• TABLE OF CONTENTS
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Review of the literature
 Results of the study
 Dsicussion of the data
 Findings, conclusions, and...
 Appendix
 Bibliography
 Biographical sketch
 Copyright














Group Title: comparison of self-concepts, peer relationships, persistence and extracurricular involvement of University of Florida freshmen with differing housing
Title: A comparison of self-concepts, peer relationships, persistence and extracurricular involvement of University of Florida freshmen with differing housing
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00089542/00001
 Material Information
Title: A comparison of self-concepts, peer relationships, persistence and extracurricular involvement of University of Florida freshmen with differing housing
Series Title: A comparison of self-concepts, peer relationships, persistence and extracurricular involvement of University of Florida freshmen with differing housing
Physical Description: Book
Language: English
Creator: Haulman, Stephen Russell
Publisher: Stephen Russell Haulman
Publication Date: 1978
 Record Information
Bibliographic ID: UF00089542
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 04522001
alephbibnum - 000071299

Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
    List of Tables
        Page vii
        Page viii
    List of Figures
        Page ix
    Abstract
        Page x
        Page xi
        Page xii
        Page xiii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Review of the literature
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
    Results of the study
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
    Dsicussion of the data
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
    Findings, conclusions, and implications
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
    Appendix
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
    Bibliography
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
    Biographical sketch
        Page 131
        Page 132
        Page 133
        Page 134
    Copyright
        Copyright
Full Text










A COMPARISON OF SELF-CONCEPTS, PEER RELATIONSHIPS, PERSISTENCE
AND EXTRACURRICULAR INVOLVEMENT OF UNIVERSITY OF FLORIDA FRESHMEN
WITH DIFFERING HOUSING ARRANGEMENTS


















By

Stephen Russell Haulman


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

UNIVERSITY OF FLORIDA


1978















This dissertation is dedicated with deep appreciation

to my parents, Colonel and Mrs. Clement Raymer Haulman;

my fiancee, Janet Minturn; and my son, Richard Scott

Haulman; who have supported me with their encouragement,

understanding and love during the most critical period of

my graduate studies.













ACKNOWLEDGEMENTS


The writer wishes to express his appreciation to

Dr. Ralph Kimbrough for his guidance as Chairman of the

writer's Supervisory Committee and particularly for his

suggestions and encouragement throughout the preparation

of this dissertation. The writer also wishes to express his

gratitude to Dr. Harold Riker for the inspiring leadership

and guidance given during the writer's graduate studies

and early years as a professional. Dr. Riker's thoughtful

and constructive comments throughout the preparation of this

dissertation are very much appreciated. The writer also

wishes to thank Dr. James Wattenbarger for the encouragement

he has given throughout the writer's doctoral studies.

Special recognition and appreciation is warmly extended

to Mrs. Rebecca Lovely for her outstanding work in typing

the drafts and final copy of this dissertation. Her patience,

throughtfulness and continued willingness to help throughout

the months of preparation of this dissertation are very much

appreciated. The writer also wishes to thank Dr. Michael

Nunnery for his guidance during the formative stages of this

dissertation. Special appreciation is extended to Mike Conlon

for his assistance with computerized statistical analyses and

for his patience in teaching the writer about certain methods

of statistical analysis.














TABLE OF CONTENTS


page
ACKNOWLEDGEMENTS . . . . . . . . . iii

LIST OF TABLES ...... . . . . . . vii

LIST OF FIGURES. . . . ... . . . . ix

ABSTRACT . . . . . . . . .. . . x

CHAPTER

I. INTRODUCTION . .... . . . . . . 1

Background and Justification for
the Study. . . . . . . . 1

Statement of the Problem. . . . .. 2

Significance of the Study . . . . 3

Scope of the Study. . . . . . . 6

Limitations of the Study. . . . . 8

Assumptions ...... .. . . . . . 8

Definition of Terms . . . . . 9

Procedures. .... . . . . 11

Description of the Sample. .. . 11

Design of the Study. . . . . .. 14

Collection of Data . . . . .. 14

Instruments ..... .... . . . 15

Analysis of the Data . . . . .. 21

Organization of the Research Report . .. 22









TABLES OF CONTENTS (CONT'D)


CHAPTER

II. REVIEW OF THE LITERATURE.


Selected Characteristics of Freshman
Students . .. . . .. . . 23

Selected Educational Benefits and the
College Experience . . . ... 29

Self-Concept . . . . . . . 29

Relationships with Other Students. . 32

Extracurricular Experiences. .. . . 35

Persistence in College .. . ... . 37

Impact of Different Types of Housing
Arrangements on College Freshmen . . 40

Research Approaches for Measuring the
Impact of Differing Living Environments. 45

Conclusions from the Literature . . .. 47


III. RESULTS OF THE STUDY . . . .

Response of Sample Participants .

Student Persistence in College. .

Changes in Students' Self-Concept

Relationships with Peers. . .

Participation in Extracurricular
Activities . . . . .

Chapter Summary . . . . .


IV. DISCUSSION OF THE DATA. . . .

Persistence . . . . .

Self-Concept. . . . . .


page

23


. . . 50

. . . 50

. . . 52

. . . 55

64

. . . 64
.. *66

. . . 75


. . . 78

. . . 78

. . . 82








TABLE OF CONTENTS (CONT'D)


CHAPTER

IV.
(CONT'D)


page

89


Relationships with Peers . . .

Participation in Extracurricular
Activities . . . . .


V. FINDINGS, CONCLUSIONS AND IMPLICATIONS .

Findings . . . . . . . .

Conclusions . . . . . . .

Implications of the Study . . . .

Suggestions for Further Research . .


APPENDICES

A SAMPLE COVER LETTERS AND ADDRESS
INFORMATION SHEET . . . . ..

B BARRETT-LENNARD RELATIONSHIP INVENTORY

C ACTIVITIES QUESTIONNAIRE . . . .


BIBLIOGRAPHY . . . .. ... . . ..


BIOGRAPHICAL SKETCH . . . . . . .


101

102

104

106

114


118

121

124


126


131












LIST OF TABLES


TABLE PAGE

1 Data on Reduction in Number of Eligible
Sample Participants . . . . . .. 51

2 Student Persistence by Residence Group ... 53

3 Persistence Among Females and Males. . . ... 53

4 Female Students' Persistence by Residence
Group . . . . . . . .. . . 54

5 Male Students' Persistence by Residence
Group . . . . . . . . . . 55

6 Estimates for Total Positive and Positive
Subscores on the Posttest of the
Tennessee Self Concept Scale. . . . .. 57

7 Results of t tests on Estimates of Selected
Parameters for the Regression Model . .. 58
2
8 R Values Calculated for the Barrett-Lennard
Relationship Inventory. . . . . .. 65

9 Results of t tests on the Estimates of
Parameters for the Regression Model . .. 66

10 Level of Participation in Campus Organiza-
tions Reported by Residence Group . . .. 67

11 Participation in Academically-Related
Organizations by Residence Group. . . . 68

12 Participation in "Other Organizations."
by Residence Group. .. ... . . .... 69

13 Students' Leadership Roles by Residence
Group . . . . . . . . . . 70

14 Leadership Roles Held by Females and Males . 70


vii








LIST OF TABLES (CONT'D)


TABLE PAGE

15 Participation in Intramural Sports
Competition on Campus by Residence
Group . . . . . . . . . . 72

16 Totals of Students by Residence Group
Who Visited the Museum or Gallery . . . 73

17 Totals of Students by Residence Group
Who Attended a Dance or Party on
Campus. . . . . . . . . 74


viii












LIST OF FIGURES


FIGURE PAGE

1 Regression Model . . . . . . ... 56

2 A Comparison of Identity Scores of
Subsample Groups. . . . . . . 59

3 A Comparison of Personal Self Scores
of Subsample Groups . . . . . ... 60

4 A Comparison of Family Self Scores of
Subsample Groups. . . . . . . .. 61

5 A Comparison of Social Self Scores of
Subsample Groups . . .. . . 62













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 Education



A COMPARISON OF SELF-CONCEPTS, PEER RELATIONSHIPS, PERSISTENCE
AND EXTRACURRICULAR INVOLVEMENT OF UNIVERSITY OF FLORIDA FRESHMEN
WITH DIFFERING HOUSING ARRANGEMENTS

By

Stephen Russell Haulman

June 1978


Chairman: Ralph B. Kimbrough
Major Department: Educational Administration


The problem of this study was to determine whether

significant differences existed in enrollment persistence,

self-concept, peer relationships, and participation in

extracurricular activities when University of Florida fresh-

men who lived on campus, at home, or in other off-campus

housing were compared. A longitudinal research design was

used during the 1976-77 academic year with a sample of 300

students selected through use of stratified systematic

sampling techniques. Over 68% of the eligible sample

participants returned completed materials.

Data on persistence were gathered from registration

records. Pretests and posttests of the Tennessee Self Concept

Scale and the Barrett-Lennard Relationship Inventory were

used to collect data on self-concepts and peer relationships







respectively. An Activities Questionnaire was designed to

gather data on extent of involvement in campus organizations,

leadership roles held by freshmen, and participation in

certain other extracurricular activities. Data on persis-

tence and participation in extracurricular activities were

analyzed using the chi-square test for significance.

Regression analysis techniques were used to study data

gathered on self-concepts and peer relationships. A regres-

sion model was constructed using three factors (sex, type

of residence and pretest scores on the instrument) and

estimates were computed for each parameter of the model

using the S.A.S. program.

The major findings are summarized as follows: 1. Signi-

ficant differences were found among males, but not females,

according to where they lived. Males who lived on campus

had a significantly higher level of persistence and males

who lived at home had a significantly lower level of persis-

tence when residence groups were compared. 2. Changes in

self-concept occurred among sample participants during their

freshman year. 3. Significant differences were found in

aspects of self-concept among groups with differing housing

arrangements. These differences were identified on the

identity, family self and social self subscales of the

Tennessee Self Concept Scale. 4. No significant differences

were found among the residence groups in students' percep-

tions of certain qualities of their relationships with

peers as measured by the Relationship Inventory. 5. Signi-

ficant differences were found among residence groups in

xi







participation in academically-related organizations and in

certain other organizations, such as the University Band and

R.O.T.C. Campus residents were far more involved in these

organizations than other students. 6. Significant differences

were found by residence group and by sex in leadership roles

held during the freshman year. Campus residents held leader-

ship roles far more often than other students. Males held

leadership roles in greater numbers than females. 7. Signi-

ficant differences were found among residence groups in

three other extracurricular activities. Students who lived

at home participated less in intramurals, were less likely

to have visited campus cultural centers during their freshman

year, and were more likely to have attended a dance or party

on campus in comparison with other students.

The results of this study indicate that a campus

residence hall was the best place for a freshman male to

live to maximize his chances of persisting at the University

of Florida. Data analysis also suggests that living at

home may be the best choice for a female to maximize her

chances of persisting, although the results were not signi-

ficant at the .05 level.

The results of this study tend to support the notion

that a diversified campus housing program, with leadership

committed to building residence hall environments that are

conducive to personal development, can offer important

support systems and opportunities for growth through


xii







interaction with peers in various activities as well as

needed assistance for students with developmental problems.






Chairman /


xiii












CHAPTER I
INTRODUCTION


Background and Justification for the Study

In the early seventies, pioneering research studies were

conducted by Arthur Chickering on the impact of differing

housing arrangements on college freshmen. Through these

studies, using longitudinal data gathered from many institu-

tions of higher education by the American Council of Education,

Chickering concluded that significant differences exist

among college freshmen who live on campus, at home, or off

campus away from home that cannot be attributed to preexisting

differences among the student groups.

The differing impact of various types of housing arrange-

ments on the self-concepts of freshman students and on their

persistence in college were among the principal findings

of Chickering (1975). Among other significant differences

found to exist among those living on campus, at home, and

off campus away from home were the extent of participation

in college life and relationships with other students.

The researcher believes that the self-concepts of

students and student persistence in college are matters of

considerable significance to students, parents, faculty and

staff, and have implications for society as well. Because of

their importance to a successful college experience, the writer

has referred to these two principal factors in his study as

educational benefits. Also, because the two other factors

1







cited from the Chickering studies appear to be related to

self-concept and/or persistence, as has been suggested in

the literature, they have also been referred to as educa-

tional benefits.


Statement of the Problem

The problem of this study is to determine, through

research conducted on one campus, whether significant

differences exist in selected educational benefits (self-

concept, persistence in college, relationships with other

students, participation in college life) arising from

differing housing arrangements. The study compares unmarried

entering freshman students at the University of Florida who

lived on campus, at home, or in other off-campus housing

during the 1976-1977 academic year. The focus of the investi-

gation is on the following questions:

1. Is there a difference in persistence in college

among students who live on campus, at home, or in

other off-campus housing?

2. For those who persist in college, is there a

change in self-concept among students who live

on campus, at home, or in other off-campus housing

and, if so, does this change differ on the basis

of residence?

3. For those who persist in college, is there a

difference in relationships with peers among

students who live on campus, at home, or in other

off-campus housing?







