4 v __ —_'——.———v w.—

 

. “.7“

A COMPARISON OF MICHIGAN STATE UNIVERSITY
STUDENTS WHO GRADUATED FROM VARIOUS
SIZE HIGH SCHOOLS ON THE BASIS OF INTELLEC‘flVE
AND AFFECTIVE VARIABLES AND ACADEMIC SUCCESS

Thesis for the Degree of Ed; D.
MICHIGAN STATE UNIVERSITY

Charles Moore :Renneisen
1-964

THESIS

IIIIlulyuzllwgwlwvzuuj”Ilium .

This is to certify that the

thesis entitled

A COMPARISON OF MICHIGAN STATE UNIVERSITY
STUDENTS WHO GRADUATED FROM VARIOUS SIZE
HIGH SCHOOLS ON THE BASIS OF INTELLECTIVE
AND AFFECTIVE VARIABLES AND ACADEMIC SUCCESS

presented by

Charles Moore Renneisen

has been accepted towards fulfillment
of the requirements for

Eh, D, degree in Educagion

12;!” ACZIL J

Major professor

Date AUgUSt 11"11961‘}

 

0-169

 

     
   
   

LIBRARY

Michigan State
University

_._____.__———__,___._
.. __ . _

 

 

 

 

9U JW {2
H” Zgl

" “19$,
.".,1 1

 

 

 

ABSTRACT

A COMPARISON OF MICHIGAN STATE UNIVERSITY STUDENTS
WHO GRADUATED FROM VARIOUS SIZE HIGH SCHOOLS
ON THE BASIS OF INTELLECTIVE AND AFFECTIVE
VARIABLES AND ACADEMIC SUCCESS

by Charles Moore Renneisen

The Problem

 

The purpose of this study was to investigate the

adequacy of the educational development of students,

who graduated from various size high schools and entered
Michigan State University in the fall of 1958. Two
general areas came under our scrutiny: the measurement
of academic preparation through the study of intellective
variables, and a survey of attitudes and values to test
for differences in this area among students from various
size schools. A study of the academic success of these
groups of students was accomplished through a comparison

of final grade point averages in their senior year.

Methodology
In the collection of data, the three instruments

used to test the intellective variables were: The College

 

Qualification Test, the Michigan State University Reading

 

Test, and the Critical Test. The affective variables were

 

Charles Moore Renneisen

tested with Prince's Differential Values Inventory,
Rokeach's Dogmatism Scale and the Inventory of Beliefs.

The statistical method employed in the study was the analy-
sis of variance. The null hypotheses were tested at the
.05 level of significance. The high school size groups
were divided into three categories based on the size of

the graduating class. The small size group represented
schools with less than one hundred in the graduating class.
The medium size schools were those with class enrollments
of one hundred to three hundred and ninty-nine and the high
schools with over four hundred in the graduating class

were listed as large schools.

Findings

The results showed no significant differences in
the size groups on any of the intellective tests when
the students were tested as entering freshmen. Significant
differences were in evidence when the freshman females
‘were tested alone on the Michigan State University

Reading Test. The Inventory of Beliefs and the Dogmatism

 

Scale did not produce significant differences in the
freshman groups but the freshman were significantly
different on the Differential Values Inventory. The

difference was traced to the female students who showed a

tendency for the small school students to be more traditional

Charles Moore Renneisen

in their beliefs over the girls from the medium or large
schools. In the follow-up study of college seniors, we
found non-significant results on all of the four tests;

the CriticalThinking Test and the three affective variable

 

instruments.

In the survey of final college grade point averages,
significant difference appeared in the study of the males,
with the males from the medium size schools succeeding best.
No significant differences were detected in the study of

the females or when the sexes were combined.

Conclusions

 

Our general conclusion is that there appears to be
very little difference among the graduates of the various
size high schools who entered Michigan State University
in 1958. Academically, the intellective tests given to
entering freshman indicated no differences in the
preparation of these students. The affective variables
showed only slight differences. This was in the area of
traditional values versus emergent values. The differences
in this area disappeared by the time the students became

graduating seniors.

A COMPARISON OF MICHIGAN STATE UNIVERSITY STUDENTS
WHO GRADUATED FROM VARIOUS SIZE HIGH SCHOOLS
ON THE BASIS OF INTELLECTIVE AND AFFECTIVE

VARIABLES AND ACADEMIC SUCCESS

BY

Charles Moore Renneisen

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF EDUCATION

College of Education

1964

DOCTORAL GUIDANCE COMMITTEE

Floyd G. Parker, Ed.D., Chairman
James w. Costar, Ed.D.
Fred Vescolani, Ed.D.

Robert H. Horwitz, Ph.D.

ii

ACKNOWLEDGMENTS

The writer would like to express his sincere thanks
to the many persons who have assisted him in completing
this work. Particular recognition should be given to
Dr. Floyd G. Parker the major advisor. His kind words
of encouragement and assistance were especially valuable.

To the members of the guidance committee, Dr. Fred
Vescolani,Dr. James w. Costar, and Dr. Robert H. Horwitz,
the author gives his heart felt thanks for their time
and gfiIdance. To Dr. Stanley Hecker and Dr. Alan Grimes,
the Author wishes to express his appreciation for their
substitutions for members of the guidance committee who
were out of the country during the oral exams. The
author wishes to thank Dr. Stanley 0 Ikenberry of
Michigan State University for the use of the test results
of the major study of the 1958 freshman class at Michigan
State University.

Finally, the writer is deeply appreciative of the

ever;present help of his wife, Phyllis.

. I t
‘4: {:3 .

iii

TABLE OF CONTENTS

Page
ACKNOWLEDGMENTS . . . . . . . . . . . . ii
VITA . . . . . . . . . . . . . . . . ' vi
LIST OF TABLES . . . . . . . . . . . . . vii
Chapters
I. PREFACE . . . . . . . . . . . . l
The Setting . . . . . . . . . l
The Problem . . . . . . . . . 5
II. REVIEW OF THE LITERATURE . . . . . . 8
High School Size and Academic Success
at College . . . . 9
Professional Opinion Concerning the
Optimum Size School . . . . 17
Research Relating High School Size to
the Development of Attitudes and
Values . . 21
Research Dealing With Attitude Change
During the College Years . . . 2A
III. METHODOLOGY . . . . . . . . . . 28
Instruments Used in the Study. . . . 28
Collection of the Data . . . . . . 37
Population Under Study . . . . . . 38
Selection of the Sample. . . . . . 39
Summary . . . . . . . . . 39
IV. ANALYSIS OF RESULTS--INTELLECTIVE
VARIABLES . . . . . . . . . 42
Survey of College Freshmen. . . . . 42
Survey of College Seniors . . . . . A8

iv

V. ANALYSIS OF RESULTS - AFFECTIVE VARIABLES.

Affective Variables at the Beginning‘
of College Careers . .

Affective Variables after Four Years of
College . . . . . . . . .

VI. ANALYSIS OF RESULTS COLLEGE GRADES
Summary of the Results . . . . .
VII. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

Summary of the Study . . .
Related Research . . . .
Methods Employed .

Findings. . .

Conclusions.

Recommendations

BIBLIOGRAPHY. . . . . . .

Page

51

51
57
6A
67
68
68
69
7O
72
73
76

78

Charles Moore Renneisen
Candidate for the Degree of
Doctor of Education
Final Examination: July 7, 1964, 10:00 A.M. Room 419
Erickson Hall
Dissertation: A Comparison of Michigan State University
Students, Who Graduated from Various Size
High Schools, on the Basis of Intellective
and Affective Variables and Academic Success

Outline of Studies:

Major Area - Administration
Minor Area - College Personnel Work and Political
Science

Biographical Items:
Born - February 4, 1930, Jasper, Indiana
Undergraduate Studies - Indiana University 1950—1954

Graduate Studies - Ball State College, Indiana
University, l955w1959

Job Experiences - Member of Armed Forces 1948-1950;
Teacher and coach Johnson TWp.
High School, Walcattville, Indiana
1954-1956; teacher, coach and
athletic director St. Mary's High
School, Anderson, Indiana, 1956-
1960; Head Resident Advisor,
Michigan State University, 1960-
1963; Dean of Men Saint Louis
University, 1963-1964.

vi

Table

LIST OF TABLES

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Subjects
Each, Representing Various Size High

Schools, Tested with the College Qualifi-
cation Test. . . . . . . . . I. . .

Analysis of Variance for Three Groups of
College Freshman, of One Hundred Students
Each, Representing Various Size High

Schools, Tested with the Critical Thinking
Test . . . . . . . . . . . . . .

Analysis of Variance for Three Groups of
College Freshmen of One Hundred Students
Each, Representing Various Size High
Schools, Tested with the Michigan State
University Reading Test. . . . . . .

 

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Students
Each, All Malee Representing Various Size
High Schools, Tested With the College
Qualification Test . . . . . . .

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Students
Each, All Males, Representing Various Size
High Schools, Tested with the Critical
Thinking Test . . . . . . .

 

 

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Students
Each, All Males, Representing Various Size
High Schools, Tested with the Michigan
StatefiUniversity Reading Test . .

 

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Students
Each, All Females, Representing Various Size
High Schools, Tested with the College
Qualification Test. . . . . . .

 

vii

Page

43

43

44

45

45

46

47

Table
8.

IO.

11.

12.

l3.

14.

15.

16.

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Students
Each, all Females, Representing Various Size
High Schools, Tested with the Critical
Thinking Test . . . . . . . . . . .

Analysis of Variance for Three Groups of
College Freshmen, of One Hundred Students
Each, all Females, Representing Various Size
High Schools, Tested with the Michigan State
University Reading Test. . . . . . .

Analysis of Variance for Three Groups of
College Seniors, Representing Various Size
High Schools, Tested with the Critical
Thinking Test . . . . . . . . . . .

Analysis of Variance for Three Groups of
College Seniors, Males only, Representing
Various Size High Schools, Tested with the
Critical Thinking Test . . . . . . . .

Analysis of Variance for Three Groups of
College Seniors, Females Only, Representing
Various Size High School, Tested with the
Critical Thinking Test . . . . . . . .

Analysis of Variance for Three Groups of
College Freshmen, of Two Hundred Students
Each, Representing Various Size High Schools
Tested with the Inventory of Beliefs .

Analysis of Vairance for Three Groups of
College Freshmen, Two Hundred Students Each,
Representing Various Size High Schools,
Tested with the Differential Values Inventory

Analysis of Variance for Three Groups of
College Freshmen, Two Hundred Students Each,
Representing Various Size High Schools, Tested
with the Dogmatism Scale . . . . . . .

