5. Recognition assesses the acceptance of the student

as a worthwhile individual by faculty and other

students as determined by their attitudes and

behaviors.

Reliability coefficients for each of the five CSSQ

scales (Form C) are reported in the manual separately for

public universities and private colleges. These reliabilities

range in public universities from .78 (Quality of Education)

to .84 (Compensation) with a median of .82. For private

schools, reliabilities range from .79 (Quality of Education)

to .84 (Recognition) with a median of .82. Total score

reliability coefficients are .94 for both normative groups

(Starr, Betz, and Menne, 1971). A more recent study (DeVore

and Handal, 1981) reported significant and uniformly high

test-retest reliability coefficients. These reliabilities

are: Working Conditions, .86; Compensation, .85; Quality

of Education, .87; Social Life, .90; and Recognition, .84.

Between scale correlations range from .46 to .70

(private colleges), the average being .50; for private

universities, scale correlations range from .35 to .62, with

the average correlation being .44. Various studies have

been conducted to investigate the validity of the CSSQ as






56


a measure of college student satisfaction (Betz, Menne,

Starr, and Klingensmith, 1971; Starr, Betz, and Menne,

1972; Betz, Klingensmith, and Menne, 1970). These studies

basically have developed out of the conceptualization which

initiated the development of the CSSQ (i.e., that student

satisfaction can be viewed as being analogous to job satis-

faction; therefore, findings from job satisfaction research

should be applicable to studies of college student satisfac-

tion).

Job satisfaction research has consistently shown a

negative correlation between job satisfaction and turnover

(i.e., higher satisfaction is associated with less turnover,

a greater likelihood that the worker will remain on the job,

rather than quit) (Starr, et al. 1971). Previous research with

the CSSQ has proven this instrument to be both a valid and

reliable measure of college student satisfaction. There-

fore, the use of this instrument will assist in the identi-

fication of the degree of college student satisfaction ex-

perienced by specially admitted students. The usefulness of

the CSSQ as a measure of college student satisfaction was

established by Betz, Klingensmith, and Menne (1970). The

results indicated that the instrument is an internally

consistent measure of several dimensions of college student

satisfaction. DeVore and Handal (1981) have also established

uniformly high test-retest reliability coefficients for each

of the five scales of the CSSQ in a private university.







57


Procedures


A coordinated summer orientation program is planned

for all incoming specially admitted students by the direc-

tors of PACT, Special Services, and other staff members at

the University of Florida. This committee of selected

personnel from the office of admissions, division of

student affairs, student mental health unit of the infirmary,

speech and hearing clinic, office of instructional resources,

and staff members from both special admittance programs,

begins meeting approximately six months preceding the

students' arrival on campus to plan the orientation program.

The names and addresses of these students are distributed

to the directors of the special admission programs as soon

as they are granted admission thus allowing immediate selec-

tion into one of the two programs. The directors then send

letters of welcome and tentative schedules of the five-day

orientation program along with an offer of assistance if

problems should arise prior to their arrival on the campus.

Once students arrive on campus, they are met by their

individual peer counselor who assists them with checking

into their assigned dormitory areas. The main function of

the peer counselor is to assist the student in making a

smooth transition from high school to college life. These

peer counselors have been trained during the previous term by

the orientation staffs of both special admittance programs.

Those students who are not planning to reside in campus







58


housing are asked (via the letter of welcome) to "sign in"

with the specific program which sent them the letter.

The first formal meeting is used to answer questions

about college life at the University of Florida, and also

during that session information is given regarding financial

aid, housing, registration, and other vital data necessary

for a successful tenure at the University. Various faculty

and staff members from within the University, as well as

prominent community leaders are on hand to personally wel-

come these students and to provide additional information

concerning their roles and the services they offer.

During this first orientation session students are also

administered various diagnostic tests and inventories.

Although the format may vary from year to year, generally

diagnostic tests of math, English, and reading are given

along with some type of personality and/or self-concept

inventory. These tests are spaced throughout the five-day

orientation period. On the first day of the five-day summer

orientation, all participants were administered the Tennessee

Self-Concept Scale and the Myers-Briggs Type Indicator,

prior to the beginning of the first session in a group setting.

Both instruments were administered by two black female

graduate students from the Counselor Education Department

at the University of Florida who were assisted by peer coun-

selors in the dissemination of materials. All students were

then allowed to proceed with the continuation of the orienta-

tion program.






59


A list of all specially admitted students who partici-

pated in the first session of the orientation and who com-

pleted the Tennessee Self Concept Scale and/or the Myers-

Briggs Type Indicator was compiled. This list was sent to

the Office of Admissions where vital information regarding

high school grade point averages and college entrance exam

scores was requested. At the end of Fall Semester, 1981,

the list also was submitted to the Registrar's Office along

with a request for cumulative grade point averages for two

terms (Summer, 1981, and Fall, 1981) at the University of

Florida. During the third week of Spring Semester classes,

each student in this study was contacted by his peer

counselor and asked to complete the College Student Satis-

faction Questionnaire.



Analysis of Data


This study focused on one dependent variable (college

grade point average) and five independent variables (high

school grade point averages, college entrance exam scores,

self-concept scores, personality type scores, and college

student satisfaction scores) as they concerned specially

admitted black students at the University of Florida. The

Pearson Product Moment Correlation was used to determine if

a relationship existed between the dependent variable and

each of the five independent variables, and further, to

determine if a relationship existed among each of the five






60


independent variables. Separate statistical analyses were

performed for both the total group (which included those

with missing data), as well as for only those with complete

data on all variables. A stepwise regression analysis was

used to determine the best model for predicting college

grade point average (that is, which variables in combina-

tion accounted for, or contributed most to individual

variance). Frequency distribution tables were established

for the Myers-Briggs Type Indicator.



Limitations of the Study


Several limitations of this study exist. One major

limitation concerns the subjects in this study. Subjects

were limited to specially admitted black freshman students

at the University of Florida and did not include regularly

admitted students, other ethnic groups or races, or students

of differing academic levels. Therefore, the results are

limited in generalizability, that is, results obtained from

this study are not generalizable to all students.

An additional limitation concerning the subjects relates

to the fact that not all students who were admitted through

special admittance guidelines during Summer, 1981, partici-

pated in the first session of the orientation program due to

late high school graduation. Therefore, all specially

admitted students did not complete the Tennessee Self Concept

Scale and the Myers-Briggs Type Indicator. This may have

affected the results.






61


A final limitation concerns the inability to collect

data on all subjects for all variables used in this study.

The College Student Satisfaction Questionnaire was given

approximately six months following the giving of the other

two instruments. All specially admitted students who com-

pleted the two initial instruments did not complete the

CSSQ. Therefore, although initially 103 subjects were

identified, only 67 could be included in the multiple re-

gression analysis.















CHAPTER FOUR
RESULTS AND DISCUSSION



Results


The purpose of this study was to investigate the

relationship between five independent variables (high school

grade point averages, college entrance exam scores, per-

sonality type scores, self-concept scores, and college

student satisfaction scores) and one dependent variable

(college grade point average) in a sample of specially

admitted black university freshmen. One hundred and three

students participated in this study. Of that number, sixty-

seven were identified as having complete data on all vari-

ables. The Myers-Briggs Type Indicator (MBTI) and the

Tennessee Self Concept Scale (TSCS) were administered to

these students during the 1981 Summer term. In addition, the

College Student Satisfaction Questionnaire (CSSQ) was ad-

ministered to these students during.the 1982 Spring term.

Information regarding each student's high school grade

point average, college entrance exam score, and college

grade point average were obtained from the Office of Admis-

sions. Data analyses were conducted as outlined in Chapter

Three.


62







63


HO1: There is no significant relationship between
college grade point averages and high school
grade point averages of specially admitted
black university freshman students.

A stepwise regression analysis was used to test this

hypothesis. The results in Table 1 indicate that there was

no significant relationship between college grade point

averages and high school grade point averages of specially

admitted black university freshman students. In addition,

the Pearson Product Moment Correlation was used to further

test this hypothesis. The correlation coefficient was

.05632. These results indicate that high school grade point

averages were not accurate predictors of college grade point

averages for the specially admitted black university fresh-

man students in this study. Therefore, hypotheses one was

not rejected.

HO2: There is no significant relationship between col-
lege grade point averages and college entrance
exam scores of specially admitted black univer-
sity freshman students.

A stepwise regression analysis was used to test this

hypothesis. The results in Table 1 indicate that when SAT-

Total was entered as the second variable in the analysis, it

accounted for an additional8 percent increment in the ex-

plained variance in college grade point average. SAT-Math

was initially entered as the ninth variable, but was replaced

in the next step by the Recognition Scale of the CSSQ. SAT-

Math was reentered to complete the best eleven variable model

found. SAT-Verbal was not entered as one of the significant

variables in the model for improvement in R2. In addition,







64

the Pearson Product Moment Correlation was used to test

this hypothesis. The correlation coefficients for SAT-Total,

SAT-Verbal, and SAT-Math were reported as .27252, .16529, and

.23262, respectively. These results indicate that college

entrance exam scores were not accurate predictors of col-

lege grade point averages for specially admitted black

university freshman students in this study. Therefore,

hypothesis two was not rejected.

HO3: There is no significant relationship between
college grade point averages and self-concepts
of specially admitted black university fresh-
man students.

A stepwise regression analysis was conducted on this

hypothesis. The results in Table 1 indicate that several

scales of the Tennessee Self Concept Scale contributed to

variance in the prediction of college grade point average.

However, the addition of these variables (Personal Self,

Identity, Self Criticism and Behavior) only increases the

prediction of college grade point average by relatively

small amounts. In addition, the Pearson Product Moment

Correlation was used to test this hypothesis. The correla-

tion coefficients for Personal Self, Identity, Self

Criticism, and Behavior were reported as -.28700, -.02440,

-.02837, and -.28297, respectively. These results indicate

that self-concept was not an accurate predictor of college

grade point averages for the specially admitted black

university freshman students in this study. Therefore, hypo-

thesis three was not rejected.






65


HO4: There is no significant relationship between
college grade point averages and personality
type preferences of specially admitted black
university freshman students.

A frequency distribution was obtained to determine the

extent of the relationship described in this hypothesis.

The results in Figure 2 indicate that there was no sig-

nificant relationship between black university specially

admitted students' college grade point averages and per-

sonality type preferences. In addition, the Pearson Product

Moment Correlation was used to test this hypothesis. The

correlation coefficient was reported as -0.03470. These

findings indicate that personality type preferences were

not accurate predictors of college grade point averages for

the specially admitted black university freshman students

in this study. Therefore, hypothesis four was not rejected.

HO5: There is no significant relationship between
college grade point averages and college
student satisfaction of specially admitted
black university freshman students.

A stepwise regression analysis was used to test this

hypothesis. The results in Table 1 indicate that three

scales of the CSSQ (Working Conditions, Compensation, and

Quality of Education) contributed to the variance in the

prediction of college grade point average although the R2

increases were only .04, .01,. and .01, respectively. In

addition, the Pearson Product Moment Correlation was con-

ducted to test this hypothesis. The correlation coefficients

for the three CSSQ scales were reported as -0.23890, -0.06574,

and 0.18766, respectively. These results indicate that





66











SENSING TYPES INTUITIVE TYPES
with THINKING with FEELING with FEELING with THINKING

ISTJ ISFJ INFJ INTJ
N= 32 N= 12 N= 3 N= 4
%= 34.0 %= 12.8 %= 3.2 %= 4.3 3
x 2.40 2.35 1.96 2.21 z
Sd .653 .983 .661 .815

-.4

im
ISTP ISFP INFP INTP
N= 1 N= 6 N= 1 N= 0 M
%= 1.1 %= 6.4 %= 1.1 %= 0 m
x 1.37 2.55 3.74
Sd .933
m


ESTP ESFP ENFP ENTP
N= 6 N= 1 N 1 N 2
%= 6.4 %= 1.1 /, 1.1 %= 2.1
x 1.93 4.00 2.00 2.93 3
Sd .685 .969 <
m
x

m
ESTJ ESFJ ENFJ ENTJ
N= 15 N= 6 N= 3 N= 1
%= 15.9 %= 6.4 %= 3.2 %= 1.1
x 2.52 2.51 2.40 2.76 0
Sd .72: .686 .971




N = 94


Figure 2

Frequency Distribution Myers-Briggs Type Indicator By
College Grade Point Average.







67


Table 1

Maximum R2 Improvements for GPA Prediction.





Order of R
Variables N R F P Increase

Personal Self 67 .08 5.84 .019 .08

SAT-Total 67 .16 6.04 .004 .08

Identity 67 .21 5.53 .002 .05

Working Conditions 67 .25 5.25 .001 .04

Self Criticism 67 .27 4.57 .001 .02

Behavior 67 .29 4.07 .001 .02

MBTI 67 .31 3.72 .002 .02

Compensation 67 .32 3.41 .003 .01

SAT-Math 67 .33 3.10 .004 .01

Quality of Education 67 .34 2.83 .007 .01

SAT-Math Replaced by 67 .34 2.91 .005 .00
Recognition

SAT-Math 67 .35 2.68 .008 .01

Family Self 67 .35 2.46 .012 .00

HSGPA 67 .36 2.26 .019 .01

Social Self 67 .36 2.07 .029 .00

Moral-Ethical Self 67 .36 1.91 .045 .00







68


college student satisfaction was not an accurate predictor

of college grade point averages for the specially admitted

black university freshman students in this study. There-

fore, hypothesis five was not rejected.

HO6: There is no significant relationship between
personality type preferences and college
student satisfaction of specially admitted
black university freshmen.

The Pearson Product Moment Correlation conducted in this

hypothesis yielded the following coefficients: Working

Conditions, 0.00273; Compensation, 0.01352; Quality of

Education, 0.12152; Social Life, 0.13616; Recognition,

0.13877, and Total Satisfaction, 0.09940. In addition, a

frequency distribution table was established and is presented

in Figure 3. These results indicate that no significant

relationship was found between personality type preferences

and college student satisfaction of the specially admitted

black university freshmen in this investigation. These re-

results imply that personality type preferences were not

accurate predictors of college student satisfaction for the

specially admitted black university freshman students in

this study. Therefore, hypothesis six was not rejected.

H0O: There is no significant relationship between
self-concept and college student satisfaction
of specially admitted black university fresh-
man students.

The Pearson Product Moment Correlation was used to test

this hypothesis. The results in Table 2 indicate that self-

concept was not an accurate predictor of college student

satisfaction for the specially admitted black freshman






69










SENSING TYPES INTUITIVE TYPES
with THINKING with FEELING with FEELING with THINKING
ISTJ ISFJ INFJ INTJ
N=26 N= 7 N= 2 N= 3
%=37.7 %= 10.1 %= 2.9 %= 4.3 o
x 219.19 234.71 191.50 221.00 z
Sd 34.579 30.379 65.76 17.52

-

ISTP ISFP INFP INTP
N= 0 N= 5 N= 1 N= 0 M
%= 0 %= 7.2 %= 1.4 %= 0.0 ,
x 220.00 206.0 -
Sd 28.439



ESTP ESFP ENFP ENTP
N= 5 N= 0 N= 0 N= 2
%= 7.2 %= 0 0 %= 2.9
x 225.40 254.5 4
Sd 10.479 43.134 m
X


ESTJ ESFJ ENFJ ENTJ
N= 11 N= 5 N= 1 N= 1
%= 15.9 %= 7.2 %= 1.4 %= 1.4 c
x 201.36 217.6 210.00 200.00 K
Sd 41.12 29.89 z




N = 69


Figure 3

Frequency Distribution Myers-Briggs Type Indicator By
Total Satisfaction.













Table 2

Person Product Moment Correlation Coefficients Between Scores on the
CSSQ and TSCS of Specially Admitted Black University Freshman Students.


Self Concept College Student Satisfaction Scales
Scales
Working Compen- Quality of Social Recog- Total
Conditions sation Education Life nition Score

Self-Criticism 0.13196 0.06319 0.27622 0.17831 0.14233 0.18647

Total P Score. 0.26737 0.18046 0.16440 0.11523 0.08672 0.18821

Identity 0.34381 0.19928 0.25076 0.19762 0.18523 0.27480

Self-Satisfaction 0.07613 0.04974 -0.05547 -0.04685 -0.09085 -0.01834
o
Behavior 0.23158 0.18662 0.21835 0.14261 0.13547 0.21199

Physical Self 0.25187 0.12987 0.20600 0.22054 0.20151 0.23847

Moral-Ethical Self 0.08598 -0.07368 -0.08355 -0.12086 -0.08361 -0.06472

Personal Self 0.22616 0.15328 0.13124 0.10335 0.07005 0.15826

Family Self 0.28203 0.30122 0.24844 0.19906 0.12031 0.26493

Social Self 0.15640 0.12833 0.06443 -0.00690 -0.00847 0.07422

Total Variability 0.03793 0.08647 0.16573 0.16407 0.13084 0.13781

Column Variability 0.09838 0.07149 0.13441 0.20525 0.16368 0.16067

Row Variability -0.11741 -0.00698 0.05230 -0.11557 -0.05874 -0.06158

N = 67






71


students in this study. Therefore, hypothesis seven was

not rejected.



Discussion


The results of this study indicate that academic

variables such as high school grade point averages and

college entrance exam scores were not significantly related

to the college grade point averages of specially admitted

black university freshman students. These results support

the findings of Dalton, Anastasiow, and Brigman (1977);

Boyd (1977); and Rovezzi-Carroll and Thompson (1980), who

reported no significant relationship between academic pre-

dictors and future college academic success. The results

of the present study also support the findings of Houston

(1980) who reported a small coefficient of multiple deter-

mination (R2) when high school rank, SAT-V, and SAT-M were

entered in a stepwise regression analysis for similar sub-

jects. These results do, however, contradict previous

findings in the literature that show a significant relation-

ship between academic predictors and college grade point

averages (Bowers, 1970; Lanning, 1977; and Price and Kim,

1976).

The finding that previous academic achievement in this

study did not accurately predict college grade point average

was not surprising. As evidenced by Stewart (1981), a sub-

statially high percentage of specially admitted students do

perform well academically at institutions of higher education.







72


One reason for this may lie in the support and encouragement

that the student receives from significant others. The

knowledge that family and friends believe in the student's

ability to succeed may cause the student to put forth that

extra effort which may mean the difference between failure

and success. Another explanation for these results could

be the change in the student's view of education upon

arrival on the college campus. High school grades and

academic achievement tests may not have been high on the

priority list of some of these students due to the lack of

interest in subject matter, motivation and encouragement.

However, once the student is exposed to a new and educa-

tionally stimulating environment, the motivation to suc-

ceed may heighten and college graduation becomes one of the

ultimate goals in life.

The self-concepts of the specially admitted black

university freshman students in this study had no significant

impact on their subsequent college grade point averages,

that is, the differences between the self-concept scores

of those students with high grade point averages and those

with low grade point averages were not significantly dif-

ferent. These findings are consistent with those of Dowdle

(1977) and Carey (1976) who reported that those students

who performed low academically did not have a lower self-

concept than those who performed high academically. One

reason for these findings.may be that the self-concept scores

were based on previous personal experiences from high school.