4. For those who persist in college, is there a

difference in participation in extracurricular

activities among students who live on campus,

at home, or in other off-campus housing?


Significance of the Study

The study is of significance in that there are three

prevailing points of view regarding college housing among

administrators in higher education. These can be described

as the traditional approach, the educational approach, and

the "no housing" approach.

The traditional approach to residence hall operation is

to provide adequate living facilities, essential services,

and a regulated lifestyle for residents supervised by college

officials. Facilities are designed to meet the basic needs

of students for satisfactory sleeping and study space. Staffing

is, for the most part, nonprofessional. Staff members

operating under this approach to housing generally regard

their function as almost exclusively in the areas of admini-

stration, upkeep of facilities, and supervision of student

behavior. The services supplied to students in this setting

are very limited. These include facilities repairs, cleaning

services, and key service; with some housing operations also

providing food service and/or linen service for residents.

Organized activities of a social, recreational or educational

nature are generally not found within the residence halls

on a campus characterized by the traditional approach.

Rather, such opportunities are provided elsewhere on or near







the campus by other service units of the institution or by

formal or informal student organizations.

The belief that college housing is a setting offering

educational benefits to campus residents is the basis for

a second approach to residence hall operation. This

educational approach reflects a commitment to building an

environment conducive to learning and personal development.

With leadership of the campus housing program provided by

professional educators, priority is given to the creation of

opportunities within the campus residential setting that

build social skills, supplement classroom learning with other

learning experiences, and provide other avenues to personal

growth. Housing facilities are designed to meet not only the

basic needs of students to study and sleep, but also to provide

appropriate common areas for the broad range of social, educa-

tional, and recreational activities that are an integral part

of life in residence communities. Leadership in articulating

this philosophy has been provided by Riker (1965). In

expounding on educational purposes, he noted that many adhere

to the

Time-honored but unwarranted assumption that
learning is a product of just the classroom,
occurring solely as the result of action and
reaction among the teacher, the student and
their subject. The accumulating indications
are, however, that many factors influence
learning and that most of them are outside
the classroom. The student society, for
example, appears to be a significant force
affecting behavior. For this reason, the
student residence becomes an important means
of working with this society. The theories
about learning are many, and the evidence
regarding the factors involved is far from
conclusive. Even so, the wise course of







action is to experiment with housing as a
factor in learning because of the over-riding
importance of finding better ways to insure
student success at college. (p. 2)

A third point of view regarding student housing that is

prevalent among administrators in higher education is that

the institution has no obligation to provide campus housing

for students. This point of view toward housing reflects an

even broader belief that the college or university should

assume little or no responsibility for the student outside

of the classroom. Such an approach to student life, partic-

ularly toward campus housing, has become prevalent among

institutions of higher education established in the 1960's

and 1970's, most of which were planned as commuter institutions

located in close proximity to population centers.

There has been little evidence from which one could

draw conclusions as to the relative merits of these three

approaches to the housing of students. One review of the

research in the area has concluded that "research designed

to study the effects and impact of different student housing

programs is lacking" (Robinson & Brown, 1961, p. 360). A

later review of the literature revealed a continuing absence

of evidence regarding the merits of campus housing in contrib-

uting to the education of students (Williams & Reilley, 1972,

p. 212), although some research results notably by Chickering,

have been reported since that time.

Much attention has been focused on the educational

impact of college housing since the 1960's, when lawsuits

challenging compulsory on-campus residence requirements were







filed against a number of colleges and universities. While

numerous educators have testified as to the merits and values

of campus living, few research studies .to either support or

refute such testimony have been available.

A study of this type was further needed because the

Division of Housing at the University of Florida has been,

in the mid-seventies, in a state of flux in its philosophical

approach to residence hall operation. Research conducted at

this particular institution on whether or not there are

significant differences in certain educational benefits

arising from differing housing arrangements may prove very

useful in helping to shape decisions on the future course of

student housing on this campus.

The study described herein is significant in light of

the prevailing differences of opinion regarding the housing

of students, court decisions on the issue of compulsory on-

campus residence requirements, and the particular situation

existing at the University of Florida in the mid-seventies.


Scope of the Study

The following contraints have been observed in conducting

this study:

1. The research was confined to a sample of a population

of unmarried freshman students who began their

college work in September of 1976 and who met admis-

sions requirements of the University of Florida.

In selecting the sample, this population was divided

into six subpopulations according to sex and







expected place of residence. Groups 1M (males)

and IF (females) included students living at home.

Groups 2M and 2F included students living off

campus away from home. Groups 3M and 3F included

campus residents. To assure adequate representation

of the first two relatively small subpopulations,

a stratified systematic sample of 300 students was

drawn. Of the total sample of 300 students who

were initially chosen, 50 were selected from each

of the six groups.

2. Measurement of self-concept was confined to pretest

and posttest scores on the positive scales of the

Tennessee Self Concept Scale administered in August

of 1976 and April of 1977.

3. Measurement of persistence in college was confined

to a count of the number of students in the sample

registering for one or more courses at the University

of Florida for the 1976-77 Fall, Winter and Spring

quarters.

4. Measurement of participation in college life was

confined to comparison of each group's after-only

scores on a researcher-developed questionnaire on

extracurricular activities, which was administered

in April of 1977.

5. Measurement of relationships with other students

was confined to pretest and posttest scores on the

Barrett-Lennard Relationship Inventory (revised)

administered in August of 1976 and in April of 1977.







Limitations of the Study

The following limitations of this study were recognized:

1. Only to the extent that other populations possess

a similar pattern of characteristics of the speci-

fied population of entering freshmen at the Univer-

sity of Florida are the results generalizable beyond

the sample studied.

2. Data-gathering materials were distributed individu-

ally to each person in the sample. The setting in

which the data for this study were collected was

not under the direct control of the writer; therefore,

some opportunities existed for variations in how

participants completed the instruments.

3. The researcher confined his comparisons among housing

groups strictly to three broad categories: those who

lived at home, those who lived off campus away from

home, and those who lived on campus. There was no

attempt made to compare differences within a single

category i.e., no attempt was made to draw compari-

sons among groups from different residence halls.


Assumptions

The writer made certain assumptions for the purposes

of this study. First, the sample participants responded

honestly when completing the instruments sent to them.

Second, the sample participants accepted as truthful the

researcher's assurances that respondents would in no way be

identified individually. Third,systematic samples drawn







from the lists of students for the three strata have yielded

representative cross sections of the subpopulations.


Definition of Terms

Admission requirements. A cumulative high school grade point

average of 2.0 or better and a Florida Twelfth Grade

Test score of 300 or better (or equivalent SAT score for

out-of-state students) were required for acceptance as a

freshman at the University.

At home student. A student who lived at home with his or

her parents or with a grandparent, aunt or uncle while

attending the University.

Campus resident. A student who resided in a campus residence

hall while attending the University. Inasmuch as those

who lived in an on-campus fraternity house, sorority

house or a residence hall reserved 'for athletes could,

at best, constitute less than 1% of the freshman

population and because their living environment might

have been very different from other freshmen, such

individuals were excluded from the sample.

Educational benefits. Selected developmental factors related

to the educational process that were chosen for study.

These include self-concept, persistence in college,

participation in college life, and relationships with

other students.

Off campus away from home student. A student who resided

off campus, but not with his or her parents, grand-

parent(s), aunt or uncle while attending the University.






10
Students in this category resided primarily in apartments

shared with one or more other students. Inasmuch as

those who lived in a fraternity or sorority house off

campus could, at best, constitute less than 1% of the

freshman population and because their living environment

might have been very different from other freshmen,

such individuals were excluded from the sample.

Participation in college life. The extent of participation

in extracurricular activities during the freshman year

as measured by a researcher-developed questionnaire.

Persistence in college. The tendency of a student to continue

his/her studies at the University as measured by a count

of the number of students who registered for one or more

courses for the Fall, Winter and Spring quarters.

Relationships with other students. The student's perceptions

of the nature of his relationships with peers as measured

by the Barrett-Lennard Relationship Inventory (revised).

Residence requirement. Any requirement for students to live

on campus (except a requirement to continue a housing

obligation voluntarily contracted for by the student)

that is imposed by an institution of higher education.

Self-concept. The attitudes one holds toward himself/herself

as measured by the positive scales of the Tennessee

Self Concept Scale.







Procedures


Description of the Sample

An admittance list was obtained from the Admissions Office

at the University of Florida on August 9, 1976, and was used

to define the population. This list was a then-current

roster of incoming freshmen admitted to the University for

the upcoming Fall quarter. It was arranged in alphabetical

order and included home addresses, high school grade point

averages, Florida Twelfth Grade Test scores, social security

number, sex, marital status, housing code, birth date, and

information concerning the status of each prospective student's

application for admission. The list was coded to eliminate

from the sample those students who did not meet admission

requirements for freshmen entering the University (those

admitted by petition) and to eliminate those students who

were married.

Each person on the list of remaining students was

assigned to one of the six subpopulation groups. If the

housing code next to a student's name indicated that he or

she had paid the campus housing prepayment or requested a

delay in payment, he or she was assigned to Groups 3M (males)

or 3F (females), which included students likely to be campus

residents. If the housing code next to a student's name

indicated that he or she would not be living on campus, then

the student's home address was checked. If the student's

home was within reasonable commuting distance of the campus

in the judgment of the researcher, then he or she was assigned

to Groups 1M or IF, which included those students likely to





12


be living at home. All .other freshmen were assigned to

Groups 2M or 2F which included those students likely to be

living off campus away from home.

A stratified systematic sample of students was drawn

from the coded list.. While it was impossible to determine

in advance the exact size of the subpopulations within the

total population, Groups 3M and 3F were expected to include

a large majority of the population and Groups 1 and 2 were

expected to be relatively small as had been the case in

previous years. To assure adequate representation of

Groups 1 and 2, stratified sampling was used. Of the total

sample of 300 students who were initially chosen, 50 were

selected from each of the six groups. Systematic sampling

procedures were used by the researcher. Since the alphabetized

admit list was not ordered in any manner associated with the

factors to be studied, systematic sampling procedures were

expected to yield a representative cross section of the

population.

Some minor changes in the number of sample participants

from each group, as well as in the total number of students

in the sample occurred prior to and during October of 1976,

when verification of place of residence was made for all of

those included in the sample. Verification proceeded according

to this plan: First, the most current alphabetical roster

of students in campus housing available from the Division of

Housing was reviewed in early October. Full names and social

security numbers were checked for all sample participants.

Those who appeared on this roster, regardless of initial








subsample assignment, were listed in Groups 3MA (males)and

3FA (females) which included those sample participants who

were actually assigned to campus housing for the fall quarter.

The "anticipated place of residence" data sheets,

included in pretest materials returned by sample participants

after being sent to their home addresses in August, were then

reviewed. Verification by telephone, mail, or visit was

attempted for those sample participants not already assigned

to Groups 3MA and 3FA. Those found to be living off campus

away from home were assigned to Groups 2MA and 2FA.

Updated local address information was then obtained

from the Registrar's Office on sample participants whose

addresses had not already been verified. When such updated

address information was found to exist, verification by

telephone, mail or visit was attempted. Sample participants

were then assigned appropriately to Group 1MA, 1FA, 2MA or

2FA.

The remaining sample participants were contacted by mail

or telephone through the permanent home addresses these

students had provided to the Registrar's Office. The new

information gained was used to assign persons to Group 1MA,

1FA, 2MA, or 2FA.

Those initially selected for the sample who did not

actually enroll for the Fall quarter or who enrolled but

did not complete pretest instruments mailed in August by

October 15 were dropped completely from the sample. The

researcher made repeated attempts where necessary to obtain

pretest instruments from enrolled sample participants by








this date. The reduction in enrolled sample participants

as a result of failure to complete pretest instruments was

kept to a minimum through exhaustive follow-up by the

researcher.


Design of the Study

The research design selected by the researcher to study

student persistence in college, self-concept and relationships

with other students can be depicted as follows:

0 X 0

O Y 0

O Z O

where X = at home living experience

Y = off-campus living experience away from home

Z = on-campus living experience

0 = testing of sample participants

The research design selected by the researcher to study

participation in extracurricular activities can be depicted

as follows:

X O

Y 0

Z 0


Collection of Data

Pretest materials were mailed to persons selected for

the sample in August of 1976, prior to enrollment. These

materials included a general information sheet requesting

assistance with the study and assuring confidentiality of

each individual's responses, a data sheet requesting








information on plans for Fall quarter residence while attending

the University; the Tennessee Self Concept Scale and answer

sheet; the Barrett-Lennard Relationship Inventory (revised);

and a stamped, addressed return envelope.

Posttest materials were distributed in April of 1977 to

those Fall enrollees who returned useable pretest materials

and who persisted at the University. These materials

included a cover letter, the Tennessee Self Concept Scale

and answer sheet, the Barrett-Lennard Relationship Inventory

(revised), a questionnaire on involvement in extracurricular

activities, and a return envelope.