Analysis of Variance for Three Groups of
College Freshmen, One Hundred Students Each,
Representing Various Size High Schools,
Tested with the Inventory offiBeliefs .

viii

Page

47

48

49

49

50

52

52

53

54

Table

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

Page

Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each
Respresenting Various Size High Schools,

Tested with the Inventory of Beliefs. . . . 54

Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each
Representing Various Size High Schools,

Tested with the Dogmatism Scale . . . . . 55

Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each,
Representing Various Size High Schools,

Tested with the Inventory of Beliefs. . . . 56

Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each,
Representing Various Size High Schools,

Tested with the Differential Values Inventory. 56

Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each,
Representing Various Size High Schools,

Tested with the Dogmatism Scale . . . . . 57

Analysis of Variance for Three Groups of

College Seniors, Representing Various Size

High Schools, Tested with the Inventory of

Beliefs . . . . . . . . . . . . . 58

Analysis of Variance for Three Groups of

College Seniors, Representing Various Size

High Schools, Tested with the Differential

Values Inventory . . . . . . . . . . 58

 

Analysis of Variance for Three Groups of
College Seniors, Representing Various Size
High Schools, Tested with the Dogmatism Scale. 59

Analysis of Variance for Three Groups of

College Seniors, All Males, Representing

Various Size High Schools, Tested with the
Inventory Beliefs . . . . . . . . . . 60

Analysis of Variance for Three Groups of

College Seniors, All Males, Representing

Various Size High Schools, Tested with the
Differential Values Inventory . . . . . . 6O

ix

Table Page

27. Analysis of Variance for Three Groups of
College Seniors, All Males, Representing
Various Size High Schools, Tested with
the Dogmatism Scale . . . . . . . . . 61

28. Analysis of Variance for Three Groups of
College Seniors, All Females, Representing
Various Size High Schools, Tested with the
Inventory of Beliefs. . . . . . . . . 62

29. Analysis of Variance for Three Groups of
College Seniors, All Females, Representing
Various Size High Schools, Tested with the
Differential Values Inventory. . . . . . 62

30. Analysis of Variance for Three Groups of
College Seniors, All Females, Representing
Various Size High Schools, Tbsted with the
Dogmatism Scale . . . . . . . . . . 63

31. Analysis of Variance for Three Groups of
College Seniors, Representing Various Size
High Schools, Compared on the Basis of
Their Final Grade Point Averages. . . . . 64

32. Analysis of Variance for Three Groups of
College Seniors, Males Only, Representing
Various Size High Schools, Compared on the
Basis of Their Final Grade Point Averages. . 65

33. Analysis of Variance for Three Groups of
College Seniors, Females Only, Representing
Various Size High Schools, Compared on the
Basis of Their Final Grade Point Averages. . 66

CHAPTER I
PREFACE
The Setting

Many assumptions are made by school patrons and
educators in regard to the fundamental causes of college
student mortality. One that is often heard is the lack of
adequate preparation provided by the secondary schools.
Particular criticism has been leveled at the small high
school. Dr. Clifford B. Smith,1 of Ohio State University,
concluded from his study of Ohio secondary schools that
schools with fewer than two hundred students are paying a
premium for an inferior program. Howard A. Dawson, of the
National Commission of School District Reorganization,
stated:

Studies of the services needed for a satisfactory
program of education through the high school show
that these cannot be provided most economically
and effectively by a district having fewer than

10,000 students. The minimum for a high school
should be 1,200 to 1,500 students.

 

lClifford B. Smith, "A Study of the Optimum Size
Secondary School" (Unpublished Doctoral Dissertation, Ohio
State University, 1960.)

2Howard A. Dawson, et al., Your School District,
National Commission of School District Reorganization,
Department of Rural Education, (Washington, D.C.; National
Education Association, 1948). i

 

 

Critics of the small school generally point to its
limited curricular offerings, overworked and poorly trained
teachers, lack of adequate supervision of instruction, and
insufficient laboratory equipment as some of the reasons
behind their efforts to consolidate these smaller high
schools. Dr. Conant,1 in his study of the nation's high
schools for the Carnegie Foundation, observed that 17,000
of the 21,000 high schools in the United States are too
small to provide an adequate variety of courses.

In spite of the strong opinions of some of our
leading educators that we eliminate the small high school,
the consolidation programs have been progressing at a
slow rate. In Michigan, in 1963-64 approximately forty-
three percent of the high schools had enrollments of four
hundred or less.2 Ninth-two schools had less than two
hundred enrolled on a regular basis. Even allowing for
the sparsely populated Upper Peninsula, there still remains
a large number of districts in the Lower Peninsula which
have desired to maintain the status quo.

Advocates of the small high schools often express the

belief that the proper development of attitudes and values

 

1James B. Conant, The American Hi h School Toda
(New York: McGraw-Hill Book Co., 19595, p. 150.
2Michigan Education Directory Service, Michigan
, Lansing

Education Directory and_§uyer's Guidey31963-6
Michigan.

depends on small classes and close personal relationships
between the teachers and the students. Saupe,l in a study
of University of Missouri students, concluded:

The students of the smaller schools may have

acquired superior habits of study which stand

them in good stead when they apply them to

their tasks as university students.

E. L. Thorndike points out some social disadvantages
as schools become too large:

Teachers do not know one another. Pupils have less
chance of becoming humanized and more danger of
becoming institutionalized. Democracy loses an
effective helper. Athletics become a question of
finance rather than play. The boys mimic college
fraternities aad men's clubs in their social
organizations.

The large schools seek to develop the "whole child"
through specialized personnel, such as guidance counselors,
deans, psychologists and homeroom teachers. Archibald
Shaw3 relates that the Greeks had another kind of teacher,
separate from the "schoolmaster". He was called the
pedagogue. He was responsible for the development of
attitudes, habits, and attributes. Today, we assign both
the role of the schoolmaster and the pedagogue to the

teacher. Shaw explains:

 

lMildred w. Saupe, "Size of High School as a Factor in
the College Success of Average and Superior Graduates,
Journal of The American Association of Collegiate Registrars

 

(October, 1941).

2E. L. Thorndike, "A Neglected Aspect of the American
High School," Educational Review, 33:245-255.

3A. B. Shaw, "A New Look at Secondary Education: A
Random Fall Idea,‘ School Executive (March, 1956), 75:47-86.

Putting more cross grids of support under the
student - vocational guidance counsellors,
psychologists, students personnel services -
may meet the need, but it is our hunch that
the very specialization of personnel has

what Toynbee calls "awkward consequences
worth attention" - and that it willlcost

as much and require as many people.

And so the arguments go. It often appears that
local conditions and sentiment are the real roadblocks
to an objective study of the question of optimum school
size. It is the purpose of this investigation to develop
a more scientific argument in our approach to consolidation
in the state of Michigan. Along with an analysis of
academic abilities and success at college of students from
various size high schools, we aim to determine if there
are measurable differences in the attitudes and values of
college students from these schools.

Since schools today are becoming increasingly con-
cerned with their responsibility in personality development
of the student, the information gathered in this investiga-
tion will be of assistance to school counselors in their
analysis of student adjustment problems. With the aid of
elaborate advisory programs in the residence halls and

counseling center, colleges seek to guide the student into

a gradual adjustment to the rigors of higher education.

 

lShaw, op. cit., p. 74.

The Problem

The problem to be investigated concerns the adequacy
of the educational development of students, from various
size.high schools, who matriculated to Basic College at
Michigan State University in 1958. The present study
sought to approach the analysis of educational development
from four aspects. First, as beginning freshmen, the
groups were compared on the basis of scores made on three
intellective tests. At the same time, the students were
given three instruments which attempt to measure the affec-
tive variables of attitudes and values. A follow-up study
was carried out four years later on those students who were
now graduating seniors. A battery of four tests were
administered at this time to collate changes which might
have taken place as a result of the college experience.
Finally, a comparison was made of the cumulative grade
point averages of the various groups, after four years of
college work.

The importance of this study is emphasized by the
need for more information about the products of our various
size secondary schools. Parents and voters are charged
with the responsibility of deciding at the polls the ques-
tions ofnconsolidation and reorganization of school dis- -
tricts.’ Many times their judgments are based on bias,

sentiment and prejudice rather than factual evidence.

The question of what is the optimum size high school
is constantly in the minds of school administrators. They
endeavor to provide the best possible education at the most
reasonable cost. As it is pointed out later in our review
of the research and literature on this subject, the opinions
of these educators vary widely when questioned as to what
they thought would be the optimum size school.

College admissions officers are continually striving
to improve their selection procedures. Is it true, as is
sometimes proposed, that the graduates of the small high
school find it difficult keeping up with their peers from
the larger high schools when in competition in the college
classroom? Should the admissions officers weigh the size
of the high school in working out a formula for comparing
the college applicants?

College guidance counselors are also interested in
the information of this kind of study. They wish to know
more about those factors that can be related to college
success. Should they give additional guidance to a certain
group of students, such as those from the small high
schools, particularly during their first semester at
college? Do students from the large high schools exhibit
attitudes which some colleges would seek to change by
organized means? These, and other similar questions, need
to be answered in our ever increasing program of guidance

for the college student.

Residence hall advisory staff members can likewise
use this type of information in the development of programs
which assist the student in achieving maximum success at
college. These guidance workers are particularly concerned
with the development of socially accepted attitudes and
habits. They point out to their advisees the handicap
that certain unfavorable attitudes would be in various
life situations.

Last, but not least, the college student himself
can directly profit by having additional information
which would assist him in understanding his possible
weaknesses and strengths. To be prOperly forewarned that
his previous experiences have not been the equal of some
of his college competitors would be an asset to the be-

ginning freshman.

CHAPTER II
REVIEW OF THE LITERATURE
Introduction

A multiplicity of approaches has been used over the
years by researchers in attempts to evaluate the effects
of high school preparation upon college achievement. In
this study, we are concerned with those investigations
which relate size of high school to future academic success
at college and, also, the relationship of size to the
development of values and attitudes.

In a democracy, the proficiency of an educational
program is to be judged by both its achievements in the
scholastic area and its concern for the training of effec-
tive citizens and public leaders. The interest of educators
in the proper development of attitudes and values is there-
fore understandable. This study will investigate whether
or not the size of the high school had an effect upon
these characteristics with those students who went on to
college.

The uniqueness of this particular investigation is
based on the following points: the sample size is large

in comparison to some of the previous studies; this is the

8

first major study with Michigan schools; a longitudinal
study with a four~year look at the sample; improved
statistical methods; and a three group comparison for a
clearer analysis and application of the statistics to
practical situations. .

The review of the literature has been divided into
four main categories:

a. Statistical research relating high school size
and academic success at college

b. Professional opinions concerning the optimum
size secondary school

c. Research relating high school size to the
development of attitudes and values

d. Research dealing with attitude change during
the college years
High School Size and Academic
Success at College

The review of previous research relating high
school size and college success reveals an interesting
pattern of results. In the earlier studies, when students
from various size high schools were compared on the basis
of college achievement, the results showed significant
differences. Four of the five studies reviewed, showed
results favoring the large size high school. One study
showed preference for the small school, while none in-

dicated the medium size to be best.

10

On the other hand, in the five more recent investi-
gations, there were no significant differences in college
grade point averages of the various high school size

groups.

Earlier Studies:

In 1911, Pittenger surveyed the freshman grades of
the within state students at the University of Minnesota.
His conclusions were that the graduates of the larger
high schools might be expected to do better than those from
the small schools, but he cautioned his readers against
making too great a generalization from these findings:

It is probable that the size of a high school
exercises only an indirect influence upon the
college efficiency of its graduates, through
other more vital factors which are correlated
with it. Size is but the sign of their
presence or absence. A large school, for
instance, is generally better equipped than a
small school, and usually employs teachers
with broader training and experience. Both
these facts, and probably others, should make
for higher efficiency of its graduates in
college.

The next major study on this question was performed
a little over a decade later, in 1922, by Thornburg2 at

the State College of Washington. His study, like

 

lBenjamin Floyd Pittenger, The Sixteenth Yearbook of
the National Society of Education Part II (Bloomington,
Illinois: Public School Publishing Company, 1919), p. 10.

2L. H. Thornburg, "College Scholarship and Size of
High Schools,” School_and Society (August, 1924), 20:189.

ll

Pittenger's, dealt only with freshman grades. He con-
cluded that students from the large high schools achieve
superior grades at college. Some technical questions
could be raised concerning the scientific findings of this
particular study because the level of significance was not
reported in the statistics. In spite of the fact that his
tables indicated higher intelligence test scores for the
larger school groups, Thornburg generalizes that the
accomplishments of the students from the larger high
schools must be due, to a great extent, to the training
which these schools gave.
Size of high school and length of school year were
both examined for any relationship to academic success
at college by Stalnaker and Remmers of Purdue. After
their 1930 investigation, they made the following
generalizations:
a. The size of high schools in Indiana is
significantly related to elimination from
Purdue University

b. Length of school term is also significantly
related to elimination from Purdue University

0. First semester freshman grade averages show
a very slight though significant relationship
with size of high school

d. First semester grades also show a tendency to be
related to length of high school term although
the relationship is not clearly statistically
significant.