73


For some students those experiences may have had a positive

effect, while for others, quite the opposite may have been

the case. However, those students with low self-concepts

may strive harder to achieve than those with high self-

concepts. Thus, the fact that they may not be as confident

in themselves as others may motivate them to study more

intensely. This may then allow them to achieve comparable,

if not better, grades than those who scored high on the

self-concept measure.

No significant relationship was found between personality

type preferences and college grade point averages. The

majority (79 or 84 percent) of the students were identified

as Sensing types. A total of only 15 or 16 percent were on

the Intuitive side of the continuum. These results support

McCaulley (1972 and 1974) who reported very similar patterns.

The fact that no statistically significant relationship was

found supports Leon (1974) and Johnson (1970), who also

reported no significant difference in regards to grade point

averages. One explanation for these results may stem from the

fact that the personality type scores were based on previous

individual experiences from high school. Regardless of

individual personality type, the key to success depended on

the flexibility of that person in being able to adjust

academically or socially to the college environment. Those

who possessed those skills prior to admittance were not that

different from those who acquired the necessary coping skills

through daily campus encounters.






74

The degree of satisfaction with the university

environment was not significantly related to the college

grade point averages of the students studied. The results

of this study somewhat support previous findings in the

literature that show a high degree of satisfaction with

college by freshmen students (Schmidt and Sedlacek, 1972;

and Madrazo-Peterson and Rodriguez, 1978). When compared

to the results reported by Lindsey (1981), the means of all

subscales of the CSSQ for the freshman group in the present

study were slightly higher than those reported by Lindsey

for all black students. One reason for these findings may

be the tendency for freshman students to be highly optimistic

about their first year in college. Therefore, many positive

aspects may be overexaggerated while negative ones may be

subconsciously overlooked.

Neither personality type preferences nor self-concepts

were identified as being significantly related to college

student satisfaction in this study. Even though freshmen

initially enter the university environment with individual

differences in regards to personality type preferences and

self-concepts, the majority of them are moderately satisfied

with the circumstances they find during their first year.

One explanation for these results is that perceptual and

self-concept changes are most evident during the freshman

year in college. Therefore after approximately six months

in the college environment, initial priorities may no longer

seem important, that is, what students expect as a one-day-

old freshmen may be quite the opposite of what they experience







75


six months later. However, because the novelty of the

freshman year experience has not worn off, the student may

still be completely satisfied.
















CHAPTER FIVE
SUMMARY, CONCLUSIONS, IMPLICATIONS,
AND RECOMMENDATIONS FOR FURTHER RESEARCH



Summary


The relationship between academic and nonacademic

variables and college grade point average has been the

source of much recent research. The establishment of a

predictive model by the use of multiple factors, however,

has not been extensively explored and determined. The

purpose of the present study was to investigate the

relationship between five independent variables (high

school grade point averages, college entrance exam scores,

personality type, self-concept, and college student satis-

faction) and one dependent variable (college grade point

average) for specially admitted black students at one ins-

titution. This study also determined the relationship

between college student satisfaction and two other vari-

ables, personality type and self-concept. Chapter One dealt

with the statement of the problem, purpose of the study,

need for the study, definition of terms, and organization

of the remainder of the study. The literature related to

college student attrition was reviewed in Chapter Two. The

sections contained in Chapter Two were academic factors

related to college attrition and nonacademic factors related

76






77


to college attrition, including specific studies dealing with

self-concept, personality characteristics and college

student satisfaction. The method of research used in this

study was described in Chapter Three. A description of the

hypotheses, population and sample, instruments, procedures,

analyses of data, and limitations of the study were included

in this chapter.

The results of the study were reported in Chapter Four,

followed by a discussion of those results. The findings

of this study indicated that academic and nonacademic

variables were not significantly related to the college

grade point averages of specially admitted black university

freshman students. The results also indicated no significant

relationship between college student satisfaction and self-

concept of personality type preferences.



Conclusions


The following conclusions were made based on the results

of this study:

1. High school grade point averages of specially admitted

black university freshman students do not accurately pre-

dict college grade point averages of these students.

2. College entrance exam scores of specially admitted black

university freshman students do not accurately predict

college grade point averages of these students.







78


3. Self-concepts of specially admitted black university

freshman students are not accurate predictors of

college grade point averages for these students.

4. Personality type preferences are not accurate predic-

tors of the college grade point averages of specially

admitted black university freshman students.

5. The degree of satisfaction with their university

environment does not affect the college grade point

averages of specially admitted black university fresh-

man students.

6. The degree of satisfaction with their university

environment is not determined by personality type

preference of specially admitted black university

freshman students.

7. The degree of satisfaction with their university

environment is not determined by the self-concept of

specially admitted black university freshman students.



Implications


Implications of this study are demonstrated in both the

selection and retention of specially admitted black univer-

sity college students. So far as the selection of students

is concerned, the results of this study imply that selection

criteria based on academic assessments are not accurate as

predictors of college grade point averages for these students.

That is, neither of the academic factors (high school grade

point averages or college entrance exam scores) were related






79


to college grade point average. Therefore, the study seems

to indicate that selection of specially admitted black

students should not be based solely upon academic variables.

University admissions committees should continue to consider

this, especially with the trend toward increasing admissions

requirements.

The results of this study indicate that nonacademic data

are somewhat related to college grade point average. There-

fore, there is a need for additional information regarding

the relationship between nonacademic variables and future

academic success. The continued search for nonacademic

baseline data is essential if research efforts are to be

expanded in this area of concern. Student personnel workers

should be encouraged to continue to conduct research in an

attempt to identify possible contributing factors to college

success.

Student personnel workers should be trained to identify

problem areas of the college environment for specially

admitted black students. These workers should then work

with administrators in an attempt to plan and implement

programs and activities to help maintain satisfaction among

these students. The results of this study did not identify

one specific variable that predicted college academic

success. Therefore, it would seem that college success may

be the result of an interaction between various factors.

Student personnel workers should be highly skilled in the

art of interpreting the college environment and the effects

that certain factors may have on the success of the student.






80


Perhaps another implication of this research can be

drawn from the finding that the students of this study

(freshmen) were moderately satisfied with their college

environment. An attempt should be made to determine the

causes of dissatisfaction which according to the literature

often seem to occur following the freshman year. There is

a need for increased retention efforts following the fresh-

man year. There are increased pressures placed on sopho-

more, junior, and senior level students to make career

decisions and to perform well academically in order to

attain career goals. However, although course difficulty

increases, there is a decrease in the use by these students

of the resources available for tutoring and academic

advisement. Student personnel workers should encourage

continued use of these services by students throughout all

levels of achievement.



Recommendations for Further Research


Based on the results of this study, several recommenda-

tions for further research seem warranted. The following

studies therefore, are suggested for use in conducting

research with specially admitted black university freshman

students:

1. A four-year longitudinal attrition study using initial

freshman academic and nonacademic baseline data should







81


be conducted. A study of this type would establish

a profile of those students most likely to succeed

in the college environment.

2. A correlational study between previous academic factors

and college grade point averages of currently enrolled

students by academic classification should be conducted.

This study would assist in the identification of the

relationship between academic factors and college grade

point averages of students as they matriculate through

college and would establish predictive validity.

3. A correlational study between college student satisfac-

tion and college grade point averages of enrolled

students by academic classification should be undertaken.

This type study would indicate the effects of student

satisfaction on college grade point average at all

academic classification levels and would

help to determine if students at different levels are

affected differently by different aspects of their

environment.

4. A college student satisfaction follow-up study should

be conducted with former students to determine the

relationship between college student satisfaction and

attrition. The results of this study would identify

problem areas in the college environment. These findings

also would assist student personnel workers in planning

and implementing programs for future students.

















APPENDIX A
INFORMED CONSENT FORM


Betty J. Stewart,
Principal Investigator
441 Little Hall
University of Florida

The purpose of this study is to determine if there is a relation-
ship between academic factors (high school grade point averages and
college entrance exam scores) and nonacademic factors (personality
"types," self-concept, and college student satisfaction) which can be
used to predict college grade point averages. In order to research
the possibility of a predictive relationship, you will be asked to
complete the following instruments: 1) the Tennessee Self-Concept
Scale; 2) the Myers-Briggs Type Indicator; and 3) the College Student
Satisfaction Questionnaire.
There are no discomforts or risks to be expected from the proced-
ures mentioned above. On the contrary, the outcome of this study could
benefit you individually by revealing how you feel about yourself
(self-concept), your likes and dislikes (personality) and how satisfied
you are with your college environment. The results also will assist
others by increasing further knowledge in this area. Your individual
"scores" will be kept confidential and will be available only to you
(the subject) and me (the principal investigator). Each subject will
be numerically coded prior to the recording of scores and grade point
averages. At the end of this study these coded sheets will be destroyed
so that information regarding individuals can be kept confidential.
Group results will be available through the published dissertation in
the University Libraries. It is hoped that this alone will inspire you
to participate in this study because I am unable to offer any type of
monetary reward for your services.
If you have any questions about this study and the procedures which
are being followed, I will be happy to answer them. You are free to
withdraw your consent to participate in this project at any time without
being penalized.

"I have read and understand the procedures described above. I
agree to participate in the procedures and I have received a copy of
this description."



Subject Date Witness Date


Relationship if other than subject Date

82
















APPENDIX B
MEANS AND STANDARD DEVIATION ON ALL VARIABLES
FOR SPECIALLY ADMITTED BLACK UNIVERSITY STUDENTS


Variable N Mean SD

College GPA 67 2.4197 0.7372
HSGPA 67 2.6118 0.3669
SAT-Total 67 666.7164 78.4198
SAT-Verbal 67 319.1791 54.4782
SAT-Math 67 347.5373 53.1605
Self-Criticism 67 31.0597 5.2365
Total Positive 67 352.5075 27.0999
Identity 67 130.8806 10.2935
Self-Statisfaction 67 108.0896 12.7395
Behavior 67 113.5373 11.8185
Physical Self 67 72.9552 7.2476
Moral-Ethical Self 67 70.1940 6.3776
Personal Self 67 68.8955 6.5508
Family Self 67 71.2836 8.0638
Social Self 67 69.3284 7.1761
Total Variance 67 52.8806 11.9374
Column Variance 67 32.5373 9.2742
Row Variance 67 20.8657 7.3976
Working Conditions 67 41.6716 8.0761
Compensation 67 43.0149 7.0032
Quality of Education 67 44.4627 7.5645
Social Life '67 45.5522 8.6134
Recognition 67 43.0299 8.3937
Total Satisfaction 67 217.7313 33.9374





83















REFERENCES



Aleamoni, L.M. Can grade point average be more accurately
predicted? Journal of Educational Psychology, 1977,
69, 227-230.

Ashcraft, C., and Fitts, W.H. Self concept change in psy-
chotherapy. Psychotherapy, 1964, 1, 115-118.

Astin, A.W. Personal and environmental factors associated
with college dropouts among high aptitude students.
Journal of Educational Psychology, 1964, 55, 219-227.

Astin, A.W. Four Critical Years. San Francisco: Jossey-
Bass Publishers, 1977.

Beal, P.E. Student retention: A case study of an action
approach. National Association of Student Personnel
Administrators Journal, 1979, 17, 9-16.

Beal, P.E., and Noel, L. What Works in Student Retention.
Iowa City: The American College TestingProgram and
the National Center for Higher Education, 1979.

Bean, A.G., and Covert, R.W. Prediction of college per-
sistence, withdrawal, and academic dismissal: A
discriminate analysis. Educational and Psychological
Measurement, 1973, 33, 407-411.

Berdie, R.F. The prediction of college achievement and
satisfaction. Journal of Applied Psychology, 1944,
28, 239-245.

Betz, E.L., Klingensmith, J.E., and Menne, J.W. The measure-
ment and analysis of college student satisfaction.
Measurement and Evaluation in Guidance, 1970, 3, 110-118.

Betz, E.L., Menne, J.W., Starr, A.M., and Klingensmith, J.E.
A dimensional analysis of college student satisfaction.
Measurement and Evaluation in Guidance, 1971, 4, 99-106.

Betz, E.L., Starr, A.M., and Menne, J.W. College student
satisfaction in ten public and private colleges and
universities. Journal of College Student Personnel,
1972, 13, 456-461.


84






85


Bowers, J. The comparison of gpa regression equation for
regularly admitted and disadvantaged freshmen at the
University of Illinois. Journal of Educational
Measurement, 1970, 7, 219-225.

Boyd, W.M. SAT's and minorities: The dangers of under-
prediction. Change, 1977, 9, 48-49.

Brown, N., Jr. A description research study of a develop-
mental plan for recruitment and retention of minority
students (Doctoral dissertation, Wayne State Univer-
sity, 1976). ERIC Document Reproduction Service No.
ED 146 254.

Carey, P. The relationship between black studies, self-
concept, and academic performance of black students
of white campuses in the southwest (Doctoral disserta-
tion, Oklahoma State University, 1975). Dissertation
Abstracts International, 1976, 37, 636A. (University
Microfilms No. 76-9639)

Carney, M., and Geis, L. Reading ability, academic per-
formance and college attrition. Journal of College
Student Personnel, 1981, 22, 55-59.

Carr, D., and Chittum, C. A study to identify non-academic
factors which may positively influence the recruitment
and retention of."other race" students at Virginia's
state-supported institutions of higher education(Final Report,
Norfolk, 1980).ERIC Document Reproduction Service No.
ED 187 166.

Cleary, T.A. Test bias: Prediction of grades of Negro and
white students in integrated colleges. Journal of
Educational Measurement, 1968, 5, 115-124.

Cole, J. The relationship of the freshman special services
program at Southern University to academic achievement
and self concept. (Doctoral dissertation, Southern
University, 1974). Dissertation Abstracts International,
1974, 35, 2567A-2568A. (University Microfilms No. 74-24,
767)

Cole, N., and Hanson, G. Racial-ethnic bias in selective
college admissions. Paper presented at American
Educational Research Association Annual Meeting New
Orleans, Louisiana, 1973.

Copeland, E.J. Peer counselor effects on the self concept
and academic adjustment of educational opportunity
students who participated in group counseling.
(Doctoral dissertation, Oregon State University, 1974).
Dissertation Abstracts International, 1974, 35, 1974A.
(University Microfilms No. 74-23, 431)






86


Cranney, A.G., and Larsen, J. Compensatory programs for
specially admitted freshmen to the University of
Florida-1968-1971. National Reading Conference 21st
Yearbook. Milwaukee: The Conference, 1972, 38-41.

Crawford, A.E., McFarland, D.E., and Rhatigan, J.J.
Special counseling programs for academic survival.
Journal of College Student Personnel, 1978, 19,
298-302.

Dalton, S., Anastasiow, M., and Brigman, S.L. The relation-
ship of underachievement and college attrition.
Journal of College Student Personnel, 1977, 18, 501-505.

Dawkins, M.P., and Dawkins, R.L. Perceptions and experiences
as correlates of academic performance among blacks at
a predominately white university: A Research note.
College and University, 1980, 55, 171-180.

DeVore, J.R., and Handal, P.J. The College Student Satis-
faction Questionnaire: A test-retest reliability
study. Journal of College Student Personnel, 1981, 22,
299-301.

DiCesare, A.C., Sedlacek, W.E., and Brooks, G.C., Jr.
Nonintellectual correlates of black student attrition.
Journal of College Student Personnel, 1972, 13, 319-324.

Dowdle, S.L. An analysis of the relationship between the
self concept and academic performance of high and low
performers in a collegiate school of business. (Doc-
toral dissertation, The University of Arizona, 1977).
Dissertation Abstracts International, 1977, 38, 1222A-
1223A. (University Microfilms No. 77-18,570)

Eckland, B., and Wisenbaker, J. A Capsule Description of
Young Adults Four and One-Half Years After High School.
Prepared for the National Center for Educational
Statistics. Research Triangle Park, North Carolina:
Research Triangle Institute, 1979.

Egerton, J. State Universities and Black Americans.
Atlanta: Southern Education Reporting Service, 1969.

Ellison, R.L., Murray, S.L., Fox, D.G., and Taylor, C.W.
Biographical data as predictors of college grades of
Negroes and whites.(Slt--Lake City, Utah, Institute
for Behavioral Research in Creativity, 1973). ERIC
Document Reproduction Service No. ED 081 358.

Fitts, W.H. Manual: Tennessee Self Concept Scale.
Nashville, Tennessee: Counselor Recordings and Tests,
1965.





87


Friedlander, J. Are College support programs and services
reaching high-risk students? Journal of College
Student Personnel, 1980, 21, 23-28.

Gadzella, B.M., and Fournet, G.P. Differences between high
and low achievers on self-perceptions. Journal of
Experimental Education, 1976, 44, 44-48.

Garber, H., and Schell, R. Differences in rates and modes
of attrition between three groups of Oswego under-
graduate students. Research report. Oswego, New York:
State University of New York, 1977.

Gibbs, J.T. Black students/white university: Different
expectations. Personnel and Guidance Journal, 1973, 51,
463-469.

Gividen, G.M. Stress in airborne training as related to the
self-concept, motivation and biographical factors. Un-
published master's thesis, Vanderbilt University, 1959.

Goodrich A. Report on the Academic Status of Minority Students
at the University of Maryland, College Park, for the
Academic Year 1975-76. College Park, Maryland: Office of
Minority Student Education, University of Maryland, 1976.

Goodrich, A., and Shuford, B. A Retention/Intervention Model
for Impacting on Minority Student Attrition at Pre-
dominantly White Colleges and Universities. College Park,
Maryland: Office of Minority Student Education,
University of Maryland, 1976.

Goodrich, A. Manual: A data-driven retention model for
improving minority study persistence in higher education
institutions. University of Illinois at Chicago Circle.
Chicago, Illinois, 1979.

Green, R.L., and Farquhar, W.W. Negro academic motivation
and scholastic achievement. Journal of Educational
Psychology, 1965, 56, 241-243.

Haettenschwiller, D.L. Counseling black students in special
programs. Personnel and Guidance Journal, 1971, 50, 29-35.

Hallenbeck, T.R. College student satisfaction: An indication
of institutional vitality. National Association of
Student Personnel Administrators Journal, 1978, 16, 19-25.

Hannah, W. Personality differentials between lower division
dropouts and stayins. Journal of College Student
Personnel, 1971, 12, 16-19.

Hart, D., and Keller, M.J. Self-reported reasons for poor
academic performance of first-term freshmen. Journal
of College Student Personnel, 1980, 21, 529-534.






88


Hathaway, S.R., and McKinley, J.C. Revised Manual:
Minnesota Multiphasic Personality Inventory.
New York: The Psychological Corp., 1951.

Herzberg, F., Mausner, B., Peterson, R.O., and Capwell,
D.F. Job attitudes: Review of Research and Opinion.
Pittsburgh, Pa.: Psychological Services, 1957.

Houston, L.N. Predicting academic achievement among
specially admitted Black female college students.
Educational and Psychological Measurement, 1980, 40,
1189-1194.

Humphries, F. Thirteen College Curriculum Program Progress
Report, 1967-1972: A Major Curriculum Effort to Reduce
Attrition Among Black College Students. Washington,
D.C.: Institute for Services to Education, 1972.

Ikenberry, S.O. Factors in college persistence. Journal
of Counseling Psychology, 1961, 8, 322-329.

Johnson, D. Personality characteristics in relation to
college persistence. Journal of Counseling Psychology,
1970, 17, 162-167.