Samples of the general information sheet and the data

sheet sent to students with pretest materials, as well as the

cover letter sent to students with posttest materials, are

included in Appendix A. A sample of the pretest of the

Barrett-Lennard Relationship Inventory (revised) is included

in Appendix B. The Activities Questionnaire, used on a

posttest-only basis, is included in Appendix C.

Extensive personal follow-up was provided for collection

of both pretest and posttest materials to secure a high

rate of return of materials.


Instruments


Tennessee Self Concept Scale

This instrument was developed by William H. Fitts in

1955. The scale was standardized by Fitts (cited in Harrell,

1976) on a total of 1091 persons representing all social,







economic, intellectual, and educational levels from sixth

grade through the Ph.D. degree.

The scale consists of 100 items. These items were

designed to guide an individual in describing himself or

herself in terms of the following: (1) this is what I

am; (2) this is how I feel about myself; and (3) this is

what I do. There are nine scales representing the

following dimensions of the self-concept: total positive,

identity, self-satisfaction, behavior, physical self,

moral-ethical self, personal self, family self, and social

self. These are referred to as the positive scales of

the Tennessee Self Concept Scale. The positive scales

as well as the distribution and total variability scores

were used in this study.

A "counseling" version and a "clinical and research"

version of the Tennessee Self Concept Scale are available.

The instrument itself does not differ, but far more sub-

scores are available on the latter version for certain

clinical and research purposes. The counseling version

of the scale was used in this study.

Two scoring methods are available for the instrument,

a hand-scored method and a computer-scored method. The

latter scoring method was most appropriate for the study,

so the optical scan version of the instrument was used.

The instrument has a Likert-type scale of five choices

ranging from completely false to completely true.








Test-retest reliability data developed by Fitts (1965)

with a sample of 60 college students over a two-week period

are as follows: Total Positive 0.92, Distribution 0.89,

and Total Variability 0.67.

Validity of the instrument was demonstrated by Fitts

(1965), using four procedures: content validity, discrimi-

nation between groups, correlation with other personality

measures, and personaltiy changes under particular conditions.

Content validity was established through use of a jury of

seven clinical psychologists to analyze individual items.

Only items on which the psychologists reached favorable

consensus regarding content were incorporated in the final

version of the instrument. Studies on a variety of groups

have established the effectiveness of the scale in discrimi-

nating between such groups as psychiatric patients and non-

patients. Comparisons of the Tennessee Self Concept Scale

with the Minnesota Multiphasic Personality Inventory (MMPI)

yielded high correlation ratios. Personality changes

resulting from psychotherapy (Ashcraft and Fitts, 1964)

and from detrimental life experiences (Gividen, 1959)

were measured by the Tennessee Self Concept Scale. "There

is considerable evidence that people's concepts of self

do change as a result of significant experiences. The

Tennessee Self Concept Scale reflected these changes in

predicted ways, thus constituting additional evidence for

the validity of the instrument" (Fitts, 1965, p. 30).







Barrett-Lennard Relationship Inventory

This instrument was developed by G. T. Barrett-Lennard

in 1962 to measure certain dimensions of a client's percep-

tions of his therapist's attitudes and responses. The

Relationship Inventory was prepared in two parallel forms

differing only in third-person pronouns (Barrett-Lennard,

1962; Hanson et al., 1966). The group form of the instrument

was used in this study following the Wiebe-Pearce revision

of Barrett-Lennard's original inventory. This revision

consisted of 32 items and 4 subscales (see Appendix B).

The group form of the instrument measures perceptions of

group facilitativeness through a total scale score. Subscale

scores measure positive regard, empathy, congruence, and

unconditionality of regard for others in the group.

The items in the revised inventory are classified

in the following groups:

Level of Regard: 1,6,8,10,14,17,21,28,29,31

Empathetic Understanding: 3,4,11,15,18,26,30

Congruence: 2,5,7,9,12,19,22,23,25,27

Unconditionality of Regard: 13,16,20,24,32

Split half reliability coefficients ranged from .82

to .93 and test-retest reliability correlations ranged

from .86 to .95 for scores analyzed by Barrett-Lennard.

Internal consistency reliability coefficients for the

four subscales have ranged from .82 to .95 in analysis

by other researchers. Lanning (1971) has reported stability

coefficients for the total score of .93 to .95.








Based on an item analysis of the instrument, Weibe

and Pearce (1973) recommended a shorter, more discriminating

instrument to include only those items correlated most

strongly with the subscale on which they appear. Toppe

(1977), using this revised inventory, found that the

computed intercorrelations for scale scores and the total

score in his investigation were all positive and significant

beyond the .001 level.

Construct validity for the scale has been established

through classification of items by five judges. Only

those 85 items on which consensus was reached regarding

classification were used in Barrett-Lennard's final form

of the inventory. Barrett-Lennard's research also estab-

lished predictive validity, demonstrating that the inventory

discriminated, as predicted, between clients who improved

and who did not improve as a result of therapy. "Studies

with less severely disturbed clients, such as students,

report significant positive relationships between perceived

facilitativeness and positive client change. Predictive

validity, therefore, has been established for relatively

normal populations" (Toppe, 1977, p. 18).


Activities Questionnaire

A survey of the theoretical writings and research

studies on participation in extracurricular activities

was conducted. Through this survey, it was determined

that there were three principal aspects of extracurricular

involvement that needed to be addressed in this research.








These aspects were level of participation in campus organiza-

tions, leadership roles in campus organizations held by

freshmen, and participation in other college-related extra-

curricular activities.

An Activities Questionnaire was developed for use on

a posttest-only basis after a search of available standardized

instruments failed to locateone suitable for this study.

Three sections of the questionnaire were developed. The

first section was designed to list categories of organiza-

tions. In this section, respondents were to indicate their

level of involvement (Not at All, To a Small Extent, Fairly

Extensive or Very Extensive) in each type of organization.

The second section was designed to determine whether or

not the respondent had been selected for or had assumed a

leadership role in any campus organization since the start

of his or her freshman year. A "yes" or "no" response was

called for and a space was provided for the respondent to

name the organizations) if an affirmative response was

made. The third section of the questionnaire was designed

to determine whether or not a respondent had participated

in certain other University-related extracurricular activi-

ties of a cultural, social or recreational nature during

the course of his or her freshman year. A "yes" or "no"

response was requested in this section of the questionnaire.

A draft of the instrument was developed, reviewed by

several professors, and revised prior to distribution with

posttest materials. Because of the nature of this information

being sought, high reliability can be expected from the







instrument. All of the items contained in the instrument

refer to different aspects of extracurricular involvement;

therefore, it can be concluded that the Activities Question-

naire is a valid measurement of participation in extra-

curricular activities. The Activities Questionnaire is

included in Appendix C


Analysis of the Data

Analyses of the research questions in the statement of

the problem were accomplished through the use of descriptive,

inferential and correlational statistics. The inferential

statistical technique of chi square was used to test the

significance of any differences that existed among the sub-

sample groups on persistence and participation in extra-

curricular activities. The .05 level of significance was

used to evaluate the chi-square tests for contingency

tables. Regression analysis was used to predict posttest

scores for students in the sample on the Tennessee Self

Concept Scale and the Barrett-Lennard Relationship Inventory.

A regression model was constructed using three factors

(sex, type of residence and pretest scores on the instrument).

Using a S.A.S. computer program, estimates were computed for

each parameter of the model. F tests were conducted to

determine how good the model was in predicting posttest

scores. The t tests were conducted on the hypothesis that

each parameter equalled zero. The .05 level of significance

was used to evaluate the results of these t tests. Descrip-

tive statistics were used in describing pretest and posttest








scores of the subsample groups where significant differences

were found on the positive scales of the Tennessee Self

Concept Scale.


Organization of the Research Report

This study is reported in five chapters. Chapter I

contains an introduction, a statement of the problem, the

significance of the study, scope of the study, limitations,

assumptions, definition of terms, and procedures. A review

of the literature is provided in Chapter II. An analysis

of the data is presented in Chapter III. Chapter IV is

devoted to a discussion of the data. In the final chapter,

Chapter V, conclusions and their implications are presented

along with suggestions for further research.













CHAPTER II
REVIEW OF THE LITERATURE


In order to provide an orderly presentation of selected

research findings and theoretical writings from the profes-

sional literature as background for this study, this chapter

is divided into four sections. The first section includes

information on selected characteristics of freshman students.

The second section reviews writings and research findings on

certain educational benefits associated with the college

experience. The results of research on the impact of

different housing arrangements on college freshmen are

reviewed in section three. The fourth section of the

chapter presents information on several research methods

that have been used to measure the impact of different

living environments on students.


Selected Characteristics of Freshman Students

Sanford (1962), Freedman (cited in Sanford, 1962) and

others, in writings about student development, have generally

preferred to refer to the adolescent stage of development

or to personal growth during the college years rather than

to the developmental changes during the freshman year. In

a book entitled College and Character, Sanford maintained

that "at no stage can the development of a man's personality

be defined simply by either his chronological age or his








academic status" (Sanford, 1964, p. 86). Instead, Sanford

(1962), Heath (1969) and others have described a stage of

late adolescence, which includes most freshmen, but also

some high school students and college sophomores, as well

as a minority of college upperclassmen.

McConnell and Heist (cited in Sanford, 1962) and others

have elaborated on the wide diversity of freshmen in terms

of academic ability, personality characteristics, interests,

values and socioeconomic background. According to Alfert

(1968), students starting college vary in initial level of

development as well as in the pace of development during

their stay. Despite the tremendous variety of individual

backgrounds represented by the freshman population, there

are some generalizations that can be gleaned from the liter-

ature that are said to apply to freshman students.

Prominent researchers in the field of student development,

including Chickering (1974) and DeCoster (1970) have referred

to The American College, edited by Nevitt Sanford, as a

major contribution to the literature and a useful source of

information on freshman characteristics. In a chapter on

the developmental status of the freshman, Sanford (1962)

draws an interesting parallel:

The freshman's psychological situation is essen-
tially a picture of an authoritarian personality
structure. Authoritarianism in personality is a
particular pattern of relationships among impulse,
conscience, and ego (Sanford, 1956b). It is a
pattern in which strong impulses are directly
opposed by an alert, rigid, and punitive conscience.
The ego has to devote so much energy to restraining
impulses that its other functions are not well
performed; it has been able to integrate little
of the primitive conscience with itself, so that
the latter continues to function more or less as
a foreign body. This state of affairs at the








core of the personality is reflected at the sur-
face in characteristic ways: in stereotyped
thinking, intolerance of ambiguity, punitive
morality, submissiveness toward the powerful and
dominance toward the weak, conventionality, anti-
intellectualism, hostility toward people perceived
to be different from oneself. The freshman inhibits
impulse by being morally strict with himself and
others. He is ready to meet stiff requirements,
to work hard, to conform with what he takes to be
the prevailing standards of behavior--and he is
inclined to be somewhat intolerant of those who do
not. An element of perfectionism, of striving
for purity of thought and action is characteristic
of the freshman. (p. 261)

Newcomb and Feldman (1969) reviewed thirty-seven studies

on changes in authoritarianism during the college years.

Results of these studies were consistent--seniors were, on

the average, less authoritarian than freshmen. In almost all

studies, these differences were relatively large and statis-

tically significant. Of the longitudinal studies comparing

freshman and sophomore students on authoritarianism, all

reported that sophomores are, on the average, significantly

less authoritarian than they were as freshmen.

Plant (1965) has found that a decrease in authoritarianism

is in evidence in this age group whether individuals have been

enrolled in college or not. Newcomb and Feldman noted that

decreases in authoritarianism in college students appears to

be more reliable and larger than for those in the same age

range who are not attending college. They concluded that:

As of now, the most that may be surmised is that
colleges in general, and particular colleges
differentially, may select or attract students
who are differentially "open" to changes in
authoritarianism; and that these colleges provide
environments that differ both among themselves
and as compared with non-college environments -
in respect to influences likely to change existing
states of authoritarianism. One way in which







college environments differ from non-college
environments is that the former appear to
facilitate decreases in authoritarianism more
than the latter. (p. 32)

Sanford (1962) noted that freshmen have not had time

to develop a system of values rooted in personal experiences--

another feature of the authoritarian personality structure.