 

1J. M. Stalnaker and H. H. Remmers, "What Kind of High
Schools Contribute to College Failures?” Bulletin of Purdue
University (March, 1930) 30:23.

 

12

By grouping the various size high schools into cate-
gories of large, medium and small, Brownl noted that the
larger high school students obtained better first semester
grades in twenty Michigan colleges in 1928. She remarked
that the trend was not consistent when additional bread-
downs were made.

The final investigation of the reported earlier
studies was one made by Saupe at the University of Missouri
in 1941. She limited her examination of freshman grade
records to average and above average students. Her findings
showed the average freshman from the smallest high schools
earned a higher mean scholarship index than the average
freshman who graduated from any of the classes of larger
high schools. She theorized from this:

The students of the smaller schools, who lack
diversified learning experiences, especially
those of the average range, may have acquired
superior habits of study which stand them in
good stead when they apply them to their tasks
as university students. It is possible that
in the smallest schools, classes are smaller
and individual students receive greater
personal attention, and thus they are directed

into practices which result in superior study
habits.2

 

1Ruth A. Brown "A Study of High School and First
Semester College Records of Freshmen Entering Twenty Colleges
in Michigan in 1928," Journal of the Michigan Schoolmasters'
Club (1930), p. 18.

2Saupe, op cit.

13

Recent Studies:

In 1949 at Purdue, White1

made a study of the
University's admission criteria. His findings revealed
that size of high school had no relationship to college
success at that state university.

Shafer,2 in a study of students entering certain
Iowa colleges, came to the same conclusions as did White.
This investigation was carried out in 1956. The size
groupings used by Shafer were as follows: 0-99; 100-199;
200-499; 500-999; and 1,000 and above.

Similar research was produced at Kansas State in
1956 by Donald Holt.3 Using the freshman class, he
attempted to answer the following questions: How does
high school size compare in terms of the ability of the
various graduates to do college work? Should the high
school rank be interpreted differently for various size
schools? His conclusions were:

1. In terms of college potential (ACE scores),

no differences existed among the various
size high school graduates.

 

lHarland w. White, "A Study of the High School Record
as a Criterion for Admission to Purdue University” (Office
of Admissions, Purdue University, 1951), p. 25.

2Dwight Shafer, "Analysis of Certain Factors in the
High School Preparation of High School Graduates Entering
Selected Iowa Colleges" (unpublished Doctoral Dissertation,
University of Iowa, Iowa City, 1956), p. 121.

3Donald P. Holt, "Size of High School and College
Grades," Personnel and Guidance Journal (April, 1959),

37:569.

 

l4

2. In terms of first year grades, no differences
existed among the various size high school
graduates.

3. For males, only the small high school graduates
had higher high school rank.

4. Wide sex differences in first year grades,
with the females higher, but since no differences
occurred in the ACE scores, this might indicate
a difference due to non-intellectual factors.

5. The high school record was a good predictor for
all groups.

6. Inconsistencies in research suggests institutional
differences.1

In 1957, Altman reviewed the research concerning high
school size and college success and concluded:

There appears to be no disagreement among writers
that rank in class or scholarship are factors
important to academic success in college, but
size of high school as a predictor of success

is still questioned.

Most of the previous researchers concerned themselves
with the study of freshman grade point averages as related
to high school size. Altman, on the other hand, decided
to investigate the long range effects by studying the grade
point averages of college seniors. He divided the high

schools into four divisions, similar to those used by the

Michigan State High School Athletic Association:

 

lIbid., p. 570.

 

2Esther Royal Altman, "The Effect of Rank in Class and
Size of High School on the Academic Achievement of Central
Michigan College Senior Class of 1957,” Journal of Educa-
tional Research (April, 1959),52:3o7.

15

glaee Enrollment Sample Size
A 900 plus 42

B 375-899 54

c 175-374 37

D under 175 10

From the results he obtained, Altman concluded that
the graduates of the larger high schools did not achieve
significantly higher college grades than did the graduates
of the smaller high schools. Criticism of this journal
report might be leveled at: the fact that the statistical
methods were not reported; the level of significance was
not given; and the sample size, particularly of the small
size schools, was definitely limited. '

The most complete research study in this field was
made by Young1 at Purdue in 1958. He included in his
investigation a statistical control of scholastic aptitude.
Another feature of Young's study was his separation of the
student groups into college majors. His subjects were the
1952 freshman class at Purdue.

In the results of his study, Young had this in-

teresting comparison:2

 

vfi , v—yi

1John Frederick Young, "A Comparison of the Academic
Preparation and Achievement of College Students from Various
Size H gh Schools in Indiana" (unpublished Doctoral
Digsgrtation, Purdue University, West Lafayette, Indiana,
19 O .

2Ibid.

l6

 

% that completed % that changed

Size of HS 8 semesters majgng
0—99 44.1% 17.6%
100—199 40.7% 22.1%
200-299 42.5% 18.9%
300-499 41.4% 21.9%
500-999 41.9% 14.4%
1000-plus 47.4% 18.6%

Young concluded that no significant relationship
existed between high school size and:

1. times on grade probation

2. number of courses failed

3' ’B‘l‘fiiirsé’fieii’é‘is‘eii TERI“

. gr e n ex

The results of this investigation showed that
students from smaller high schools did not do as well as
those from the large high schools in first semester grade
averages in the Engineering School, but in the Schools of
Science, Education, Agriculture and Home Economics, no
differences were in evidence.

In 1962, Harmon2 reported on the relationship be-
tween doctorate productivity and size of high school
graduating class. The author pointed out that apparently
something was happening in the high schools to differen-

tiate the people who, more than a decade later, will earn

doctorate degrees in various scholarly fields. He

 

11bid., p. 99.

2Lindsey R. Harmon, "On Decision Making in High
School," National Association of Secondary School Princi-
pals Bulletin (November, 1962), 46:277.

17

concluded that size of high school, as reflected in size
of graduating class, has a profound effect on the
probability of an individual's going on to college, to

graduate school, and eventually to the doctorate degree.

Professional Opinion Concerning the

Optimum Size SchooI

While a wide range of suggestions have been made
in this field over the years, consensus of professional
opinion, as to the adequacy of the various size high
schools, seems to favor the medium size.

Gaumnitz and Tompkins of the U.S. Office of Education
wrote:

There seems to be some recognition that the very
large high schools tend to a mass production type
of education which is undesirable. No clear
maximum or minimum size high school has been
established as a result of research, but definite
ideas are emerging to suggest the high school
ranging in enrollmenf from 300 to 1,200 tend to
include the optimum.

Dr. James B. Conant, a noted authority in this area
of education, after a nationwide studybe secondary schools
and their problems, concluded that the small high school
was not providing the rich, educational experiences that

are available in the larger schools. Although he did not

 

1Walter H. Gaumnitz and Ellsworth Tompkins, "How
Large Are Our Public Schools?" Federal Security Agency,
Office of Education, Circular No. 304.

18

suggest the optimum size school, he proposed that the
minimum size of "one-hundred in the graduating class" be

established.1

One can also find strong feelings among educational
writers that the large high schools are not the ideal
solution. In their well-known book, Educational Adminis-
tratipn, Concepts, Practices and Issues, Morphet, Johns
and Reller wrote:

The school should be large enough to make available
necessary specialized competencies and services at
reasonable cost. It should be small enough to
facilitate the development of competencies that
may result from genuine involvement in the life of
a society. The desirable size of schools has

not been determined through any rigidly scientific
procedures. However, it would appear that there
is rather widespread agreement concerning approx-
imate desirable ranges of population of schools.
Senior high schools with a populatipn of 400 to
1,500 would appear to be desirable.

The school superintendents themselves seem to
favor a lower minimum and a slightly lower maximum. In
an editorial survey conducted by the Nation's Schools
magazine in 1954,3 it was reported that the largest

percentage of administrators, thirty-seven per cent,

fi—

1Conant, op cit.

2Edgar L. Morphet, Roe L. Johns and Theodore L. Reller,
Educational Administrationy Concepts, Practices and Issues
IEnglewccd Cliffs, N.J.: Prentice—Hall, 1959)

3"What Size School is Best?" Opinion Poll, Nation's
Schools (October, 1954), 54:59.

19

favored the size category of 150 to 400 enrollment.
Combining categories in their study, we find that the
vast majority, ninty-two percent, believed the optimum
size fell in the range of 150 to 1,200. The groupings

and percentages were as follows:

 

 

School Size Percentage Approving
Less than 150 3%
150—400 37%
400—750 31%
750-1,2oo 24%
l,2OO-2,OOO 5%
over 2000 0%

The greatest significance of this survey might be
found in the fact that none of the superintendents voted
for the category of "over 2,000," which is the size of the
vast majority of schools in metropolitan areas.

In 1949, Oliver1 surveyed school superintendents and
state department field consultants on the quest of the
most desirable size high school. He reported that 79.5%
of these educators indicated preference within the range
of 500 to 1,000 enrollment.

Through the years, various state departments of
education have conducted studies to determine the optimum

size high school, or the minimum size allowable for an

 

1A. I. Oliver, "How Big Should the Small School Be?”
School and Society (1949), 69:128.

2O

adequate education. The Indiana School Study Commission1
recommended that the minimum size high school have a
membership of approximately 300.

A complete analysis of size, cost and educational
opportunity in secondary schools of New York was compiled
by Kowitz and Sayres. They concluded:

The results of the present study would seem to
indicate that schools with a secondary enrollment
of less than 500 pupils are operating at a
disadvantage in terms of certain indices of
educational opportunity and are operating
uneconomically compared with schools somewhat
larger. The most economical size for the
secondary school would seem to lie between 600
and 800 pupils. In this interval also, the
indices of educational opportunity showed 2
generally to greatest advantage relative to cost.

Shannon3 argued against the small school approach
to education on the basis of the limited program Offerings
and the excessive cost. He believes that the only way to
keep adequate administrative and supervisory services
with disproportionate costs is to have comprehensive high

schools.

 

lIndiana School Study Commission. "An Evaluation of
the Indiana Public Schools,” Commission Report (Indianapolis
Indiana, 1949), p. 332.

2Gerald T. Kowitz and William C. Sayres, Size, Cost
and Educational Opportunity in Secondary Schools, Division
of Research, New York State Educatior Department, Albany,
New York, p. 71.

3J. R. Shannon, "Criteria for Determining Local
Units of School Administration," American School Board
Journal (October, 1949), 64:28.

21

Schools below a certain limited number of students
have been censured by various professional organizations.
The American Association of School Administrators'
Yearbook, 1958,1 emphasized the lack of well-prepared
teachers as the main disadvantage of high schools with
under 300 enrollment. The exodus of good teachers to the
larger schools was attributed to salary increases and the
desire to teach full-time in their preferred areas.

Research Relating_High School Size to the
DevelOpment of Attitudes and Values

Psychological considerations must be recognized in
determining the effectiveness of various size high schools.
Small school advocates have consistently argued against the
impersonal approach they contend is inevitable with large
enrollments.

Livingston2 emphasized this point in his article in
Progressive Education in 1956:

Students of adolescent development have stressed
the need for a wide variety of educational
experiences on the part of adolescents, the

need for such youth to identify and become a
part of their school, and the need for personal
guidance of other than a vocational nature.

The general business of educating children

and youth can best be carried out in an atmos-
phere of friendly cooperation. Some of the

 

lEster Royal Altman, ”The High School in a Changing
World," American Association of School Administrators'
Yearbook 1958 (Washington, D.CT).

 

2A. Hugh Livingston, ”Is There an Optimum Size
High School? Progressive Education (September, 1956).

 

22

considerations discussed previously allude

to the notion that a very large school tends

to limit rather than extend these opportunities.
It is felt that a large school may contribute
to personality disintegration and unsuccessful
participation by failing to meet these needs.

This same conclusion was reinforced by a research
memo issued by the National Education Association Research
Division. They had this to say about the psychological
climate of the various size schools:

Although it is agreed that a minimum number of
pupils is necessary for an adequate educational
program, there is a growing reccgnition of the
fact that it may be difficult, and probably
impossible, to maintain the quality of desirable
human relations and pupil development in schools
that are too large.