Jones, F.D. The effects of personal growth group counseling
on the self-concepts and academic achievement of
college freshmen. (Doctoral dissertation, Lehigh Univer-
sity, 1975). Dissertation Abstracts International,
1975, 36, 2631A-2632A. (University Microfilms No.
75-23,994)

Jones, J.C., Harris, L.J., and Hauck, W.E. Differences in
perceived success of academic difficulties; Black
students in predominately black and predominately white
colleges. Journal of Negro Education, 1975, 44, 519-529.

Kallingal, A. The prediction of grades for black and white
students at Michigan State University. Journal of
Educational Measurement, 1971, 8, 263-265.

King, H., and Walsh, W.B. Change in environmental expecta-
tion and perceptions. Journal of College Student
Personnel, 1972, 13, 331-336.

Lanning, W. Factors related to college student persistence
and withdrawal. National Association of Student
Personnel Administrators Journal, 1977, 15, 34-38.

Leon, G.R. Personality change in the specially admitted
disadvantaged student after one year in college. Journal
of Clinical Psychology, 1974, 30, 522-528.






89


Lindsey, M.D. Satisfaction and university involvement
among black and white undergraduate students.
(Doctoral dissertation, University of Florida, 1981).
Dissertation Abstracts International 1982, 42
University Microfilms No. 8203,689).

Lokitz, B.D., and Sprandel, H.Z. The first year: A look
at the freshman experience. Journal of College
Student Personnel, 1976, 17, 274-279.

Madrazo-Peterson, R., and Rodriguez, M. Minority students'
perceptions of a university environment. Journal of
College Student Personnel, 1978, 19, 259-263.

Martin, R. Freshman satisfaction with college. Journal of
College Student Personnel, 1968, 9, 382-383.

Maudal, G.R., Butcher, J.N., and Mauger, P.A. A multi-
variate study of personality and academic factors in
college attrition. Journal of Counseling Psychology,
1974, 21, 560-567.

Mayhew, L.B. Non-test predictors of academic achievement.
Educational and Psychological Measurement, 1965, 25,
39-46.

Maynard, M. Can universities adapt to ethnic minority
students' needs? Journal of College Student Personnel,
1980, 21, 398-401.

McCaulley, M.H. Type table for EEOP freshmen entering
summer 1972. The Freshman Studies. Center for
Applications of Psychological Type, Inc. Unpublished
Data, 1972.

McCaulley, M.H. Type table for university students in a
Special Services Program. The Freshman Studies.
Center for Applications of Psychological Type, Inc.
Unpublished Data, 1974.

McDavis, R.J., Mingo, G.W., Stewart, B.J., and Hough, T.A.
A summer orientation program for students in special
services programs: A coordinated approach. Journal
of College Student Personnel, 1980, 21, 88-89.

Menning, A.J., Bradley, L.J., and Cochran, J. Attrition.
College and University, 1975, 50, 753-756.

Merton, R.K. Social theory and social structure (rev. ed.).
New York: Free Press, 1957.






90


Millott, R.F. Reading performance as a correlate of the
personality type of college freshmen. (Doctoral
dissertation, University of Florida, 1974).
Dissertation Abstracts International, 1974, 36, 712A.
(University Microfilm No. 75-16,420).

Mornell, E. The program of special directed studies:
A five-year summary. Claremont University, California,
1973. (ERIC Document Reproduction Service No. ED 088
638).

Morrisey, R.J. Attrition in probationary freshmen.
Journal of College Student Personnel, 1971, 12, 279-285.

Morrison, T.L., Thomas, M., and Weaver, S.J. Self-esteem
and self-estimates of academic performance. Journal
of Consulting and Clinical Psychology, 1973, 41, 412-415.

Morstain, B.R. An analysis of students' satisfaction with
their academic program. Journal of Higher Education,
1977, 48, 1-15.

Muskat, H. Educational expectations and college attrition.
National Association of Student Personnel Administra-
tors Journal, 1979, 17, 17-22.

Myers, I. Manual: The Myers-Briggs Type Indicator.
Princeton, New Jersey: Educational Testing Service,
1962.

National Advisory Committee on Black Higher Education and
Black Colleges and Universities. Access of Black
Americans to Higher Education: How Open is the Door?
Washington, D.C.: Department of Health, Education,
and Welfare, 1979.

Pandey, R.E. Personality characteristics of successful,
dropout, and probationary black and white university
students. Journal of Counseling Psychology, 1972, 5,
382-386.

Pantages, T.J., and Creedon, C.F. Studies of college
attrition: 1950-1975. Review of Educational Research,
1978, 48, 49-101.

Pazandak, C.H. Self-estimate of level .f effort as a pre-
dictor of college grades. Measurement and Evaluation in
Guidance, 1975, 8, 43-50.

Pervin, L.A. Satisfaction and perceived self-environment
similarity. Journal of Personality, 1967, 35, 623-634.





91


Pfeifer, C.M., Jr., and Sedlacek, W.E. The validity of
academic predictors for black and white students at
predominately white universities. Journal of
Educational Measurement, 1971, 8, 253-261.

Pfeifer, C.M., Jr., and Sedlacek, W.E. Predicting black
student grades with non-intellectual measures. Journal
of Negro Education, 1974, 43, 67-76.

Pfeifer, C.M. Relationship between scholastic aptitude,
perception of university climate, and college success
for black and white students. Journal of Applied
Psychology, 1976, 61, 341-347.

Price, F., and Kim, S. The association of college perform-
ance with high school grades and college entrance
test scores. Educational and Psychological Measurement,
1976, 36, 965-970.

Robertson, J.A. Black student satisfaction in the deep
south. Journal of College Student Personnel ,1980, 21,
510-513.

Rossmann, J.E., and Kirk, B.A. Factors related to persistence
and withdrawal among university students. Journal of
Counseling Psychology, 1970, 17, 56-62.

Rovezzi-Carroll, S., and Thompson, D. Forecasting college
success for low-income students. Journal of College
Student Personnel, 1980, 21, 340-343.

Samuel, N., and Laird, D.S. The self concepts of two groups
of black female college students. Journal of Negro
Education, 1974, 43, 228-233.

Schmidt, D.K., and Sedlacek, W.E. Variables related to
university student satisfaction. Journal of College
Student Personnel, 1972, 13, 233-237.

Schroeder, C.C., Warner, R., and Malone, D.R. Effects of
assignment to living units by personality types on
environmental perceptions and student development.
Journal of College Student Personnel, 1980, 21, 243-249.

Scott, J.E. An investigation of black and white student
expectations and perceptions of a predominately white
public institution. (Doctoral dissertation, Eastern
Michigan University, 1978). Dissertation Abstracts
International, 1978, 39, 3410A. (University Micro-
films No. 7823004).

Sedlacek, W.E., and Brooks, G.C., Jr. Black freshmen in
large colleges: A survey. Personnel and Guidance
Journal, 1970, 49, 307-312.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E4TJRVBJF_F5ZO3T INGEST_TIME 2015-01-14T19:18:49Z PACKAGE AA00026570_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

ACADEMIC AND NONACADEMIC FACTORS RELATED TO THE ATTRITION RATE OF SPECIALLY ADMITTED BLACK UNIVERSITY FRESHMAN STUDENTS BY BETTY JEAN STEWART A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1982

PAGE 2

— 'o 00 o Copyright 1982 by Betty Jean Stewart

PAGE 3

k?rJ> s'l-.ji't**--**.THIS PROJECT IS DEDICATED TO MY MOTHER, MRS. DOROTHY HICKMAN, AND THE PAST, PRESENT, AND FUTURE BLACK STUDENTS AT THE UNIVERSITY OF FLORIDA

PAGE 4

*-. n<-i-m*lii r '#i M > _i it*w)B | Wi a ACKNOWLEDGMENTS All glory goes to God for the great thing he has done through the many persons in my life who made this endeavor possible. I know that I will not be able to include all of the "significant" others who offered words of encouragement and support during the course of my educational experience, but I will attempt to list a few of these individuals by name, To Dr. Joseph Wittmer, my chairman, consultant, and most of all, friend, I am forever grateful for your everpresent support, guidance and faith in my ability to succeed. Thank you from the depths of my heart for your concern and sensitivity. Special words of thanks are also extended to the other members of my supervisory committee. Dr. Roderick McDavis and Dr. Robert Ziller. Thank you Dr. McDavis for your firm and loving guidance in completing this project. I am also especially appreciative of the inspiration you gave me throughout my tenure as a graduate student. Dr. Ziller, your continuous support, helpful suggestions and friendship greatly aided me in achieving this goal. There are many true friends to whom I am sincerely igrateful because of their patience and understanding during all of the "crazy" moments of my life. I am indebted to my typist, Vita Zamorano (who is faster than the speed of iv

PAGE 5

light), and my computer analyst, Alecia Schmitt (who made it all so very plain and simple). To Gloria, Suzan Liz, Juanita, and Lainee, I appreciate your moral support, although you were facing the same pressures. I know that each of you will likewise succeed. To Pat, Brenda, Ruthell, Elmira, and Charlene, who allowed me to "steal" time from our friendships in order to pursue and complete this project; I thank you. To the staff of the College of Liberal Arts and Sciences, Dr. Charles F. Sidman, Dr. Harry B. Shaw, G.W. Mingo, Deloris Slater, John Watson, and the PACT and Special Services peer counselors, without whose faith, support, and cooperation, I would not have been able to succeed; I am greatly indebted. In conclusion, a special word of thanks is extended to my family for their moral support, prayers, love, and patience. Thank you Mom, Dad, Rita, Damon, Graylyn, Thelma, Julius and Sister for your sacrifices., understanding, and assistance throughout my educational experience.

PAGE 6

TABLE OF CONTENTS ACKNOWLEDGMENTS ABSTRACT CHAPTER Page iv viii ONE TWO THREE FOUR INTRODUCTION ................. 1 Statement of the Problem ........... 1 Purpose of the Study ....... 7 Need for the Study 8 Definition of Terms ............. 12 Organization of the Study .......... 13 REVIEW OF RELATED LITERATURE .... 14 Introduction ............ 14 Academic Factors Related to College Attrition 14 Nonacademic Factors Related to College Attrition 21 Self-concept and College Student Attrition 24 Personality Characteristics and College Student Attrition ............. 29 College Student Satisfaction and College Student Attrition 33 METHODOLOGY .......... 41 Introduction ................. 41 Hypotheses .................. 41 Population and Sample ............ 42 Instruments ................. 44 Tennessee Self Concept ........... 44 Myers-Briggs Type Indicator ........ 49 College Student Satisfaction Questionnaire 52 Procedures ......... .... 57 Analysis of Data ............... 59 Limitations of the Study ........... 60 RESULTS AND DISCUSSION 62 Results 62 Discussion ........... 71 VI

PAGE 7

CHAPTER Page FIVE SUMMARY, CONCLUSIONS, IMPLICATIONS, AND RECOMMENDATIONS FOR FUTURE RESEARCH ..... 76 Summary ................... 76 Conclusions ................. 77 Implications ................ 78 Recommendations for Further Research .... 80 APPENDICES A INFORMED CONSENT FORM ............ 82 B MEANS AND STANDARD DEVIATIONS ON ALL VARIABLES FOR SPECIALLY ADMITTED BLACK UNIVERSITY STUDENTS ........ 83 REFERENCES ...................... 84 BIOGRAPHICAL SKETCH ........... 94 Vll

PAGE 8

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ACADEMIC AND NONACADEMIC FACTORS RELATED TO THE ATTRITION RATE OF SPECIALLY ADMITTED BLACK UNIVERSITY FRESHMAN STUDENTS By Betty Jean Stewart May 1982 Chairman: Dr. Paul Joseph Wittmer Major Department : Counselor Education The purpose of this study was to investigate the relationship between five independent variables (high school grade point averages, college entrance exam scores, personality, self-concept, and college student satisfaction) and one dependent variable (college grade point average) concerning specially admitted black students at the University of Florida. Further, this study examined the relationship between college student satisfaction scores for these black students and two other variables, personality and self-concept. A total of 103 specially admitted University of Florida black freshmen participated in this study. Each was present and participated in the first day of the five-day orientation program conducted by the two special admission programs during Summer, 1981, The Tennessee Self Concept Scale (TSCS) and the Myers-Briggs Type Indicator (J£BTI) were administered to each subject as part of the orientation program. During Vlll

PAGE 9

the third week of Spring, 1982, the College Student Satisfaction Questionnaire (CSSQ) was also given to each student. Vital information regarding high school grade point averages, college entrance exam scores, and cumulative college grade point averages was requested and obtained from the University of Florida offices of Admissions and the Registrar. A Pearson Product Moment Correlation indicated no significant relationship between college grade point averages and any of the five independent variables. Correlation coefficients also indicated no significant relationship between college student satisfaction and the variables of personality type and self-concept. The results of a stepwise regression analysis yielded 2 a multiple coefficient of determination (R ) of only .25 (p<.001) when the four subscales of Personal Self, Scholastic Aptitude Test-Total, Identity, and Working Conditions were entered into the prediction model. IX

PAGE 10

CHAPTER ONE INTRODUCTION The gap between the percentage of black and white students who make it to graduation at Florida universities is. widening Almost half of the black students enrolled in the nine state schools during the last four years dropped out before completing a bachelor's program. The Independent Florida Alligator September 22, 1981 Statement of the Problem The above statement is only one of many in recent months which has focused on the attrition/retention/graduation problem being experienced throughout the State University System of Florida. Although the main objective of colleges and universities is to graduate students, there is a high attrition rate for black students. A recent study by Egerton (1969) reported that of every 100 white 18 year olds in the United States, approximately 75 finish high school, about 45 enter college, and about 15 earn a baccalaureate degree. However, the persistence and graduation rates show a marked decline for the black population of 18 year olds. Of every 100 black 18 year olds, approximately 65 finish high school, about 30 enter college, and only about 5 actually earn a baccalaureate degree.

PAGE 11

An extensive follow-up study of 20,000 1972 high school graduates was conducted by Eckland and Wisenbaker (1979). While the overall college dropout rate for both white men and women was found to be 34 percent, the rates for black men and women were 43 and 46 percent, respectively. An examination of the graduation trends for the two groups revealed the following rates: white men, 36 percent; white women, 46 percent; black men, 27 percent; and black women, 32 percent. Goodrich (1976) conducted a similar in-depth study on the academic status of minority students. The results of this study indicated that of the first-time enrolled freshmen in 1974, a year later 14 percent of the black students had been dismissed for academic reasons, as compared to 3 percent for white students and 4 percent for the university as a whole. The reported overall dropout rate (for both academic and nonacademic reasons) was 27 percent for blacks, 18 percent for whites and 21 percent for the overall university population. However, another follow-up study of black and white students (White and Suddick, 1981) found that the graduation rates for both groups over a period of five years were very similar. There was also very little difference in the academic standing (cumulative grade point averages) of those students who withdrew by race. In a comprehensive ten-year study Astin (1977) however concluded that although the absolute persistence rate for blacks was lower than that for whites, the persistence rate for black women was actually higher than that for white women of comparable preparation

PAGE 12

and ability. Therefore, not only were differences found to exist between races, but also between sexes within the same race. While the majority of the above studies suggests high attrition rates for some black students, those initially admitted under special guidelines which were established in compliance with federal requirements face additional problems (Garber and Schell, 1977). Although special admission criteria may vary among similar institutions, they are usually based on minimal standardized test scores and/or high school grade point averages, A recent report by the National Advisory Committee on Black Higher Education and Black Colleges and. Universities recommended that the specific needs of black students at predominately white institutions should be defined. This report also suggested identification of ways in which these institutions can increase their response to the access to and completion of programs by blacks (N.A.C.B.II.E. 1979). In a majority of cases, a great deal of emphasis is placed on the recruiteraent and admission of specially admitted students (Carr and Chittum, 1980). These students are overwhelmed with letters of welcome and congratulations from various offices prior to entering the university or college. Their initial face-to-face contact with university personnel usually comes in the form of some type of formal or informal orientation program (McDavis, Mingo, Stewart, and Hough, 1980). Haettenschwiller (1971) suggests that the establishment

PAGE 13

of early contact is fundamental to the counseling of black college students in special programs. Although these contacts are basically positive in nature, they are not often maintained at the same level of intensity over an extended period of time. This relationship is usually not reestablished until the student incurs academic difficulty. Too often the student's dissatisfaction with the college environment is made apparent through low academic achievement It is at this point that concern is shown for these students and assistance is offered. Many times this concern and assistance is too late to be of any benefit to the students (Friedlander 1980). Other specially admitted students, although performing well academically, also may be experiencing dissatisfaction with the college environment and choose not to return to the institution for one reason or another. Satisfaction with the college environment, including the helpfulness of faculty, staff, advisors, and counselors, seems to be an important determinant in the attrition rate of these specially admitted students. In contrast, the same set of circumstances may result in a very different outcome for regular students because of their tendency to generalize high school experience to the college situation (Haettenswiller 1971). During the course of previous daily interactions, they have internalized the rules of the game and know the roles that they are expected to play. When the expectations of the regular student are not realized, the resources to make the appropriate

PAGE 14

adjustment are readily available. Merton (1957) uses the term "anticipatory socialization" to describe this process. Since the beginning of special admission policies, researchers have attempted to identify factors related to academic success. The most common factors that have been associated with college attrition are academic and nonacademic. There is a wealth of research concerned with academic factors such as high school grade point averages and standardized college entrance test scores (Aleamoni, 1977; Dalton, Anastasiow, and Brigman, 1977; Price and Kim, 1976; Rovezzi-Carroll and Thompson, 1980; Pfeifer and Sedlacek, 1971). There have also been attempts to identify nonacademic factors such as personality characteristics, self -concept socioeconomic status, biographic background, and satisfaction (Green and Farquhar, 1965; Hannah, 1971; Starr, Betz, and Menne 1972; Samuel and Laird, 1974). The University of Florida is part of a State University System whose policy permits 10 percent of the freshman class to be admitted without having met minimum admission requirements. At present, these requirements are a high school grade point average of 2,0 and a total of 800 on the SAT (Scholastic Aptitude Test) or 17 on the ACT (American College Testing Program). Except in very few cases, the students are required to enroll for the term immediately following high school graduation. The rationale behind this requirement is that during the summer a more intensive orientation program can be implemented and individual attention can be

PAGE 15

aifgnnn w mi. fij i ww-^^iaw>j>jri 6 given to these students since there is generally a lower enrollment during this term. Presently there are two programs at the University of Florida primarily responsible for the overall coordination of the orientation program, and subsequent retention and graduation of these students. These two programs are Special Services and the Program for Academic Counseling and Tutoring (PACT) The Special Services Program is a federally funded program designed to assist students from "culturally and/or educationally" deprived backgrounds. This program began in 1971 and provides supportive services and alternative grading and test-taking practices to insure the participants' success in the academic as well as social environment. Currently there are hundreds of such programs being conducted throughout the United States with 17 based on college campuses within the State of Florida, The PACT, which began summer 1979, is state funded and provides essentially the same type of assistance that the Special Services Program provides. These programs work closely in the planning and implementation of an orientation program and other activities which involve specially admitted students at the University of Florida. These other activities include tutoring, counseling, general information, financial assistance information, academic advisement and career information. Another important aspect of these programs is the use of peer counselors who serve as social and academic role

PAGE 16

models. These former "program" participants are a valuable asset providing assistance to students throughout the critical freshman year. Although the type of assistance which these two programs offer varies according to individual student needs and resources available, they both have as their main goal graduation of program participants. While most of these students seem to make the initial adjustment to college life, problems arise for them after this introductory period. Even though the special selection criteria are essentially used to predict which "marginal' students will most likely succeed if admitted, predictions based solely on standardized test scores and/or high school grade point averages have not proven to be accurate. Purpose of the Study The purpose of the present study was to investigate the relationship between five independent variables (high school grade point averages, college entrance exam scores, personality type scores, self concept scores, and college student satisfaction scores) and one dependent variable (college grade point average). This study also attempted to determine the relationship between college student satisfaction scores, personality type scores, and selfconcept scores of specially admitted black university freshman students. Although this relationship has been the source of much recent research, the establishment of a

PAGE 17

8 predictive model by the use of multiple factors has not been extensively explored and determined. This was the focus of this study. Specifically, this study proposed to answer the following questions concerning specially admitted black university freshman students: (1) Does a relationship exist between college grade point averages and high school grade point averages? (2) Does a relationship exist between college grade point averages and college entrance exam scores? (3) Does a relationship exist between college grade point averages and self-concept? (4) Does a relationship exist between college grade point averages and personality type preferences? (5) Does a relationship exist between college grade point averages and college student satisfaction? (6) Does a relationship exist between personality type preferences and college student satisfaction? (7) Does a relationship exist between self-concept and college student satisfaction? Need for the Study Institutions of higher education, along with other segments of society, are facing difficult times due to financial cutbacks and decreasing enrollments. These impending

PAGE 18

9 budgetary difficulties have caused institutions to raise admission standards for regularly adraitted students and to scrutinize even more closely those students who are admitted through special admission policies. Numerous national surveys have been conducted in an attempt to monitor admission and retention trends for minority students attending large, predominately white universities in the United States (Sedlacek and Brooks, 1970; Sedlacek, Brooks, and Horowitz, 1972; Sedlacek, Brooks, and Mindus, 1973; Sedlacek, Lewis, and Brooks, 1974; Sedlacek and Webster, 1978), The results of these studies indicated that there was gradual progress towards admitting more black freshmen and searching for better methods of selection. Also, the number of special programs demonstrated a decrease as well as the number of schools utilizing different admissions criteria for minority students. The attrition/graduation rate of all students, and especially those admitted through special guidelines, is a major concern to university administrators. Rovezzi-Carroll and Thompson ( 1980) concluded that college grade point average and college graduation may result from the interaction of different sets of antecedent and intervening variables. It also has been suggested that a variety of factors should be considered in making decisions about prospective students (Mayhew, 1965). However, the problem exists in attempting to operationally define some of those factors.

PAGE 19

10 In recent years there have been questions raised concerning the effectiveness of special admission programs and/ or the academic success of former participants in terms of grade point averages and graduations. Past research in this area has produced ambiguous and mixed results (Cleary, 1968; Temp, 1971; Boyd, 1977; Pfeifer and Sedlacek, 1971), However, the decision as to whether a student is admitted or denied admission is still most often based on the academic factors of standardized test scores, high school grade point averages, and/or letters of recommendation regarding their academic potentials. It has been shown that decisions based on the above mentioned factors are not foolproof in most cases, and admission committees cannot be absolutely certain of the accuracy of their decisions. One researcher (Gibbs, 1973) suggested that in order to reduce the casualties of the high risk admissions programs, admissions officers should be more attentive to alternative indicators that might be more predictive of a black student's success than traditional indicators. Examples of these nontraditional criteria may include positive selfconcept, leadership potential, goal oriented activities, and personal maturity, Tinto (1975) also cited the failure of past research to define more clearly the possibility of multiple characteristics of dropouts. This researcher hypothesized that dropout occurs because of academic and social malintegration and is a longitudinal process of. interactions between the individual and the academic and social systems of the college.

PAGE 20

11 Most certainly administrators and others responsible for admitting, retaining, and graduating specially admitted students would desire to admit those individuals with a high probability of success based on as many factors as are available. However, many questions still go unanswered as to which factors combine to produce high retention and graduation rates. Do some students suceeed because entrance examination scores accurately predict their scores? Why do other students with those same test scores fail? Is there something unique about individuals who score low on standardized tests and still manage to achieve academic success? Answers to these and other questions would provide invaluable information to admissions committees, counselors, and students. During spring 1979 at the University of Florida, a retention fact-finding study was conducted (Stewart, 1981). The purpose of this study was to establish retention rates for specially admitted students. The findings of this study indicated that of the 762 students admitted between the stimmers of 1974 and 1979, 498 or 64 percent were still enrolled, while 36 percent were no longer enrolled. Of those students still enrolled, 38 or 8 percent had 3.0-4.0 grade point averages; 320 or 65 percent had grade point averages of 2.02.9; 66 or 13 percent were on academic warning; and 65 or 13 percent were on academic probation. This fact-finding study provided evidence that not all specially admitted students achieve at the same level and that nonacademic as

PAGE 21

12 well as academic type factors may be predictors for the success of these students. It is apparent that individuals differ on characteristics other than high school grade point averages and standardized test scores. Thus, other factors which may affect the academic performance of college students should be studied. What makes one student succeed and another student with identical or very similar academic background fail? This question haunts university admissions committees and administrators and was addressed by the present study. Definition of Terms The following terms have been operationally defined for the purpose of this study: Attrition The tendency of students to leave an institution of higher education prior to receiving a degree due to the failure of maintaining cumulative grade point averages of 2.0 or better and/or nonacademic factors. College entrance exam scores The college entrance scores of specially admitted black university freshman students as measured by ACT (American College Testing) or SAT (Scholastic Aptitude Test). Grade point average The quotient achieved when total grade points are divided by total hours carried. Personality scores The personality scores of specially admitted black university freshman students as measured by the Myers-Briggs Type Indicator.

PAGE 22

13 Retention The tendency of students to remain at an Institution of higher education and receive a baccalaureate degree Satisfaction scores The satisfaction scores of specially admitted black university freshman students as measured by the College Student Satisfaction Questionnaire. Self concept scores The self concept scores of specially admitted black university freshman students as measured by the Tennessee Self Concept Scale. Specially admitted students First-time-in-college freshman students admitted to the University of Florida through special admission criteria during siimmer quarter, 1981. Organization of the Study The remainder of this study is divided into four chapters. A review of the literature on specially admitted students and the academic and nonacademic factors relating to college student attrition is presented in Chapter Two. The hypotheses, population and sample, instruments, procedures, analysis of data, and limitations of the study are explained in Chapter Three. The results of the study are reported and discussed in Chapter Four. Chapter Five includes conclusions, implications for further research, a summary of the study, and recommendations for further research.

PAGE 23

CHAPTER TWO REVIEW OF RELATED LITERATURE Introduction There are many factors which affect the college grade point averages and attrition of university specially admitted students. Basically, these factors can be divided into two categories: academic and nonacademic. For the purpose of the present study, the review of literature is discussed in terms of those two major sections. However, due to the extensiveness of the research in the area of nonacademic factors and their relationship to college attrition, the following subheadings are used: self-concept, personality characteristics, and college student satisfaction, Academic Factors Related to College Attrition The results of a recent national survey entitled. What Works in Student Retention (Beal and Noel, 1979) suggest that retention efforts for both fourand two-year institutions need improvement. Various authors have attempted to identify potential sources of problems associated with low retention rates (Carney and Geis, 1981; Cleary, 1968; Price and Kim, 1976; Stanley, 1971). The focus of the majority 14

PAGE 24

15 of these studies has been toward academic factors and their relationship to college attrition. Most of the criteria for selection into college are based on previous academic performance in the form of college entrance exam scores or high school grade point averages. One author (Aleamoni, 1977) concluded that if grades are to be used to predict college grade point averages, then they should be used as the criterion measure in the selection process, Beal (1979) suggests that a combination of factors, including high school grade point average and class rank, first semester college grades, study habits, motivational level and commitment, student-faculty relationships, and the fit between the college and student should be considered when addressing the problem of student retention. Recent research has revealed a significantly strong relationship between students who receive below average first-term grade point averages and attrition (Pantages and Creedon, 1978). Therefore it would seem that identification of the reasons for poor academic achievement is a necessity in the attempt to reduce attrition rates. One such study (Hart and Keller, 1980) was designed to obtain self-reported reasons for poor academic achievement among first-term freshmen. The findings indicated that these students placed the greatest responsibility for their low grades on themselves (i.e., their own lack of motivation, improper study habits, and inattention to school work).

PAGE 25

16 The freshman college year experience has been identified as being quite different from any other year's experience. Lokitz and Sprandel (1976) reported that freshman students demonstrate a movement from academic to social concerns during the first year. Following the establishment of academic status, emphasis is focused toward building and maintaining interpersonal relationships. The results of another study (Madrazo-Peterson and Rodriguez, 1978) seem to add credence to the idea that freshman perceptions are different from those expressed by older and more experienced students. Freshman students reported fewer feelings of isolation and significantly greater satisfaction than other students These authors suggested that initially freshmen perceive the campus environment in an idealistic and optimistic light but as matriculation continues idealism is replaced by realism and optimism by pessimism. The outcome is stress and dissatisfaction. Scott (1978) also reported a significant change between the initial expectations of black freshman students and their later perceptions about the campus environment The research in the area of academic factors and their relationship to college attrition is quite extensive. Many researchers (Kallingal, 1971 ; Pfeifer, 1976; Bean and Covert, 1973; Thomas and Stanley, 1969) have attempted to correlate previous academic achievements (SAT scores, high school rank, and high school grade point averages) and attrition. The findings, however, have yielded inconclusive and often conflicting results.

PAGE 26

17 Ikenberry (1961) cited the following limitations of previous studies seeking to identify factors related to college persistence: 1) too few variables had been included; 2) no differentiation was made between students who withdrew (in terms of achievement level or sex); 3) no multivariate statistical techniques had been used to analyze data; and 4) there had been little indication of the interrelationships among variables. One of the first studies which attempted to compare grade prediction equations of Negro and white students was conducted by Cleary (1968). This study reported mixed results; no significant difference was found in two integrated colleges in the East, but in a southwest college, the Negro students' scores were overpredicted by the use of the white regression equation. Temp (1971) replicated Cleary 's (1968) research and obtained similar results. The author further suggested that since a predictive system based on majority students serves to penalize black applicants by underpredicting their potential college performance, institutions using SAT and other admissions tests and predictors should conduct a validity analysis at their own institution. Another researcher (Bowers, 1970) also compared GPA regression equations for regular and specially admitted freshmen. High school percentile rank and verbal quantitative scores on the Cooperative School and College Ability Tests (SCAT) were found to be useful predictors of college GPA for both groups in this study.

PAGE 27

18 One group of researchers conducted a multivariate study of personality and academic factors in college attrition (Maudal, Butcher, and Mauger, 1974). They concluded that academic variables such as SAT, high school grade point average and rank, etc, are useful predictors of academic achievement with transfers, persisters, and dropouts. Lanning (1977) also investigated the relationship between numerous variables (age, high school rank, SAT scores, previous semester's grade point averages and number of hours completed) and college persistence and voluntary withdrawal. A consistently significant relationship was found to exist between prior grade point average and prior semester course load and withdrawal and persistence, that is, prior to withdrawing, withdrawals had lower grade point averages and took fewer class hours. Other research conducted in this area has produced inconclusive results. Price and Kim (1976) in a study designed to 1) identify the specific factors most likely to be associated with success of college education and 2) test the significance of those factors on college performance, found college entrance examination scores (ACT) to be more significant and important as predictors than high school grades. However, just as some studies have found a significant relationship between prior academic achievement as evidenced by college entrance exam scores or high school grade point averages and college grade point averages, many others have found little or no significant relationship. Pfeifer and Sedlacek (1971)

PAGE 28

19 obtained mixed results suggesting that while high school grades correlated highly with college grades for black females, white males and females, the Scholastic Aptitude Test-Verbal was the best predictor for black males. In an attempt to determine the relationship between underachievement and college attrition Dalton, Anastasiow, and Brigman (1977) studied three variables (predicted grade point average, Scholastic Aptitude Tests, and relative high school rank). No significant relationship was found between persisters, nonacademic dropouts and academic dropouts on these three variables. Boyd (1977) in yet another study found SAT scores to be ineffective as predictors of college graduation for minority students at selected institutions. The relative effectiveness of four variables, SATVerbal, SAT-Math, high school GPA, and family income as predictors for college graduation among low-income students was investigated by Rovezzi-Carroll and Thompson (1980). None of the four variables were found to be effective either singly or in any combination as predictors of college graduation, withdrawal, or academic dismissal. These authors concluded that perhaps college grade point average and college graduation result from the interaction of different sets of antecedent variables (self-concept, study habits) and intervening variables (college environment, perceived isolationism, provision of tutorial and support services).

PAGE 29

20 Houston ( 1980) evaluated the relationship between college grade point average (as the dependent variable) and selection predictor variables (high school rank, and the Verbal and Mathematics scores on the College Board Scholastic Aptitude Test) in a sample of specially admitted black female undergraduates. Results indicated that after 8 semesters, 19 percent of the students had withdrawn or were transferred to another institution, 52 percent had graduated, and 27 percent had been dismissed for academic reasons. Stepwise regression analysis using the predictor variables revealed a coefficient of multiple 2 determination (R ) of .22, that is, only 22 percent of the variance in college grade point average could be attributed to the three variables. High school rank con2 tributed to the greatest proportion of the variance to R (.14). SAT-V and SAT-M accounted for approximately 7 and 1 percent of the variance respectively. Houston stressed the need to identify predictors other than high school rank and SAT-V or SAT-M scores because of possible underprediction in some cases. The above section of this literature review cited various studies concerned with academic factors such as previous high school grade point average and/or rank and college entrance exam score and their relationship to college attrition. Although agreement has been reached on the relevance of adequate academic factors and their relationship to student attrition, agreement on specific selection criteria has not proven conclusive. Therefore,

PAGE 30

21 previous academic achievement is still a critical factor in the overall issue of college student attrition. Nonacademic Factors Related to College Attrition Research studies concerning nonacademic factors related to college student attrition are profuse. Although this section of the literature review cites research investigations regarding numerous nonacademic factors related to college student attrition, emphasis is placed on studies relating to self-concept, personality, and college student satisfaction Mornell (1973) in a study designed to access the success of a special admissions program reported a 60 percent retention rate. Mornell suggested that the traditional cognitive measures used to predict academic success for all students are valid only for the student with traditional credentials J that is, the success of the program depends on the selection process. Another researcher (Mayhew, 1965) also suggested that a variety of factors should be taken into consideration when making decisions about prospective students. Factors mentioned by Mayhew include past academic performance, academic aptitude, motivation, personality, special interests or abilities, extra-curricular skills, economic status, and physical characteristics. Two other authors (Dawkins and Dawkins 1980) similarly suggested that because of the inconclusive results in the past,

PAGE 31

22 additional noncognitive criteria such as racial perceptions and experiences, goals and values, social participation in campus activities and social background should be examined in the context of academic performance. Extensive research in the area of nonacademic factors and their relationship to college student attrition has been conducted (Menning, Bradley, and Cochran, 1975; Morrisey, 1971; Muskat 1979; Wessell, Engle and Smidchens, 1978). The various factors that have been researched cover a wide spectrum of variables. One group of researchers (Ellison, Murray, Fox, and Taylor, 1973) explored the effectiveness of biographical inventory data as possible predictors of college performance. The results indicated that biographical data were, in general, as effective or in some cases more superior than previous high school perfoimance in predicting college grade point average. Carney and Geis (1981) used data from a standardized reading test and student background information to correlate college academic performance and retention. Both college entrance test scores and reading test scores were found to be related to academic performance and attrition. These authors suggested that the use of reading scores could be helpful to colleges who seek to identify those students who are in need of special assistance. Other research (DiCesare, Sedlacek, and Brooks, 1972) has explored the ways, if any, in which black students returning to college are different from those not returning

PAGE 32

23 on demographic and attitudinal variables. They reported that blacks who returned 1) have more self-confidence and higher expectations; 2) feel more strongly that the university should influence social conditions; 3) see more racism at the university; and 4) are more likely to have lived on campus and made use of its facilities than did nonreturning blacks. Students were also asked to identify possible causes that would lead to their dropping out prior to receiving a degree. The authors concluded that persisters have strong self -concepts take a more realistic look at the university, and adapt to it to achieve their own goals In a four-year longitudinal study of 6,660 high aptitude students, Astin (1964) contended that students who drop out of college demonstrated the following characteristics: 1) come from lower socioeconomic backgrounds; 2) have lower ranks in high school; 3) plan initially to get lower college degrees; and apply for relatively fewer scholarships. When asked to identify and rate possible sources of academic problems, black students in both predominately black and white schools rated their high school preparation as less adequate (Jones, Harris, and Hauck, 1975). In an attempt to identify and describe the extent to which certain influence sources affect attendance and retention in college. Brown (1976) conducted a survey at the end of the college student's freshman and sophomore years. The top three influences were 1) students' own career

PAGE 33

24 plans; 2) the mother; and 3) the academic advisor, Webster, Sedlacek, and Miyores (1979) in a comparison of the problems perceived by minority students (Hispanic, American Indian, Asian American, and black American) and white American students found that black students expressed more serious problems in feeling like victims of racism and discrimination because of race. Other problems identified by all students included problems regarding vocational decisions, managing time, studying, and negotiating the university system. One author suggested specific program strategies beneficial for both students and the university (Maynard, 1980). Problems arise because of the inability of many universities to respond to the special academic, sociocultural and economic needs of minority students (Maynard, 1980). The conclusion reached by Cole and Hanson (1973) following a study into the extent of racial ethnic bias in selective college admissions was that college admissions personnel should give consideration to the relation of selection procedures to the values and goals of their college. Self-concept and College Student Attrition One of the most researched factors associated with college student attrition is self-concept. Mixed results have been reported in the literature regarding the selfconcept and its relationship to college student attrition. Astin (1977) reported positive changes in self-esteem and suggested that college attendance may increase the individual's sense of competence and self -worth. The longitudinal

PAGE 34

25 analyses performed in this study demonstrated that students undergo a variety of changes in self-concept, attitudes, and values after they enter college. These changes are evidenced by a more positive self-image and demonstrated in a greater sense of interpersonal and intellectual competence. Carey (1976) investigated the nature of the relationship between the self-concept and academic performance of black students on white campuses who participated in black studies and those who did not. No significant correlation between self-concept and academic performance was reported. Dowdle (1977) also reported no significant relationship between self-concept (as measured by the Tennessee Self -Concept Scale) and academic achievement of freshman students, that is, students who performed low academically did not have a lower self-concept than those who performed higher academically. Similarly, Samuel and Laird (1974) found no significant differences in the selfconcepts of black females on a predominately black campus as measured by the Tennessee Self-Concept Scale when compared to black females on a predominately white campus. In a study designed to determine the effectiveness of the Special Services Program in a southern university. Cole (1974) compared the grade point averages and self -concepts of program participants with nonparticipants Even though the Special Services Program appeared to have had a positive effect on academic achievement, the two groups did not differ significantly with respect to self -concept

PAGE 35

26 During pretesting, 82 percent of the students had initial total positive self-concept scores within plus or minus one standard deviation of the normative population. Cole hypothesized that the initial level of self-concept probably accounted for the small change score and the fact that there was no significant difference between the two groups as a result of the program treatment. Still another researcher (Copeland, 1974) studied peer counselor effects on the self-concept and academic adjustment of a group of Special Services students who participated in group counseling, The researcher used the Tennessee Self-Concept Scale and the College Inventory of Academic Adjustment as measures of self-concept and academic adjustment. Three groups of students were used in this study. Group one consisted of students with peers as group counselors. Group two consisted of students with Educational Opportunity Program staff counselors as leaders. Group three consisted of students who did not participate in the Educational Opportunity Program. Groups one and two both changed positively, but not significantly on all subscores on the Tennessee Self-Concept Scale. Peer counselors therefore were found to be as effective as group counselors when Special Services students were compared on the two variables. In contrast to the above mentioned studies, other researchers have reported significant relationships between self-concept and college attrition. Gadzella and Fournet (1976) found a significant difference between high and low

PAGE 36

27 achievers, with high achievers rating themselves significantly higher than low achievers. Green and Farquhar (1965) investigated the relationship of personality and cognitive factors with academic achievement of eleventh grade Negro and white students. These researchers used the Michigan State M Scales ( as a measure of academic motivation), the verbal score of the School and College Ability Test and the Verbal Reasoning score of the Differential Aptitude Test (as measures of aptitude), and current grade point average (as a measure of school achievement). No correlation was found between verbal aptitude and achievement for Negro males, However, a significant correlation between verbal aptitude and achievement was found for Negro females. Among the four subsets of the M scales, the single best predictor of achievement for the Negro sample was the self-concept. Therefore, a strong relationship between the students' selfperception and school achievement was reported. Jones (1975) researched the effect of personal growth group counseling on the self-concept and academic achievevement of Special Services Program participants. This researcher defined self-concept as a multi-dimensional phenomenom and therefore hypothesized that change may take place in one or morefacets without that change being evident in the total self-picture. The effect of personal growth group counseling on five scales (the physical, personal, social, family and the moral-ethical) of the Tennessee Self Concept Scale was compared to grades at the end of the summer

PAGE 37

28 session in two classes (sociology and English composition). Students were randomly assigned to one of three groups: 1) those who received group counseling; 2) those who received tutoring; and 3) those who received no treatment. The findings indicated that while the personal growth group counseling appeared to improve the total self-concept of males, the same did not hold true for females. However, overall, the students who received group counseling had more positive self-concepts and greater general academic achievement than those not receiving group counseling. One study (Morrison, Thomas, and Weaver, 1973) sought to determine the relationship between self-esteem and selfestimates of academic performance. These researchers hypothesized that students with low self-esteem would predict receiving low grades and students with high self-esteem would predict receiving high grades on an examination. Mixed results were obtained for two separate measures of self-esteem. The hypothesis was confirmed for the Coopersmith Self -Esteem Inventory, but not for the Ziller Social Self-Esteem scale or for the subscale of the Coopersmith inventory specifically relevant to school self-esteem. Another researcher (Pazandak, 1975) also investigated the role of self-estimation as a moderator variable in the prediction of college academic achievement. Valine (1976) studied the effects of dropping out of school on the selfconcept of students initially identified as underachievers The findings indicated that those who had attained senior

PAGE 38

29 status generally had a statistically more positive selfconcept (as measured by the Tennessee Self-Concept Scale) than those, who for various reasons dropped out of college. The studies cited above have produced inconclusive results regarding the relationship between various nonacademic factors including self-concept and college student attrition. Therefore, it is apparent that there is needed additional research in this area. Personality Characteristics and College Student Attrition Various attempts have been made to relate both academic and nonacademic factors to college student attrition. One nonacademic factor which has been researched is the personality of the students involved. In a historical three-year follow-up study on the fate of minority group freshmen who were admitted to the University of Florida under a 5 percent exemption quota, Cranney and Larsen (1972) concluded that "early identification of college potential and appropriate programs to develop it may further enhance success rates of marginally admitted freshmen" (p. 40). Therefore, it would seem that any information which could lead to improving this early identification process would be welcomed, including information regarding personality characteristics. Pfeifer and Sedlacek (1974) utilized two personality measures (the California Psychological Inventory and the Holland Vocational Preference Inventory) in an attempt to determine the relationship between these two variables and ?B=;''6^SiT*iqii,ft iT*r-* '
PAGE 39

30 college grades. The results indicated some relation between the measured college grades. The researchers therefore suggested that nonintellectual measures should be used in combination with intellectual measures to predict which ones uniquely and appropriately reflect the experiences of blacks in the society. Hannah (1971) investigated the relationship between personality traits and college entrance test scores of dropouts and "stayins." The personality traits of dropouts were found to differ significantly from "stayins" on several scales of the Omnibus Personality Inventory. Persisters also had obtained higher college entrance test scores. Another study (Tillman, Millott, and Larsen 1974) found no significant differences in mean reading test scores (as measured by the McGraw-Hill Reading Test) and personality type (as measured by the Myers-Briggs Type Indicator) when student preferences were compared on the four indices of the Myers-Briggs Type Indicator. However, higher end-of-thequarter grade point averages were obtained by those students who preferred Feeling than those who preferred Thinking. Pandey (1972) utilized the Minnesota Multiphasic Personality Inventory in an attempt to differentiate personality characteristics of successful, dropout, and probationary students. This researcher reported few significant differences between the three groups of students. Johnson (1970) also compared personality characteristics and grade point averages of persisters and nonpersisters and found distinct differences

PAGE 40

31 in regards to personality characteristics (as measured by the Minnesota Counseling Inventory). However, both persisters and nonpersisters had above 2.0 cumulative grade point averages and thus, no significant difference was reported in regards to grade point averages. Rossmann and Kirk (1970) examined the differences in ability (as measured by the Omnibus Personality Inventory), and attitude (as measured by questionnaire data) and found that voluntary withdrawals had higher verbal ability and were more intellectually oriented than those required to withdraw for academic violations. Leon (1974) hypothesized that personality and college grade point averages were significantly related. However, no significant differences were found and Leon suggested that the reason may lie in the role that academic success may play in the lives of specially admitted students. Millott (1974) reported significant correlations ranging from .065 to .186 between the MBTI proference scores and the McGraw-Hill Reading Test scores of 2,514 University of Florida freshmen.. However, although the correlations with the MBTI were in the direction predicted by type theory, this correlation was not large. Introverted-Intuitive (IN) types tended to be more successful in academic endeavors and scored higher on achievement tests than types in the other three quadrants of the MBTI. These types (IN) were also found to have higher scores than the other types on scholastic aptitude tests. This study reported no statistically significant differences by sex between type and reading.

PAGE 41

32 McCaulley (1972) researched Expanded Educational Opportunities Program freshmen at the University of Florida in an attempt to identify personality types „ The findings indicated that out of the total sample of 135, the majority (105 or 77 percent) of the students were identified as Sensing types. A total of only 30 or 22 percent were on the Intuitive side of the continuum. The three type preferences which yielded the highest number of students were ESTJ (N = 29 or 21.5 percent); ESFJ (N = 20 or 14.8 percent), and ISFJ (N = 17 or 12.6 percent). The three type preferences which yielded the lowest number of students were : ENFJ and ENTP (both with Ns of 1 and percentages of 7 percent each) and ESTP and INFP (both with Ns of 2 and percentages of 1.5 percent each). McCaulley (1974) conducted another research study with Special Services Program participants and obtained results very similar to those reported above. The majority (N = 129 or 84 percent) of the 154 students in this study were identified, as Sensing types. A total of only 25 or 17 percent were on the Intuitive side of the continuum. The three type preferences which yielded the highest number of students were ESTJ (N = 32 or 20,8 percent) and ISFJ and ESFJ (both with Ns of 24 and percentages of 15.6 percent). The lowest percentages were recorded by the following type preferences: ENTP (N = 0), ESTP (N = 1 or .6 percent), and INFJ, INTJ, and INTP, all with Ns of 2 each and percentages of 1.3 percent each.

PAGE 42

33 In a recent follow-up study Schroeder, Warner, and Malone (1980) utilized the Myers-Briggs Type Indicator in conjunction with retention-attrition data and reported the following retention rates by personality type: Intuitive, 71 percent; Feeling, 71 percent; Sensing, 78 percent; and Thinking, 57 percent. The results of these studies indicate that although attempts have been made to correlate specific personality characteristics with certain levels of college academic achievement, attrition, and subsequent graduation, research in this area is still needed. While the evidence generally indicates that some type of relationship does exist between personality characteristics and college student attrition, the extent and preciseness of this relationship is still a topic of much debate in higher education. College Student Satisfaction and College Student Attrition Until recently, the least researched phenomercn relating to college student attrition has been college student satisfaction. However, college student unrest and the turbulence of the Vietnam era ushered in the need for research in this almost forgotten area. Thus, there has been a tremendous increase in the number of investigations concerning college student satisfaction during the last ten years. Numerous attempts have been made to correlate several factors with college student satisfaction. These include age (Sturtz, 1971), tenure in college (Starr, Betz, and Menne, 1972),

PAGE 43

34 institutional assessment (Hallenbeck, 1978), and studentcollege congruence (Pervin, 1967, Morstain, 1977), Prior to the disturbances of the 1960's, Berdie's (1944) was the only recorded report on the relationship between college academic achievement and student satisfaction, Berdie investigated engineering students' satisfaction v/ith the curriculum as it related to academic achievement. This study utilized first-year honor points, high school grades, and scores on a series of ability tests as measures of performance. Even though satisfaction was found to be significantly related to academic achievement, no single factor was found to relate to a student's satisfaction with his/her curriculum, Berdie suggested a need for a more complete measure of student satisfaction. Martin (1968) attempted to evaluate freshman students' perception and their degree of satisfaction by using a modified Q-sort which consisted of college-oriented items. Students were assumed to be satisfied with college if their real and ideal Q-sorts were similar. If Q-sorts were dissimilar, students were assumed to be dissatisfied with college. The results indicated that initially freshmen were moderately satisfied with college but that this satisfaction decreased by the end of the year. Results also indicated that there was no statistically significant relationship between initial freshman satisfaction with college and academic achievement at the end of the year. King and Walsh (1972) utilized the College and University Environmental

PAGE 44

35 Scales (CUES) to investigate change in environmental expectations and perceptions of college students. The findings from this study indicated that freshman year experiences have a direct impact on student perceptions of the environment. Astin (1977) reported that students, in general, maintained relatively high satisfaction with most aspects of their college experience as measured by a follow-up questionnaire. Women students were found to be slightly more satisfied than men with their collegiate experience and whites were found to be relatively more satisfied than blacks. The most important institutional characteristics affecting student satisfaction were academic selectivity, prestige, and institutional size. Scott (1978) attempted to determine if black and white freshman students had different initial expectations and later perceptions of their college environment. Two of the many variables that this researcher hypothesized would influence student expectations and actual perceptions of the campus atmosphere were high school scholastic achievement and freshman year scholastic achievement. Students were asked to complete the College and University Environment Scales, Second Edition (CUES 2), on two occasions with a seven-month interval between administrations. During the first administration, students were asked to complete the CUES 2 in terms of what they expected to be true or untrue of their college environment. On the second occasion, students were requested to complete the CUES 2 in terms of

PAGE 45

36 their actual perceptions of the campus atmosphere. Scott hypothesized that there would be a significant difference between high school grade point averages and student expectations and freshman year accumulative grade point averages and student expectations and/or actual perceptions „ However, the findings indicated that neither of the two variables seemed to influence or contribute to student expectations and perceptions. In spite of the tremendous gains in knowledge that have come from these recent studies, the question still remains : V/hat is the relationship between college student satisfaction and attrition? It seems clear that more data are needed both as a needs assessment tool for planning and implementing institutional policy as well as for a genergil indication of college students' attitudes and satisfaction with regard to their college experiences. Betz, Menne, Starr, and Klingensmith (1971) attempted to systematically investigate college student satisfaction based on a factor analytic study. Their study produced consistent results across two groups of college undergraduates who were administered the questionnaire independently during the fall and winter terms. In yet another study, Betz, Starr, and Menne (1972) proposed to investigate the degree of student satisfaction with colleges and universities. In a comparison of large public university students with small private colleges, public university students appeared to be more satisfied with social life and working

PAGE 46

37 conditions in their institutions than did private college students. On the other hand, private college students appeared to be more satisfied with recognition, quality of education, and compensation for effort. However, both groups were found to be satisfied with various aspects of their college experience as measured by the CSSQ. Hallenbeck (1978) views satisfaction indicators as primary sources of information regarding institutional assessment. Thus, the information obtained from satisfaction studies can be used as baseline data to guide in program planning, implementation, and change. Schmidt and Sedlacek (1972) used an activities and attitude inventory to survey variables related to university student satisfaction. Their findings reflected the optimism and idealism expressed by freshman students. In a study designed to investigate the relationship between age and college student satisfaction, Sturtz (1971) reported that older women (25 or above) were found to be generally more satisfied than younger women (18 to 24). In a comparison of black and white student satisfaction, Robertson (1980) utilized the College Student Satisfaction Questionnaire. Black student satisfaction was found to be significantly lower than that of white students only with respect to the university's system of rewards (compensation). On the scales of social life, working conditions, and quality of education, the satisfaction level of black students was found to be slightly higher than that of their white

PAGE 47

38 counterparts. Although black student satisfaction on the recognition scale was slightly lower than that of white students, the difference was not statistically significant. Also, even though the overall level of black student satisfaction was lower than that of white students, no significant differences were found. In an effort to identify differences in college student satisfaction among academic dropouts and nonacademic dropouts, Starr, Betz, and Menne (1972) utilized the College Student Satisfaction Questionnaire. These researchers hypothesized that 1) students who remained in college would be more satisfied than those who dropped out; and 2) dropouts who left for nonacademic reasons would be more satisfied than academic dropouts. There were three scales of the College Student Satisfaction Questionnaire which were found to discriminate between students who remained and those who dropped out. These were compensation, recognition, and quality of education, that is, dropouts were consistently less satisfied on these three scales. In a recent study Lindsey (1981) utilized the College Student Satisfaction Questionnaire (CSSQ) along with the University Involvement Questionnaire (UIQ) in an attempt to investigate black and white undergraduate university students' degree of satisfaction with various aspects of their university environments and their levels of involvement in those environments. A total of 800 undergraduate students from two predominately white (Florida State and the University

PAGE 48

39 of Florida) and one predominately black (Florida Agricultural and Mechanical University) institution were included in the study. Lindsey hypothesized that 1) there would be no differences between black and white undergraduate university students' degree of satisfaction with their university environment; 2) there would be no differences between the degree of satisfaction of black undergraduate students attending predominately white universities and those attending a predominately black university; and 3) there would be no differences between the degree of satisfaction of white undergraduate students attending a predominately black university and those attending predominately white universities. The means for the 373 blacks in the study on the five scales of the CSSQ were reported as follows: Working Conditions, 40.87; Compensation, 42.65; Quality of Education 37.72; Social Life, 40.32; and Recognition, 42.64. The total score mean was 204.22. These means compared to 42.17, 45.08, 39.44, 39.83, 41.77, and 208.29, respectively, for the 427 whites included in the study. The total score mean for black undergraduate students attending the University of Florida was reported as 200.95. Although mean total satisfaction scores for black undergraduate students at the two predominately white institutions were slightly lower than that of their white counterparts, the differences were not statistically significant. Lindsey therefore concluded that black undergraduate students attending predominately

PAGE 49

40 white universities were as satisfied with their university environments as black and white students attending predominately black and white universities. However, black students were found to be less satisfied than white students on the scales of working conditions, compensation, and quality of education in their university environments as measured by the CSSQ. Recent studies in the area of college student satisfaction have proven to be beneficial in the attempt to identify factors related to college student attrition. The ability to identify exit-prone students prior to their leaving seems a necessity for those persons delegated the responsibility for increasing graduation and retention rates.

PAGE 50

CHAPTER THREE METHODOLOGY Introduction The purpose of this study was to investigate the relationship between five independent variables (high school grade point averages, college entrance exam scores, personality type scores, self-concept scores, and college student satisfaction scores) and one dependent variable (college grade point average) in a sample of black specially admitted university freshmen. The hypotheses, population and sample, instruments, procedures, analysis of data, and limitations of the study are discussed in this chapter. Hypotheses The following hypotheses were investigated: HO-j^: There is no significant relationship between college grade point averages and high school grade point averages of specially admitted black university freshman students. HO^: There is no significant relationship between college grade point averages and college entrance exam scores of specially admitted black university freshman students. 41

PAGE 51

42 HO^ : There is no significant relationship between college grade point averages and self-concepts of specially admitted black university freshman students H0_^: There is no significant relationship between college grade point averages and personality type preferences of specially admitted black university freshman students. H0_ : There is no significant relationship between college grade point averages and college student satisfaction of specially admitted black university freshman students. HOg : There is no significant relationship between personality type preference and college student satisfaction of specially admitted black university freshman students. HO^ : There is no significant relationship between self-concept and college student satisfaction of specially admitted black university freshman students Population and Sample The target population to which the results of this study is generalizable to includes black freshman students who have been admitted to four-year public universities under special admission guidelines. These students enter the university as f irst-time-in-college freshmen directly

PAGE 52

43 from high school and participate in special admittance programs such as those described in the statement of the problem section of this study. These students have not met the minimum academic admission standards set by universities which would allow them to be admitted as regular students and are thus considered "high risk." However, they are granted admission through special procedures which may include a special subcommittee which reviews their records along with recommendations from counselors, teachers, and administrators who feel they have the potential for success in college. The majority of these students receive some type of financial aid from either federal and/or state assistance programs. From a total population of 142 specially admitted students who were admitted to the University of Florida during summer 1981, a sample of 103 black freshmen (69 females, 34 males) were chosen for this study. All of the subjects were present and participated in the first day of the fiveday orientation conducted by the two special admission programs (PACT and Special Services). The SAT (Scholastic Aptitude Test) or ACT (American College Testing) scores of these students were either below the minimum admission requirement of 800 and 17, respectively, and/or their HSGPA (high school grade point averages) were below 2.0. m^ac* f -> g — nM rg^n .. n: tas.fc^ 7y yfi,y^>...^)yy.ti^ng

PAGE 53

44 Instruments The Tennessee Self Concept Scale, the Myers-Briggs Type Indicator, and the College Student Satisfaction Questionnaire were the instruments used in this study. The Tennessee Self Concept Scale is a 100-item instrument which is also self-descriptive and was used to identify "self-perceived" attributes of specially admitted black students. The Myers-Briggs Type Indicator is a selfdescriptive instrument and was used to "type" and identify personality characteristics of specially admitted black students. The College Student Satisfaction Questionnaire consists of 70 items relating to various aspects of college and university life and was used to assess college student satisfaction of specially admitted black students. Tennessee Self Concept Scale The Tennessee Self Concept Scale consists of 100 selfdescriptive, self-administering items by which responses are recorded in a Likert-type scale of five choices ranging from completely false to completely true. The manner in which individuals perceive themselves influences their behavior and their relationships and is related to their overall personality. Therefore, being aware of how people view themselves is both useful and necessary in order to assist them. Two forms of the TSCS are available for use, a Counseling Form and a Clinical and Research Form. The Counseling Form

PAGE 54

45 was used for the purpose of this study since it is less time consuming and the only differences between the two forms is the method of scoring and the profiling system. Both forms can be scored either by hand or machine. The time required to complete the scale varies from 10 to 20 minutes The Counseling Form generates five scores which provide information regarding an individual's self-concept. The following is a brief description of each of these scores: 1. The Self-Criticism Score (SC) is composed of 10 items taken from the L-Scale of the Minnesota Multiphasic Personality Inventory (Hathaway and McKinley, 1951). These are statements (even though negative to some extent) which most persons agree to as being true for them. In general, low scores (denial) indicate defensiveness and high scores indicate a normal capacity for self-criticism. 2. The Positive Scores (P) consist of three horizontal categories (Row 1, Row 2, and Row 3) which represent an internal frame of reference and five vertical categories (Column A, Column B, Column C, Column D, and Column E) which represent an external frame of reference. In addition, a Total P score is also calculated. a. Total P Score is the most important single score on the Counseling Form because it reflects the overall level of self-esteem.

PAGE 55

46 b. Row 1 P Score (Identity) indicates "what I am." The individual describes what he perceives to be his basic identity. c. Row 2 P Score (Self-Satisf action) indicates "how I feel about myself." This score reflects how satisfied an individual is with himself. d. Row 3 P Score (Behavior) indicates "what I do or how I act." This score reflects the individual's perception behavior. e. Column A (Physical Self) indicates how an individual views his body and overall physical well-being. f. Column B (Moral-Ethical Self) describes how an individual views himself in terms of moral worth, religion, relationship to God, etc. g. Column C (Personal Self) measures the individual's feelings of adequacy without regard to his body or relationships to others. h. Column D (Family Self) measures the individual's feelings of value and usefulness as a family member i. Column E (Social Self) describes the individual's feelings of adequacy and usefulness in relation to other people in general 3. The Variability Scores (V) reflect the inconsistency which may exist among different areas of the individual's self-perception. Three (V) scores are measured: Total V, Column Total V, and Row Total V.

PAGE 56

47 a. Total V measures the sum of all variability. b. Column Total V measures the sum of the variations within the columns. c. Row Total V measures the sum of the variations across the rows 4. The Distribution Score (D) summarizes how an individual varies answers across the five available response categories. This score may also indicate how certain an individual is about the choices made 5. The Time Score measures the amount of time it requires the individual to complete the scale. Norms for the scale were developed from a group of 626 people from various parts of the United States and varying in age from 12 to 68. Subjects included members from high school and college classes, and employers at state institutions in addition to other sources. The sample included both black and white subjects, representatives from all social, economic, and intellectual levels and educational levels from sixth grade through the Ph.D. degree, and an approximately equal number of males and females. The testretest reliability coefficients based on a group of 60 college students over a two-week period (Fitts, 1965) are as follows: Self-Criticism, .75; Total Positive, .92; Total Variability, .67; Distribution, .89; and Time .89. The four procedures used by Fitts (1965) to establish validity were content validity, discrimination between

PAGE 57

48 groups, correlation with other measures, and personality changes under particular conditions. Content validity was established by having clinical psychologists analyze each item. Only unanimous agreement by the judges that it was classified correctly allowed an item to remain a part of the scale. Various groups were used to demonstrate that the scale is a reliable discriminator between psychiatric patients and nonpatients, delinquents and nondelinquents and the average person and a psychologically integrated person. Further assessment of validity was established by comparing the scale with other personality measures. In comparing the scale with the Minnesota Multiphasic Personality Inventory (MMPI), it was found that most of the scores of the scale correlated with MMPI scores. Correlations with the Edwards Personal Preference Schedule indicate a nonlinear relationship between scores on the two tests (Sundby, 1962). In order to validate personality changes under particular conditions, the effects of positive and negative life experiences were studied. Gividen (1959) reported a significant decrease in scores following a stressful situation. Ashcraft and Fitts (1964) studied changes in patients' scale scores due to psychotherapy. They reported that the therapy group changed significantly and in the expected direction on 18 of the 22 variables studied, while the control group changed in only two variables.

PAGE 58

49 The TSCS was used to identify self-concept because the development of this instrument is based on the premise that the way a person perceives himself influences his behavior. Therefore, the ability to obtain and identify self-reported self-concept scores would be of invaluable use for the purpose of this research, Myers-Briggs Type Indicator The Myers-Briggs Type Indicator is a 166-item, forcedchoice, self-administering, self-validating instrument, first published in 1962 by Educational Testing Service and later in 1975 by Consulting Psychologists Press. It was designed to facilitate use of that part of Jung's theory which describes psychological types. In essence the theory proposes that variations in human behavior which appear to be random at first glance are in actuality orderly and planned. Perceived differences only exist in the manner by which people choose to make use of their perception and judgment. The purpose of the MBTI is to identify and "type" individuals in terms of four preferences: ExtraversionIntroversion (EI); Sensing-Intuition (SN); Thinking-Feeling (TF); and Judging-Perception (JP). The kinds of perception are sensing and intuition; thinking and feeling are the two kinds of judgment. Myers (1962) postulates that if people differ systematically in what they perceive and the conclusions they come to, they may show corresponding differences in their reactions, interests, values, needs, i^i (gr i rTi i^ T iHi >. tf riT ..• m i w wj ^rtti

PAGE 59

50 motivations, in what they do best, and in what they like best to do. The indicator aims to obtain people's basic preferences as they relate to perception and judgment so that the outcome of the preferences and their combinations may be established by additional research and put to practical use. Each of the four independent preferences is "scored" on a continuum and yields a total of 16 possible "type" combinations. The following is a brief description of the four basic preferences : DIRECTION OF INTEREST Extraversion E Prefers the outer world of actions, objects, and people. PERCEPTION Sensing S Present-oriented; relies on immediate reality, known facts, and direct experience. JUDGMENT Thinking T Views events objectively, scientifically; uses the scientific method to make decisions; logical order — cause and effect Introversion I Prefers the inner world of ideas and concepts. Intuition N Future-oriented; looks into inferred meanings and relationships. Feeling F Views events subjectively; uses personal values and importance to make decisions, LIFE STYLE Judging J Prefers planned, decided orderly way of life; looks for closure in decision making. Perceptive P Prefers a flexible, spontaneous way of life; openended, tentative schedule, Figure 1 Myers-Briggs Type Indicator Preferences •i^iMCmS-ilj^ K^ti(

PAGE 60

51 Studies with academic populations have yielded fairly low intercorrelations for EI and TF, with the median absolute interrelation ranging from .03 for males and .06 for females. However, the JP index has been shown to correlate consistently with SN. The range of this correlation is from ,20 to .47. Split-half reliability correlations were obtained by computing tetrachoric r's and applying the Spearman-Brown Prophecy Formula, The correlations for types are EI, .74 to .88; SN, .77 to ,88; TF, .66 to ,90; and JP, ,76 to .93, The results of the extensive validity studies that have been conducted are difficult to summarize. Even though theoretically the most important correlates for the MBTI are the "types" themselves, evidence is available to support the validity of the theory and the indicator. This evidence is found in the ability of the "type" preferences to correlate positively with interests, values, and needs identified by other tests, or to correlate with any other external indication of internal differences. The Gray-Wheelwright (which also proposes to identify the Jungian types) and the MBTI yield correlations of .79, .58, and .60 for EI, SN, and TF, respectively. The Gray-Wheelwright has no scale for JP. The MBTI correlates with 103 of the 180 correlations of the Strong Vocational Interest Blank. These correlations are significant at the .01 level. When compared to the AllportVernon-Lindzey study of values, two-thirds of the 24 correlations with the MBTI are significant at the .01 level (Myers, 1962).

PAGE 61

52 All instructions are contained on the cover page of the test booklet, and there is no time limit. Two forms of the MBTI are available, Form F and Form G, both with standard IBM answer sheets that may be either hand or machine scored. For the purposes of identifying the "type" preferences of each subject in the present study, Form G was used. Continuous scores transformed from preference scores were computed. A brief description of this transformation process is explained by Myers (1962) in the MBTI manual (p. 9). For an I, N, F, or P score, the continuous score is the preference score plus 100. For an E, S, T, or J score, the continuous score is 100 minus the preference score. A table is provided to speed the transformation process (p. 10). The MBTI was used to identify personality types because it seeks to identify how individuals use perception and judgment in making decisions. Myers (1962) suggests that identifying whether an individual is influenced by intellectual or economic values could contribute to a more accurate matching of applicants to the college of their choice. College Student Satisfaction Questionnaire The College Student Satisfaction Questionnaire (CSSQ) was designed to measure five selected dimensions of college student satisfaction. The theory behind its development stems from research based on the satisfaction of employees in business and industry (Herzberg, Mausner, Peterson, and

PAGE 62

53 Capwell, 1957), Job satisfaction research has shown a consistently negative relationship between job satisfaction and job turnover, that is, the greater the satisfaction, the less the turnover. Therefore, if the college environment can be viewed as a place of "employment" and the student as an "employee," then student satisfaction with college should be negatively related to turnover (dropping out of college). It has been the purpose of much of the research with the CSSQ to test the validity of this analogy. The initial form of the CSSQ (Form A) consisted of a 139-item instr^iment The model for the format of the instrument is the Minnesota Satisfaction Questionnaire which was developed by Weiss, Dawis, England, and Lofquist (1967) to measure job satisfaction. The 139-item instrument proposed to measure six selected satisfaction dimensions: Policies and Procedures, Working Conditions, Compensation, Quality of Education, Social Life and Recognition, An analysis of data following the administration of the measure to 643 Iowa State University students yielded internal consistency reliability coefficients for each of the six scales. These reliabilities ranged from .85 to .92 with a median of .88. Relatively normal scale score distributions were also found. Between scale correlations ranged from .39 (Social Life and Compensation) to ,77 (Policies and Procedures and Quality of Education), with a median correlation of .54. I ji II jimi ii>ii>iii

PAGE 63

54 The present form of the CSSQ (Form C) was developed on the basis of the above analyses and consists of a 70item, five-choice, Likert-type scale. There are five response alternatives offered which range from "Very Dissatisfied," through "Satisfied," to "Very Satisfied," and are scored one to five points, respectively. Scale scores are derived from the sum of each of the 14 selected items. A total satisfaction score is obtained by summing all 70 item responses. The five dimensions of college student satisfaction which are outlined by Starr, Betz, and Menne (1971), can be summarized as follows: 1. Working Conditions assesses the student's physical environment such as the cleanliness and comfort of his living space, adequacy of study areas on campus., quality of meals, facilities for relaxing between classes. 2. Compensation judges the amount of input (e.g., study) required relative to academic outcomes (e.g., grades) and the effect of study requirements on the student's satisfaction of his other needs and goals 3 Quality of Education surveys the various academic conditions relative to the student's intellectual and career development such as the competence and helpfulness of faculty and staff, including advisors and counselors, and the adequacy of curriculum requirements, teaching, methods, assignments, etc,

PAGE 64

55 4. Social Life refers to the opportunities to satisfy socially relevant goals such as dating, meeting compatible or interesting people, making friends, participating in campus events and informal social activities. 5. Recognition assesses the acceptance of the student as a worthwhile individual by faculty and other students as determined by their attitudes and behaviors Reliability coefficients for each of the five CSSQ scales (Form C) are reported in the manual separately for public universities and private colleges. These reliabilities range in public universities from .78 (Quality of Education) to .84 (Compensation) with a median of .82. For private schools, reliabilities range from .79 (Quality of Education) to .84 (Recognition) with a median of .82. Total score reliability coefficients are .94 for both normative groups (Starr, Betz, and Menne, 1971). A more recent study (DeVore and Handal 1981) reported significant and uniformly high test-retest reliability coefficients. These reliabilities are: Working Conditions, .86; Compensation, .85; Quality of Education, .87; Social Life, .90; and Recognition, .84. Between scale correlations range from .46 to .70 (private colleges), the average being .50; for private universities, scale correlations range from ,35 to .62, with the average correlation being .44. Various studies have been conducted to investigate the validity of the CSSQ as

PAGE 65

56 a measure of college student satisfaction (Betz, Menne, Starr, and Klingensmith, 1971; Starr, Betz, and Menne, 1972; Betz, Klingensmith, and Menne, 1970). These studies basically have developed out of the conceptualization which initiated the development of the CSSQ (i.e., that student satisfaction can be viewed as being analogous to job satisfaction; therefore, findings from job satisfaction research should be applicable to studies of college student satisfaction) Job satisfaction research has consistently shown a negative correlation between job satisfaction and turnover (i.e., higher satisfaction is associated with less turnover, a greater likelihood that the worker will remain on the job, rather than quit) (Starr, et al. 1971). Previous research with the CSSQ has proven this instrument to be both a valid and reliable measure of college student satisfaction. Therefore, the use of this instrument will assist in the identification of the degree of college student satisfaction experienced by specially admitted students. The usefulness of the CSSQ as a measure of college student satisfaction was established by Betz, Klingensmith, and Menne (1970). The results indicated that the instrument is an internally consistent measure of several dimensions of college student satisfaction. DeVore and Handal (1981) have also established uniformly high test-retest reliability coefficients for each of the five scales of the CSSQ in a private university. vfl^^*tfi _vr iwa r.J .. cr^Pi. a y iiaB a
PAGE 66

57 Procedures A coordinated summer orientation program is planned for all incoming specially admitted students by the directors of PACT, Special Services, and other staff members at the University of Florida. This committee of selected personnel from the office of admissions, division of student affairs, student mental health unit of the infirmary, speech and hearing clinic, office of instructional resources, and staff members from both special admittance programs, begins meeting approximately six months preceding the students' arrival on campus to plan the orientation program. The names and addresses of these students are distributed to the directors of the special admission programs as soon as they are granted admission thus allowing immediate selection into one of the two programs. The directors then send letters of welcome and tentative schedules of the five-day orientation program along with an offer of assistance if problems should arise prior to their arrival on the campus. Once students arrive on campus they are met by their individual peer counselor who assists them with checking into their assigned dormitory areas. The main function of the peer counselor is to assist the student in making a smooth transition from high school to college life. These peer counselors have been trained during the previous term by the orientation staffs of both special admittance programs. Those students who are not planning to reside in campus

PAGE 67

58 housing are asked (via the letter of welcome) to "sign in" with the specific program which sent them the letter. The first formal meeting is used to answer questions about college life at the University of Florida, and also during that session information is given regarding financial aid, housing, registration, and other vital data necessary for a successful tenure at the University. Various faculty and staff members from within the University, as well as prominent community leaders are on hand to personally welcome these students and to provide additional information concerning their roles and the services they offer. During this first orientation session students are also administered various diagnostic tests and inventories. Although the format may vary from year to year, generally diagnostic tests of math, English, and reading are given along with some type of personality and/or self-concept inventory. These tests are spaced throughout the five-day orientation period. On the first day of the five-day summer orientation, all participants were administered the Tennessee Self-Concept Scale and the Myers-Briggs Type Indicator, prior to the beginning of the first session in a group setting. Both instruments were administered by two black female graduate students from the Counselor Education Department at the University of Florida who were assisted by peer counselors in the dissemination of materials. All students were then allowed to proceed with the continuation of the orientation program.

PAGE 68

59 A list of all specially admitted students who participated in the first session of the orientation and who completed the Tennessee Self Concept Scale and/or the MyersBriggs Type Indicator was compiled. This list was sent to the Office of Admissions where vital information regarding high school grade point averages and college entrance exam scores was requested. At the end of Fall Semester, 1981, the list also was submitted to the Registrar's Office along with a request for cumulative grade point averages for two terms (Summer, 1981, and Fall, 1981) at the University of Florida. During the third week of Spring Semester classes, each student in this study was contacted by his peer counselor and asked to complete the College Student Satisfaction Questionnaire. Analysis of Data This study focused on one dependent variable (college grade point average) and five independent variables (high school grade point averages, college entrance exam scores, self-concept scores, personality type scores, and college student satisfaction scores) as they concerned specially admitted black students at the University of Florida. The Pearson Product Moment Correlation was used to determine if a relationship existed between the dependent variable and each of the five independent variables, and further, to determine if a relationship existed among each of the five ^< w' M > >^ Havj W'^ iy BtrjBifia M3an ^ *ja i wj. i i < i ukJu

PAGE 69

60 independent variables. Separate statistical analyses were performed for both the total group (which included those with missing data) as well as for only those with complete data on all variables. A stepwise regression analysis was used to determine the best model for predicting college grade point average (that is, which variables in combination accounted for, or contributed most to individual variance). Frequency distribution tables were established for the Myers-Briggs Type Indicator, Limitations of the Study Several limitations of this study exist. One major limitation concerns the subjects in this study. Subjects were limited to specially admitted black freshman students at the University of Florida and did not include regularly admitted students, other ethnic groups or races, or students of differing academic levels. Therefore, the results are limited in generalizability that is, results obtained from this study are not generalizable to all students. An additional limitation concerning the subjects relates to the fact that not all students who were admitted through special admittance guidelines during Summer, 1981, participated in the first session of the orientation program due to late high school graduation. Therefore, all specially admitted students did not complete the Tennessee Self Concept Scale and the Myers-Briggs Type Indicator. This may have affected the results.

PAGE 70

61 A final limitation concerns the inability to collect data on all subjects for all variables used in this study. The College Student Satisfaction Questionnaire was given approximately six months following the giving of the other two instruments. All specially admitted students who completed the two initial instruments did not complete the CSSQ. Therefore, although initially 103 subjects were identified, only 67 could be included in the multiple regression analysis.

PAGE 71

CHAPTER FOUR RESULTS AND DISCUSSION Results The purpose of this study was to investigate the relationship between five independent variables (high school grade point averages, college entrance exam scores, personality type scores, self-concept scores, and college student satisfaction scores) and one dependent variable (college grade point average) in a sample of specially admitted black university freshmen. One hundred and three students participated in this study. Of that number, sixtyseven were identified as having complete data on all variables. The Myers-Briggs Type Indicator (MBTI) and the Tennessee Self Concept Scale (TSCS) were administered to these students during the 1981 Summer term. In addition, the College Student Satisfaction Questionnaire (CSSQ) was administered to these students during the 1982 Spring term. Information regarding each student's high school grade point average, college entrance exam score, and college grade point average were obtained from the Office of Admissions. Data analyses were conducted as outlined in Chapter Three 62

PAGE 72

63 HO-j^ : There is no significant relationship between college grade point averages and high school grade point averages of specially admitted black university freshman students. A stepwise regression analysis was used to test this hypothesis. The results in Table 1 indicate that there was no significant relationship between college grade point averages and high school grade point averages of specially admitted black university freshman students. In addition, the Pearson Product Moment Correlation was used to further test this hypothesis. The correlation coefficient was .05632. These results indicate that high school grade point averages were not accurate predictors of college grade point averages for the specially admitted black university freshman students in this study. Therefore, hypotheses one was not rejected. HO2: There is no significant relationship between college grade point averages and college entrance exam scores of specially admitted black university freshman students. A stepwise regression analysis was used to test this hypothesis. The results in Table 1 indicate that when SATTotal was entered as the second variable in the analysis, it accounted for an additional 8 percent increment in the explained variance in college grade point average. SAT-Math was initially entered as the ninth variable, but was replaced in the next step by the Recognition Scale of the CSSQ. SATMath was reentered to complete the best eleven variable model found. SATVerbal was not entered as one of the significant variables in the model for improvement in R In addition,

PAGE 73

64 the Pearson Product Moment Correlation was used to test this hypothesis. The correlation coefficients for SAT-Total, SATVerbal, and SAT-Math were reported as .27252, .16529, and .23262, respectively. These results indicate that college entrance exam scores were not accurate predictors of college grade point averages for specially admitted black university freshman students in this study. Therefore, hypothesis two was not rejected. HOo : There is no significant relationship between college grade point averages and self-concepts of specially admitted black university freshman students. A stepwise regression analysis was conducted on this hypothesis. The results in Table 1 indicate that several scales of the Tennessee Self Concept Scale contributed to variance in the prediction of college grade point average. However, the addition of these variables (Personal Self, Identity, Self Criticism and Behavior) only increases the prediction of college grade point average by relatively small amounts. In addition, the Pearson Product Moment Correlation was used to test this hypothesis. The correlation coefficients for Personal Self, Identity, Self Criticism, and Behavior were reported as -.28700, -.02440, -.02337, and -.28297, respectively. These results indicate that self-concept was not an accurate predictor of college grade point averages for the specially admitted black university freshman students in this study. Therefore, hypothesis three was not rejected.

PAGE 74

^ ^ 65 HO^: There is no significant relationship between college grade point averages and personality type preferences of specially admitted black university freshman students. A frequency distribution was obtained to determine the extent of the relationship described in this hypothesis. The results in Figure 2 indicate that there was no significant relationship between black university specially admitted students' college grade point averages and personality type preferences. In addition, the Pearson Product Moment Correlation was used to test this hypothesis. The correlation coefficient was reported as -0.03470, These findings indicate that personality type preferences were not accurate predictors of college grade point averages for the specially admitted black university freshman students in this study. Therefore, hypothesis four was not rejected. HO^: There is no significant relationship between college grade point averages and college student satisfaction of specially admitted black university freshman students. A stepwise regression analysis was used to test this hypothesis. The results in Table 1 indicate that three scales of the CSSQ (Working Conditions, Compensation, and Quality of Education) contributed to the variance in the prediction of college grade point average although the R^ increases were only .04, .01, and .01, respectively. In addition, the Pearson Product Moment Correlation was conducted to test this hypothesis. The correlation coefficients for the three CSSQ scales were reported as -0.23890, -0.06574, and 0.18766, respectively. These results indicate that ^g— TfJi S 'g fe r r j tyjsi tfi

PAGE 75

66 SENSING TYPES INTUITIVE TYPES with THINKING with FEELING with FEELING with THINKING iSTJ ISFJ INFJ INT J N= 32 N= 12 N= 3 N= 4 4_ C % =34.0 %= 12.8 %= 3.2 %= 4.3 o o X 2.40 2,35 1.96 2.21 z Sd .653 .98c .661 .815 o z ISTP iSFP INFP INTP < m — 1 N= 1 N= 6 N= 1 N= % = 1.1 % = 6.4 %= 1.1 %= m X 1.37. 2.55 3.74 r 1 m Sd .933 -4 i ESTP ESFP ENFP ENTP N= 6 N=l N= 1 N= 2 m % = 6.4 %= 1.1 %= 1.1 %= 2.1 n X 1.93 4.00 2.00 2.93 "0 Sd .685 .969 < m m X •H > ESTJ ESFJ ENFJ ENTJ < m N= 15 N= 6 N= 3 N= 1
PAGE 76

67 Table 1 2 Maximum R Improvements for GPA Prediction, Order of Variables Personal Self SAT-Total Identity Working Conditions Self Criticism Behavior MBTI Compensation SAT-Math Quality of Education SAT-Math Replaced by Recognition SAT-Math Family Self HSGPA Social Self Moral-Ethical Self N r2 F P I r2 ncrease 67 .08 5.84 .019 .08 67 .16 6.04 .004 .08 67 .21 5.53 .002 .05 67 .25 5.25 .001 .04 67 .27 4.57 .001 .02 67 .29 4.07 .001 .02 67 .31 3.72 .002 .02 67 .32 3.41 .003 .01 67 .33 3.10 .004 .01 67 .34 2.83 .007 .01 67 .34 2.91 .005 .00 67 .35 2.68 .008 .01 67 .35 2.46 .012 .00 67 .36 2.26 .019 .01 67 .36 2.07 .029 .00 67 .36 1.91 .045 .00

PAGE 77

68 college student satisfaction was not an accurate predictor of college grade point averages for the specially admitted black university freshman students in this study. Therefore, hypothesis five was not rejected. HOg : There is no significant relationship between personality type preferences and college student satisfaction of specially admitted black university freshmen. The Pearson Product Moment Correlation conducted in this hypothesis yielded the following coefficients: Working Conditions, 0.00273; Compensation, 0.01352; Quality of Education, 0.12152; Social Life, 0.13616; Recognition, 0,13877, and Total Satisfaction, 0.09940. In addition, a frequency distribution table was established and is presented in Figure 3 These results indicate that no significant relationship was found between personality type preferences and college student satisfaction of the specially admitted black university freshmen in this investigation. These reresults imply that personality type preferences were not accurate predictors of college student satisfaction for the specially admitted black university freshman students in this study. Therefore, hypothesis six was not rejected. HO-,: There is no significant relationship between self-concept and college student satisfaction of specially admitted black university freshman students. The Pearson Product Moment Correlation was used to test this hypothesis. The results in Table 2 indicate that selfconcept was not an accurate predictor of college student satisfaction for the specially admitted black freshman

PAGE 78

69 X Sd X Sd X Sd X Sd SENSING TYPES with THINKING with FEELING INTUITIVE TYPES with FEELING with THINKING 1ST J ISFJ INFJ INT J N = 26 % =37.7 219.19 34.579 N= 7 %= 10.1 234.71 30.379 N= 2 %= 2.9 191.50 65.76 N= 3 %= 4.3 221.00 17.52 ISTP ISFP INFP INTP N= %= N= 5 % = 7.2 220.00 28.439 N= 1 %= 1.4 206.0 N= %= 0.0 ESTP ESFP ENFP ENTP N= 5 •/'= 7.2 225.40 10.479 N= %= N= %=o N= 2 ''•=2.9 254.5 43.134 ESTJ ESFJ ENFJ ENTJ N= 11 %= 15.9 201.36 41.12 N= 5 % = 7.2 217.6 29.89 N= 1 %= 1.4 210.00 N= 1 %= 1.4 200.00 O < m TO —i v > < N = 69 Figure 3 Frequency Distribution Myers-Briggs Type Indicator By Total Satisfaction.

PAGE 79

70 (M +-> C O 03 +J C CD p CO CO c 01 as ^ e O J2 o CO CO u -p -P -H (D CC 03 0) CQ > 4J -H C C a in 00 CM CO CD t> o in CM 00 CM CI a 00 CO CO p -a CO 00 o I in 00 o o o I m oo CO o m in o o I o CO tH CD 1> O c o •H P o

as s: Q) CQ 00 CO CM in rH O CM in o CM CM O o CO o CM 00 C3 C CO o p •H X2 as •H ;-l OS > OS r-l CO o CO 00 CO CO CO in CM in o CM CO o o o Eh 05 r-t o 00 CO 00 05 o p X3 aS •H U a > o 00 in r-i CO O O I [> 00 in o o I t> in in o CO CM in o 00 05 CD O O O I o I p aS •H U as > O OS CD

PAGE 80

71 students in this study. Therefore, hypothesis seven was not rejected. Discussion The results of this study indicate that academic variables such as high school grade point averages and college entrance exam scores were not significantly related to the college grade point averages of specially admitted black university freshman students. These results support the findings of Dalton, Anastasiow, and Brigman (1977); Boyd (1977); and Rovezzi-Carroll and Thompson (1980), who reported no significant relationship between academic predictors and future college academic success. The results of the present study also support the findings of Houston (1980) who reported a small coefficient of multiple deter2 mination (R ) when high school rank, SAT-V, and SAT-M were entered in a stepwise regression analysis for similar subjects. These results do, however, contradict previous findings in the literature that show a significant relationship between academic predictors and college grade point averages (Bowers, 1970; Lanning, 1977; and Price and Kim, 1976). The finding that previous academic achievement in this study did not accurately predict college grade point average was not surprising. As evidenced by Stewart (1981), a substatially high percentage of specially admitted students do perform well academically at institutions of higher education.

PAGE 81

72 One reason for this may lie in the support and encouragement that the student receives from significant others. The knowledge that family and friends believe in the student's ability to succeed may cause the student to put forth that extra effort which may mean the difference between failure and success. Another explanation for these results could be the change in the student's view of education upon arrival on the college campus. High school grades and academic achievement tests may not have been high on the priority list of some of these students due to the lack of interest in subject matter, motivation and encouragement. However, once the student is exposed to a new and educationally stimulating environment, the motivation to succeed may heighten and college graduation becomes one of the ultimate goals in life. The self-concepts of the specially admitted black university freshman students in this study had no significant impact on their subsequent college grade point averages, that is, the differences between the self-concept scores of those students with high grade point averages and those with low grade point averages were not significantly different. These findings are consistent with those of Dowdle (1977) and Carey (1976) who reported that those students who performed low academically did not have a lower selfconcept than those who performed high academically. One reason for these findings, may be that the self-concept scores were based on previous personal experiences from high school.

PAGE 82

73 For some students those experiences may have had a positive effect, while for others, quite the opposite may have been the case. However, those students with low self-concepts may strive harder to achieve than those with high selfconcepts. Thus, the fact that they may not be as confident in themselves as others may motivate them to study more intensely. This may then allow them to achieve comparable, if not better, grades than those who scored high on the self-concept measure. No significant relationship was found between personality type preferences and college grade point averages. The majority (79 or 84 percent) of the students were identified as Sensing types. A total of only 15 or 16 percent were on the Intuitive side of the continuum. These results support McCaulley (1972 and 1974) who reported very similar patterns. The fact that no statistically significant relationship was found supports Leon (1974) and Johnson (1970), who also reported no significant difference in regards to grade point averages. One explanation for these results may stem from the fact that the personality type scores were based on previous individual experiences from high school Regardless of individual personality type, the key to success depended on the flexibility of that person in being able to adjust academically or socially to the college environment. Those who possessed those skills prior to admittance were not that different from those who acquired the necessary coping skills through daily campus encounters.

PAGE 83

74 The degree of satisfaction with the university environment was not significantly related to the college grade point averages of the students studied. The results of this study somewhat support previous findings in the literature that show a high degree of satisfaction with college by freshmen students (Schmidt and Sedlacek, 1972; and Madrazo-Peterson and Rodriguez, 1978). When compared to the results reported by Lindsey (1981), the means of all subscales of the CSSQ for the freshman group in the present study were slightly higher than those reported by Lindsey for all black students. One reason for these findings may be the tendency for freshman students to be highly optimistic about their first year in college. Therefore, many positive aspects may be overexaggerated while negative ones may be subconsciously overlooked. Neither personality type preferences nor self-concepts were identified as being significantly related to college student satisfaction in this study. Even though freshmen initially enter the university environment with individual differences in regards to personality type preferences and self-concepts, the majority of them are moderately satisfied with the circumstances they find during their first year. One explanation for these results is that perceptual and self-concept changes are most evident during the freshman year in college. Therefore after approximately six months in the college environment, initial priorities may no longer seem important that is what students expect as a one-dayold freshmen may be quite the opposite of what they experience

PAGE 84

75 six months later. However, because the novelty of the freshman year experience has not worn off, the student may still be completely satisfied.

PAGE 85

CHAPTER FIVE SUMMARY, CONCLUSIONS, IMPLICATIONS, AND RECOMMENDATIONS FOR FURTHER RESEARCH Summary The relationship between academic and nonacademic variables and college grade point average has been the source of much recent research. The establishment of a predictive model by the use of multiple factors, however, has not been extensively explored and determined. The purpose of the present study was to investigate the relationship between five independent variables (high school grade point averages, college entrance exam scores, personality type, self-concept, and college student satisfaction) and one dependent variable (college grade point average) for specially admitted black students at one institution. This study also determined the relationship between college student satisfaction and two other variables, personality type and self-concept. Chapter One dealt with the statement of the problem, purpose of the study, need for the study, definition of terms, and organization of the remainder of the study. The literature related to college student attrition was reviewed in Chapter Two. The sections contained in Chapter Two were academic factors related to college attrition and nonacademic factors related 76

PAGE 86

77 to college attrition, including specific studies dealing with self-concept personality characteristics and college student satisfaction. The method of research used in this study was described in Chapter Three. A description of the hypotheses, population and sample, instruments, procedures, analyses of data, and limitations of the study were included in this chapter The results of the study were reported in Chapter Four, followed by a discussion of those results. The findings of this study indicated that academic and nonacademic variables were not significantly related to the college grade point averages of specially admitted black university freshman students. The results also indicated no significant relationship between college student satisfaction and selfconcept of personality type preferences. Conclusions The following conclusions were made based on the results of this study: 1, High school grade point averages of specially admitted black university freshman students do not accurately predict college grade point averages of these students. 2. College entrance exam scores of specially admitted black university freshman students do not accurately predict college grade point averages of these students.

PAGE 87

78 3. Self-concepts of specially admitted black universityfreshman students are not accurate predictors of college grade point averages for these students. 4. Personality type preferences are not accurate predictors of the college grade point averages of specially admitted black university freshman students. 5. The degree of satisfaction with their university environment does not affect the college grade point averages of specially admitted black university freshman students. 6. The degree of satisfaction with their university environment is not determined by personality type preference of specially admitted black university freshman students. 7. The degree of satisfaction with their university environment is not determined by the self-concept of specially admitted black university freshman students. Implications Implications of this study are demonstrated in both the selection and retention of specially admitted black university college students. So far as the selection of students is concerned, the results of this study imply that selection criteria based on academic assessments are not accurate as predictors of college grade point averages for these students That is, neither of the academic factors (high school grade point averages or college entrance exam scores) were related

PAGE 88

79 to college grade point average. Therefore, the study seems to indicate that selection of specially admitted black students should not be based solely upon academic variables. University admissions committees should continue to consider this, especially with the trend toward increasing admissions requirements The results of this study indicate that nonacademic data are somewhat related to college grade point average. Therefore, there is a need for additional information regarding the relationship between nonacademic variables and future academic success. The continued search for nonacademic baseline data is essential if research efforts are to be expanded in this area of concern. Student personnel workers should be encouraged to continue to conduct research in an attempt to identify possible contributing factors to college success. Student personnel workers should be trained to identify problem areas of the college environment for specially admitted black students. These workers should then work with administrators in an attempt to plan and implement programs and activities to help maintain satisfaction among these students. The results of this study did not identify one specific variable that predicted college academic success. Therefore, it would seem that college success may be the result of an interaction between various factors. Student personnel workers should be highly skilled in the art of interpreting the college environment and the effects that certain factors may have on the success of the student.

PAGE 89

80 Perhaps another implication of this research can be drawn from the finding that the students of this study (freshmen) were moderately satisfied with their college environment An attempt should be made to determine the causes of dissatisfaction which according to the literature often seem to occur following the freshman year. There is a need for increased retention efforts following the freshman year. There are increased pressures placed on sophomore, junior, and senior level students to make career decisions and to perform well academically in order to attain career goals. However, although course difficulty increases, there is a decrease in the use by these students of the resources available for tutoring and academic advisement. Student personnel workers should encourage continued use of these services by students throughout all levels of achievement. Recommendations for Further Research Based on the results of this study, several recommendations for further research seem warranted. The following studies therefore, are suggested for use in conducting research with specially admitted black university freshman students : 1. A four-year longitudinal attrition study using initial freshman academic and nonacademic baseline data should

PAGE 90

81 be conducted. A study of this type would establish a profile of those students most likely to succeed in the college environment. A correlational study between previous academic factors and college grade point averages of currently enrolled students by academic classification should be conducted. This study would assist in the identification of the relationship between academic factors and college grade point averages of students as they matriculate through college and would establish predictive validity. A correlational study between college student satisfaction and college grade point averages of enrolled students by academic classification should be undertaken. This type study would indicate the effects of student satisfaction on college grade point average at all academic classification levels and would help to determine if students at different levels are affected differently by different aspects of their environment A college student satisfaction follow-up study should be conducted with former students to determine the relationship between college student satisfaction and attrition. The results of this study would identify problem areas in the college environment. These findings also would assist student personnel workers in planning and implementing programs for future students.

PAGE 91

APPENDIX A INFORMED CONSENT FORM Betty J. Stewart, Principal Investigator 441 Little Hall University of Florida The purpose of this study is to determine if there is a relationship between academic factors (high school grade point averages and college entrance exam scores) and nonacademic factors (personality "types," self-concept, and college student satisfaction) which can be used to predict college grade point averages In order to research the possibility of a predictive relationship, you will be asked to complete the following instruments: 1) the Tennessee Self-Concept Scale; 2) the Myers-Briggs Type Indicator; and 3) the College Student Satisfaction Questionnaire. There are no discomforts or risks to be expected from the procedures mentioned above. On the contrary, the outcome of this study could benefit you individually by revealing how you feel about yourself (self-concept) your likes and dislikes (personality) and how satisfied you are with your college environment. The results also will assist others by increasing further knowledge in this area. Your individual "scores" will be kept confidential and will be available only to you (the subject) and me (the principal investigator). Each subject will be numerically coded prior to the recording of scores and grade point averages. At the end of this study these coded sheets will be destroyed so that information regarding individuals can be kept confidential. Group results will be available through the published dissertation in the University Libraries. It is hoped that this alone will inspire you to participate in this study because I am unable to offer any type of monetary reward for your services. If you have any questions about this study and the procedures which are being followed, I will be happy to answer them. You are free to withdraw your consent to participate in this project at any time without being penalized. "I have read and understand the procedures described above, I agree to participate in the procedures and I have received a copy of this description." Subject Date Witness Date Relationship if other than subject Date 82

PAGE 92

APPENDIX B MEAJSTS AND STANDARD DEVIATION ON ALL VARIABLES FOR SPECIALLY ADMITTED BLACK UNIVERSITY STUDENTS Variable N Mean SD College GPA 67 2.4197 0.7372 HSGPA 67 2.6118 0.3669 SAT-Total 67 666.7164 78.4198 SAT-Verbal 67 319.1791 54.4782 SAT-Math 67 347.5373 53.1605 Self-Criticism 67 31.0597 5.2365 Total Positive 67 352.5075 27.0999 Identity 67 130.8806 10.2935 Self -Stat isf act ion 67 108.0896 12.7395 Behavior 67 113.5373 11.8185 Physical Self 67 72.9552 7.2476 Moral-Ethical Self 67 70,1940 6.3776 Personal Self 67 68.8955 6.5508 Family Self 67 71.2836 8,0638 Social Self 67 69.3284 7.1761 Total Variance 67 52.8806 11.9374 Column Variance 67 32.5373 9.2742 Row Variance 67 20.8657 7.3976 Working Conditions 67 41.6716 8.0761 Compensation 67 43.0149 7.0032 Quality of Education 67 44.4627 7.5645 Social Life "67 45.5522 8.6134 Recognition 67 43.0299 8.3937 Total Satisfaction 67 217.7313 33,9374 83