The values of the entering freshman are supported,
typically, by his family and home community; he is
dependent upon this support, but, at least in the
very beginning of the college career, the support
is adequate enough so that the freshman is able to
present a picture of organization and self-contain-
ment. (p. 262-3)

Data from the 1976 Student Information Forms, published

by the American Council on Education, have indicated that

many entering freshmen have unrealistic expectations regarding

their chances of changing major field, failing one or more

courses, and completing degree requirements on time. Morstain

(1972) commented positively on the ultimate outcome once these

expectations are shattered for many:

Undoubtedly, there is much confusion and ambiguity
if students are in situations that might be incon-
gruent with their expectations about education and
their role in it, but this confusion can be helpful.
If it enables students to think about their education
and to establish personally relevant goals, the
initial confusion has served its purpose. (p. 286)

This shattering of initial expectations and the resulting

confusion can have a major impact on a student's confidence

in his or her abilities. According to Larson and Laramee

(1976),

There is often an initial shock to students who
come to college having been leaders in their high
schools, churches, and communities. The competi-
tion for extracurricular leadership roles, higher
standards for academic work, and the more hetero-









generous value systems of the new environment may
leave them without the familiar support systems
from which they had drawn recognition, encourage-
ment, and purpose. The hope that "things will get
better" is realized by some students, but others
do not soon regain the confidence they once felt
in a more familiar environment. (p. 48)

Against this background of high academic expectations,

increased competition, greater exposure to differing value

systems and (for many) the loss of support systems, Sanford

(1964) elaborated on the freshman's initial view of self

in College and Character:

Closely related to the freshman's authoritarian
propensities is his uncertain self-esteem. He
does not know what he can do, how good he is,
or what to think of himself. On the one hand he
is happy to remember his parents' faith in him,
the accomplishments of high school and the plaudits
received there, and, perhaps particularly, the fact
that he was chosen for admission to his college.
On the other hand, he suspects that he is now
playing in a different league, and he knows that
the major tests of life still await him. In his
uncertainty, he vacillates between overestimation
and underestimation of himself. His inclination
is to stick to patterns of behavior that have
been rewarded in the past and to display such
confidence as he can muster. (p. 87-8)

Freshman students can also be characterized as demon-

strating increasing tolerance and flexibility of thinking.

Gough (1958) has found that college freshmen score between

high school freshmen and graduate students on these factors,

using the California Psychological Inventory.

Newcomb and Feldman (1969), on the basis of a review of

the literature on student characteristics and how these change

during the college years, concluded that developmental change

is gradual over the college years rather than pronounced in

any particular year. They found that studies on patterns of








change reflected fairly consistently a steady decline in

authoritarianism as well as a decline in political, economic

and social conservatism.

In commenting further on patterns of change, Newcomb

and Feldman supported the widely held view that the impact

of the college experience is greatest upon students during

their freshman and sophomore years. Yet, they also cited

some reasons why this pattern is not consistently true:

For some dimensions the early college years may
indeed provide the greatest impetus of change,
but other areas of potential change may not become
salient or relevant to students until their later
college years. Likewise, at some colleges the
challenges of the early years may be greater than
those of the later years, whereas structural arrange-
ments of other colleges may create greater pressures
for change on upper division than on lower division
students. (p. 101)

Newcomb and Feldman also noted that individual differences

play an important role in determining when change takes place

during the college years.

For some,'change may be almost immediate, for others
there may be a longer period of "working through,"
with observable change being evidenced only in
later college years. It is even possible that some
students find the challenges of their freshman year
so heavy that they become resistant to change, only
to become less defensive and more likely to change
in their junior or senior year. (p. 101)

In The Student in Higher Education, the writers concluded

that the freshman student "generally has an open mind, is eager

to work, and willing to learn . the freshman year is of

critical importance because it is the time when the student's

enthusiasm, curiosity, and willingness to work can be snuffed-

out or reinforced" (The Committee on the Student in Higher

Education, 1968, p. 45).







Selected Educational Benefits
and the College Experience


Self-Concept

One of the major developmental needs of freshman students,

as was identified by The Committee on the Student in Higher

Education, is the enhancement of self-concept (p. 11).

Erikson (1950) also identified the achievement of a personal

sense of identity as the main growth task of the adolescent

stage of development. He described achievement of identity

as a gradual process involving exploration and trying out of

alternative values and goals and resulting in a commitment

to a particular set of values. Segal (1967) has commented

on the implications of various residence settings for achieve-

ment of identity in maturing college students.

But different settings allow different kinds of
testing, trying, and doing, and the choice of
one or another setting suggests different strat-
egies and tactics in terms of the primacy of
different needs with the person. (p. 309)

Segal has pointed out that the living arrangements of

college students differ in the extent to which the process

of achieving identity is "worked through directly within

the concept of family, with greater or less emphasis on

peer culture as a critical factor, with variation in who sets

the rules and for what" (Segal, 1967, p. 309). He noted

that each setting offers possibilities for personal growth

toward maturity as well as opportunities to avoid the

trials inherent in achieving identity.








Powell, Plyler, Dickson and McClellan (1969), Sanford

(1964), Chickering (1969) and others have studied the concept

of self as it relates to the college experiences of students

and have acknowledged its significance in terms of impact on

student development. The self-concept consists of "all of

the beliefs, attitudes and opinions which an individual

holds about himself" (Purkey, 1967, p. 3)

Rogers (1951) described self-concept in a more detailed

manner:

The self concept may be thought of as an organized
configuration of perceptions of the self which
are admissible to awareness. It is composed of
such elements as the perceptions of one's charac-
teristics and abilities; the percepts and concepts
of the self in relation to others and to the
environment; the value qualities which are perceived
as associated with experiences and objects; and
goals and ideals which are perceived as having
positive or negative valence. (p. 136)

This organized configuration of perceptions of the self,

according to Rogers, unifies the individual's personality

and regulates his behavior. So long as perceptions of self

remain firmly organized, and no contradictory material is

even dimly perceived, then the self is seen as "worthy and

acceptable, and conscious tension is minimal" (p. 191).

Wylie (.1961), in a comprehensive survey of the literature

on self-concept, concluded that self-concept theories are in

many ways ambiguous, incomplete, and overlapping. As a result

of her review of pertinent research literature, she noted

that no one theory has received a large amount of systematic

empirical exploration.








In a research bulletin entitled The Self and School

Achievement, published in 1970, Purkey summarized numerous

studies on the topic and concluded that there is a strong

relationship between a student's concept of self and his

performance in school--and in life:

Students who feel good about themselves and have
confidence in their abilities are those who are
most likely to succeed. Conversely, it appeared
that those who see themselves and their abilities
in a negative fashion usually fail to achieve
good grades. Academic success or failure appears
to be as deeply rooted in concepts of self as it
is to measured mental ability, if not deeper.
(p. 14)

In research in the area of student development at

Haverford College, it was found that "no student reported

that he had not changed his conception of himself as a

consequence of his college experience" (Heath, 1968, p. 222).

Heath, who provided a model of the maturing process for

the college years and beyond, had this to say about the

freshman experience:

The freshman is "shocked" by the demands of
the college. He becomes unstable and dis-
organized, though, in a "good" college that
provides adequate supports, not enough to
regress seriously or withdraw. That is, a
good college helps its students to remain
educablee." Disorganization and confusion
are painful; without them, students may
never make efforts to understand. The
freshman becomes more aware of his own
inadequacies, questions his competence and
values, and despairs. By the end of his
freshman year, his self-concept has shifted
from being "God's gift to humanity" to being
a person of little value. (p. 175)

Sanford (1967), in writing about student change in

college, noted that students did not rate themselves high







on personal stability and integration (as reflected by

student comments about self-concept) in ratings made during

the first semester of college. He attributed this situation

to the many adjustment problems faced by freshmen during this

period. His data reflected a sharp increase in personal

stability and integration from the first to the second semester

of the freshman year. In an earlier text, Sanford (1964)

also noted how vulnerable freshmen are to others' perceptions

of them. Freshmen are particularly subject to the influence

of peers, whose approval is needed to bolster self-confidence.


Relationships with Other Students

According to a number of studies reviewed by Newcomb and

Feldman (1969), students about to begin college lifP have high

expectations that academic and intellectual experiences will

be an extremely important source of satisfaction for them.

However, in a longitudinal study, King (cited in Newcomb and

Feldman, 1969) found that by their senior year, most Harvard

students in reflecting upon their college experiences, felt

that their greatest satisfactions had come from interpersonal

activities and personal growth rather than from academic and

intellectual activities. Wallace (cited in Newcomb and Feldman,

1969) reported similar results in a study of freshman students.

He found that while getting the highest possible grades and

learning as much as possible were the most important and

most widely held goals among students entering college, the

importance of these goals declined sharply before the end

of the first year of college attendance, while the importance

of relationships with peers received much greater emphasis.








Maxon and Malone (1977) noted that the increase in the

influence of the peer group is a function of a changing

culture. In describing the increasing power of the peer

group in shaping values, they have said, "Today, however,

due to his frequent contact with peers, plus the increasing

absence of his parents from the home, the peer group has

provided the model that has become as important a socializing

agent as the home" (Maxon and Malone, 1977, p. 191).

Relationships with other students can profoundly affect

individual patterns of development. The importance of these

relationships with peers was recognized by The Committee on

the Student in Higher Education (1968):

Just as the friendship group controls production
on a factory assembly line and cohesiveness in a
military squad, so the student friendship group
helps determine what is learned in the college,
how it is learned, and what effect both knowledge
and the learning experience have on the student's
total personality. While our knowledge of how
the friendship group can contribute positively
to the educational process is still meager, the
importance of peer group influence is so obvious
that we must rapidly acquire more knowledge of
how it works and integrate it into the educational
experience--hopefully, without attempting to
manipulate it. (p. 13)

Powell, et al. (1969), Newcomb (1964) and others have

also recognized the strong influence of peers. Newcomb

described the isolation from contact with more mature adults,

which the freshman who is away from home typically experi-

ences through greatly reduced contact with parents and

infrequent contact with faculty outside the classroom.

Against this background, Newcomb commented on peer group

formation and the impact of such groups on students:








Such a combination of circumstances is hardly
calculated to aid the. student in his search for
identity, precisely at the time when he is least
certain about it. Small wonder, then, that
students tend to be drawn together; their common
problems and their relative isolation from non-
students make them ready material for the forma-
tion of strong peer groups. Membership in a peer
group is more likely to influence directly students'
capacities, or basic personality traits. (p. 141)

Sherif and Sherif (1964) have described the manner in

which relationships with peers are established and maintained

by adolescents. They cited two major reasons for peer group

formation. First, social ties with peers are formed to aid

individuals in developing self-concepts that are more con-

sistent and well-defined. Second, social relationships are

formed with peers to provide support systems that can help

individuals cope with the problems of everyday living.

Sherif and Sherif have described the factors which determine

the degree of influence which a peer group has on an

individual:

The extent to which such an informal formation
becomes the center of the universe for an
individual member is proportional to the degree
of disruption of other ties (family, school,
and other social establishments), and propor-
tional to the degree that the group and its
activities serve as vehicles for joint action
towards the fulfillment of goals he feels
denied otherwise (social, financial, sexual,
and so on). To this extent, the individual's
concerns over social acceptance or rejection,
his concern to prove himself as a person who
counts, his very conception of the kind of
person he is, revolve in no small part around
"being somebody" in this group of his own
choosing.








Extracurricular Experiences

In an historical account tracing the development of

student extracurricular activities, Koos (1940) described

the then-current attitude of educational administrators

which was "to recognize positive educational values in

the extra-curriculum, to promote pupil-initiated projects,

and to capitalize for educational growth the natural, social

and creative propensities of youth." Although a later review

of the literature by Gilligan (1967) reflected periodic

changes in emphasis and support given to student activities,

an appreciation of the value of, and commitment to, such

activities remained in evidence.

The values of extracurricular experiences have been

noted by many, including Koos (1940), Frederick (1965) and

Lloyd-Jones, et al. (1938). Lloyd-Jones has articulated

these values in a concise manner:

SExtra-curricular activities offer the opportunities
for students to develop good qualities-of leaderships
and fellowships. They offer the opportunity to
serve the institution; to experience and to help
create good fellowship and social good will; they
further self-realization and all-round growth. They
assist students in adjusting to their student world
and in learning the qualities of good citizenship.
(p. 186)

Erikson (1950) has cited both the search for identity and

for peer support as factors motivating students to participate

in extracurricular activities in college. Weston and Stein

(1977) have also pointed out the role that involvement in

extracurricular activities can play in the development of a

person's self-concept. They have described the college years








as one of the final steps of adolescents in the process of

identity development and have noted that participation in

extracurricular activities can provide opportunities for a

student to test his or her abilities, values and interests.

This testing can play an important part in developing or

.clarifying a person's self-concept.

In research on participation in extracurricular activities,

Weston and Stein (1977) found that participation in college

activities was definitely related to female students' self-

concept. They noted that participation should be seen as

having at least three dimensions: number of organizations

in which membership was held, degree of involvement within

those organizations, and leadership functions. Extent of

involvement was more of a predictor of identity achievement

in college women than other dimensions of participation,

according to Weston and Stein.

Iffert (1957), Goble (1956), Harnett (1965), Vaughan

(-1968) and others have studied the relationship between

-academic success and participation in extracurricular activ-

iities. Research has indicated not only that the two are

/ mutually-supportive, but has also demonstrated that partici-

pation in extracurricular activities appeared to favorably

"influence chances of academic success in college.

Astin (1975) studied the relationship between persistence

and student participation in extracurricular activities.

He found that participation in extracurricular activities is







significantly related to staying in college. Astin concluded

that his findings supported the theory that student persistence

to some extent depends on the degree of personal involvement

in campus life.


Persistence in College

Studies by Iffert (1957), Astin and Panos (1969) and

others have found no significant differences in attrition

rates for college men and women, although they do not withdraw

for the same reasons. "Researchers have found consistently

that the reasons given for dropping out differ between the

sexes . men tending to cite internal and academic reasons

while women more frequently mentioned external and nonacademic

ones" (Cope and Hannah, 1975, p. 16).