The effects of the large school environment are
so interwoven with similar consequences brought on by
the impersonal relationships experienced by the students
from the large city that it is impossible to say with
any degree of certainty which was the major contributing
factor. In spite of this handicap, there is strong
evidence to indicate the end products of these developmental
conditions, the large city and the large school, are some-

what different in their attitudes and values than the

students who live in small towns and attend small schools.

 

r——

1National Education Association Research Division.
"Efficient School Size”, National Education Association
Bulletin (Washington, D.C., January, 1959).

23

One guage of any individual's attitude and value
system is his choice of vocation. Elsbree and Reutter
point out in their book, Staff Personnel in the Public
Schools:

The rural areas of the United States have
contributed a much larger proportion, relatively,
of teachers than have the urban centers.
Similarly, in the past, graduates of small
high schools have been more favorably disposed
toward teaching than graduates of large high
schools. Whether or not the interest in
teaching exhibited by pupils in rural areas
and those graduating from small high schools
is due to limited vocational opportunities

or to some phase of the educational program
cannot definitely be stated.1

The argument for the large school is sometimes
based on the logic that in order to provide the proper
guidance facilities it is necessary to establish large
schools which can support such programs. The rationale
being that with better counseling programs the large
schools are better able to perform the work of
attitude and value development. The logic is clear
but the facts do not always verify that this is what
happens in reality. In a survey of thirty-three of the
nation‘s largest school systems, a teachers' group found
that better than one-third admitted that their present

counseling facilities did not meet their needs. All the

 

1Willard S. Elsbree and E. Edmund Reutter, Jr.
Staff Pepsonnel in the Public Schools, (Englewood Cliffs,
N. J.: Prentice-Hall, Inc., 19547, p. 23.

24

secondary schools in this survey had well over 1,000

in enrollment.

Research Dealing With Attitude Change
iDuring the College Years

 

It is well to note at the beginning of our review
of the research in this area the dangers involved in
generalizing from one campus to others. As Nevitt Sanford
has pointed out, the educational philosophy of each
school will determine the degree of emphasis each school
places on character building. He states:

It appears that much educational theory is based
upon the notion that personality is well established
by the age of seventeen, particularly that which
assumes that the student will choose his experiences
in accordance with notions that are now basic to

his personality and the chief job Of the college is
to help him acquire the means for the successful
pursuit Of his purposes. The contrasting educational
view is that colleges should be concerned, not so
much with teaching students how to realize their
values, as with helping them decide what values to
have. That instead of permitting them to become
confirmed in patterns of motives whose determinants
have been childish or accidental, the college

should encourage the developmeni toward patterns
that are valued in our society.

In this area, the present study will attempt to
answer the following questions: Are there differences in
the attitudes and values of the students entering Michigan
State University from various size high schools? Do these

differences remain after four years of college life?

 

f w

lNevitt Sanford, "Personality Development During the
College Years," Pergonnel and Guidance Journal (October,

1956i 35:64—80.

25

If there are changes in attitudes and values of the
students during their college years, Dr. Edward D. Eddy
believes it represents a move toward conformity with the
group. So that, if differences are found among various
groups of freshmen, he thinks that these same differences
are not likely to be present among the seniors. He

states: .
In most colleges which we have visited, we
found that student and faculty conformity
was so much the rule that we questioned
whether the genuine, sincere individualist
could ever be accepted. We were tempted to
wonder, for instance, what might happen to
a student with ability who Yefused completely
to go along with the group.

This same conclusion was reached by Philip E.
Jacob in his well-known book, Changing Values in College.
Jacob presents the view that American college students
today tend to think alike, feel alike and behave alike.
He writes:

Most American students share many values in
common. There is a striking homogeneity of
basic values throughout the country. On
issues where students do differ, they split
in about the same proportions at most
institutions. The patterns of value tent to
be similar at American colleges, regardless
of location, administration, size and back-
ground of the student bogy, or character of
the educational program. .

 

v—wrfiw

1Edward D. Eddy, The College Influence on Student
Character, American Council on Education, (Washington, D.C.,

1959;:90 137--

2Philip E. Jacob, Chan in Values in Colle e, (New
Haven, Connecticut, Hazen Foundafionl’I956), p. in.

26

A study Of University Of Buffalo students by
Bugelski and Lester also revealed the tendency for college
students to think alike. Their conclusions were:

(1) a significant change in attitudes between
freshman and senior years on an "Opinion
test," which concerned national and social
Optimism, labor problems, discipline,
social life, conventions and religious

(2) the change was toward greater liberality

(3) a test given three years after graduation
showed little change in aititudes from
those in the senior year.

A number of studies found significant changes in
attitudes occurring during the four years of college.
Kuhlen2 discovered increased sensitivity to human qualities
with such characteristics as tolerance, cooperativeness,
broadmindedness and democratic attitudes. Plant3 concluded
that significant Changes took place resulting in less of a
readiness to accept ethnocentric ideology.

Webster“ also reported findings that did not support

the view that college students become more alike in their

 

lRichard Bugelski and Olive P. Lester, "Changes in
Attitudes in a Group of College Students During Their Col-
lege Course and After Graduation," Journal of Spcial Psy-

chology (1940),12:319-332.

2Raymond G. Kuhlen, "Changes in the Attitudes of
Students and Relations of Tests Responses of Judgments of
Associates," School and_§opiety (April, 1941), 53:514-9.

3Walter J. Plant, "Changes in Ethnocentrism Associated
with a Four-Year College Education," Journal of Educational

Psychology (1958),49:162-5.

“Harold Webster, "Changes in Attitudes During College,"
Journal of Educational Psychology (1958), 49:109.

27

general attitudes. His results reinforced those personality
theories which emphasize the increased complexity, differ-
entiation, ability and independence of students during

their late adolescence.

1 reported, in 1954, that the effectiveness of

Dressel
college experience in inducing change is still in doubt.
He suggests that colleges form programs which encourage

substantial interest changes.

1Paul L. Dressel, "Interests - Stable or Unstable,"
Journal of Educational Research (1954), 48:95-102.

CHAPTER III
METHODOLOGY
Instrumentsipsed in the Stpdy

The following battery of tests was given to every
freshman, except foreign and transfer students, entering
Michigan State University in the fall term of 1958. The
three non-intellective measures, along with the Critical
Thinking Test, were also given to the graduating seniors
in 1962 as a follow-up study.

College Qualification Test (CQT)1 This test consists
of three parts; seventy-five items designed to measure
verbal ability; fifty items which measure skill in handling
numerical concepts; and seventy-five items which test
general information. The summation of these scores is
compiled and the individuals are ranked on a percentile
basis.

The reliability coefficient of the College Qualifica-

tion Test was estimated by the authors to be .97 for the

r .7. v v r Vi v—f y #

1Stanley 0. Ikenberry and Irvin J. Lehman, Critical
Thinking, Attitudes and Valuesin Higher gducation,(Michigan
State University,IEast Lansing, Michigan 1959),"p. ll.

28

29

males and .96 for the females.1 A check of the reliability

2 using a restricted sample of the

was made by Ikenberry
population of this study. It revealed a reliability
coefficient of .93.

The College Qualification Test is used as a predictor
of collegiate success, as based on the grade point average.
The validity of this test, when applied to the sample
population of this study, was investigated by Ikenberry.3
He found correlation coefficients ranging from .34 to .66,
in comparing the relationship of CQT scores to freshman
grade point averages. He noted that the magnitude of the
correlation coefficients tended to decrease from the fall
term of the freshman year to the spring term. Ikenberry
speculated that this decrease might be "due to several
factors including a narrower range of ability in the
latter terms due to acceleration and withdrawal of some
students, and possible fluctuation and change in student

ability from the time of measurement in the fall."4

 

V v

1George K. Bennett, Marjorie G. Bennett, Winburn L
Wallace and Alexander C. Wesman, College Qualification
Tests, Manual, (New York: The Psychological Corporation,

1957)-

ZStanley O. Ikenberry, "A Multivariate Analysis of
the Relationship of Academic Aptitude, Social Background,
Attitudes and values to Collegiate Persistence" (unpublished
Doctoral Dissertation, Michigan State University, East
Lansing, Michigan 1960), p. 50.

31bid., p. 49.
“lbid., p. 50.

30

Michigan State University Reading Test.--The construc-
tion of this test was the produce of the Office of Evalua-
tion Services of Michigan State University. Along with the
College Qualification Test, it was used as a predictor of
academic success at Michigan State. The Basic College of
the University divided the freshman class into groups
based on their scores on these two tests. At the end of
the term, the students were ranked in accordance with
their grade point averages. For example, if a student
ranked in the upper ten per cent on the test scores and
was achieving a GPA which placed him much lower in
percentile rank, that was an indication that he might be
underachieving and in need of counseling.

_In§_Mfiu_Beading_Iest, which contains forty-five
items, measures reading speed and comprehension. The
validity of the test is based on its ability to correlate
highly with the first year grades at Michigan State
University. A study of this correlation was done by
Ikenberry,l who found a range of .34 to .52 for females
in freshman grades and a range of .36 to .57 for the males.
In a breakdown of particular subjects, he found the
highest correlation between the Reading Test and freshman
communication skills grades for the males. This correlation

was .65. The lowest correlation was the male average for

 

lIbid., p. 53.

31

natural science in the third term. This correlation was
.35. We can readily see that the Reading Test is a fair
predictor of future college success at Michigan State.
After a check of the scores of a restricted sample, the
authors estimated the reliability of their test to be
.80.

Criticgl Thinking Test, Form G.—-This is a problem
solving type of test that was constructed by Dressel and
Mayhew.l Its fifty-two items are intended to measure
five factors of critical thinking:

'(1) define a problem
(2) select pertinent information

(3) recognize stated and unstated assumptions

(4) formulate and select relevant and promising
hypotheses

(5) draw valid conclgsions and Judge the validity
of the inference

Ikenberry and Lehmann3 tested the validity of the
Critical Thinking Test with the following freshman grades:

communication skills, natural science and term grade point

 

_‘,

lPaul Dressel and Lewis B. Mayhew. General Education:
Explorations in Evaluation, (Washington, D. 0.: American
Council on Education, 1954).

21bid.

3Ikenberry and Lehmann, op. cit., p. 14.

32

averages. The Pearson product-moment correlation
coefficients were significant in all phases at the .05
level. The range of these correlations was from .21 to
.57.

Reliability checks on this test were reported to
range from .71 to .89, according to the manual for

1 Ikenberry2 found a reliability coefficient

instructors.
of .79 when he used the Kuder-Richardson formula on a
small sample of his study.

Inventory of Beliefs, Form I.--This instrument,
developed by Dressel's Committee on Student Values and
Attitudes,3 consists of one-hundred and twenty forced-
choice, pseudorational statements to which the responent

replies as follows:

Strongly agrees . . . . . . . . 4 points
Agrees . . . . . . . . . . . . 3 points
Disagrees . . . . . . . . . . . 2 points
Strongly disagrees. . . . . 1 point

The high scores were listed as being independent,

adaptive and non-stereotypic in beliefs. The low scores

 

._v_.fi. 1 yw—fi

1Paul L. Dressel, Cooperative Study of Evaluation in
General Education, Instructor's Manual for the Test of
Critical Thinking, Form G. ThewAmerican Council on
Education, Committee on Measurement and Evaluation (1953)
(Mimeograpl.)

 

2Ikenberry, op. cit., p. 52.
3Dressel and Mayhew, op. cit., p. 20.