PAGE 93

REFERENCES Aleamoni, L.M. Can grade point average be more accurately predicted? Journal of Educational Psycho logy. 1977 69, 227-230. ~~ Asfacraft, C, and Fitts, V7.H. Self concept change in psychotherapy. Psychotherapy 1964, 1_ 115-118. Astin, A.W. Personal and environmental factors associated with college dropouts among high aptitude students. Journal of Educational Psychology 1964, 55, 219-227. Astin, A.W. Four Critical Years San Francisco: JosseyBass Publishers, 1977. Beal, P.E. Student retention: A case study of an action approach. National Association of Student Personnel Administrators Journal 1979, 17. 9-16^ Beal, P.E., and Noel, L. What Works in Student Retention Iowa City: The American College Testing Program and the National Center for Higher Education, 1979. Bean, A.G., and Covert, R.W, Prediction of college persistence, withdrawal, and academic dismissal: A discriminate analysis. Educational and Psychological Measurement 1973, 33, 407-411. Berdie, R.F. The prediction of college achievement and satisfaction. Journal of Applied Psychology 1944, 28, 239-245. ~~ Betz, E.L., Klingensmith, J.E., and Menne J.W. The measurement and analysis of college student satisfaction. Measurement and Evaluation in Guidance 1970, 3, 110-118, Betz, E.L., Menne, J.W., Starr, A.M., and Klingensmith, J.E. A dimensional analysis of college student satisfaction. Measurement and Evaluation in Guidance 1971, 4, 99-106. Betz, E.L,, Starr, A.M., and Menne, J.W. College student satisfaction in ten public and private colleges and universities. Journal of College Student Personnel 1972, 13, 456-46r; — 84

PAGE 94

85 Bowers, J. The comparison of gpa regression equation for regularly admitted and disadvantaged freshmen at the University of Illinois. Journal of Educational Measurement 1970, 7, 219-225. Boyd, W.M. SAT's and minorities: The dangers of underprediction. Change 1977, 9, 48-49. Brown, N. Jr. A description research study of a developmental plan for recruitment and retention of minority students (Doctoral dissertation, Wayne State University, 1976). ERIC Document Reproduction Service No. ED 146 254 Carey, P, The relationship between black studies, selfconcept, and academic performance of black students of white campuses in the southwest (Doctoral dissertation, Oklahoma State University, 1975). Dissertation Abstracts International 1976, 37, 636A. (University Microfilms No. 76-9639) Carney, M,, and Geis, L. Reading ability, academic performance and college attrition. Journal of College Student Personnel 1981, 22, 55-59, Carr, D. and Chittum, C. A study to identify non-academic factors which may positively influence the recruitment and retention of "other race" students at Virginia's state-supported institutions of higher education (Final Report, Norfolk, 1980). ERIC Document Reproduction Service No. ED 187 166 Cleary, T.A, Test bias: Prediction of grades of Negro and white students in integrated colleges. Journal of Educational Measurement 1968, 5, 115-124. Cole, J. The relationship of the freshman special services program at Southern University to academic achievement and self concept. (Doctoral dissertation. Southern University, 1974). Dissertation Abstracts International 1974, 35, 2567A-2568A. (University Microfilms No. 74-24, 767) Cole, N. and Hanson, G. Racial-ethnic bias in selective college admissions. Paper presented at American Educational Research Association Annual Meeting New Orleans, Louisiana, 1973. Copeland, E.J. Peer counselor effects on the self concept and academic adjustment of educational opportunity students who participated in group counseling. (Doctoral dissertation, Oregon State University, 1974). Dissertation Abstracts International 1974, 3^, 1974A. (University Microfilms No. 74-23, 431)

PAGE 95

86 Cranney, A.G., and Larsen, J. Compensatory programs for specially admitted freshmen to the University of Florida-1968-1971. National Reading Conference 21st Yearbook Milwaukee: The Conference, 1972, 38-41. Crawford, A.E., McFarland, D.E., and Rhatigan, J.J. Special counseling programs for academic survival. Journal of College Student Pers onnel. 1978, 19, 298-302. ^ — Dalton, S., Anastasiow, M. andBrigman, S.L. The relationship of underachievement and college attrition. Journal of College Student Personnel 1977, 18, 501-505. Dawkins, M.P., and Dawkins, R.L. Perceptions and experiences as correlates of academic performance among blacks at a predominately white university: A Research note. College and University 1980, 55, 171-180. DeVore, J.R., and Handal P.J. The College Student Satisfaction Questionnaire: A test-retest reliability study. Journal of College S tudent Personnel, 1981, 22, 299-301. — DiCesare, A.C., Sedlacek, W.E., and Brooks, G.C., Jr. Nonintellectual correlates of black student attrition. Journal of College Student Personnel 1972, 1^, 319-324. Dowdle, S.L. An analysis of the relationship between the self concept and academic performance of high and low performers in a collegiate school of business. (Doctoral dissertation, The University of Arizona, 1977). Dissertation Abstracts International 1977, 38, 1222A1223A. (University Microfilms No. 77-18,570) Eckland, B., and Wisenbaker, J. A Capsule Description of Young Adults Four and One-Half Years After High School Prepared for the National Center for Educational Statistics. Research Triangle Park, North Carolina: Research Triangle Institute, 1979. Egerton, J. State Universities and Black Americans Atlanta: Southern Education Reporting Service, 1969. Ellison, R.L., Murray, S.L., Fox, D.G., and Taylor, C.W. Biographical data as predictors of college grades of Negroes and whites (Salt -Lake City, Utah, Institute for Behavioral Research in Creativity, 1973). KRIC Document Reproduction Service No. ED 081 358 Fitts, W.H. Manual: Tennessee Self Concept Scale Nashville, Tennessee: Counselor Recordings and Tests, 1965.

PAGE 96

87 Friedlander, J. Are College support programs and services reaching high-risk students? Journal of College Student Personnel 1980, 21^, 23-28. Gadzella, B.M., and Fournet G.P. Differences between high and low achievers on self-perceptions. Journal of Experimental Education 1976, 44_, 44-48. Garber, H., and Schell, R. Differences in rates and modes of attrition between three groups of Oswego undergraduate students. Research report Oswego, New York: State University of New York, 1977. Gibbs, J.T. Black students/white university: Different expectations. Personnel and Guidance Journal 1973, 51, 463-469. ~" Gividen, G.M. Stress in airborne training as related to the self-concept, motivation and biographical factors. Unpublished master's thesis, Vanderbilt University, 1959. Goodrich A. Report on the Academic Status of Minority Students at the University of Maryland, College Park, for the Academic Year 1975-76 College Park, Maryland: Office of Minority Student Education, University of Maryland, 1976. Goodrich, A., and Shuford, B. A Retention/Intervention Model for Impacting on Minority Student Attrition at Predominantly White Colleges and Universities College Park, Maryland: Office of Minority Student Education, University of Maryland, 1976. Goodrich, A. Manual: A data-driven retention model for improving minority study persistence in higher education institutions University of Illinois at Chicago Circle. Chicago, Illinois, 1979. Green, R.L., and Farquhar, W.W. Negro academic motivation and scholastic achievement. Journal of Educational Psychology 1965, 56, 241-243. Haettenschwiller, D.L. Counseling black students in special programs. Personnel and Guidance Journal 1971, 50_, 29-35. Hallenbeck, T.R. College student satisfaction: An indication of institutional vitality. National Association of Student Personnel Administrators Journal 1978, 16, 19-25. Hannah, W. Personality differentials between lower division dropouts and stayins. Journal of College Student Personnel 1971, 12, 16-19. Hart, D. and Keller, M.J. Self-reported reasons for poor academic performance of first-term freshmen. Journal of College Student Personnel, 1980, 21, 529-534.

PAGE 97

88 Hathaway, S.R., and McKinley, J.C. Revised Manual: Minnesota Multiphasic Personality Inventory New York: The Psychological Corp., 1951. Herzberg, F. Mausner, B., Peterson, R.O. and Capwell, D.F. Job attitudes: Review of Research and Opinion Pittsburgh, Pa,: Psychological Services, 1957. Houston, L.N. Predicting academic achievement among specially admitted Black female college students. Educational and Psychological Measurement 1980, 40, 1189-1194. ~~ ~ — Humphries, F. Thirteen College Curriculum Program Progress Report, 1967-1972: A Ma.jor Curriculum Effort to Reduce Attrition Among Black College Students Washington, D.C.: Institute for Services to Education, 1972. Ikenberry, S.O. Factors in college persistence. Journal of Counseling Psychology 1961, 8, 322-329. Johnson, D. Personality characteristics in relation to college persistence. Journal of Counseling Psychology 1970, 17, 162-167. ~ Jones, F.D. The effects of personal growth group counseling on the self-concepts and academic achievement of college freshmen. (Doctoral dissertation, Lehigh University, 1975). Dissertation Abstracts International 1975, 36, 2631A-2632A. (University Microfilms No. 75-23,994) Jones, J.C, Harris, L. J. and Hauck, W.E. Differences in perceived success of academic difficulties; Black students in predominately black and predominately white colleges. Journal of Negro Education 1975, 44, 519-529. Kallingal, A. The prediction of grades for black and white students at Michigan State University. Journal of Educational Measurement 1971, 8^, 263-265. King, H. and Walsh, W.B. Change in environmental expectation and perceptions. Journal of College Student Personnel 1972, 1^, 331-336. Lanning, W. Factors related to college student persistence and withdrawal. National Association of Student Personnel Administrators Journal 1977, 15, 34-38. Leon, G.R. Personality change in the specially admitted disadvantaged student after one year in college. Journal of Clinical Psychology, 1974, 30, 522-528.

PAGE 98

89 Lindsey, M,D. Satisfaction and university involvement among black and white undergraduate students. (Doctoral dissertation, University of Florida, 1981). Dissertation Abstracts International 1982 42 University Microfilms No. 8203,689). Lokitz, B.D., and Sprandel, H.Z. The first year: A look at the freshman experience. Journal of College Student Personnel 1976, 17, 274-279. Madrazo-Peterson, R. and Rodriguez, M. Minority students' perceptions of a university environment. Journal of College Student Personnel 1978, 19, 259-263. Martin, R. Freshman satisfaction with college. Journal of College Student Personnel 1968, 9, 382-383. Maudal, G.R., Butcher, J.N., and Mauger, P. A. A multivariate study of personality and academic factors in college attrition. Journal of Counseling Psychology 1974, 21, 560-567. Mayhew, L.B. Non-test predictors of academic achievement. Educational and Psychological Measurement 1965, 25, 39-46. ~ Maynard, M. Can universities adapt to ethnic minority students' needs? Journal of College Student Personnel 1980, 21, 398-401. McCaulley, M.H. Type table for EEOP freshmen entering summer 1972. The Freshman Studies Center for Applications of Psychological Type, Inc. Unpublished Data, 1972. McCaulley, M.H. Type table for university students in a Special Services Program, The Freshman Studies Center for Applications of Psychological Type, Inc. Unpublished Data, 1974. McDavis R.J. Mingo, G.W. Stewart, B.J., and Hough, T.A. A summer orientation program for students in special services programs: A coordinated approach. Journal of College Student Personnel 1980, 21, 88-89. Menning, A.J., Bradley, L.J., and Cochran, J. Attrition. College and University 1975, 50, 753-756. Merton, R.K. Social theory and social structure (rev. ed. ) New York: Free Press, 1957,

PAGE 99

90 Millott, R.F. Reading performance as a correlate of the personality type of college freshmen. (Doctoral dissertation, University of Florida, 1974). Dissertation Abstracts International 1974, 36, 712A. (University Microfilm No. 75-16,420). Mornell, E. The program of special directed studies: A five-year summary. Claremont University, California, 1973. ( ERIC Document Reproduction Service No. ED 088 638). Morrisey, R.J. Attrition in probationary freshmen. Journal of College Student Personnel 1971, 12, 279-285. Morrison, T.L. Thomas, M. and V/eaver S.J. Self-esteem and self-estimates of academic performance. Journal of Consulting and Clinical Psychology 1973, 41, 412-415. Morstain, B.R. An analysis of students' satisfaction with their academic program. Journal of Higher Education 1977, 48, 1-15. Muskat, H. Educational expectations and college attrition. National Association of Student Personnel Administrators Journal 1979, 17, 17-22. Myers, I. Manual: The Myers-Brlggs Type Indicator Princeton, New Jersey: Educational Testing Service, 1962. National Advisory Committee on Black Higher Education and Black Colleges and Universities. Access of Black Americans to Higher Education: How Open is the Door ? Washington, D.C.: Department of Health, Education, and Welfare, 1979. Pandey R.E. Personality characteristics of successful, dropout, and probationary black and white university students. Journal of Counseling Psychology 1972, 5, 382-386. Pant ages, T. J. and Creedon, C.F. Studies of college attrition: 1950-1975. Review of Educational Research 1978, 48, 49-101. Pazandak, C.H. Self-estimate of level .of effort as a predictor of college grades. Measurement and Evaluation in Guidance 1975, 8, 43-50. Pervin, L.A. Satisfaction and perceived self -environment similarity. Journal of Personality, 1967, 35, 623-634.

PAGE 100

91 Pfeifer, CM., Jr., and Sedlacek, W.E. The validity of academic predictors for black and white students at predominately white universities. Journal of Educational Measurement 1971, 8, 253-261. Pfeifer, CM., Jr., and Sedlacek, W.E. Predicting black student grades with non-intellectual measures. Journal of Negro Education 1974, 43, 67-76. Pfeifer, CM. Relationship between scholastic aptitude, perception of university climate, and college success for black and white students. Journal of Applied Psychology 1976, 61, 341-347. j! Price, F., and Kim, S. The association of college perform' ance with high school grades and college entrance test scores. Educational and Psychological Measurement 1976, 36, 965-970. Robertson, J. A. Black student satisfaction in the deep i south. Journal of College Student Personnel ,1980, 21, '1 510-513. ~ Rossmann, J.E., and Kirk, B.A. Factors related to persistence and withdrawal among university students. Journal of Counseling Psychology 1970, r?, 56-62. Rovezzi-Carroll, S., and Thompson, D. Forecasting college success for low-income students. Journal of College Student Personnel 1980, 2_1, 340-343. I Samuel, N. and Laird, D.S. The self concepts of two groups of black female college students. Journal of Negro Education, 1974, 43, 228-233. Schmidt, D.K., and Sedlacek, W.E. Variables related to university student satisfaction. Journal of College Student Personnel 1972, 13^, 233-237. I Schroeder, CC, V/arner, R. and Malone D.R. Effects of assignment to living units by personality types on environmental perceptions and student development. Journal of College Student Personnel 1980, 21, 243-249. Scott, J.E. An investigation of black and white student expectations and perceptions of a predominately white public institution. (Doctoral dissertation, Eastern Michigan University, 1978). Dissertation Abstracts International 1978, 39, 3410A. (University Microfilms No, 7823004) Sedlacek, W,E., and Brgoks ,. G.C. Jr. Black freshmen in large colleges: A survey. Personnel and Guidance Journal, 1970, 49, 307-312.

PAGE 101

92 Sedlacek, W.E., Brooks, G.C., Jr., and Horowitz, J.L. Black admissions to large universities: Are things changing? Journal of College Student Personnel 1972, 13, 305-310. Sedlacek, W.E. Brooks, G.C., Jr., and Mindus L.A. Black and other minority admissions to large universities: Three-year national trends. Journal of College Student Personnel 1973, 14, 16-21. Sedlacek, W.E., Lewis, J. A., and Brooks, G.C., Jr. Black and other minority admissions to large universities: A four-year national survey of policies and outcomes. Research in Higher Education 1974, 2, 221-230. Sedlacek, V/. and Webster, D. Admission and retention of minority students in large universities. Journal of College Student Personnel 1978, 19, 242-248. Stanley, J.C. Predicting college success of the educationally disadvantaged. Science 1971, 171 640-647. Starr, A.M., Betz, E.L., and Menne, J.W. Manual : The College Student Satisfaction Questionnaire Ames, Iowa: Central Iowa Associates, 1971. Starr, A., Betz, E.L., and Menne, J. Differences in college student satisfaction: Academic dropouts, non-academic dropouts and non dropouts Journal of Counceling Psychology 1972, 19, 318-322. Stewart, B.J. Attrition, retention, and graduation of "special admit" students at the University of Florida. Journal of the Society of Ethnic and Special Studies 1981, 5, 30-35. Sturtz, S.A. Age differences in college student satisfaction, Journal of College Student Personnel 1971, 12, 220-222. Sundby, E.S. A study of personality and social variables related to conformity behavior. (Doctoral dissertation, Vanderbilt University, 1962). Dissertation Abstracts International 1963, 23, 3470. (University Microfilms No. 63-1848) Temp, G. Validity of the SAT for blacks and white in thirteen integrated institutions. Journal of Educational Measurement 1971, 8, 245-251. Thomas, C.L., and Stanley, J.C. Effectiveness of high school grades for predicting college grades of Black students: A review and discussion. Journal of Educational Measurement, 1969, 6, 203-215.

PAGE 102

93 Tillman, C.E. Millott, R.F., and Larsen J.J. Using personality test data in individualized reading instruction: Brief reports of five studies of personality and reading. National Reading Conference 23rd Yearbook Milwaukee: The Conference, 1974, 125-130. Tinto, V. Dropout from higher education: A theoretical synthesis of recent research. Review of Educational Research 1975, 45, 89-125. Valine, W.J. Four-year follow-up study of underachieving college freshmen. Journal of College Student Personnel 1976, 17, 309-312. Webster, D.W., Sedlacek, \i and Miyores, J. A comparison of problems perceived by minority and white university students. Journal of College Student Personnel 1979, 20, 165-170. Weiss, D.J., Dawis, R.V. England, G.W., and Lof quist L.H. Manual for the Minnesota Satisfaction Questionnaire. Minnesota Studies in Vocational Rehabilitation 1967, 22. Wessel, T.R. Jr., Engle, K. and Smidchens, U. Reducing attrition on the college campus. National Association of Student Personnel Administrators Journal 1978, 16, 26-32. White, A.J. and Suddick, D.E. Five years after matriculation: A comparison of grades and attrition of black and white college students. Journal of College Student Personnel, 1981, 22, 177.

PAGE 103

BIOGRAPHICAL SKETCH Betty Jean Stewart was born in Gainesville, Florida, on February 10, 1950, to James and Dorothy Stewart. She attended public school in Alachua County and received many awards upon graduation including valedictorian of the senior class. She began her undergraduate career at Bethume-Cookman College in Daytona Beach, Florida, and later transferred to the University of Florida in Gainesville, Florida, and received a B.A. in psychology. She continued her education at the University of Florida and received a Master of Arts degree in 1974. In August 1974, she was hired as the Counselor Coordinator for Special Services/Upward Bound at the University of Florida. She was accepted into the doctoral program in the Department of Counselor Education at the University of Florida in January 1978. In July 1979, she was hired as the Director of the Program for Academic Counseling and Tutoring (PACT) at the University of Florida. Betty became a candidate for the Doctor of Philosophy program in December 1980 94

PAGE 104

I certify that T u sepFwHtmiF ssor of ^airma. .n I =tify that I h J^ "" Counselor Education a ..s.e„atlo„1o-\\^ ---, teiS-"ierick Associate Professor of Counselor Education J. certify that r u P?S„?aj---™| LF--\^-1.-^a.. that 1. ., a .i3se„atlo.1o-\\^ --J. -.jS^^rStV^ "^ ot-g2h T 1 osophy Professor of Psychology May 1982