Summerskill (1962) reviewed the literature published

between 1913 and 1957 on student attrition in higher education.

Based on this review of 35 studies, Summerskill reached this

conclusion:

Colleges lose, on the average, approximately half
their students in the four years after matriculation.
Some 40% of college students graduate on schedule
and, in addition, approximately 20% graduate at
some college, some day. (p. 631)

Kauffman (1966) also reviewed the literature on student

attrition, including studies published more recently than

those reviewed by Summerskill. Kauffman was not able to locate

a single conclusive study of attrition in higher education

and found no precise data available on the causes of attrition.

Among the reasons for attrition noted by Kauffman were these:

motivation and expectations and their interaction with institu-

tional environments and values; academic incompetence or






38
unsuitability related to type of institution or major field

of study; financial circumstances; personal or emotional

adjustment problems.

Both Astin (1975) and Cope and Hannah (1975) have

concluded that student attrition and student persistence

in college are primarily a product of experiences of students

in the college environment. Cope and Hannah have concisely

stated their conclusions based on reviews of the literature

on persistence and their own research:

Most quantitative research (such as admissions
data) is without value in either predicting
withdrawal or understanding the reasons for
discontinuance, since such research considers
a limited number of variables, usually easily
quantified, such as high school grade point
averages, college entrance test scores, parents'
income, socioeconomic class background, partici-
pation in high school activities, and the like.
In addition, the reasons for discontinuance are
usually complex, overlapping, often have nothing
to do with the student, and in some instances
may not even be recognized by the student.
Changed vocational choice, poor choice of the
college in the first place, meeting a loved one
and transferring to be with him or her, dissatis-
faction with the college, fulfilling less than
degree expectation, and other multifaceted issues
are involved in withdrawal.

It seems clear too, that there is no dropout per-
sonality, only individual personalities interacting
with different campus environments, at various
times in their mutual and changing lives. (p. 102)

Based on a review of previous studies, Astin (1975)

identified the following student background characteristics

that may be used in predicting which freshmen persist and

which do not. These factors are ability, high school grades,

socioeconomic status, educational aspirations, and the

student's own predictions on chances of his or her completing






39
college. Astin's own research findings supported the theory

that student involvement is a key factor in persistence.

Astin has developed a theory of college persistence which

holds that:

A student's tendency to drop out of college is
inversely related to the degree of direct involve-
ment in the academic and social life of the insti-
tution. Backing this notion are the positive
effects on persistence of participation in extra-
curricular activities, work-study and other on-
campus employment, participation in ROTC and
honors programs, and dormitory living. The
strong relationship between academic performance
and persistence is also, in a sense, additional
support for this theory, given the assumption
that getting good grades is a sign of student
involvement in the academic life and environment
of the institution. (p. 176)

Chickering (1974) has forseen ways in which we can better

deal with the problem of attrition in preventive ways through

use of improved mental measurement techniques:

It will not be long before it will be possible to
assess and to chart individual cognitive maps that
characterize in rich, complex and comprehensive
ways, major modes of perceptual, cognitive, social
and interpersonal functioning. The contours of
these maps will reflect major areas of individual
strength and weakness and by so doing will allow
each individual to conceptualize more clearly
his own unique characteristics, and more importantly,
these contours will provide guidelines by which
each person can take more effective charge of his
own education and personal development. (p. 101)

Chickering has concluded that more definitive information

on the nature of the person, as well as a better understanding

of the consequences of particular academic and environmental

alternatives for persons with particular "contours," may

allow a significant raising of the "success rate" of higher

education beyond the mere 50% of those who enroll in college

and actually reach graduation.








Impact of Different Types of Housing Arrangements
on College Freshmen

Astin (1975), in a multi-campus study using data

collected from 1968-72 through the Cooperative Institutional

Research Program, reported his findings on the impact of

residence on student persistence in his book Preventing

Students from Dropping Out. Astin found that "living in

a dormitory as a freshman is associated with reduced dropout

probabilities" (p. 91). For students in public universities,

Astin found that living in a campus residence hall rather

than other accommodations appears to reduce the dropout rate

among freshman students by about 12%. In analyzing the

effects of living at home with parents and living in other

off-campus facilities, Astin concluded that residence at home

with parents had a negative impact on persistence while

residence in other off-campus accommodations is associated

with increased persistence for men and decreased persistence

for women (p. 93).

Perhaps the most significant research studies conducted

prior to the mid-seventies on the impact of various types

of housing arrangements on college freshmen were reported

by Chickering (1974) in his book entitled Commuting versus

Resident Students. Chickering studied the characteristics

of freshman students who live on campus, those who live at

home, and those who live in other off-campus accommodations.

He compared the groups as to college experiences and sought

to discover what differences, if any, existed in learning

and personal development among the groups.








In studying student characteristics, Chickering used

multiple regression analyses on student responses to a

questionnaire developed by the American Council on Education

(ACE). A random sample of 5351 students was selected from

a population of 38,000 students from a wide variety of

institutions of higher education. The population consisted

of those who had responded to a follow-up questionnaire at

the end of their freshman year. Chickering found that

within the public universities:

Substantial differences occur between commuters and
residents in parental occupation, income and educa-
tional background; in high school achievements,
experiences and activities; and in long-range aspira-
tions; but the two groups are similar in degree
plans. (p. 49)

Multiple regression analyses were undertaken on responses

by students to items on college experiences contained in the

ACE questionnaire. 'For these analyses, a random sample of

5351 students was drawn from a population of 26,806 students

who completed both the initial and the end-of-freshman-year

questionnaires. From his analyses, Chickering concluded that

campus residents have more social experiences with college

peers, show greater persistence in college, and have more

frequent contact with faculty both inside and outside of the

classroom setting than students who lived.at home.

Chickering also reported marked differences in partici-

pation in extracurricular activities between campus residents

and students who lived at home. Of the 27 items analyzed

on the ACE questionnaire related to extracurricular experiences

in college, campus residents scored higher on 24. Students








who lived at home played more chess and watched television

more frequently. Campus residents participated in R.O.T.C.

programs more than those who lived at home. Chickering,

writing about students who live at home, notes:

Compared with dormitory residents, substantial pro-
portions never in the course of their college career
attend a meeting of some college organization, par-
ticipate in student government, attend political
meetings or lectures, or attend a concert, play or
art film. (p. 63)

Data from the questionnaire relating to self-concept

were also studied using stepwise multiple regression analyses

for students living at home. Chickering found that:

In general, therefore, after differences in back-
ground characteristics and difference in self-
perceptions at entrance are taken into account,
students who live at home during the freshman year
rate themselves lower on many important characteris-
tics and abilities, than students who live under
other conditions. Thus, during the freshman year,
the self-esteem of these commuting students suffers
in comparison with their residential peers. (p. 67)

He has found this group significantly lower on self-perceptions

of leadership ability, social self-confidence and popularity

in comparison with freshmen living on campus, who are also

dealing with the problems of transition to college.

Chickering also reported studies that have found major

differences between students who lived at home and campus

residents in their relationships with other students. In

one study, campus residents scored higher than students who

lived with parents on every questionnaire item concerning

social relationships with other students. Based on another

study, he concluded that students who lived at home "were not







as widely acquainted with other students and had fewer close

friends at the college" (Chickering, 1974, p. 64).

Graff and Cooley (1970) investigated whether or not

students living at home and on-campus students differed on

adjustment to college. The College Inventory of Academic

Adjustment was used in this study. The sample consisted of

185 campus residents and 116 students who lived at home.

After-only measures at the conclusion of one semester's

enrollment, holding ability levels constant, revealed one

major difference-that students who lived at home had lower

self-concepts than campus residents. No differences between

the two groups were found to exist in ability or achievement

levels, or their relationships with peers and faculty.

Arbuckle (1957) studied whether there were any significant

differences in the changes which occur among college women

with differing housing arrangements. He used matching

techniques to equate two groups (27 nonresidents and 35

residents) by age, intelligence, socioeconomic background

and religion. Among his findings were that positive changes

occurred in self-confidence among women in both housing groups.

These changes were significant at the .05 level. He also

found significant positive change among those in the nonresident

group, but not in the resident group, in their relationships

with others using the Hestan Personal Adjustment Inventory and

the Gordon Personal Profile.

Astin (1973), using data collected in a multi-campus

study from 1966-70, concluded that campus residence exerts








a consistently positive effect on the self-concepts of

students. He also found through this study that

Living in a dormitory clearly increased the
chances that students would be satisfied with
their overall undergraduate experience, par-
ticularly in the area of interpersonal contacts
with faculty and other students. (p. 207)

Stewart (1969) compared freshmen on one campus who

lived in campus housing and freshmen who lived at home

using a pretest, posttest design. During the first week

of classes in the fall, 281 beginning freshman students,

divided into categories by on-campus or at home residence

and by sex, completed Part I of the College Student Question-

naire. Family social status was indicated to be a major

difference, with campus residents found to have higher

family social status than students who lived at home.

Part II of the College Student Questionnaire was

administered to the sample during the last few weeks of

the Spring Quarter. Significant differences were found

to exist between the groups who lived on campus and at

home in extracurricular involvement, with campus residents

experiencing more involvement. No differences were found

between these groups on a number of other factors, including

satisfaction with relationships with other students.

Barton (1972), using a cross-sectional design, studied

the effects of place of residence upon value development in

college students. With a random sample of 108 freshmen and

149 seniors, Barton also found differences on the Extracurricular

Involvement Scale of the College Student Questionnaire (Part II)

among students according to where they lived. He concluded








that campus residents were more involved in college activ-

ities than students who lived at home, but that no generalized

differences existed between the values of students with

differing housing arrangements.

Stark (1965) and Sprague (1969) found no significant

difference in extracurricular participation among students

with differing housing arrangements. Sprague attempted to

determine whether or not differences existed in certain

intellectual and nonintellectual factors among university

freshmen based on place of residence. His study involved

a sample of 108 freshmen who had been enrolled for a full

academic year. A personal data questionnaire also found no

significant difference in occupational level of parents or

in self-concept among students as a function of either sex

or place of residence while at college.


Research Approaches for Measuring the Impact
of Differing Living Environments

Research studies cited in the previous segment of the

review of literature contain many instances of conflicting

results. Such findings do not provide sufficient basis

for predicting the outcome of further research. Theories

are often useful for predicting the results of further

investigation, yet the conflicting results of prior research

may cause a researcher to be reluctant to make predictions

based on a particular theory. Researchers have recognized

that it is unnecessary for a study of this type to base

anticipated outcomes on any explicit theory. As noted by

Feldman, "The investigator may say something like the








following: 'Here are some interesting dimensions that may

(or may not) be affected by the college experience; lets

compare college class levels to find out'" (Feldman, 1972,

p. 61).

A number of research designs are available for use in

measuring the impact of differing living arrangements on

college students. The cross-sectional design is one "which

evaluates change over time by comparing at the same point in

time different people representing different stages of

development" (Fox, 1969, p. 440). Data can normally be

collected and analyzed more simply and within a short period

of time with this approach. Cross-sectional studies are also

free of an external threat to validity, interaction of testing

and the residence experience. The major weakness of the

cross-sectional approach is its assumption that different

groups are,in fact, comparable.

Another way to measure the impact of differing living

arrangements on students is through the use of a longitudinal

design, which involves studies of the same persons at two

or more points in time. Fox (1969) concisely noted the

advantages and disadvantages of this approach:

The major strength of the longitudinal design
is that it avoids the assumption of compara-
bility of different groups by using the same
respondents at every data-collection interval.
Its weaknesses are the length of time required
to obtain the complete set of data and the
difficulty in maintaining contact with, and
reassembling, the respondents throughout the
length of the study. Some researchers also
add the weakness that continued exposure to
the data-collection instruments and the
research produces a degree of sophistication
or test-wiseness in the respondents which is
reflected in the data. (p. 442)








Feldman (1972) described the two-stage, input-output

approach used by Astin, Chickering and others to study the

impact of different environments:

The background of students entering college
and their values, orientation, and personality
characteristics are considered as input. An
"expected output" based on these input charac-
teristics is computed. The effects of this
expected output is then statistically removed
from students' "observed output," producing a
residual output now independent of input charac-
teristics. Measures of the characteristics of
institutions of higher education are then
related to this residual output to determine the
extent to which they explain variation in the
output beyond that explained by the input
characteristics, thus determining the nature
and strength of college influences. A related
procedure is the following: a stepwise regres-
sion is carried out in which all input variables
are entered into the regression, after which
the environmental variables are permitted to
enter. (p. 211)


Conclusions from the Literature

Freshman students are a diverse group in terms of

academic ability, personality characteristics, interests,

values, and socioeconomic background. Although research

findings cited are not consistent, some evidence has

indicated that campus residents participated more in

activities in high school and may have advantages in

academic ability and socioeconomic background as a group

when compared to college students living at home.