33

were described as being defensive, conforming, resistent

l

and stereotypic. Recognizing the difficulty in measuring

the validity of instruments dealing with attitudes and

values, Ikenberry states:

We can conclude that although characteristics
measured by the Inventory of Beliefs are not
independent of the phenomenon measured by the
cognitive instruments, the Inventory appears
to be measuring characteristics more clearly
relates to factors measured by the Dogmatism
Scale. ‘

The reliability estimates, as reported in the
instructor‘s manual, are in the range from .69 to .95,
with the median being .84.3

In a comparison of Inventory of Beliefs scores and
freshman grades in natural science and communication skills
courses at Michigan State University, Dr. Ikenberry came to
the following conclusion:

Although these relationships are slight
(highest correlation coefficient is .33),
they are generally consistent and indicate
that students who are less stereotypic in
beliefs tend to receive higher instructor,
exam, and total grades in the two basic
courses.

lIkenberry and Lehmann, op. cit., p. 9.

2Ikenberry, op. cit., p. 55.

3Cooperative Study of Evaluation in General Education,
Paul L. Dressel, Director. Instructorls Manual for the
Inventory of Beliefs, The American Council on Education,
Committee on Measurement and Evaluation (1953).(Mimeograph.)

”Stanley 0. Ikenberry, and Irvin J. Lehmann. Critical
Thinking, Attitudes and Values in Higher Education, A
Preliminary Re ort,(Michigan State University, East Lansing,
Michigan 1959 , p. 57.

 

34

Rokeach's Dogmatism_Scale, Form E.1--The scale con-
sists of forty statements, such as, "It is better to be a
dead hero than a live coward." The subject is forced to
agree or disagree and decide to what degree or extent he
does either on a six point scale, ranging from strongly
agree to strongly disagree. The higher the score, the
more dogmatic the subject would be rated.

Author Rokeach defines dogmatism as follows:

A total ideological defense against threat and

at the same time a cognitive framework for satisfying
one's need to know and comprehend the world one

lives in. In other words, dogmatic thinking and
believing makes it possible to ward off threatening
aspects of reality and at the same time gives_one
theesatisfaction of feeling that one understands

it.

A high score on the Dogmatism Scale correlates highly
with a low score on the Inventory of Beliefs. Recent
research has found this negative correlation to be as high
as -.63 for males and -.61 for females.3

Rokeach has outlined a framework for this study of

dogmatism as being:

a. a relatively closed cognitive system of beliefs
and disbeliefs about reality

 

v—‘f fir a 17 —v—~

lRokeach, Milton J. "Political and Religious Dog-
matism: An Alternative to Authoritarian Personality,”
Psyghological Monographs (1956), 70:425.

21bid., p. 5.

3Ikenberry, "A Multivariate Analysis. . .," op. cit.
pg. 56.

35

b. organized around a central set of beliefs
about absolute authority which, in turn
c. provides a framework for patterns of
intollerance and qualified tolerance
toward others.l
Reliability coefficient of .76 was obtained by
Ikenberry2 in using a restricted sample of his study.
Rokeach reported a median reliability of .78 using the
odd-even method.3
Qifferentialealges Inventory.--This instrument,
consisting of sixty—four forced-choice items, was con-

“ Its purpose is to differentiate

structed by Prince.
between the subject who holds a traditional value system
as opposed to one who maintains an emergent value orien—
tation. Needless to say, there is no right and wrong
determination in this type of testing. An attempt is
made to categorize groups of individuals according to
the kinds of values that to them are the more important.

Prince describes the traditional subject as

holding in high esteem the following values:

 

*7— ._V. v...

lMilton Rokeach, "The Nature and Meaning of Dogmatism,l
,Psychologicalvfieview (1954), 61:203.

I

QIkenberry
3Rokeach,

“Richard Prince, "A Study of the Relationship Between
Individual Values and Administrative Effectiveness in the
School Situation" (unpublished Dissertation, University of
Chicago, Chicago, Illinois, 1957).

36

personal respectability, thrift, self—denial,
respect for elders, as valuing hard work as
good in itself and necessary for success, as
placing his individual desires and ideas

equal to or above the desires and ideas of

the group, and being oriented toward the
future to the extent that present needs should
be sacrificed for future satisfaction and
reward.

A person with a high emergent value system
places a great deal of importance on getting
along with others and achieving group harmony.
Absolutes are questiOnable and morality is
group-determined rather than predetermined by
any particular source. Everything one does
should be done in consideration of others and
in regard for feelings and reactions of others.
Emphasis is on the present.

2 reported that Catholic college

Ikenberry and Lehman
students held the more traditional value approach, as
outlined above, in comparison to their Jewish counterparts.
Students from the urban areas scored high in emergent
values while the students with rural backgrounds were
high in the traditional values. Those students who
expressed an interest in graduate school scored higher on
the traditional values than those who were satisfied with
a bachelor's degree.

The reliability coefficients for this test were .61

for the males and .60 for the females. These were found

on a test-retest situation with college freshmen.3

W W 7 fi w— W v.1 V. a?

1Prince, 0p. cit., p. 42.
2Ikenberry and Lehmann, op. cit., p. 122.

31bid., p. 74.

37

Collection of the Data

This study is a supplement to a major research
project conducted at Michigan State University by their
Office of Evaluation Services. The following battery of
tests were given to the freshman class of 1958:

The College Qualificatlon Test

The Michigan State University Reading Test
The Test of Critical Thinking, Form G

The Inventory of Beliefs

Rokeach's Dogmatism Test 1
The Differential Values Inventory, Form I

A follow-up study was carried out in the spring term
of 1962, using as subjects those seniors who were nearing
graduation. The instruments used in this retesting were:

The Critical Thinking Test

The Inventory of Beliefs
Rokeach's Dogmatism Scale

The Differential Values Inventory

Also included in the follow-up study were the final
grade point averages of the graduating seniors. Their
point averages were based on the scale of 4.0 for A, 3.0
for B, 2.0 for C, and 1.0 for D. An F is failing and no
points are given for this grade.

Background information in regard to the size of the
high school graduating class was used as a criterium for
determining the approximate size of the high school. The
following figures indicate the breakdown of the high school

size groups.

 

1Ikenberry and Lehmann, op. cit., p. 9.

38

SEPARATION OF SUBJECTS OF POPULATION ACCORDING
TO SIZE OF HIGH SCHOOL AND SEX

m

High School Range of Size of Males Females Total

 

Size Group Graduating Class N - % N - % N - %

A 400 or more 472-33 338-26 810-30

B 100 - 399 692-48 688-53 1380-50

C 99 or less 268-19 282-21 550—20
Totals 1432 1308 2740

 

The breakdown into three basic size groups, large,
medium and small, was selected by the author because this
appears to be the logical separation used in most litera-
ture on this question. Dr. Conant's recommended minimum
was one-hundred in the graduating class. Tyelve to
fifteen hundred enrollment is often used as the separation

of the medium and large size schools.

Population Under Study
The population of this study consists of 2,746 fresh-
men students at Michigan State University in the fall term
of 1958. Transfer students, foreign students and those
freshmen who failed to show at the time of testing were
not included in the study. The group consisted of 1,436
males and 1,310 females. The percentage breakdown was 52

per cent to 48 per cent.

39

Selection of Sample
A random sample on one-hundred males and one-hundred
females was selected from each of the three high school
size groups. Of this original total of six—hundred, one—
hundred and eighty-seven completed their college work at
Michigan State University in four years and were retested

in the follow-up study.

Statistical Method

The statistical method employed in this study is
the analysis of variance. Since we are using three
groups of students and we desire to test for any signifi-
cant difference among these groups, the analysis of
variance techniques would be simpler and more convenient
than using the t test.

As the name implies, analysis of variance uses

variances as a measuring tool. Edwards explains what

this means:1

The rationale of the analysis of variance is that
the total sum of souares of a set of measurements
composed of several groups can by analysed or broken
down into specific parts, each part identifiable with
a given source of variation. In the simplest case,
the total sum of squares is analyzed into two parts:
a sum of squares based upon variation within the
several groups, and a sum of squares based upon

the variation between the group means. Then, from
these two sums of squares, independent estimates

of the population variance are computed.

 

'v—v— 1" . vy— —.——7

1Allen L. Edwards, Statistical Methods for the
Behavioral Sciences (New York: Holt, Rinehart and Winston,
1961): p- 315-

40

In analysis of variance, we accept the assumption
that the groups are random samples of a common population,
and our estimates of the population variance may be expected
to vary only within the limits of random sampling. The
null hypothesis is tested by dividing the mean square score
between groups by the mean square score within groups.

This gives an F score which is then compared to the tables
of F scores, developed by Snedecor. If the observed F,
the variance ratio, equals or exceeds the value found in
the tables, then the null hypothesis is rejected. That
is, the judgment is made that the samples do not come

from a common population.

The five per cent level of significance was used in
this experiment to test the null hypothesis. That is, we
will accept the null hypothesis as tenable if the score

falls within the limits of the confidence interval.

W1

Two basic groups of instruments were used in
this study. The first group was composed of three
intellective tests: the College Qualification Test, the
Michigan State University Reading Test, and the Critical
Thinking Test. The second battery is made up of three in-
struments designed to survey student attitudes and values.
They are: the Inventory of Beliefs, the Differential

Values Inventory and Rokeach's Dggmatism Scale.

 

41

The collection of data in this longitudinal type
study was carried out over a four year period. In the
initial battery, all six tests were administered to the
entering freshman class at Michigan State University in the
fall of 1958. The follow-up study was composed of a re-
test on four of the instruments used in the original
battery. A check was made of the cumulative grade point
averages of the graduating seniors as part of the follow-
up study.

The subjects were separated according to the size
of the high school from which they graduated, and also be
sex. The school size breakdown was as follows: Large--
over 400 in the graduating class: Medium--100 to 399 in
the class; and Small-~99 or less in the high school class.

The population of the study was the 1958 freshman
class at Michigan State University, not counting transfers,
foreign students and those not taking the complete battery.
A random sample of one-hundred males and one~hundred females
was selected from each high school size group.

The statistic; used was the analysis of variance.
The hypotheses were all listed as null hypotheses. The
five per cent level of confidence was employed in rejecting

the null hypothesis.

CHAPTER IV
ANALYSIS OF RESULTSe-INTELLECTIVE VARIABLES

The findings reported in this chapter are divided
into two basic sections: a survey of the freshmen as
they enter college and a follow-up study of the same
groups after four years of college. The freshmen are
separated into sub-groups according to the size of the
high school from which they graduated, large, medium
or small. The sexes are studied together and also
individually.

The three intellective instruments used to
test the freshmen were the following: the College
nglification Tbst, the Critical Thinking Tgpp, and the
Michigan State University Reading Test. In the follcw-up
study of the college seniors, only the Critical Thinking

Test was given.

Survey of College Freshmen

Males gnd Females Combined

 

lfiypothesis No. 1.--There are no differences among

 

the various high school size groups on the College Quali-
fication Test, administered at the beginning of their

freshman year of college.

2+2

43

TABLE l.--Ana1ysis of Variance for Three Groups of College
Freshmen, of One Hundred Subjects Each, Representing Various
Size High Schools, Tested with the College gpalifigatign

 

 

 

Test
Source of Variance Sum of Squares Df Mean Square F
Between Groups 1,581.40 2 790.70 1.15
Within Groups 409,977.00 597 686.39
Total 411,558.40 599

—r W W

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. 2.--There are no differences among

 

the various high school size groups on the Critical
Thinking Test, administered at the beginning of their

freshman year of college.

TABLE 2.--Ana1ysis of Variance for Three Groups of College
Freshman, of One Hundred Students Each, Representing Various
Size High Schools, Tested with the Critical Thinking Test

‘v v—v

Source of Variance Sum of Squares df Mean Square F

 

~—v—v—— w—V— fl.

Between Groups 69-75 2 34.88 .66

Within Groups . 31.3é5.62 597 52.17
Total 31,395.37 599

w—vv vrv

Results: Non-significant at the .05 level of confidence
Conclusions: Null hypothesis accepted

44

Hypothesls No._3.--There are no differences among
the various high school size groups on the Michigan State
University Reading Test, administered at the beginning of

their freshman year of college.