Most freshmen are in the adolescent stage of develop-

ment, a stage which also includes some high school students

as well as some college-age students beyond the normal

freshman age range. This stage is characteristized in part

by inexperience in many areas and decreasing authoritarian





48

behaviors. Increasing tolerance and flexibility of thinking

develops as individuals progress through this stage of

development. Freshmen are said to be, for the most part,

motivated and open to new ideas and challenges.

The self-concepts of college students may often suffer

a slump during the first part of the freshman year, then

may frequently improve during the latter part of that year.

Research studies on self-concept and its relationship to

place of residence are conflicting, but the weight of the

evidence seems to show campus residents as generally posi-

tively affected by the residence experience. The self-

concepts of students living at home seem to be negatively

affected by their continued residence at home while

attending college.

Peer groups are important to individual development.

Research cited is not at all in agreement on whether or

not differences exist among students on campus and students

at home in their relationships with peers. While inter-

action patterns may be the same, students who lived at

home appeared to associate less with other students at

the college and more with noncollege persons than is the

case for campus residents.

Participation in extracurricular activities in college

seems to have a positive effect on academic success and

persistence in college. There is no consensus reported in

the research on whether or not levels of participation are

the same among groups having differing housing arrangements.







Persistence appears to be essentially a product of the

experience of students in the college environment. Research

cited constantly indicated that there is a greater persis-

tence among campus residents than among students living at

home. There are indications that living off campus away

from home has a positive influence on persistence for men,

but the reverse may be true for women.

Based on a review of research approaches, a study of

longitudinal design appears appropriate for a study of

selected educational benefits for student groups representing

three types of housing arrangements on one campus.














CHAPTER III
RESULTS OF THE STUDY


The purpose of this study was to determine whether

there are significant differences in certain educational

benefits (persistence, self-concept, peer relationships,

and extracurricular involvement) for beginning University

of Florida freshmen who live on campus, at home, or in

other off-campus housing. This chapter includes a review

of participant response and findings on the impact of

differing housing arrangements on student persistence at

the University, changes in self-concept during the freshman

year, changes in students' perceptions of their relation-

ships with peers, and involvement in extracurricular

activities.


Response of Sample Participants

The initial size of the sample was 300 students,

including 50 females and 50 males who lived at home, 50

females and 50 males who lived off campus away from home,

and 50 females and 50 males who lived in campus residence

halls. Presented in Table 1 are figures on the number

of persons dropped from the sample and the reasons why

these persons were dropped.







Table 1
Data on Reduction in Number of
Eligible Sample Participants



Students Who
Lived Off Students Who
Students Who Campus Away Lived On TOTAL
Lived At Home From Home Campus
Females Males. Females Males Females Males
Initial Size of Subsamplesa
and Total Sample 50 50 50 50 50 50 300
Persons Dropped from the Sample
for the Following Reasons:
Did Not Enroll at the
University 11 11 12 15 2 1 52
Did Not Remain Enrolled for
the Fall, Winter and
Spring Quarters of the
1976-77 Academic Year 3 9 8 4 7 2 33
Ineligible by Place of
Residence 3 1 0 5 2 6 17
Ineligible by Marital
Status 0 0 1 0 0 0 1
Total Number of Persons
Dropped From Sample 18 21 21 24 11 8 103
Persons Enrolled for all Three
Quarters of the 1976-77
Academic Year and Eligible
by Residence and Marital
Status to Remain in the
Sample 30 27 29 27 37 47 197

aThis row of data indicates the size of the subsample that was selected based on preliminary
classification of the population by residence group. Some reclassification of participants
to other subsample groups occurred once address verification was completed in the Fall of 1976.







Although the sample was drawn from an updated roster

furnished by the University's Office of Admissions shortly

before the start of the 1976 fall quarter, over one-sixth

of the persons in the sample failed to enroll for the fall

quarter and were dropped from the sample. A substantial

majority of those who did not enroll were from the subsample

groups expected to live at home or off campus away from home.

Practically all of the persons in the sample who had made

arrangements for on-campus housing did enroll. Persons

whose places of residence were not included in the definitions

of the three types of residence settings to be compared, as

defined in Chapter One of this study, were dropped from the

sample. Since only unmarried beginning freshmen were to be

studied, the one person who was identified during the course

of the study as being married was also dropped.

Of the 197 students remaining in the sample, 41 (21%)

failed to return pretest materials and another 22 (11%) failed

to return posttest materials. Complete pretest and posttest

data were collected from the remaining 134 students, who

represent 68% of the 197 persons in the sample.

The following sections of Chapter Three are devoted to

the analysis of the research questions listed in the statement

of the problem. In each section, the research question is

restated and pertinent findings are described.


Student Persistence in College

The first research question posed in this study concerned

student persistence: "Is there a difference in persistence at







53

the University among students who lived on campus, at home, or

in other off-campus housing?" The total number of students

who enrolled in the fall, the number who persisted through

the spring and the number who did not persist are presented

in Table 2 by sex and by type of residence.


Table 2
Student Persistence by Residence Group


Students
Living Off
Students Campus Away Students Living
Living At Home From Home On Campus
Females Males Females Males Females Males
Total Enrolled in
September 33 36 37 31 44 49
Total Who Persisted
Through Spring
Quarter 30 27 29 27 37 47
Total Who Did Not
Persist Through
Spring Quarter 3 9 8 4 7 2
Note. N=230


The chi-square statistic was used in the analysis of

these data to determine whether or not there was a significant

difference among these groups on persistence at the University.

Table 3 contains data on persistence comparing males and

females in the sample.


Table 3
Persistence Among
Females and Males


Females Males
Students Who Did Persist
Actual Number 96 101
Expected Number 97.64 99.36
Students Who Did Not Persist
Actual Number 18 15
Expected Number 16.36 16.64








The chi-square statistic was .381 with 1 degree of

freedom, which indicates that there was not a significant

difference in persistence between females and males.

Table 4 provides information concerning student persis-

tence among females in each residence group. The chi-square

statistic was 2.066 with 2 degrees of freedom, which indicates

that there was no significant difference in persistence among

female students that can be attributed to differing housing

arrangements.


Table 4
Female Students' Persistence
by Residence Group


Females
Living Off Females
Females Campus Away Living On
Living At Home From Home Campus
Students Who Did Persist
Actual Number 30 29 37
Expected Number 27.79 31.16 37.05
Students Who Did Not Persist
Actual Number 3 8 7
Expected Number 5.21 5.84 6.95


Table 5 contains information concerning student persis-

tence among males in each residence group. The chi-square

statistic was 8.058 with 2 degrees of freedom, which indicates

that there was a significant difference in persistence among

male students according to where they lived.








Table 5
Male Students' Persistence
by Residence Group


Males
Living Off Males
Males Living Campus Away Living On
At Home From Home Campus
Students Who Did Persist
Actual Number 27 27 47
Expected Number 31.34 26.99 42.66
Students Who Did Not Persist
Actual Number 9 4 2
Expected Number 4.66 4.01 6.34


One may show from the data that, for this sample of

freshman males, those who lived on campus had a signifi-

cantly higher level of persistence at the University and

that those freshman males who lived at home had a signifi-

cantly lower level of persistence among the three groups.

In this study, persistence at the University differed

significantly for males, but not for females, according to

type of residence. Male students who lived at home had a

significantly higher attrition rate, while attrition was

significantly lower for male students who lived on campus.


Change in Students' Self-Concept

The second question posed in this study was: "For

those who persist in college, is there a change in self-

concept among students who lived on campus, at home, or

in other off-campus housing and, if so, does this change

differ on the basis of residence?"

Regression analysis techniques were used to determine

whether or not there was a change in self-concept among

those studied and if so, whether this change differed on








the basis of residence. Three factors (sex, type of residence

and score on the pretest of the Tennessee Self Concept Scale)

were used to predict posttest scores on the Tennessee Self

Concept Scale. The regression model used in this study

appears in Figure 1.


Y=00 + 01Xi + 02X2 + 83X3 + 4X1X3 + 05X2X3
+ Before [6g + 7X1 + 88X2 + 09X3

+ B10X1X3 + B11X2X3] +4

where y = actual value of the posttest score
X1 = 1 if off campus away from home, 0 if otherwise
X2 = 1 if on campus, 0 if otherwise
X3 = 1 if respondent is female, 0 if otherwise
00"-11 are parameters of the model
E is random error

Figure 1
Regression Model





F tests indicated that this model was significant at the

.05 level. The R2 figures calculated for the total positive

and the positive subscales of the Tennessee Self Concept

Scale were high, indicating that the regression model was a

good predictor of posttest scores. R2 figures for the positive

scales ranged from .993 to .997.

If there was no change in self-concept among sample

participants, then posttest scores would remain essentially

the same as pretest scores. The estimates of slope computed

using regression analysis techniques would be close to 1.00

if these scores were equal. As shown in Table 6, none of

the estimates of the eight positive scales computed in this








study was close to 1.00, indicating that there were changes

in self-concept among sample participants during their

freshman year.


Table 6
Estimates for Total Positive and Positive Subscores
on the Posttest of the Tennessee Self Concept Scale


Total Positive .77
Identity .61
Self-Satisfaction .65
Behavior .84
Physical Self .83
Moral-Ethical Self .71
Personal Self .84
Family Self .72
Social Self .87


For each of the sections of the Tennessee Self Concept

Scale used in this study, t tests were conducted on the

hypothesis that each parameter contained in the regression

model equalled zero. The results of t tests for each

parameter for certain subscales of the Tennessee Self Concept

Scale are reported in Table 7. Parameters are listed in this

table in the order of inclusion in the model. Tests are

for single elimination against the full model. These tests

yielded significant results at the .05 level on certain

parameters of four positive subscales: identity, personal

self, family self and social self.








Table 7
Results of t tests on Estimates of Selected Parameters
for the Regression Model


Name of Personal Family Social
8 Parameter Identity Self Self Self
B8 Males at home .003 .367 .034 .433
81 Males off campus .580 .059 .032 .124
82 Males on campus .981 .321 .424 .926
63 Females at home .125 .036 .696 .009
B4 Females off campus .412 .075 .555 .030
B5 Females on campus .038 .066 .363 .197
86 Before/Males at home .0001 .0001 .0001 .0001
87 Before/Males off campus .515 .073 .037 .105
88 Before/Males on campus .884 .414 .563 .988
69 Before/Females at home .147 .047 .747 .013
810 Before/Females off campus .431 .075 .484 .021
811 Before/Females on campus .039 .090 .360 .176

Note. p<.05.


No significant differences in scores could be attributed

to differences in sex or in type of residence as a result

of testing parameters on the total positive section or the

other four positive subscales of the instrument. Similarly

no differences in the total variability score or distribution

score were found that could be attributed to differences in

sex or type of residence.

The first subscale yielding significant results on

certain parameters was the identity scale. The data in

Table 7 show that the t tests for the parameters for this

subscale were significant for o0, 65, 6 and 811.

Figure 2 illustrates the finding that female sample

participants who lived on campus and had low identity scores

on the pretest did not perform as well on the identity scale

of the posttest as other sample participants who had low

identity scores on the pretest. Females who lived on








campus and who had high identity pretest scores performed

almost as well on the identity section of the posttest as

other sample participants with high identity pretest scores.



160


Other sample
participants -


- Females Who Lived
on Campus


20 40 60 80 100 120 140 160 180 200

Pretest Scores


Figure 2
A Comparison of Identity Scores
of Subsample Groups


The second subscale yielding significant results on

certain parameters was the personal self scale. The data

in Table 7 show that the t tests for the paramters for

this subscale were significant for 83, 6g, and B9. Figure 3

illustrates the finding that female sample participants


140


120


a) 100
0

80
U)
00
4-1
-J 60
4O
0

40


20


0


-20 -









with low personal self pretest scores had significantly

higher personal self posttest scores than males with low

personal self pretest scores. For sample participants with

high personal self pretest scores, however, males scored

higher on the personal self posttest than females.


Males -


- Females


40 60

Pretest Scores


100


Figure 3
A Comparison of Personal Self Scores
of Subsample Groups


100






80 -






o 60

U)
U)
440 -
-p
4J







20 -





0








The third subscale yielding significant results on

certain parameters was the family self scale. The data

show in Table 7 that the t tests for the parameters on

this subscale were significant for 0B, 1, Be and 87.

Figure 4 illustrates the finding that sample participants

who lived off campus away from home and who scored low on

the family self pretest scored higher on the family self

posttest than sample participants who lived at home or on

campus and who had low family self pretest scores. For

students with high family self pretest scores, however,

those who lived at home or on campus had higher family

self posttest scores than those who lived off campus away

from home.

80 Students Who Lived
at Home or On -
Campus

Students Who
60 / Lived Off
/ Campus Away
Q/ From Home

n4





0
20-









20 40 60 86 100
Pretest Scores
Figure 4
A Comparison of Family Self Scores
of Subsample Groups








The fourth subscale yielding significant results on

certain parameters was the social self scale. The data in

Table 7 show that the t tests for the parameters on this

subscale were significant for 83, 4, 86, B9 and B10o

Figure 5 illustrates the finding that of the females

with low social self pretest scores, those who lived at home

and on campus scored higher on the social self posttest than

females who lived off campus away from home. Results on the

social self posttest were similar for females who lived off

campus away from home who had low social self pretest scores

and for men who had low social self pretest scores regardless

of residence. Of the students with high social self pretest

scores, females who lived off campus away from home scored

higher on the social posttest than females who lived at

home or on campus and who had high social self pretest scores.



100

m Females Who Lived
k 80 Off Campus Away
0 All Males
o From Home
60 Females Who Lived
-P 60
(U At Home or On
W Campus
W 40
40
o

20


0
20 40 60 80 100 120
Pretest Scores

Figure 5
A Comparison of Social Self Scores
of Subsample Groups





63

The following results can be drawn from the data collected

and analyzed on self-concept:

1. Changes in self-concept occurred among sample participants

during their freshman year.

2. When compared with other students who had low identity

scores on the pretest, females who lived on campus had

lower identity scores on the posttest. However, when

compared with other students who had high identity

scores on the pretest, females who lived on campus

scored about as high on the post test.

3. When those students who had low personal self scores

on the pretest were compared, females had higher

personal self scores than males on the posttest.

4. When compared with students in other residence groups

who had low family self scores on the pretest, students

who lived off campus away from home had higher family

self scores on the posttest. However, when compared

with other students who had high family self scores

on the pretest, students who lived off campus away from

home had lower family self scores on the posttest.

5. When compared with other students who had low social

self scores on the pretest, females who lived on campus

and at home had higher social self scores on the posttest.

Of the students with high social self scores on the

pretest, females who lived on campus and at home had

the highest social self scores on the posttest, followed

by males, who were followed by females who lived off

campus away from home.








Relationships with Peers

The third research question posed in this study was:

"For those who persist in college, is there a difference in

relationships with other students among students who live

on campus, at home, or in other off-campus housing?"

Regression analysis techniques were used to determine

whether or not there was a change in relationships with

peers among sample participants and, if so, whether this

change differs on the basis of residence. Three factors

(sex, type of residence and score on the pretest of the

Barrett-Lennard Relationship Inventory) were used to predict

posttest scores on the Barrett-Lennard Relationship Inventory.

The regression model used in these analyses is the same as

that used in predicting posttest scores on the Tennessee

Self Concept Scale and is illustrated in Figure 1.

F tests of the regression model were conducted. The

R2 figures calculated for the four scales of the inventory

and the inventory as a whole are reported in Table 8. This

table indicates that the regression model was not as good

a predictor of posttest scores on the Barrett-Lennard

Relationship Inventory as it was for posttest scores of the

Tennessee Self Concept Scale.








Table 8
R Values Calculated
for the Barrett-Lennard Relationship Inventory


2
Scales R
Scale 1 .705
Scale.2 .280
Scale 3 .477
Scale 4 .311
Total Scale .538


Table 9 reports the results of t tests conducted on

the hypothesis that each parameter in the regression model

equalled zero. Tests are for single elimination against the

full model. No significant differences (p<.05) in scores

could be attributed to differences in sex or in type of

residence as a result of testing parameters on the four

scales of the inventory as well as the inventory as a whole.

A description of the instrument is included in Appendix B.

The pretest scores for the first and third scales were

highly significant predictors of the posttest scores for

each respective scale. The pretest score for the second

scale, however, was of no value as a predictor of the post-

test score for that scale. Pretest scores for the fourth

scale and for the total scale were significant at the .04

level as predictors of the respective posttest scores.









Table 9
Results of t tests on the Estimates of Parameters
for the Regression Model


Name of
Parameter
Males at home
Males off campus
Males on campus
Females at home
Females off campus
Females on campus
Before/Males at home
Before/Males off campus
Before/Males on campus
Before/Females at home
Before/Females off campus
Before/Females on campus

Note. p<.05.


Scale 1
.9289
.0453
.5415
.0576
.5282
.6985
.0003
.2183
.5710
.1299
.9777
.7246


Total
Scale 2 Scale 3 Scale 4 Scale
.1558 .8570 .3895 .7832
.6035 .6124 .2384 .9847
.6295 .3266 .4452 .1524
.1553 .1044 .8517 .5133
.3987 .5574 .7189 .9650
.0439 .7465 .6386 .6245
.9861 .0090 .0446 .0448
.3648 .6432 .2174 .6891
.9899 .4389 .8020 .2240
.1269 .0788 .8793 .4797
.6561 .7744 .2311 .4137
.0673 .1450 .5253 .1315


Except for protests, no factors or combination of factors

contained in the model as constructed had a significant effect

on the prediction of posttest scores on the scales of the

Barrett-Lennard Relationship Inventory.


Participation in Extracurricular Activities

The fourth research question posed in this study was:

"For those who persist in college, is there a difference in

participation in extracurricular activities among students

who live on campus, at home, or in other off-campus housing?"

A researcher-designed Activities Questionnaire, administered

as a posttest, was returned by 134 sample participants.

The chi-square statistic was used in the analysis of these

data to determine whether or not there was a significant

difference among those in the subsample groups according to

type of residence.








Table 10 contains information from the first section

of the questionnaire, which measured level of participation

in campus organizations. No significant difference among

the groups was found to exist in the level of participation

in student government-related organizations, social frater-

nities or sororities, campus service organization or campus

religious organizations.


Table 10
Level of Participation in Campus Organizations
Reported by Residence Group



Students
Living
Students Off Campus Students
Type of Level of Living At Away From Living
Organization Participation* Homea Home On.Campusc
Student 0 33 26 28
government- 1 11 3 8
related 2 3 2 8
organizations 3 2 4 6
Social 0 43 30 39
fraternity 1 3 2 4
or sorority 2 3 2 2
3 0 1 5
Campus 0 42 28 38
service 1 4 4 9
organization 2 1 3 3
3 2 0 0
Academically- 0 46 26 36
related 1 3 8 7
organization 2 0 1 5
3 0 0 2
Campus 0 38 31 40
religious 1 4 2 2
organization 2 1 2 3
3 6 0 5


*O=not at all,l=to a

Note. N=134
an=49
b
n=35
Cn=50


small extent,2=fairly extensive,3=very extensive








A significant difference was found among the groups in

the level of participation in academically-related organizations,

such as professional clubs and honor societies. Table 11

illustrates the findings that students who lived at home

participated less than was expected (expected levels of parti-

cipation were calculated using chi square). Students who

lived off campus away from home participated more than was

expected (though this was at the "to'a small extent" level

of participation). Campus residents in the sample exceeded

expectations at all levels of participation, particularly at

the "fairly extensive" and "very extensive" levels.


Table 11
Participation in Academically-Related Organizations
by Residence Group


Students
Living
Students Off Campus Students
Level of Living At Away From Living
Participation Response Homea Homeb On Campusc
Actual 46 26 36
Not at all Expected 39.5 28.2 40.3
To a small Actual 3 8 7
extent Expected 6.6 4.7 6.7
Fairly Actual 0 1 5
extensive Expected 2.2 1.6 2.2
Very Actual 0 0 2
extensive Expected 0.7 0.5 0.7
Note. N=134
n=49

n=35
n=50


A significant difference was also found in participation

in the "other organizations" category. Most responses in

this category specified R.O.T.C. or the Gator Band. There








was less participation than expected in such organizations

among students living at home or off campus away from home.

Campus residents greatly exceeded the response that was

expected for participation in such organizations, as is

shown in Table 12.


Table 12
Participation in "Other Organizations"
by Residence Group


Students
Living
Students Off Campus Students
Living At Away From Living
Response Homea Home On Campus
Did not Actual 46 32 39
participate Expected 42.8 30.6 43.7
Actual 3 3 11
Participated Expected 6.2 4.4 6.3
Note. N=134
an=49
b
n=35
n=50


The second section of the questionnaire was designed to

determine whether or not persons in the sample had been

selected for or assumed a leadership role in a campus

organization during the course of their freshman year.

Significant differences were found among the groups by

type of residence on this section. Table 13 shows that

while students who lived off campus away from home held

leadership roles in the numbers expected, students who

lived at home fell short of expectations and students who

lived on campus exceeded expectations.








Table 13
Students' Leadership Roles
by Residence Group


Students
Living
Students Off Campus Students
Living At Away From Living
Response Homea Home On Campus
Was not selected
for or did not Actual 42 23 26
assume a leader- Expected 33.3 23.8 34.0
ship role
'Was selected for
or assumed a Actual 7 12 24
leadership role Expected 15.7 11.2 16.0
Note. N=134
an=49
b
n=35
Cn=50



While all categories of the questionnaire were tested

for differences between females and males, this is the only

category where such differences were found. Table 14 reports

the responses to this section of the questionnaire by sex.

As shown in Table 14, freshman males were selected for or

assumed leadership roles in significantly greater numbers

than freshman females in the sample.


Table 14
Leadership Roles Held
by Females and Males


a b
Response Femalesa Malesb
Was not selected for or did Actual 57 34
not assume a leadership role Expected 50.3 40.7
Was selected for or assumed Actual 17 26
a leadership role Expected 23.7 19.3
Note. N=134
n=74
n=60







The final section of the Activities Questionnaire was

designed to assess participation in other extracurricular

activities, such as intramurals, attendance at campus

cultural and social events, visits to cultural centers on

the campus, employment while enrolled as a student and other

University-related extracurricular activities. The chi-

square statistic was calculated for each of the six

categories in this section of the instrument. A description

of the instrument is included in Appendix C. Significant

differences among the residence groups were found for three

of these categories. The chi-square statistic was also

calculated for females and males on each of the six cate-

gories in this section of the instrument, but no significant

differences were found to exist between the sexes.

The first category that yielded significant results

among the residence groups was the response to the question:

"Do you participate in intramural sports competition on

campus?" The chi-square statistic was 6.115 with 2 degrees

of freedom, which is significant at the .05 level. Table 15

contains the actual and estimated numbers of sample partici-

pants in each residence group who participated in intramural

sports on campus. Data in this table indicate that sample

participants who lived off campus away from home and those

who lived on campus participated in intramural sports

competition on campus in numbers greater than expected.

Participation was less than expected among those who lived

at home.









Table 15
Participation in Intramural Sports Competition on Campus
by Residence Group


Students
Living
Students Off Campus Students
Living At Away From Living
Response Homea Homeb On Campus
Actual 44 35 49
Participated Expected 46.8 33.4 47.8
Did Not Actual 5 0 1
Participate Expected 2.2 1.6 2.2
Note. N=134
a
n=49
n=35
cn=50


The second question in this section in which significant

differences were found asked: "Have you visited the Florida

State Museum or the University Gallery since the start of the

1976 Fall Quarter?" The chi-square statistic was 23.781 with

2 degrees of freedom, which is significant at the .0001 level.

Table 16 contains the actual and estimated numbers of sample

participants in each residence group who visited the museum

or the gallery. The data in this table show that sample

participants who lived on campus and who visited either of

these cultural centers greatly exceeded the number expected.

Those who lived at home who visited the museum or the gallery

fell far short of the number expected. The actual number of

sample participants who lived off campus away from home and

who had visited the museum or gallery equalled the

expected number.








Table 16
Totals of Students by Residence Group
Who Visited the Museum or Gallery


Students
Living
Students Off Campus Students
Living At Away From Living
a b c
Response Homea Home On Campus
Actual 28 27 49
Visited Expected 38.0 27.2 38.8
Actual 21 8 1
Did Not Visit Expected 11.0 7.8 11.2
Note. N=134
a
n=49
b
n=35
Cn'50


The final question in this section in which significant

differences were found asked: "Have you attended a dance or

party on the campus since the start of the 1976 Fall Quarter?"

The chi-square statistic was 7.336 with 2 degrees of freedom,

which is significant at the .026 level. Table 17 contains

the actual and estimated numbers of sample participants in

each residence group who attended a dance or party on campus.

The data in this table indicate that sample participants who

lived at home exceeded the number expected to attend these

social functions and that sample participants who lived off

campus away from home or on campus fell short of the number

expected to participate in these activities.





74

Table 17
Totals of Students by Residence Group
Who Attended a Dance or Party on Campus


Students
Living
Students Off Campus Students
Living At Away From Living
Response Homea Home On Campus
Actual 19 6 9
Attended Expected 12.4 8.9 12.7
Actual 30 29 41
Did Not Attend Expected 36.6 26.1 37.3
Note. N=134
n=49

n=35
Cn='50



The results gleaned from the analysis of responses to

the Activities Questionnaire can be summarized as follows:

Differences among those sampled by residence group occurred

in level of participation in academically-related organizations,

such as professional societies and honor societies. Partici-

pation was highest and most extensive for campus residents in

the sample. Those who lived off campus away from home parti-

cipated more than those at home, but the level of involvement

of both of these groups was low. Differences in participation

in "other organizations," such as R.O.T.C. and the University

Band, were also found among the residence groups. Campus

residents participated in such organizations far more than

was the case among others in the sample.

Students who lived off campus held leadership roles in

the numbers expected (as calculated using chi square).

Students who lived at home fell short of expectations and

students who lived on campus exceeded expectations. Freshman








males were selected for or assumed leadership roles in

significantly greater numbers than freshman females in

the sample.