TABLE 3.--Analysis of Variance for Three Groups of College

Freshmen of One Hundred Students Each, Representing Various

Size High Schools, Tested with the Michigan State Universlgy
ReadinggTest

 

Source of Variance Sum of Squares df Mean Square F

 

a Y fi—vv

Between Groups 156.88 2 78.44 1.99

Within Groups 23,524.12 597 39.40
Total 23,681.00 599

 

v. v71 f v7 FY

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

.Mfil§§_in¥

Hypothesis No. 4.--There are no differences among the

 

males from the various high school size groups on the

College Qualification Test, administered at the beginning

 

of their freshman year of college.

’45

TABLE 4.--Analysis of Variance for Three Groups of College
Freshmen, of One Hundred Students Each, All Males Represent-
ing Various Size High Schools, Tested With the College

' Qualification Test

 

 

 

' Y 11.1 .. m
Source of Variance Sum of Squares df Mean Square F
Between Groups 564.70 2 282.35 .43
Within Groups 194.222.30 297 653.94

Total 194,787.00 299

 

“—7 '7 V ‘ . ~

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No.l5.--There are no differences among the
males from the various high school size groups on the
Critigal-Thinking Tes , administered at the beginning of
their freshman year of college.

TABLE 5.--Analysis of Variance for Three Groups of College
Freshmen, of One Hundred Students Each, All Males, Represent—

ing Various Size High Schools, Tested With the Critical
Thinking Test

 

 

M L , M
Source of Variance Sum of Squares df Mean Square F
Between Groups 126.75 2 63.38 1.31
Within Groups 14,320.64 297 48.22

Total 14.447.39 299

 

V v.7 r a _,

Results: Non-significant at the 0.5 level of confidence
Conclusion: Null hypothesis accepted

46

Hypothesis No. 6.—-There are no differences among
the males from the various size high school groups on the
Michigan State University Reading Test, administered at

the beginning of their freshman year of college.

TABLE 6.--Ana1ysis of Variance for Three Groups of College

Freshmen, of One Hundred Students Each, All Males, Repre-

senting Various Size High Schools, Tested with the Michigan
State UniversityReading Test

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups .51 2 .26 .007
Within Groups 11,598.48 297 39.02

Total 11,598.99 299

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Females Only

 

Hypgthesis No. 7.--There are no differences among
the females from the various high school size groups on

the College Qualification Teet, administered at the

 

beginning of their freshman year of college.

47

TABLE 7.--Ana1ysis of Variance for Three Groups of College

Freshmen, of One Hundred Students Each, All Females, Rep-

resenting Various Size High Schools, Tested with the Col-
lege Qualification Test ""

 

W

‘1‘ "v...—
*V f

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 1,144.90 2 572.45 .86
Within Groups 197,973.80 297 666.58

Total 199,118.70 299

 

fi Y—v if 7 w W

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No, 8.v-There are no differences among
the females from the various high school size groups on
the Critical Thinking Test, administered at the beginning

of their freshman year of college.

TABLE 8.--Analysis of Variance for Three Groups of College

Freshmen, of One Hundred Students Each, All Females, Rep-

resenting Various Size High Schools, Tested with the Criti—
gal Thinking Test

 

 

 

======
Source of Variance Sum of Squares df Mean Square F
Between Groups 30.45 2 15.23 .27
Within Groups 16,783.10 297 56.51
Total 16,813.55 299

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

us

Hypothesis No. 9.--There are no differences among the
females from the various high school size groups on the
Michlgan State University Reading Test, administered at the

beginning of their freshman year of college.

TABLE 9.--Analysis of Variance for Three Groups of College

Freshmen, of One Hundred Students Each, All Females, Rep-

resenting Various Size High Schools, Tested with the Mich-
lgan State University Reading Test

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 319.76 2 159.88 4.07
Within Groups 11,671.21 297 39.30

Total 11,990.97 299

f i“ W

Results: Significant at the .05 level of confidence
Conclusion: Null hypothesis rejected

Sgrvey of Cpllege Seniors

Males and Females Combined
Hypothesis No. lO.--There are no differences among the
various high school size groups on the Critical Thinking

Test, administered during their senior year of college.

49

TABLE 10.--Analysis of Variance for Three Groups of
College Seniors, Representing Various Size High Schools,
Tested with the Critical Thinking Test

“V v—' we —7 Y ——.r

rv r Y fi

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 419.70 2 209.85 .40
Within Groups 99,215.80 187 530.57

Total 99,635.50 189

 

w—r ——r __ —f f

Result: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Males Only

Hypothesis No. 11.--There are no differences among the
males from the various high school size groups on the
Critical Thinking Test, administered during their senior
year of college.
TABLE 11.--Ana1ysis of Variance for Three Groups of

College Seniors, Males Only, Representing Various Size
High Schools, Tested with the Critical Thinking Test

 

h

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 589.36 2 294.68 .52
Within Groups 60,631.94 108 561.41

Total 61,221.30 110

 

T

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

50

Females Only
Hypothesis No. l2.-~There are no differences among the
females from the various high school size groups on the

Critical ThinkingATest, administered during their senior

 

year of college.

TABLE l2.--Analysis of Variance for Three Groups of

College Seniors, Females Only, Representing Various

Size High Schools, Tested with the Critical Thinking
Test '

 

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 332.68 2 166.34 .33
Within Groups 37,925.52 .76 499.02

Total 38,258.80 (78

v—'

 

Results: Non-significant at the .05 level of confidence
Conclusions: Null hypothesis accepted

CHAPTER V
ANNALYSIS 0F RESULTS--AFFECTIVE VARIABLES

There are factors other than intellectual variables
that need to be taken into consideration in the study of
high school preparation as it relates to future success
at college. Attitudes and values, in many cases, are the
determining factors in the college students achieving
academic success. This investigation of the affective
variables is divided into two basic sections, similar to
those in the preceding chapter. The first section is an
analysis of the testing of college freshmen and the second
deals with a follow-up study of the group when they were
college seniors. The following instruments were used in
the study of attitudes and values: Inventory of Beliefs,
Form I, Differential Values Inventory, and Rokeach's
Dogmatism Scale, Form E.

Affective Variables at the Beginning
of College Careers
Meles and Females Combined

Hypothesie No. 13.--There are no differences among
the various high school size groups on the Inventory of
Beliefs, administered at the beginning of their freshman

year of college.
51

52

TABLE 13.--Ana1ysis of Variance for Three Groups of

College Freshmen, of Two Hundred Students Each, Rep-

resenting Various Size High Schools, Tested with the
Inventory of Beliefs

 

 

 

W

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 26.00 2 13.00 .08
Within Groups 101,669.50 597 170.30

Total 101,695.50 599

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. l4.--There are no differences among
the various high school size groups on the Differential

Velues Inventory, administered at the beginning of their

 

freshman year of college.

TABLE l4.--Analysis of Variance for Three Groups of

College Freshmen, Two Hundred Students Each, Rep-

resenting Various Size High Schools, Tested with the
Differential Values Inventory

1

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 409.49 2 204.74 4.17
Within Groups 29,326.88 597 49.12

Total 29,736.37 599

 

.~_._._ .,

Results: Significant at the .05 level of confidence
Conclusion: Null hypothesis rejected

53

Hypothesis No. l5.--There are no differences among

 

the various high school size groups on the Dogmatism
Scale, administered at the beginning of their freshman
year of college.

TABLE 15.--Analysis of Variance for Three Groups of
College Freshmen, Two Hundred Students Each, Rep-

resenting Various Size High Schools, Tested with the
Dogmatism Scale

 

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 353.00 2 176.50 .29
Within Groups 361,835.00 597 606.09

Total ‘ 362.188.00 599

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

,Males Only

Hypothesis No. l6.—-There are no differences among

 

the males from the various high school size groups on the

Inventory of Beliefs, administered at the beginning of

 

their freshman year of college.

54

TABLE l6.--Ana1ysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each, Rep-

resenting Various Size High Schools, Tested With the
Inventoryyoleeliefs

 

 

Source of Variance Sum of Squares df Mean Squares F
Between Groups 84.65 2 42.33 .23
Within Groups 55,634.25 297 187.32

Total 55,718.90 299

 

*—

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. 17.--There are no differences among

the males from the various high school size groups on the

Differential Values Inyentpry, administered at the beginning

of their freshman year of college.

TABLE l7.--Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each, Rep-

resenting Various Size High Schools, Tested with the
Differential Values Inventogy

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 186.85 2 93.43 1.91
Within Groups 14,515.99 297 48.88

Total 14,702.84 299

 

Results: Non-significant at the .05 level of confidence
Conclusions: Null hypothesis accepted

55

Hypothesis No. 18.--There are no differences among
the males from the various high school size groups on the
Dogmatism Scale, administered at the beginning of their

freshman year of college.

TABLE 18.--Ana1ysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each, Rep—

resenting Various Size High Schools, Tested With the
Dogmatism Scale

”
C“.

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 1,959.70 2 979.85 1.54
Within Groups 187,881.30 297 632.60

Total 189,841.00 299

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Females Only

 

Hypothesis No. 19.--There are no differences among

 

the females from the various high school size groups on

the Inventorylof Beliefs, administered at the beginning of

 

their freshman year of college.

56

TABLE l9.--Analysis of Variance for Three Groups of

College Freshmen, One Hundred Students Each, Rep-

resenting Various Size High Schools, Tested with the
Inventory of Beliefs

 

 

m
Source of Variance Sum of Squares df Mean Squares F
Between Groups 11.19 2 5.59 .04
Within Groups 45,706.81 297 153.89

Total 45,718.00 299

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. 20.--There are no differences among
the females from the various high school size groups on
the Differential Values Inventory, administered at the
beginning of their freshman year of college.

TABLE 20.--Analysis of Variance for Three Groups of
College Freshmen, One Hundred Students Each, Rep-

resenting Various Size High Schools, Tested With the
Differential Values Inventooy

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 352.65 2 175.33 3.68
Within Groups 14,247.39 297 47.97

Total 14,600.04 299

Results: Significant at the .05 level of confidence
Conclusion: Null hypothesis rejected

57

Hypothesis No. 21.--There are no differences among
the females from the various high school size groups on
the Dogmatism Scale, administered at the beginning of their
freshman year of college.
TABLE 21.--Ana1ysis of Variance for Three Groups of
College Freshmen, One Hundred Students Each, Rep-

resenting Various Size High Schools, Tested with the
Dogmatism Scale

 

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 317.30 2 158.65 .28
Within Groups 169,156.40 297 569.55

Total 169,473.70 299

_,'

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Affective Variables After Four
Years of College

 

Males and Females Combined

Hypothesis No. 22.--There are no differences among

 

the various high school size groups on the Inventory of

Beliefs, administered during their senior year of college.

58

TABLE 22.--Analysis of Variance for Three Groups of
College Seniors, Representing Various Size High
Schools, Tested with the Inventory oleeliefs

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 40.63 2 20.31 .51
Within Groups 7,392.64 187 39.53

Total 7,433.27 189

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. 23.--There are no differences among

 

the various high school size groups on the Differential
Values Inventory, administered during their senior year

of college.

TABLE 23.-~Analysis of Variance for Three Groups of
College Seniors, Representing Various Size High
Schools, Tested with the Differential Values Inventogy

j
__

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 133.72 2 66.86 1.32
Within Groups 9,491.55 187 50.76

Total 9,625.27 189

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

59

Hypothesis No. 24.--There are no differences among
the various high school size groups on the Dogmatism Scale,

administered during their senior year of college.

TABLE 24.--Ana1ysis of Variance for Three Groups of
College Seniors, Representing Various Size High Schools,
Tested with the Dogmatism Scale

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 88.63 2 44.32 .23
Within Groups 36,693.27 187 196.22

Total 36,781.90 189

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

MalesfiOnly
Hypothesis No.l25.--There are no differences among
the males from the various high school size groups on the

Inventory of Beliefs, administered during their senior

 

year of college.

60

TABLE 25.--Ana1ysis of Variance for Three Groups of
College Seniors, All Males, Representing Various Size
High Schools, Tested with the Inventory of Beliefs

 

 

 

m m L
Source of Variance Sum of Squares df Mean Square F
Between Groups 79.66 2 39.83 1.22
Within Groups 3,531.82 108 32.70

Total 3,611.48 110

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. 26.--There are no differences among
the males from the various high school size groups on the
Differential Values Inventory, administered during their
senior year of college.

TABLE 26.-~Analysis of Variance for Three Groups of

College Seniors, All Males, Representing Various Size
High Schools, Tested with the Differential Values

 

 

 

 

Inventory
Source of Variance Sum of Squares df Mean Square F
Between Groups 174.01 2 87.00 1.79
Within Groups 5,243.85 108 48.55
Total 5,417.86 110

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

61

Hypothesis No. 27.--There are no differences among
the males from the various high school size groups on
the Dogmatism Scale, administered during their senior

year of college.

TABLE 27.--Analysis of Variance for Three Groups of
College Seniors, All Males, Representing Various Size
High Schools, Tested with the Dogmatism Scale

v —___7

W f—_ V

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 449.14 2 224.57 1.04
Within Groups 23,269.57 108 215.46

Total 23,718.71 110

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Females Only

Hypothesis No. 28.--There are no differences among

 

the females from the various high school size groups on

the Inventory of Beliefs, administered during their

 

senior year of college.

62

TABLE 28.--Ana1ysis of Variance for Three Groups of
College Seniors, All Females, Representing Various
Size High Schools, Tested with the Inventory of

 

 

 

Beliefs
:=={, _ will
Source of Variance Sum of Squares df Mean Square F
Between Groups 40.28 2 20.14 .41
Within Groups 3,768.10 76 49.58
Total 3,808.38 78

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

Hypothesis No. 29.--There are no differences among
the females from the various high school size groups on
the Differential Values Inventory, administered during
their senior year of college.

TABLE 29.-~Ana1ysis of Variance for Three Groups of
College Seniors, A11 Females, Representing Various

Size High Schools, Tested with the Differential
Values Inventory

 

 

 

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 15.20 2 7.60 .16
Within Groups 3,512.47 76 46.22

Total 3,527.67 78

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

63

Hypotheeis No. 30.--There are no differences among
the females from the various high school size groups on the
Dogmatism Scale, administered during their senior year of
college.

TABLE 30.--Ana1ysis of Variance for Three Groups of

College Seniors, A11 Females, Representing Various
Size High Schools, Tested with the Dogmatism Scale

1

 

 

W
Source of Variance Sum of Squares df Mean Square F
Between Groups 152.27 2 76.14 .48
Within Groups 12,058.77 76 158.67

Total 12,211.04 78

 

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

CHAPTER VI
ANALYSIS OF RESULTS-'COLLEGE GRADES

In addition to the study of affective and intellective
variables, this investigation surveyed the final college
grade point averages of students from the various size high
schools.

Hypothesis No. 3l.--There are no differences among the
various high school size groups on their final grade point
averages at colleges.

TABLE 31.--Analysis of Variance for Three Groups of College

Seniors, Representing Various Size High Schools, Compared
on the Basis of Their Final Grade Point Averages

 

Y_ fiv—

 

Source of Variance Sum of Squares df Mean Square F
Between Groups 8,850.27 2 4,425.13 2.80
Within Groups 518,571.06 244 2,125.29

Total 527,421.32 246

 

Results: Non-significant at the .05 level of confidence
Conclusion: Nul hypothesis accepted

64

65

Males Ooly

Hypothesis No. 32.-~There are no differences among
the males from the various high school size groups on
their final grade point averages in college.
TABLE 32.--Analysis of Variance for Three Groups of
College Seniors, Males Only, Representing Various

Size High Schools, Compared on the Basis of Their
Final Grade Point Averages

 

 

 

 

W ==
Source of Variance Sum of Squares df Mean Square F
Between Groups 14,688.60 2 7,344.30 3.33
Within Groups 237,837.00 108 2,202.00

Total 252.525.60 110

 

Results: Significant at the .05 level of confidence
Conclusion: Null hypothesis rejected

Females Only

 

Hypothesis No. 33.--There are no differences among
the females from the various high school size groups on

their final grade point averages in college.

66

TABLE 33.--Analysis of Variance for Three Groups of

College Seniors, Females Only, Representing Various

Size High Schools, Compared on the Basis of Their
Final Grade Point Averages

 

 

 

m =—L—— j
Source of Variance Sum of Squares df Mean Square F
Between Groups 317.50 2 158.75 .08
Within Groups 273,372,30 133 2,055.43

Total 273,689.80 135

 

~7—

Results: Non-significant at the .05 level of confidence
Conclusion: Null hypothesis accepted

A summary of the grade point averages of the various
high school size groups is included in the following
table.

TABLE 34.--The Final College Grade Point Averages of
Seniors From Various Size High Schools

 

W . g a

 

Males and Females
School Size Females Males Only Only
Large 2.61 2.53 2.68
Medium 2.75 2.80 2.71

Small 2.64 2.60 2.68

 

67

Suemary of Resulte

Iptellectlve Variables:

College Freshmen

 

All

assesses
College Qualification Test N.S.
Critical Thinking Test N.S.
MSU Reading Test N.S.

Intelleotive Variables: lgollege Seniors

All
sealers.

Critical Thinking Test N.S.

Affectlve Variables: College Freshmen

A11

W .
Inventory of Beliefs N.S.
Differential Values Inventory Sign.
Dogmatism Scale N.S.

Affective Variables: College Seniors

All

We
Inventory of Beliefs N.S.
Differential Values Inventory N.S.
Dogmatism Scale N.S.

Notes:
groups; large, medium and small.
N.S.--non-significant at the

Sign.--significant at the

Males

Only
NOS.

N.S.
N.S.

Males
Only

N.S.

Males
Only

N.S.
N.S.

N.S.

.05 level

Females
Only

N.S.
N.S.

Sign.

Females
Only

NOS.

Females

Only
NOS.

Sign.

N.S.

Females

Only
N.S.

N.S.

N.S.

All comparison are among the three high school size

.05 level of confidence

CHAPTER VII
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

Summary of the Study

In recent years, many writers and educators have
argued that the small high school was not providing an
adequate education for its students, especially those who
went on to college. At the same time, criticism was being
leveled at the large, metropolitan high schools because they
were becoming too big and impersonal. The question there-
fore arises, what is the optimum size for a high school
in order to provide the most economical and effective
educational program.

In our approach to this problem, we analyzed first
the various complaints against our nation's high schools,
both large and small. Generally, criticism of the small
high school was focused on their limited course offerings
and poorly trained teachers. They also come under repre-
hension for the high cost of operation and administration.

Censure of the large school stems mainly from an
alleged impersonal approach to education. In many cases,
there seems to be a lack of personal identification by the

individual student with the school and his fellow classmates.

68

69

This atmosphere causes the student to feel small and
unimportant. Likewise, the large groups make it difficult
to transmit the teachings of proper attitudes and values

The primary purpose of this investigation is to
study the graduates of the various size high schools who
went on to college, in an effort to determine if real
differences do exist. Three crucial questions arev
prOposed for empirical investigation:

1. Is there any relationship between high school
size and academic preparation for college?

2. Is there any relationship between high school
size and academic achievement at college?

3. Is there any relationship between high school
size and student attitudes and values?
Relatedjfigsearch

After reviewing the literature in the field, we
found the results are inconclusive as to whether or not
high school size is a significant factor in determining
future academic success at college. Over the years, the
studies performed on this subject have produced, in the
main, either results showing the small high schools
academically inferior or results indicating no significant
differences in the comparison of the high schools
of various size.

In the state of Michigan, very little work has been

done in the area of a scientific investigation of high

70

school size and college achievement. This is the first
attempt made to study this question with the students
of Michigan State University.

A number of articles can be found in professional
jOurnals which emphasize the importance of psychological
considerations in determining the effectiveness of various
size high schools. In spite of this, there has been no
previous research, to our knowledge, relating high school
size to the development of attitudes and values.

Research dealing with attitudes and value change
during the college years has produced a variety of
conclusions. Those investigators who found significant
changes, as indicated in Chapter II, generally concluded
that the tendency was toward more conformity in thought
and behavior. A few researchers found an increase in
openrmindedness as the student progressed through college.
Other studies found no change in attitudes, of any degree
of significance, taking place during the college years.
This particular study_concentrates on the question
whether there has been significant attitude changes during
the college years of students who graduated from various

size high schools.

Metheds Employed
The methodology of this study contains a description

of the instruments used, the selection of the sample of the

71

population under study, and an explanation of the means of
collecting and analyzing the data.

The instruments used in this study, as described in
Chapter III, were the following: the College Qualification
Eepp, Critical Thinking Test, Form G, the Michigan State
University Reading Teet, lnventory oleeliefs, Form Ia

Rokeach's Qogmatism Scale; Ferm E, and the Differential

Values Inventory.

 

 

The above six tests were given to the entering
freshman at Michigan State University in the fall of 1958.
After four years of college, a follow-up battery composed
of four of these same instruments was given to the graduating
seniors.

The group's academic achievement at college was
measured by surveying the final grade point averages of
the graduating seniors.

The student groups were compared on the basis of the
size of their high school graduating class. The high
schools were divided into three ranges: large, medium and
small.

The statistical techniques of the analysis of
variance were used in making the comparisons of the high

school size groups.

72

Findin 3

Tha analysis of the first results concerned the
intellective tests. We found no differences among the
students from the various size high schools in the area
of academic preparation. When the freshman males were
studied separately, there were also no significant
differences on their scores on these tests. The only
significant difference among the freshman girls occurred
or the MWW- on this
test, the highest scores were attained by the female
students from the medium size high schools.

After four years of college, the graduates from the
various size high schools were again compared on the
Critigal_1nipkinggmegt. The results showed no significant
differences among the three groups either when studied
separately by sex or when grouped together.

In the survey of attitudes and values among the
students from large, medium and small high schools, no
significant differences were found on the Inventory of
Beliefs or on the ngmatism Scale, both in the testing of
freshmen and also in the study of college seniors.

Significant differences were noted among the college
freshmen, when compared in relationship to the size of the
high school from which they graduated, on the leferential

Values Inventory. This difference also showed up when the

73

freshman girls were compared but not in the comparison of
the males. After four years of college these differences
disappeared.

In comparing the high school size groups on the basis
of the final college grade point averages, there were no
significant differences among the groups when the males and
females were studied together. Significant differences did
occur when the males were compared separately, but not in

the comparison of the females.

Conclusione

Our general conclusion from these findings is that
there appears to be very little difference among the
graduates of the various size high schools who entered
Michigan State University in 1958. Academically, the
intellective tests, which were given to the entering
freshmen, indicated no differences in the preparation of
these students in the areas of reading, critical thinking
and general academic knowledge.

After four years of college, a survey of the student
groups points out the fact that high school size also was
not a factor in the final grade point averages of the
college seniors.

Our investigation of student attitudes and values
was based on the belief that there are other worthwhile

considerations in the determination of the optimum size

74

high school. Our conclusions from this investigation

were that the size of the graduating class, which was our
indication of the size of the high school, does not have a
profound effect on these college students in the area of
attitudes and values.

The one test of values on which a significant dif-
ference was detected, the Differential Values Inventopy,
indicated that freshmen from the small high schools were
more traditional in their views. That is, they tended to
regard more highly the values of puritan morality, the
meaning of work and responsibility, and the ideas of indivi-
dualism. Since these differences did not occur when the
males were tested separately, most of the differences can
be attributed to the scores made by the female students.
After four years of college life, a re-test with the
Differential Values Inventory showed no differences among
the high school size groups when the seniors were studied.

The initial differences in this area among the fresh-
men from the large and small schools is in agreement with a

similar study using the DVI, which was performed by Iken-

 

berry.1 He compared students on the basis of home back-
ground and found: "Students who lived a major portion of
1Ikenberry and Lehmann. Critical Thinking. . ." A

 

Preliminary Report., p. 34.

75

their life on a farm or in a small community have a
significantly higher traditional value score than those
who come fron an urban center."

The implications of this study for school adminis-
trators can be found first in the need for more knowledge
concerning the optimum size high school. Opinions should
give way to a scientific analysis of the advantages and
disadvantages of various size units. In school consolida-
tion studies, the general public looks to the school admin-
istrator for professional advice in future schoolhouse
planning.

The school administrator, who is charged with the
responsibility for the development of curricular and co—
curricular activities, must select those programs with the
objective of developing proper attitudes and values in the
students' minds. He must be aware of the factors, both
internal and external to the school system, which effect
the development of these values and attitudes. This study
revealed that size is not a major factor in this area.

In our generalizations from the results of this study,
we propose it as additional evidence for the arguments
that high school size is not a leading factor in determining
future academic success at college. This study does not
rule out other advantages of school consolidations, but it
is contending that the academic arguments as concerning those

students who went on to college, have not been verified by

76

research statistics. This group of college bound students
is only a portion of the total school population. In
determining the optimum size high school, consideration
must also be given to those students who do not aspire to
a college education.

In the past, most of the research concerning high
school size limited itself to the investigation of grade
records and economic factors. While this study is far from
conclusive, it does show that we use additional insight into
the development of attitudes and values at this stage of
the student's educational career. This study might provide
a beginning to the building of empirical data which would
lead to some conclusions in this field.

The results of this study might also be of interest
to college counselors and admission officers. These
educators have sometimes generalized that the small high
school graduates operated at a disadvantage in competing in
college classes with students from the large high schools.
The results of the present study do not verify this

assumption.

Recommendations

 

As a suggestion for addition research, it is proposed
that a similar study be carried on at a private, metropolitan

college. The character of the student population at this

77

type of institution might be radically different from that
of the large, stage university, such as Michigan State
University.

The results of this type of investigation might also
vary from state to state, depending upon the amount of
state aid given to the rural school systems. In poor
states, where the amount of money provided for education
is relatively low, the small, rural schools are the
first to suffer. This is particularly true in the areas
where the state law allows an eight months school term.

Our final recommendation is that each school
administrator should examine his own local situation, in
light of the present information, to judge whether or
not the graduates of his high school are achieving

reasonably well in their college endeavors.

BIBLIOGRAPHY

78

BIBLIOGRAPHY

Altman, Ester Royal. "The Effect of Rank in Class and Size
of High School on the Academic Achievement of Central.
Michigan College Senior Class of 1957," Journal of
Educetional Research 8:307, April, 1959.

. "The High School in a Changing Work," American
Association of School Admlnistrators' Yearbook 1958,
Washington, D. 0.

Bennett, George K., Bennett, Marjorie G., Winburn L. Wallace,
and Alexander G. Wesman. College Qualification Tests,
Mappal, 1957. New York: The Psychological Corporation,

195757

Brown, Ruth A. "A Study of High School and First Semester
College Records of Freshmen Entering Twenty Colleges
in Michigan in 1928," Journal of the Michigan School-
masters' glub, 1930, p. 18. Y

Bugelski, Richard and Olive P. Lester. "Changes in
Attitudes in a Group of College Students During Their
College and After Graduation, Journal of Social
Psychology 12:319-332, 1940. ‘77

Conant, James B. The American High School Today. New York:
McGraw-Hill Book Co., 1959, p. 14.

"Educator Says High Schools Too Small," Indianapolis
Star, May 10, 1950, p. 1.

. "The Size of High School Considered to be the Most
Efficient Scholastically and E onomically,” Corpus
Christi Public Schools, May 4, 1959.

Dawson, Howard A., et a1. Your School District (National
Commission of School District Reorganization, Depart-
ment of Rural Education.), Washington, D. 0.:
National Educational Association, 1948.

Dressel, Paul. Director of Cooperative Study of Evaluation
in General Education. "Instructor's Manual for the
Inventory of Beliefs," The American Council on
Education, Committee on Measurement and E aluation,

1953.
79

80

Dressel, Paul. Director, Cooperative Study of E aluation
in General Education, "Instructor's Manual for the
Test of Critical Thinking, Form G," The American
Council on Education, Committee on Measurement and
Evaluation, 1953.

. "Interest--Stab1e or Unstable," Journal of
Educational Research, 48:95-102, 1954.

Dressel, Paul and Lewis B. Mayhew. General Education:
Explorations in Evaluation. Washington, D. C.:
American Council on Education, 1954.

Dwyer, P. S. "Some Suggestions Concerning the Relationship
Existing Between Size of High School Attended and
Success in College," Journal of Educational Research
32:271, December, 1938.

Eddy, Edward D. The College influence on Student Character,
Washington, D. C.: American Council on Education,

1959. p. 137.

Edwards, Allen L. Statistical Methods for the Behavioral
Sciences: New York, Holt, Rinehard and Winston, 1961

Elsbree, Willard S. and E. Edmund R utter, Jr. Staff
Personnel in the Public Schools, Englewood Cliffs,
J. J.: Prentice-Hall, Inc., 1954, p. 23.

Faunce, Roland C. and Theodore D. Rice. "The Michigan
Secondary School," A Report of the Michigan Study
of Secondary School Curriculum, 1937-1945. Lansing,
Michigan, 1945{

Gaumnitz, Walter H. and Ellsworth Tompkins. "How Large
Are Our Public High Schools," Federal Security
Agency, Office of Education, Circular No. 304.

Harmon, Lindsey R. "0n Decision Making in High School"
National Association of Secondary School Principals
Bulletin, 46:277, November, 1962.

Holland, John. "The Prediction of College Success from
the California Psychological Inventory and Scholastic
Aptitude Test," Journal of Educational Psychology,

August. 1959. p- 135-

Holt, Donald P. "Size of High School and College Grades,"
Personnelwend Guidance Journal, 37:569, April, 1959.

81

Indiana School Study Commission. "An Evaluation of the
Indiana Public Schools," Commission Report, 1959,
Indianapolis, Indiana, p. 332.

Ikenberry, Stanley 0. "A Multivariate Analysis of the
Relationship of Academic Aptitude, Social Background,
Attitudes and Values to College Persistence.”
unpublished Doctoral Dissertation, Michigan State
University, East Lansing, Michigan, 1960, p. 50.

Ikenberry, Stanley 0. and Irvin J. Lehmann. Critical
Thinking, Attitudes and Values in High EducatiorL
East Lansing, Michigan, Michigan State University,

1959-

Ikenberry, Stanley 0. and Irvin J. Lehmann. Critical
Thinking, Attitudes and Values in Higher Education.
A Preliminary Report, Michigan State University
East Lansing, Michigan, 1959, p. 57.

Jacob, Philip E. ghanging Values in College, New Haven,
Conn., Hazen Foundation, 1956.

Kowitz, Gerald T. and William C. Sayres. ”Size, Cost
and Educational Opportunity in Secondary Schools,”
Division of Research, New York State Education
Department, Albany, New York, p. 71

Kuhlen, Raymond G. "Changes in the Attitudes of Students
and Relations of Test Responses to Judgments of
Aseociates," School and Society, 53:514-9, April,
19 l.

Livingston, A. Hugh. "Is There an Optimum Size High
School?" Progressive Education, September, 1956.

McTarnaghan, Roy E. "An Analysis of Certain Personality
Factors and Academic Success in the State University
College, Geneseo, New York.” Unpublished Doctoral
Dissertation, Michigan State University, East Lansing,
Michigan, 1962.

Standards for School Planning. Metropolitan
Planning Commission of Pulaski County, Arkansas,
September, 1957.

Michigan Educational Directory Service. Michigan Education
Directory and Buyer's Guide, 1960-61, Lansing, Michigan.

82

Morgan, H. H. "A Study of Achieving and Nonachieving
Students of High Ability," Journal of Consultipg
Psychology, 16:292-8, 1952.

Morphet, Edgar L., Roe L. Johns and Theodore L. Reller,
Educational Administration, Cencepts, Practices
and Issues, Englewood Cliffs, N. J.: Prentice-
Hall, 1959.

 

National Education Association Research Division.
Efficient School Size, National Education Association
Bulletin, Washington, D. 0., January, 1959.

 

Oliver, A. I. "How Big Should the Small School Be?" School
and Society, 69:128, 1959.

 

Pittenger, Benjamin Floyd. "The Efficiency of College
Students as Conditioned by Age at Entrance and Size
of High School," The Sixteenth Yearbook of the
National Society for the Study of Education, Part II,
Bloomington, Illinois: Public School Publishing
Company, 1919.

Plant, Walter J. "Changes in Ethnocentrism A sociated
with a Four-Year College Education," Journal of
Educational Psychology, 49:109, 1958.

 

 

Prince, Richard, "A Study of the Relationship Between
Individual Values and Administrative Effectiveness
in the School Situation." Unpublished Doctoral
Dissertation, University of Chicago, 1957.

Rokeach, Milton J. "Political and Religious Dogmatism:
An Alternative to Authoritarian Personality,"
Psychological Monographs, 70:425, 1956.

 

H

. "The Nature and Meaning of Dogmatism, Psychological

 

Review 61:203, 1954.

. The Open and Closed Mind, Basic Books, Inc., New
York, 1960.

 

Rust, Ralph M. and F. J. Ryan. "The Relationship of Some
Rorshack Variables to Academic Behavior,” Journal
of Personality, 21:411-56, June, 1953.

 

Sanford, Nevitt. ”Personality Development During the
College Years," Personnel and Guidance Journal,

35:74-80, October, 1956.

83

Saupe, Mildred W. "Size of High School as a Factor in
the College Success of Average and Superior
Graduates," Journal of the American Association
of Collegiate Registrars, October, 1941, p. 45.

 

Shafer, Dwight. "Analysis of Certain Factors in the
High School Preparation of High School Graduates
Entering Selected Iowa Colleges." Unpublished
Doctoral Dissertation, University of Iowa, Iowa
City, 1956.

Shannon, J. R. "Criteria for Determining Local Units
of School Administration," American School Board
Journal,64:28, October, 1949.

Shaw, A. B. "A New Look at Secondary Education: A
Random Fall Idea," School Executive,75:47-86,
March, 1956.

Smith, Clifford B. ” Atudy of the Optimum Size
Secondary School.‘ Unpublished Doctoral Dissertation,
Ohio State University, Columbus, Ohio, 1960.

Stalnaker, J. M. and H. H. Remmers. "What Kind of High
Schools Contribute to College Failures?" Bulletin
of Purdue University,30:23, March, 1930

State Department of Education of Connecticut. Report on
Schools, 1956.

Thornburg, L. H. "College Scholarship and Size of High
School,” School and Society,20:l89, August, 1924.

 

Thorndike, E. L. "A Neglected Aspect of the American High
School,” Educational Review,33:245-255.

 

Tulare County Planning Commission. Tulare City School
District: A Planning Analysis, 1954.

Tulsa Metropolitan Area Planning Commission. Public
School Needs, April, 1960.

 

U. S. Office of Education. Circular No. 634, Fall, 1960.

Webster, Harold. "Changes in Attitudes During College,”
Journal of Educational Psychology,49:109, 1958.

"What Size School is Best?" _Qpinion Poll, Nations Schools,
October, 1954, 54:59.

 

 

84

White, Harland W. "A Study of the High School Record
as a Criterion for Admission to Purdue University,"
Office of Admissions, Purdue University, 1951, p. 25.

Young, John F. "A Comparison of the Academic Preparation
and Achievement of College Students from Various
Size High Schools in Indiana." Unpublished Doctoral
Dissertation, Purdue University, Lafayette, Indiana,

1958.

I L ‘(t.
- his)“;