In other extracurricular activities, a greater propor-

tion of students in the sample who lived off campus away

from home or on campus participated in intramural sports

competition than those who lived at home. Those who lived

on campus visited the Florida State Museum or the University

Gallery in greater numbers than expected, while fewer of

those who lived at home than was expected had visited

the museum or gallery. Of those who lived on campus or

off campus away from home, fewer than was expected had

attended a dance or party on campus; however, a greater

number of those who lived at home than was expected had

attended such a campus social event.


Chapter Summary

This chapter includes a review of participant response

and findings on the impact of differing housing arrangements

on student persistence at the University, changes in self-

concept during the freshman year, changes in students'

perceptions of their relationships with peers, and involve-

ment in extracurricular activities.

An initial sample size of 300 entering freshman students

was reduced by 52 who were admitted, but did not enroll at

the University. This was further reduced by 18 persons in-

eligible by place of residence or marital status to remain

in the sample. The 33 students who did not remain enrolled







for all three quarters of their freshman year were included

only in the study of persistence among residence groups.

Both pretest and posttest data were collected from 134

students, who represent 68% of the 197 persons enrolled for

all three quarters.

In this study, persistence at the University differed

significantly for males but not for females, according to

type of residence. Male students who lived at home had

a higher rate of attrition which was significant at the

.05 level, while attrition was significantly lower for males

who lived on campus. There was nosignificant difference in

persistence between females and males.

The following are among the findings of the analyses

of data on self-concept:

Changes in self-concept occurred among sample participants

during their freshman year. Significant differences at the

.05 level were found to exist according to type of residence

among students who had low identity scores or low personal

self scores on the pretest of the Tennessee Self Concept Scale.

Students also differed on the basis of residence on the

family self scale and the social self scale of the Tennessee

Self Concept Scale.

No significant differences were found in relationships

with other students when comparing those sample participants

who persisted in college by residence group.

Differences were found to exist among students with

differing housing arrangements in participation in extra-

curricular activities. Significant differences at the .05







level were found among the residence groups in level of

participation in academically-related organizations, such

as professional clubs and honor societies; and in "other

organizations," such as R.O.T.C. and the University Band.

Campus residents were involved far more extensively in

college activities than students in other residence groups.

Freshmen who lived on campus held a greater number of

leadership positions in college organizations than freshmen

in other residence groups. This difference was significant

at the .05 level. Freshmen who lived at home held signifi-

cantly fewer leadership positions. Freshman males were

selected for or assumed leadership roles in significantly

greater numbers than freshman females in the sample.

Significant differences at the .05 level were found among

the residence groups on participation in certain other

extracurricular activities, including intramural sports,

visits to campus cultural centers, and attendance at social

functions on campus.













CHAPTER IV
DISCUSSION OF THE DATA


This chapter contains a discussion of the data presented

in Chapter III. These data are examined in relation to the

previous research and theoretical writings included in the

review of the literature.

This chapter is divided into four sections. Each

section is devoted to a discussion of the findings relating

to a particular research question posed in Chapter I.

In the first section, the results of comparing levels of

persistence at the University among subsample groups are

discussed. The second section is devoted to a discussion of

the impact of various residence settings on the self-concepts

of students in the sample. The third section provides a

discussion of the findings of that part of the study which

dealt with relationships with peers. The results of com-

paring subsample groups on their extent of involvement in

extracurricular activities is discussed in the fourth section

of this chapter.


Persistence

The first research question contained in the statement

of the problem asked: "Is there a difference in persistence

in college among students who lived on campus, at home, or

in other off-campus housing?" There was a consensus in the








literature on this question. Research by Astin (1975),

Chickering (1974), and others suggest that campus residence

enhances persistence.

In this study, significant differences in persistence

were found for males, but not for females, according to

type of residence. Male students who lived at home had a

higher attrition rate which was significant at the .05

level, while attrition was significantly lower for male

students who lived on campus. The rate of attrition of

males who lived off campus away from home was in between

the other two residence groups.

For men, the results of this study agree totally with

research by Astin and others. One may show from the data

that the best place for freshman males to live in order to

minimize attrition is in campus residence halls. One may

also show from the data that the chances for dropping out

of college are significantly higher for freshman males who

live at home.

Differences were found in persistence among females

according to type of residence. Although these differences

did not reach the .05 level of significance, they neverthe-

less suggest that living at home during the freshman year

may increase the likelihood that a female student will

persist at the University. Incidence of attrition was

found to be greatest among freshman females who lived off

campus away from home. Results of this study agree in part

with research by Astin (1975) and others, who have found the








environment off campus away from home to be associated with

the highest levels of attrition and campus residence to be

associated with the lowest levels of attrition among freshman

females.

Although the impact of residence setting was shown

through this study to be different for females and males,

this research found no overall difference in persistence

between the sexes. This finding is consistent with previous

research by Panos and Astin (1969), Iffert (1957) and

others, who found no differences in persistence levels

between females and males.

What factors may be operating that would explain the

differences between the sexes on levels of persistence in

various residence settings? Astin (1975) has speculated

that differences in the degree of personal autonomy and

independence given to females and males during their high

school years may be a factor. Parents allow male students

greater opportunities for exercising independent judgment

and experiencing personal autonomy than they allow female

students, according to Astin. He concluded that these

differences are at least partially responsible for the

differing levels of persistence associated with types of

residence for female and male students.

One might expect females and males to react differently

in various residence settings in terms of persistence based

on findings of a review of the literature by Cope and Hannah

(1975). "Researchers have found consistently that the reasons

given for dropping out differed between the sexes . men







tending to cite internal and academic reasons while women

more frequently mentioned external and nonacademic ones"

(Cope and Hannah, 1975, p. 16).

Astin (1975) cited student involvement as a key factor

in persistence. He proposed a theory of college persistence

which holds that a student's tendency to drop out of college

is inversely related to his or her degree of direct involve-

ment in the academic and social life of the institution.

Based on Astin's theory of college persistence, one would

expect that the higher level of persistence found to exist

among campus residents would have resulted from their greater

involvement in campus organizations and activities.

The present study provided this theory to be true at

the University of Florida. Significant differences at the

.05 level were found when students who lived on campus and

students who lived at home were compared in levels of

participation in academically-related organizations, such

as professional clubs and honor societies. Table 11 illus-

trates the finding that students who lived at home partici-

pated less than was expected, while those who lived on campus

exceeded expectations at all levels of participation,

particularly at the "fairly extensive" and "very extensive"

levels. Similar results were found in the "other organiza-

tions" category, which included R.O.T.C. and the University

Band. Table 13 also shows similar results in comparing

these two groups on the number of persons who were selected

for or assumed a leadership role in a campus organization

during their freshman year.







Thus, this study provides additional evidence to support

the proposition that type of residence does have a significant

impact on persistence of college freshmen, specifically male

freshmen. The findings of this study also provide additional

support for Astin's theory of college persistence.


Self-Concept

The second research question contained in the statement

of the problem was: "For those who persist in college, is

there a change in self-concept among students who lived on

campus, at home or in other off campus housing and, if so,

does this change differ on the basis of residence?"

Although there is no consensus in the literature avail-

able on the effects of differing housing arrangements on

the self-concepts of students, most of the research tends

to indicate that campus residents differ significantly

from other students in positive self-concept change.

Graff and Cooley (1970), in comparing campus residents

with students who lived at home, found that those who lived

at home "tended to be more beset by lack of self-confidence"

(p. 56). Astin (1973) also found significant differences

between these two groups and concluded that residence on

campus has a consistently positive impact on self-concept.

Chickering (1974) found that students who lived at home,

in comparison with students who lived on campus, rated them-

selves significantly lower on social self-confidence, popu-

larity, and public speaking and leadership abilities.








Studies by Sprague (1969) and Arbuckle (1957) found no

significant differences in self-concept for students with

differing housing arrangements.

This study found significant differences in certain

aspects of self-concept among students who lived at home,

off campus away from home and on campus. These aspects were:

identity, family self and social self, which are three of

the positive subscales of the Tennessee Self Concept Scale.

Although the same instrument was used in Sprague's study,

the data of the present study failed to support Sprague's

findings.

The identity subscale describes the basic identity

of sample respondents, or who a person is as he or she sees

himself or herself. Scores on the identity subscale differed

significantly for females who lived on campus as compared

with other subsample groups. Females who lived on campus

who had high identity pretest scores scored almost as well

on the identity section of the posttest as other sample

participants with high identity pretest scores. Females

who lived on campus who had low identity scores on the pre-

test, however, did not perform as well on the identity scale

of the posttest as other sample participants who had low

identity scores on the pretest. This particular finding

does not support Astin's conclusion that residence on

campus has a consistently positive impact on self-concept.

Rather, the results of this study indicate that a female

entering college with a poor identity seems to be affected

more positively by living at home or off campus away from

home.





84


One interpretation of the findings of this study is

that peers in the residence setting may have a negative

effect on females with low identity scores on the pretest.

As Segal (1967) noted in his description of the residence

hall setting, "The pressure in this setting is the

confrontation of the students with an intense peer-culture

experience . ." (p. 309). This strong peer influence

may not be what is needed for an entering freshman female

who already has poor perceptions of herself as a person.

Another explanation is that exposure to a wider variety

of persons and values than may have previously been encoun-

tered is a source of inner conflict for a person whose

identity is not yet well-developed. Still another explana-

tion is that the absence of parents meant loss of a critical

support system for which no adequate substitute was avail-

able in the residence halls. This absence of an adequate

substitute may influence the behavior patterns of these women

with poor identity scores. Those students with low initial

identity scores who lived off campus away from home may have

an adequate support system in the small, personally-chosen

peer group with whom they share their apartment.

The personal self subscale of the Tennessee Self

Concept Scale describes persons' sense of self-worth, their

feelings of personal adequacy, and their perception of their

personality apart from their body image or interpersonal

relationships. Significant differences at the .05 level

were found between males and females on this subscale.

Males had higher personal self scores than females. No








differences were found, however, that could be attributed

to differences in place of residence. The fact that males

and females differed significantly on this subscale indicates

that comparisons by sex should be made when analyzing the

impact of differing housing arrangements on students.

The family self subscale describes persons' perceptions

of self in relation to their closest circle of associates.

For students with low family self pretest scores, those

who lived off campus away from home scored higher on the

family self posttest than those who lived at home and those

who lived on campus. For students with high family self

pretest scores, however, those who lived at home or on

campus had higher family self posttest scores than those

who lived off campus away from home.

In order to interpret these findings, the reader must

understand that for those who lived off campus away from

home and for those who lived on campus, the pretest and

posttest instruments for this subscale reflected relation-

ships with two different "circles of associates." For almost

all students in the sample, the pretest involved perceptions

of self in relation to their immediate family prior to the

start of the freshman year. On the posttest, however, the

instrument was completed by campus residents in terms of

their relationships with those living in close proximity

to their rooms in the residence hall. For those off campus

away from home, the posttest instrument was completed in

terms of their relationships with persons with whom they

shared an apartment.







These findings can be interpreted as follows: For

those students who lived off campus away from home who had

low family self pretest scores, perceptions of self in

relation to those with whom they shared an apartment were

much improved over their pretest perceptions of self in

relation to family members at home. For those with low

initial perceptions of self in relation to family members,

the act of having joined a small, personally-chosen peer

group in sharing an apartment off campus proved to be

better than either remaining at home or living on campus

insofar as improvement in scores on the family self post-

test was concerned.

For students with high family self pretest scores,

those who remained at home did better on family self post-

test scores than those off campus away from home. Scores

of campus residents were not significantly different on

the posttest than students who lived at home. Those who

had strong initial positive perceptions of self in relation

to family members did even better on the.posttest of the

subscale if they remained at home or lived on campus.

For those at home, the family support system seemed to

strengthen self-concept as the freshman year progressed.

For those on campus, it may be that this initially strong

self-concept arising from family relationships, or other

factors, enables these students to seek out particular

persons (from the many in their living group) with whom

to build a new, strong peer support system. Those off








campus away from home may have found the choice of the few

persons they shared apartments with to be too limited,.

hampering their ability to build as strong a support system

or "circle of associates" as they found in their family

group.

The social self subscale of the Tennessee Self Concept

Scale describes a person's sense of personal adequacy in his

social interaction with others in a broader manner than in

the family self subscale. Females with low social self

pretest scores who lived at home or on campus did better

on the social self posttest than females with similar pre-

test scores who lived off campus away from home.

This is an interesting finding in light of the previous

discussion of family self scores. Fitts (1965) described

the high intercorrelation of positive scores of the

Tennessee Self Concept Scale and reported a correlation

of .73 between family self and social self scores. Many

students with low family self pretest scores also had low

social self pretest scores. One might expect the living

situation off campus away from home to have a similar impact

on both family self and social self posttest scores of this

group of students. However, this is not the case.

This finding can be interpreted to mean that many of

those living off campus away from home have made a choice

to grow toward maturity largely within the context of a small,

personally-chosen peer group rather than the family setting

at home or the more intense and diverse peer relationships

characteristic of residence halls. This idea of choice




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs