I
70-15 , i m
SPROULL, Natalie Loraine, 1928THE RELATIONSHIP OF RESEARCH PREPARATION TO
DURATION, ATTRITION AND RESEARCH COMMUNICA­
TION OF DOCTORAL CANDIDATES IN EDUCATION,
MICHIGAN STATE UNIVERSITY.
Michigan State University, Ph.D., 1969
Education, higher

University Microfilms, A XEROXC om pany, Ann Arbor, Michigan

THE RELATIONSHIP OP RESEARCH PREPARATION TO
DURATION, ATTRITION AND RESEARCH COMMUNICA­
TION OF DOCTORAL CANDIDATES IN EDUCATION,
MICHIGAN STATE UNIVERSITY

Natalie L f. Sproull

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Counseling, Personnel Services,
and Educational Psychology
1969

\

PLEASE NOTE:
Not original copy. Some
pages have indistinct print.
Filmed as received.
UNIVERSITY MICROFILMS

ABSTRACT
THE RELATIONSHIP OF RESEARCH PREPARATION TO
DURATION OF PROGRAM, RESEARCH COMMUNICATIONS
STYLE AND ATTRITION OF DOCTORAL CANDIDATES
IN EDUCATION, MICHIGAN STATE UNIVERSITY
By
Natalie L. Sproull

The Problem
Among the problems associated with doctoral study In
Education are those of:
doctoral program,

(1) prolonged duration of the

(2 ) attrition of doctoral candidates at

the "all but dissertation stage", and (3 ) research communi
cation that is less effective than considered desirable.
Therefore, it was the purpose of this study to
Investigate the relationship of the amount of research
p r e p a r a t i o n , through coursework, that doctoral candidates
in Education complete to:

(1) duration of the doctoral

program from admission to Ph.D.,
stage, and

(2) attrition at the ABD

(3 ) effective communication of research through

the dissertation abstract.
In the model from which the hypotheses were derived
was posited that the more task preparation an individual
receives, the better will be his performance.

Thus, the

hypotheses stated that the more research preparation a
doctoral candidate completes during the doctoral program:

Natalie L. Sproull
(1 ) the shorter will be the elapsed time from admission to
doctoral coursework to completion of the doctoral program,
(2 ) the more effective will be the research communication
In the dissertation abstract and (3 ) the less likely it is
that the candidate will drop out at the ABD stage.
Procedures
The Population and Setting
The population selected for this study included all
doctoral candidates who:

(1 ) took their doctoral program

at the College of Education, Michigan State University,
(2 ) transferred no more than nine credit hours from another
institution, and

(3 ) were enrolled in one of four depart­

ments in the College of Education:

Elementary and Special

Education, Secondary Education and Curriculum, Administra­
tion and Higher Education and Counseling, Personnel Services
and Educational Psychology,

(4) completed comprehensive

examinations in the two year period from Fall term 1963
through Summer term 1965.
Candidates who met the above criteria and received
their degree by the end of Summer term 1968 are termed
Graduates.

Candidates who had not graduated by the end

of Summer term 1968 exceeded the university required time
*

limit of completion of the degree within three years of
completing comprehensive examinations, and thus are con­
sidered ABD's.

Natalie L. Sproull
Of the 204 doctoral candidates who met the above
criteria, 180 were Graduates and 24 were considered A B D 1s.
Data Collection and Instrumentation
Data were collected from the doctoral candidate's
university records and dissertation abstract.

Duration

of the doctoral program was measured by number of terms
of elapsed time from admission to doctoral coursework
to graduation

(Ph.D./Ed .D. ).

Attrition rate was computed

on the number of doctoral candidates
graduated by Summer term 1968.

(ABD's) who had not

Effective research com­

munication was defined by the amount of research informa­
tion communicated in the dissertation abstract, as rated
on the Research Information Index; the higher the score
on the R I I , the more effective is the research communica­
tion.

Interrater reliability on the RII was

.87.

Amount

of research preparation was defined as the total number
of credits completed by doctoral candidates in courses
categorized by faculty members as research preparation
courses.
The hypotheses were tested and the data analyzed by
computing Pearson Product-Moment correlations, point
biserial correlations, analyses of variance and chi
squares.

T-he .05 level of significance was preselected

as the criterion for rejecting the null hypotheses.

Natalie L. Sproull
Major Findings
Hypotheses
The findings supported two of the three hypotheses
derived from the premise of the model.

That is, the more

research preparation these doctoral candidates in Educa­
tion completed during the doctoral program, the more
likely it was that they:

(1) completed the doctoral pro­

gram in a shorter amount of elapsed time

(r = -.21) and

(2) communicated research more effectively
However,

(r = .24).

the premise did not hold for attrition.

The

correlation of -.06 indicated that the amount of research
preparation completed was not related to attrition at the
ABD stage.
Duration
1.

Mean elapsed times were:
(1) B.A.-Ph.D., 13 1/2 years,
(2) M.A.-Ph.D., 8 1/2 years, and (3) admission to Ph.D.
5 1/2 years.

2.

Of the 204 doctoral candidates, 43^ (88) extended one
or both of the university time criteria of:
(1) com­
pletion of comprehensive examinations 5 years after
admission or (2) graduation 3 years after comprehensive
exams.

3.

Of the 88 candidates who extended the time criteria,
76/S (6 7 ) extended admission to comprehensive exams
past 5 years.

4.

The three time periods considered extended are:
(1)
B.A. to M.A., 5 years, (2) M.A. to admission, 3 years,
and (3) admission to completion of coursework, 3 3/4
years.

Natalie L. Sproull
5.

Shorter duration of the doctoral program from
admission to graduation was associated with:
(1)
a greater amount of research preparation, (2)
enrolling for a Ph.D. rather than an Ed.D., (3)
holding an assistan tsh ip, (4) higher junior-senior
undergraduate and doctoral GPA, (5) fewer total
program credits, and (6) a behavioral science under­
graduate major.

6.

Duration from admission to comprehensives of four
years or less is similarly related to the variables
listed above in addition to:
(1) having a cognate in
Psychology, and (2) being male.

7.

Affiliation with the department of Counseling,
Personnel Service and Educational Psychology was
associated with: (1) shorter duration from admission
to Ph.D., (2) a greater amount of research prepara­
tion, (3) a behavioral science undergraduate major,
and (4) a cognate in Psychology.

8.

There were no departmental differences in:
(1) pro­
portion of Ph.D.'s, (2) proportion of ass ist ant shi ps,
(3) junior-senior undergraduate and doctoral GPA,
(4) total number of program credits, and (5) p r o ­
portion of males.

Attrition
1.

The attrition rate was 12% for doctoral candidates
at the ABD stage.

2.

ABD's completed more total program credits than
Graduates.

3.

ABD's did not differ from Graduates in any other
elapsed time measures except extending the time after
comprehensives past three years.

Effective Research Communication
1.

Average score on the Research Information Index was
11.27 of a possible total score of 16 points.

2.

More effective research communication was associated
with:
(1) a greater amount of research preparation,
and (2) affiliation with the department of Counseling,
Personnel Services and Educational Psychology.

Natalie L. Sproull
Amount of Research Preparation
1.

Average number of courses, categorized as research
preparation, completed by the doctoral candidates
was 5 1/2 courses.

2.

The average number of courses in each kind of
research preparation course were approximately:
(1)
two courses in t h e o r y , theory construction or logic,
(2) one course in research m e t h o d s , (3) one course in
measurement or evaluation and ( * 0 one and a half
courses in statistics or mathematics.

3.

The average number of research preparation courses
ranged from approximately three to eight courses over
the four departments.

*1.

Candidates from the department of Counseling, Personnel
Services and Education Psychology and candidates with
a cognate in Psychology had a greater amount of
research preparation in theory, measurement and
s t a t is tic s.

5.

Candidates from the department of Secondary Education
and Curriculum and candidates with a cognate in
Sociology had a greater amount of research prepara­
tion in research methods.

ACKNOWLEDGMENTS
l

I am deeply grateful to the people listed below,
for without their encouragement and advice I would never
have known the challenges and joys of the academic pro­
fession.

Therefore,

I would like to express my sincere

gratitude to:
Dr. Robert Craig, committee chairman, whose
consultation throughout the past years was generous,
thorough,

intellectually stimulating and professionally

knowledgeable;
Dr. Wilbur Brookover,

committee member and "signifi­

cant other," who first introduced me to self-concept and
to the rewards and pitfalls of research which stimulated
my interest in higher education;
Dr. Erwin Bettinghaus,

committee member, who sug­

gested programming which has been most beneficial and
provided the professional advice, consultation and
encouragement which enabled me to continue with the
program.
These three individuals in particular held expectat Ions for my performance which made the past years
challenging and productive.

li

D r . William M e h r e n s , committee member, whose
assistance throughout the dissertation process is par­
ticularly appreciated;
Dr.

Paul Campbell, whose consultation over the

years was most generous and greatly beneficial;
My husband, Ken Sproull, who had much patience
and forbearance amidst the occasional chaotic conditions;
My sons, Scott and Kelly, who helped in so many
ways but particularly in the care of their sister,
Shannan.
A research study of this magnitude involves the
cooperation and help of many individuals.
kind assistance,

For their

1 wish to express my appreciation to:

Dr. Walter Scott, Mrs. Virginia Wiseman, Mr.
Richard Brandt and Mr. Marvin Hist from the Graduate
Student Affairs Office;
D r s . Robert Craig, Robert Ebel and James Costar,
who categorized courses for Education;
Drs. James Phillips, Jeanne Gullahorn and Charles
Kenoyer who categorized courses for Psychology;
Drs. Wilbur Brookover and Donald Olmsted who
categorized courses for Sociology;
Drs. Erwin Bettinghaus and Gerald Miller who
*

categorized courses for Communication;
Dr.

Paul Dressel, whose positive remarks encouraged

me to continue with this particular research;
iii

Mrs. Mary Thorsen, who rated the dissertation
abstracts, and her husband, Dr. Jack Thorsen, who helped
through many rewarding "communication11 sessions; and
Dr. Andrew Porter and David Wright, who advised on
many aspects of the research.

iv

TABLE OP CONTENTS

ACKNOWLEDGMENTS

..................................

Page
ii

LIST OF TABLES

..................................

v ii

LIST OF F I G U R E S .....................................

ix

Chapter
I . I N T R O D U C T I O N ................................

1

Purpose of Study
........................
The Problem .
The Duration Issue
.....................
The ABD I s s u e ...........................
Doctoral Study in Education
. . . .
The Dissertation Process
..............
Amount of Research Preparation
. . .
S u m m a r y ..................................

1
1
3
5
7
9
10
13

II.

RELATED RESEARCH AND LITERATURE

.

.

.

.

The Duration I s s u e .....................
The ABD Issue
. . . . . . . . .
The Role of Academic Achievement . . .
The Role of the Dissertation
. . . .
Amount of Research Preparation
. . .
Effective Research Communication . . .
M e t h o d o l o g y ...............................
Content Analysis
........................
S u m m a r y ..................................
III.

THEORY AND O B J E C T I V E S .....................
Amount of Research Preparation
. . .
intervening V a r i a b l e s .................
A General Model for Studying Doctoral
Performance
,.. .........................

v

14
14
16
21
22
23
24
28
30
34
36
36
36

Chapter

Page
Hypotheses
. . . . . . . . . .
Exploratory Questions
.................
Operational Definitions .................

IV.

P R O C E D U R E S ........................

50

The P o p u l a t i o n ............................
Data and I n s t r u m e n t a t i o n ..............
The Research Information Index
. . .
V.

VI.

43
45
- 4 6

50
51
56

F I N D I N G S ......................................

60

Hypotheses
. . . . . . . . . .
Exploratory Questions and Post-Hoc
A n a l y s i s ...............................
Summary
........................

60

CONCLUSIONS AND RECOMMENDATIONS

.

.

.

.

Duration of the Doctoral Program . . .
Attrition at the ABD Stage . . . . .
Research Communication
. . . . . .
Research Preparation . . . . . . .
The Theoretical Model
. . . . . .
Summary of the Findings and Conclusions.
Summary of Recommendations ..............

65
84
89
89
96
100
103
107
109
113

R E F E R E N C E S ....................

115

A P P E N D I C E S .........................................

120

vi

LIST OP TABLES

Table
1.

2.

3.
4.
5.
6.
7.
8.

9.

10.

11.

Page
Correlations of total credits in research
preparation with number of terms admission
to Ph.D. and total score on R I I ..............

6l

Amount of research preparation by groups
based on median elapsed time from admission
to Ph.D. and median score on the RII . . . .

63

Amount of research preparation completed by
graduates and A B D ’s
............................

64

Distribution of number of research preparation
courses completed by graduates and ABD's
. .

64

Correlates of elapsed time from admission to
Ph.D. for graduates
. . . . . . . . .

66

Elapsed time from admission to Ph.D. by
department and major area for graduates

67

. . .

Score on the RII by department and major area
for g r a d u a t e s ...................................

68

Means of variables related to groups based on
median score RII and median elapsed time from
..............
admission to Ph.D. for graduates

70

Distributions of variables related to groups
based on median score RII and median elapsed
time from admission to Ph,D. for graduates . . .

71

Mean number of total program credits completed
for the doctoral program by graduates and
A B D ' s .............................................

72

Mean Aumber of courses in each kind of
research preparation completed by graduates and
A B D ' s .............................................

73

vii

Table
12.

13.

14.
15.
16.

17.

18.

19.

20.
21.
22.

Page
Mean number of credits completed in each
kind of research preparation course by
department and major area for graduates and
A B D ' s .............................................

74

Mean number of credits completed in each kind
of research preparation course by cognate area
for graduates and A B D ' s .........................

75

Mean score on RII by kind of undergraduate
major for graduates
. . . . . . . .

76

.

Distribution of kind of undergraduate major
by departments for graduates and ABD's
. . .

77

Mean elapsed time in terms and years for
various periods of post-bachelor's education
for graduates and A B D ' s .........................

78

Number of graduates and ABD's who extended
duration from admission to comprehensives past
5 years, comprehensives to Ph.D. past 3 years
and admission to Ph.D. past 8 years
. . . .

80

Correlates of ideal duration from admission to
comprehensives 4 years or less and admission to
Ph.D. 5 years or less for graduates and ABD's .

81

Distributions of cognate area, departmental
affiliation and undergraduate major by ideal
duration of admission to comprehensive exams
4 years or less for graduates and ABD's . . .

82

Means of demographic and institutional variables
for graduates and A B D ' s ........................

85

Frequencies of demographic and institutional
variables for graduates and ABD's ..............

86

Correlations of major variables with duration,
attrition and effective research communication.

87

23.

Total scores on the Research Information Index
used *for interrater reliability prior to the
s t u d y ................................................. 139

24.

Total scores on the Research Information Index
used for interrater reliability for the study .

vlii

140

Table

Page

25.

Correlation matrix of major variables for
graduates and A B D ' s ............................ 1^2

ix

LIST OP FIGURES
Figure
1.
2.

Page
A general model for studying doctoral
....................
performance

42

A specific model for studying doctoral
p e r f o r m a n c e .................

43

x

CHAPTER I
INTRODUCTION

Purpose of the Study
The primary objective of this study is to investigate
hypothesized relationships between the amount of research
p r e p a r a t i o n , as measured by number of course hours, co m­
pleted during the doctoral program by doctoral candidates
in Education and:
1.

duration of the doctoral program from admission to
graduation;

?.

completion of the doctoral program (Graduates) or
attrition at the dissertation stage (ABD's);

3.

effective research communication, as measured by the
amount of research information communicated in the
dissertation abstract.
The Problem
To produce the number of doctorate holders required

by societal demands
universities.

is a task of increasing concern to the

Government and business employers, who will

continue to need large numbers of highly educated people,
have also expressed concern about the limited production
%

of doctorate holders
well

founded.

[33:V].

The concern Is apparently

For example, Cartter [14:232] estimates

that one Federal agency alone could absorb the projected

1

2
total output of doctoral graduates In engineering from
1968 to 1972.
One result of this growing demand for doctoral
graduates is that universities are being forced to more
closely examine their own processes and products.

As

the universities attempt to identify ways of increasing
the supply of doctorate holders, attention is being
focused not only on means of attracting individuals to
doctoral study but also on means of improving the
graduation rate of candidates already enrolled in
doctoral study.
This focus on improving the doctoral graduation
rate stems largely from results of an extensive survey
of graduate education conducted by Berelson [9:156-181]
who concluded that:
1.

duration of doctoral study is a major issue in
graduate education;

2.

the number of A B D ’s (All But Dissertation)
increasing.

is

Despite the general recognition of the problems, a
paucity of data exists regarding factors related to
duration and attrition at the doctoral level.

In order

to even partially furnish the number of doctorate holders
required now and in the future, the universities should
m

begin to initiate procedures which would encourage
doctoral candidates to complete the doctorate, and to
complete it within an optimal time span.

In order for

such procedures to be maximally effective, knowledge of

3
the major factors which are associated with extended
duration and attrition is desirable.

Procedural

decisions based on "hard data" are usually more ef fec ­
tive in producing the desired outcomes than are
procedural decisions based on "best guesses".
The Duration Issue
Ideal Time and Actual Time
There seems to be general consensus [35:2] that
three or four calendar years beyond the b a c h e l o r ^ degree
are sufficient to complete all requirements

for the

doctorate.

35] average

Yet, in two recent studies

[31»

elapsed time from entry into post-masters work to com­
pletion of the doctorate ranged from four to nine years
in four academic fields.
The National Academy of Science report.
Recipients

Doctorate

from United States U n i v e rsi tie s, 1958-1966

[23:69-70] shows similar figures.
from M.A. to Ph.D.

Elapsed time medians

for over 30 fields which had doctorate

recipients between 1964 and 1966, ranged from 3*7 to 8.8
years.

Elapsed time medians of over 4 years were reported

in all fields except

five in the physical sciences.

Time in Attendance
However,

it should be noted that the actual time

spent in attendance averages between one-third and
one-half of the total elapsed time and does not vary

much from the normal

time which most educators think a

doctoral program should require

[21:933.

Berelson

[10:129] reports actual time spent in doing work for
the degree in full-time equivalents o f 2.8 years in
Education and 3.5 years in the Arts and Sciences.

Wilson

[35:273 reports mean calendar year equivalents in
attendance which range from 2.9 to 4.0 in four academic
fields.
The data indicate that total elapsed time between
entry into a doctoral program and completing the
doctorate, rather than time in attendance,
lengthy.

is excessively

Thus, as Grigg [21:93] points out, efforts

should be directed toward reducing total elapsed time of
the doctoral program so that it more nearly corresponds
with time needed to complete the program.

To reduce the

total time span o f the degree process would not only
increase the supply of doctorate holders

(more graduates

per year) but would also lengthen the span o f the
scholar*s productive life.
Candidate Control of Time
Although some universities have set time limits at
certain stages of the doctoral program (e.g., that orals
be within 3„years after comprehensive exams),

there usually

is no inviolate time requirement for duration of doctoral
study.

Thus, duration of doctoral study reflects the rate

5
of progress of the student in a program which has neither
a fixed attendance pattern nor a rigid time schedule.
Duration of doctoral study then, is largely con­
trolled by the individual's actions.

In progress toward

the doctorate the candidate may attend full-time or
part-time, on-campus or off-campus; he may enter the
doctoral program at an early age or delay entry until
he is older; he may be employed full-time, part-time or
not at all; and there may be long periods of time in
which he does no work, directly related to completing
the degree.
This loosely structured process of doctoral study
does not lend itself easily to either change or analysis.
Yet, systematic study and analysis is needed before
decisions can be made about probable effective approaches
to reduction of. the time taken by doctoral candidates to
complete their program of study.
The ABD Issue
According to Berelson there is a large group of
doctoral candidates

(estimated to exceed 10,000) who have

completed all requirements for the doctorate except the
dissertation.

Berelson [9:171] states:

They are so numerous and so visable that they have
been given a 'degree' of their own.
They are the
ABD's— 'All But Dissertation'.
Estimates of attrition of doctoral candidates range
from 31 per cent [3 1 ] to *10 per cent

[9 ].

Although these

6
estimates are similar to undergraduate attrition figures,
the results of attrition at the doctoral level, and
particularly at the ABD level, are relatively more serious
in terms of more immediate production of doctorates.
Concern about attrition at the ABD level is related
to the swelling of graduate enrollments which has begun
to create the many problems

concomitant with expansion.

According to Cartter [14:227] expansion in all fields is
expected to continue.

Under such conditions, faculty

time and university resources would be more effective
toward attaining the desired outcome if devoted to
doctoral candidates who would complete the degree rather
than drop o u t .
Chase [15:31] estimated a "rock bottom figure” of
4,500 ABD's for all graduate schools in the Mation in
I960.

This is indeed a rock bottom figure because he

defined ABD's as those candidates who had completed all
course requirements at least 3 years prior to the study,
whom the graduate deans would recommend for a one year
fellowship to complete the dissertation.

Not only was

the figure of 4,500 ABD's an admitted underestimate in
i960 but also, with increasing graduate enrollments, the
number o f ABD's has undoubtedly increased.
There is little question about the need for infor­
mation about, and attention to, the ABD problem.
[15:30] points out,

As Chase

"the supply of ABD's is the most

7
obvious potential source for increased numbers of
doctorates in the shortest possible time".
Doctoral Study in Education
The problems of extended duration and attrition are
particularly crucial for doctoral candidates in the field
of Education.

This is partially because doctoral pro­

duction in the field of Education is larger than any
other field except for the Physical Sciences
and, of the non-science areas,

[2:14,

23:4],

is the most rapidly

expanding field [14:229]*
A comprehensive series of studies conducted by the
American Association o f Colleges

for Teacher Education

focused on conditions affecting pursuit of the doctoral
degree in the field of Education.

Results of these

studies were published in four volumes

[1,2,3 j 4], which

provide a major source of information about doctoral
study in Education.
Information from the AACTE studies suggests reasons
why duration and attrition are widespread in the field of
Education.

In particular, many doctoral candidates in

Education are likely to be:
1.

over 30 years old at admission to the doctoral
program [3:993;

2.

employed during their doctoral program [3:100];

3.

off-campus residents during their doctoral program
[3:993;

8
4.

married with families during their doctoral
program [3:10];

5.

self-supporting during their doctoral program
[3:99].
Duration of doctoral study is lengthy for doctoral

candidates in Education.

Time lapse medians reported by

the MAS [23:69-70] indicate that doctoral graduates in
Education take longer to complete their doctoral study
than do doctoral graduates in any other of over thirty
academic fields.
The number of ABD's is large in Education.

Chase

[15:31] reported that Education has the largest number
of ABD's of ov er sixty fields.

Berelson [10:130]

concludes that Education has a much more serious problem
with ABD's than do the Arts and Sciences.
There is little doubt that administrators and
faculty in the field of Education are concerned about
the problems of extended duration and attrition of their
doctoral candidates.

The AACTE Conference Report

[3:7*0

states that "there is no question that the overall time
span now current can and must be shortened".

In

addition, recommendations of the Conference focused on
the need for a study of the reasons for drop-outs in the
doctoral program [3:28].

The report emphasized that

"the major concern should be directed toward those
factors, other than academic Incompetence, v/hich caused
students to fail to complete requirements" [3:66-68],

9
The Dissertation Process
Berelson [9:156-181] concluded from his study that
the dissertation is one of the major causes of too much
elapsed time.

This conclusion is supported by Wilson's

[35: *13- ^6 ] report that both graduate faculty and recent
graduates cited the dissertation as a major factor in
increasing the duration of doctoral study.
to Grigg [21:61],

According

"the dissertation has been a

stumbling block for many students and has contributed
more than is warranted to an extended length of
elapsed time between matriculation and graduation".
The dissertation process may also be a factor in
attrition at the ABD stage.
Gottlieb and

For example, Tucker,

Pease [31:278] studied attrition of

doctoral students

from four academic fields and found

that a large number of drop-outs indicated that the
research requirement was one of the main reasons for
their attrition.

Grigg [21:93] also emphasizes the

crucial role of the dissertation in the plight of the
ABD.

He adds

[21:9*0 that although "the dissertation

is now considered more of a training device, the length
of time to complete the dissertation has been increased".
Thus,

It appears that one of the pivotal processes
*

involved in both excessive duration of doctoral study and
attrition of doctoral candidates is the ponducting and
reporting of reasearch for the dissertation.

10
Amount o f Research Preparation
As increasing concern is being focused on the role
of the dissertation in duration and attrition of doctoral
candidates,

there is concurrent questioning of the

adequacy of the programming of doctoral candidates
for
s
the dissertation task.

For example, one memorandum to

a college faculty [20:10-11] expressed concern about the
adequacy of the programming of doctoral candidates for
the dissertation tasks of:
research design,

(1) writing skills,

(2)

(3) data analysis and interpretation,

(4) logic and (5) statistics and mathematics.
Apparently the concern about adequate research
preparation holds not only with faculty but also with
recent graduates and drop-outs.
studies

Results of two recent

[31,353 indicate that graduate faculty, recent

recipients of the doctorate, and drop-outs of the
doctoral program all cite the amount o f research
preparation a candidate receives as a factor related to
both attrition and duration of doctoral candidates.
Suggestions for time reduction made by doctoral
graduates in W i l s o n ’s [35:1553 study Included:
research.preparation,

(1) more

(2) better research preparation

and (3) earlier initiation of research preparation.
*

Effective Research Communication
As the adequacy of research preparation for the
dissertation task is being questioned,

it is apparent

11
that, apart from the worth and relevance of the problem
selected, the dissertation is expected to reflect, and
is partially judged upon, some degree of research
sophistication.

That is, at the end of his program of

study, the doctoral candidate is expected to demonstrate
his research prowess and sophistication not only in
conducting but also in reporting the r e s e a r c h .
Research reporting is an important aspect of the
research prowess and, according to the American Insti­
tute for Research [5:1], effective communication of
research is an essential function of researchers.
Effectiveness of the communication is defined as the
amount of information conveyed to the reader of the
research report [5:1].
In an American Institute for Research study

[5]

in which research performance was evaluated by rating
dissertations,

the results indicated that research

communication in dissertations is not as effective as
might be desired.

However, the results of this study

were somewhat tenuous because of the small amount of
agreement between raters.

It is possible that the low

interrater agreement was partially a result not only
of disagreement on what constitutes a "good" product
but also the' difficulties concomitant with evaluating
unique and lengthy dissertations which contain few
commun a li t ie s .

That is, dissertations can vary In many

ways such as format, length, wording and content.

12

However, because almost all doctoral graduates
conduct and report research for their dissertations,
common research report, the dissertation abstract,
available for these doctoral graduates.

a

is

The dissertation

abstract is a product of the doctoral candidate which
summarizes the research problem, procedures and findings.
The abstract also has the advantage of being short.

It

requires more sophistication of the writer to include
the information considered essential to understanding
the research in a short report than in a long one.

The

results of the AIR study, as well as their position on
the importance of research communication,

indicate that

the amount of information conveyed to the reader of the
dissertation abstract would be an acceptable measure of
effective research communication.

Furthermore,

it is

expected that an individual who is adequately prepared
in research will know the kinds of information necessary
to include in the research report in order to communicate
effectively to the reader.

Therefore,

it is posited

here that the more research preparation a doctoral
candidate receives,

the more effective will be the

research communication.
In view of the concerns about attrition,

extended

duration and research communication of doctoral ca n di ­
dates in Education, and the apparently crucial role of
the research requirement in these problems, the objective

13
of this study is to investigate the amount of research
p reparation, and other institutional and demographic
variables, which may distinguish b e t w e e n :
1.

doctoral graduates in Education who spend a short
amount of elapsed time to complete their doctoral
program and doctoral graduates who spend an extended
amount of elapsed t i m e ;

2.

graduates of the doctoral program in Education and
A B D ’s (candidates who have completed all requirements
for the doctorate except the dissertation);

3.

graduates who indicate more effective research
communi cat ion and graduates who indicate less
effective res"earch c o m m u n ic a ti o n.
Summary
Doctorate production by graduate schools is not

increasing as rapidly as the demand for individual's with
doctorates.
reduction of:

Thus,

the universities are focusing on

(1) time spent by candidates in progress

toward the doctorate and (2) attrition rates of doctoral
candidates at the ABD level, as means of improving the
graduation rate of doctoral candidates.

The problems of

extended duration and attrition are particularly crucial
in the field of Education.
Concern has also been expressed about the effec­
tiveness of research communication of doctoral candidates.
The amount of research preparation for the dissertation
requirement has been suggested as a major factor in
extended duration of doctoral study, attrition of doctoral
candidates and effective research communication by
doctoral c a n d i d a t e s .

CHAPTER II
RELATED RESEARCH AND LITERATURE

The Duration Issue
Time Lapse Figures
A major study on duration of doctoral work was
conducted by Kenneth Wilson [35] who gathered data
from:

(1) over 1900 recipients of a doctorate,

(2)

25 graduate deans and (3) 100 representatives of
graduate departments from 23 southern graduate insti­
tutions.

Wilson [35:6] reports that, although a small

number of studies bear somewhat on doctoral time lapse,
none has focused exclusively on the duration issue.
In Wilson's regional study, the selection of
academic fields was limited to Biosciences,
Sciences, Social Sciences and Humanities.

Physical
The field

of Education, which Wilson [35:9] termed "large", was
excluded.
For the fields included In Wilson's study, mean
B.A.-Ph.D.

time lapse ranged from 7.2 to 14.1 years,

with an average overall fields of 9*2 years.
M.A.-Ph.D.

Mean

time lapse ranged from 4.4 to 9.0 years with

an average overall fields of 5*9 years.

14

15
These time lapse figures are similar to those found
by Tucker et a l . for P h . D . ’s In the same four fields.
Tucker [31:68] reported mean B.A.-Ph.D. time lapse ranging
from 7 . 3 to 1 1 . 7 years with an average of 8.9 years.
Mean P o s t - M a st e r- P h. D . time lapse was reported as ranging
from *4.1 to 6.0 years with an average of 4.8 years.
Duration of doctoral study Is substantial in all
fields.

However, duration is longer in the field of

Education than In other fields.
Science report

The National Academy of

[23:69-70] indicates that, duri ng the

period 1964-1966, median B.A.-Ph.D.
Education was, at 13.8 years,

time lapse in

the highest reported for

over thirty f i e l d s .
The M a s t e r s - P h . D . median time lapse of 8.4 years
in Education was the highest of all fields except for
Pine Ai*ts and Music

(8.8 years).

The field of Education

ranked either second highest or highest for all the time
lapse indices reported by the NAS.
Factors Related to Duration
In his study, Wilson contrasted faster and slower
duration groups on several v a r i a b l e s .

The faster group

was defined as those P h . D . ’s with a B.A.-Ph.D.
lapse of les,s than the field median.

time

The following are

some of the factors which Wilson [35:145] found to be
associated with membership in the faster group:

16
1.

lower incidence of pre-doctoral employment;

2.

less total time and less part-time attendance;

3.

lower incidence of delayed entry;

4.

earlier development of interest in doctoral work;

5.

greater continuity of major field;

6.

broader base of financial support;

7.

higher incidence of research-related graduate
appointments;

8.

earlier completion of preliminary examinations;

9.

earlier approval of dissertation topic and more
expeditious completion of the dissertation;

10.

-.

higher incidence of research or research-related
duties in post-doctoral employment;
Duration also varies by sex.

The NAS report

[23:113] shows median B.A.-Ph.D. time lapse as 13*3 years
for males in Education and 16.0 for females in Education.
However, registered time between B.A. and Ph.D.

is the

same for both males and females in Education.
It is apparent

from the reported figures that

duration of time spent by doctoral candidates in progress
toward the degree is:

(1) long in all fields,

for people in the field of Education and

(2) longest

(3) far from the

ideal time lapse of three to four years.
The ABD Issue
Estimated Attrition Figures
Little information is available on either the extent
of, or factors related to, doctoral attrition.

This

17
paucity of data particularly holds concerning attrition
at the ABD level.

Information which is available

consists primarily of estimated figures arid general
impressions.
Berelson [9 ] has estimated the number of ABD's to
exceed 10,000.

Chase [15:31] concluded that a reasonable

"rock bottom" estimate for all graduate schools in the
Nation for 1959-1960, would be approximately 4,500.
However,

Chase [15:30] emphasizes that the ABD's were

defined in restricted terms in order to arrive at the
"rock bottom" estimate.

His definition of ABD's

included those doctoral candidates who had completed
all formal course requirements at least three years
prior to the study, whom the graduate deans would be
willing to recommend for a one-year fellowship to
complete the dissertation.
A major study by Tuck er et a l . [31] provides some
Interesting data concerning ABD's.

Tucker's study,

claimed to be one of the first serious studies of
attrition on a national basis

[31 f o r e w o r d ] ,

included

information on approximately 23*000 post-master students
for who m data was obtained from 24 universities.

The

study was restricted to the fields o f Biosciences,
Physical Sciences,

Social Sciences and Humanities.

Despite the exclusion of the field o f Education, Tucker's
results are relevant to this study.

According to Tucker

et a l . [31:64], attrition of post-master students ranged

18
from 14 per cent to 67 per cent with an average of 38
per cent.

Of these drop-outs,

20 per cent had completed

all but the dissertation [31:67].
Overall attrition rate at the ABD level was reported
as 8 per cent.
expected.

This figure is belcw what might be

However, it should be noted that Tucker

defined ABD's as individuals who were enrolled for
doctoral work between 1950 and 1953 and had completed
all requirements but the dissertation by 1962.
period not only spans twelve years, but,

This time

in addition,

23

per cent of the persons Tucker identified as ABD's began
their post-master study prior to 1949*

Therefore,

the

time lapse between initiation of post-master study and
the ABD stage could range from fourteen years upward
for these people.

If the ABD's had b ee n defined in terms

of a time ,lapse which more nearly corresponds to the
program requirements

(e.g., as those candidates who had

not graduated within three years of completing all
requirements but the dissertation) the attrition rate at
the ABD level may have been different.

In other words,

reported attrition rates of ABD's can vary according to
the definition of an ABD.
Among Tucker's findings was that attrition rate
was lower for students who:
1.

were male [31:57];

2.

attended high quality schools

3.

majored in the sciences

[31:67];

[31:67];

19
4.

did all of their doctoral work at one university
[31:131];

5*

took doctoral work in the same area of study as
at the bachelor's or m a s t e r ’s level [31:1331;

6.

decided before or during high school that they
would enter Ph.D. programs [31:1331;

7.

were definitely committed to a specific field of
study at entrance into the doctoral program [31:1331;

8.

knew a large proportion of the faculty [ 3 1 :20 2 ];

9.

received encouraging advice from a faculty member

[31 :202 ];
10.

had enough money for necessary expenses

[31:2291;

11.

held assistantships [31:2291;

12.

considered doctoral study an excellent investment
[31:2291;

13.

are Caucasian [31:246];

14.

are Jewish or indicated no religious background
[31:246];

15.

had either no children or one child [31:2471*
In Interpreting his data, Tucker et a l . [31:292]

stated as a major conclusion that "it appears that most
of the students who dropped out of Ph.D. programs did
not complete the requirements for the degree mainly
because they lacked sufficient motivation".
Some differences between the findings of Tucker
and Berelson should be noted.
survey,

In B e r e l s o n ’s [9:1691

the two most frequent reasons for attrition
*

cited by graduate deans, graduate faculty and recent
recipients o f the doctorate were lack of finances and
lack in intellictual ability.

Yet Tucker [31:2791

20
reported that some Individuals having less ability than
what might be considered necessary for Ph.D. students
"can and do complete requirements

for the doctorate if

they are strongly motivated".
T u c k e r ’s findings on finances also differ from
expectations.

Less than 20 per cent of drop-out

respondents in Tucker's study indicated that lack of
adequate finances was the main reason for their
attrition [31:229].
As mentioned earlier, the ABD problem is p a rt i c­
ularly widespread in the field of Education.

Chase

[1 5 :3 1 ] surveyed 139 doctorate-granting institutions
and found that Education has the largest number of A B D ’s
reported in any one of over 60 fields.
in Education,

The 66*1 A B D ’s

reported by 57 universities having ABD's,

comprised 16.5 per cent of the total ABD's for the
academic year 1959-1960.

Berelson [10:130] concludes

that because only *10 per cent of dissertations in
Education are completely done at the university,

the

field o f Education has a much more serious problem with
ABD's than is the case in the Arts and Sciences.
Thus, although there is some evidence of factors
which are related to attrition of doctoral candidates,
relatively little is known about factors associated with
attrition at the ABD level.
concensus that:

There is, however, general

(1) there is a substantial number of

21
ABD's and (2) the number of ABD's is greater in the
field of Kducation than in other f i e l d s .
The Bole of Academic Achievement
What little information is available on the
relationship of academic achievement to attrition of
doctoral candidates and duration of doctoral study is
somewhat contradictory.

In Berelson's survey [9:l69 3s

one of the most frequent reasons given for attrition
was lack in intellectual ability.

Yet, Tucker et al.

[31:206] found that 60 per cent of those admitted to
graduate study with undergraduate grade point averages
of less than 3-0 were able to complete the requirements
for a Ph.D. degree.
Tucker [31:27^] adds that although undergraduate
grade point average appears to be a fairly good
predictor of the ability of a doctoral student to
obtain satisfactory grades for graduate course work,
it is not as good a predictor as to whether a student
will successfully complete the research requirement of
a Ph.D. program.

He concludes

[31:210] that Ph.D.

level grade point average Is a better Indicator than
master's or bachelor's grade point average of a
student's 4 potential

for completing requirements for a

Ph.D. degree.
Wilson [35:1^6] feels that not controlling for
"functional abilities" associated with performance in

\

22
Ph.D,. programs complicated his interpretation of observed
differences between faster and slower duration groups.
He adds that little is known about the correlates of
"successful performance" in Ph.D. study.
The Hole of the Dissertation
The dissertation requirement for the doctorate is
undoubtedly one important factor in both attrition of
doctoral candidates and extended duration of doctoral
study.

For example, Tucker [31:278] found that only a

few of the drop-outs reported that one of the main
reasons

for their attrition was unsatisfactory grades.

A considerably

larger number reported that the

research requirement was one of the main reasons *for
their attrition.
Results of Wilson's

[35] study indicate concern

of both recent graduates and graduate faculty about the
role of the dissertation requirement in duration of
doctoral study.

Twenty-five per cent of 1,226 recent

doctoral graduates who offered suggestions for timereductlon of the doctoral program,

suggested modifications

relating to the dissertation [35:1551*

In addition,

graduate deans and faculty members ranked variables
related tq the dissertation second only to continuity of
study as a factor which increases the duration of
doctoral programs

[35:^31*

23
It appears then, that the dissertation is viewed not
only as a lengthening factor in duration of doctoral study
but also as a factor involved in dropping out at the ABD
level.
Amount of Research Preparation
Reports from graduate faculty, recent recipients of
the doctorate and drop-outs of the doctoral program seem
to indicate that the amount of research preparation a
candidate receives is related to both duration and
attrition of doctoral candidates.

As an example, 55

per cent of the 4,7^7 Ph.D. graduates and drop-outs whom
T u c k e r '[31:163] surveyed, considered valid the criticism
that their department "didn't provide enough training
for research and scholarly activities” .

Tucker et a l .

[31:175] consequently found that attrition rates were
higher among those who felt that their department did not
provide enough training for research and scholarly activ­
ities than among those who felt that It did.
Also supporting the need for adequate research
preparation were the modifications relating to the
dissertation suggested by doctoral graduates in Wilson's
[35] study.

These suggestions included:

(1) better

preparation for research and (2) that training in research
techniques be initiated earlier.
faculty in Wilson's [35: M ]

Graduate deans and

study also indicated the

24
importance of research preparation when they suggested
prior experience in research as an "important" variable
related to the dissertation and duration of doctoral
study.
Effective Research Communication
In the late 1950's, the American Institute for
Research initiated a series of studies dealing with the
evaluation and measurement of research performance.
The expressed objective of this project was to evaluate
research through the report.
In the first publication of this series of studies
[53, the importance of the research report as a means of
effective communication among research workers is empha­
sized and the following remarks are stated [5:1]:
Effective communication among research workers Is
an essential and time honored aspect of research.
It is therefore surprising that the effectiveness
of this c o m m u ni c at i on , that i s , the amount of
information conveyedT to different readers by
scientific reports, has not been investigated.
Such an assessment should provide valuable knowledge
concerning the effectiveness with which written
reports communicate to research w o r k e r s . For an
important function of scientific report writing is
to communicate to the qualified reader the contri­
bution which has been made by the reported
research.
Acc ording to the AIR report

[5 ] an additional

advantage* o f examining a written research report is that
It provides a means of evaluating the p r o d u c t .
^Underl ining is added

Many

25
persons specializing in evaluation feel that the product
is the most important

factor to use in evaluation of

educational processes.

For example, Sorenson [28:3]

states that the proper way to evaluate the educational
process is to find out whether they are in fact producing
the hoped-for product.

According to Tyler [32:410-412],

it is more important to evaluate the product than the
process.
The primary problem in evaluation of an unique
product such as the dissertation is that any two or more
evaluators are likely to disagree about the purpose of
the task and thus disagree about the "goodness" of the
product.

A relevant example of disagreement between

evaluators occurred when the AIR used evaluators to rate
dissertations, and a "consistently small amount of
agreement between two independent evaluations of a given
piece o f research through the report" was found [5:39]•
However,

it was emphasized that, despite the poor agree­

ment between raters,

evaluation of research through the

report is an important problem which needs careful
attention [5:40].
Evaluation of the dissertation is a discouraging
task for at least three reasons:
communalftIes among dissertations,

(1) there are few
(2) there is too

little agreement among evaluators and

(3) it requires

a considerable amount of time to evaluate dissertations
which can vary from 40 to over 500 pages.

Since these

26
difficulties in evaluating the dissertation usually exist
to some degree, there is some question about the utility,
particularly when viewed in terms of time and cost factors,
of evaluating the dissertation itself.

Yet, some me asure

of the research performance o f the doctoral candidate is
desirable.
The Dissertation Abstract
One bit of communication which is common to all
dissertations is the abstract.

To a degree,

the abstract

of the dissertation can be likened to a journal article
since both must contain a maximum amount of information
in limited space.

(It is interesting to note that,

according to Berelson [9:183], a growing number of faculty
members advocate a dissertation of journal-article length.)
Although a journal article is usually more detailed than
an abstract,

the purpose of both is to communicate infor­

mation to the reader.

In Guide to a Graduate Degree [2*1],

theses abstracts are discussed as follows:
The major purpose of the abstract is to give infor­
mation which will enable the scholar to decide
whether he wishes to read the complete work.
The
following information is generally included:
(a) A brief statement of the problem
(b) A description of the methods,
and data used

techniques,

(c) The major findings of the study
The nature of these portions of the abstract would
vary with the type of project reported and in some
cases might take quite different form.
In general,

27
however, useful abstracts will contain these three
types of information.
There? are many advantages to using the dissertation
abstract to evaluate research communication.

The major

advantage is that the abstract is a piece of the doctoral
candidate's work.

The candidate writes it as a summary of

what the dissertation supposedly contains.

Therefore,

it

is a product of the doctoral candidate which is amenable
to measurement and evaluation by o t h e r s .
In addition,

the abstract has the four advantages

mentioned in the AIK report [5:3]:
The Report:
1.

Is a permanent record of performance;

2.

Can be made available to several judges or
raters;

3.

Can be evaluated by persons who do not know
the author, thus decreasing personal bias;

1\. Can be evaluated in a single and relatively
brief period, rather than re quiring a series
of observations of job performance.
The AIR studies

indicate that:

(1) communication

of research through the report is an important aspect of
the research process,

(2) the effectiveness of thi3

communication can be measured by the amount of infor­
mation conveyed, and

(3) the research report can be used

to evaluate research performance.
implications,

Prompted by these

in this study the amount of research

information communicated In the dissertation abstract

28
will be used as a measure of effective research
communication of the doctoral graduate.
Methodology
One of the reasons that universities sometimes
lag in research concerning their own processes and
products is that, at times,
such research exists.

considerable resistance to

This resistance can occur not

only within the university setting itself but also from
outside of the institution from individuals who are
concerned with university policies and procedures.
Factors which create resistance to institutional research
and evaluation have been identified as cost [13:288,
27:1019],

confidentiality of material

"no-difference" outcomes

[17:^26], fear of

[13:289], social effects of the

research activity itself [27:1019] and political
vulnerability [13:288].
Institutional Records
Means of coping with resistance based on cost or
social effects have been considered by Campbell
making the following recommendations:
existing institutional records,

[13] in

(1) make use of

(2) incorporate new

measures into institutional records and (3) keep records
on all kinds of institutional experimentation.
In discussing these points, Campbell [13:260-262]
emphasizes that records are nonreactive measures which

29
do not intrude upon the situation and thus alleviate
the problems connected with the artificiality of the
laboratory situation.

The use of nonreactlve or

unobtrusive measures also rescues the researcher from
charges of harmful effects upon the subjects.
Unobtrusive measures are also advocated by Webb
et a l . [3^:vii] who point out that use of such measures
can provide relevant data without identifying individuals
and without manipulating them in any way.

These authors

also note the generally lower cost of data collection of
nonreactive measures as compared with other types of
data collection methods

[3*t:l8 0 ].

Additional advantages of using nonreactive
measures and institutional records are:
generalizabllity

[13:291533:1731*

to the act of measurement

optimal

control of error due

[3^:1751* and the possible

opportunity to expand the usual trivial research base
to larger groups of people and wider settings

[13:258].

Disadvantages of using institutional records occur
when records are:

(1) inaccurate,

(3) out-of-date or (4) nonexistent.

(2) lnaccesible,
As with all documents,

the possibility of selective deposit and selective
retention exists.

Thus universities, a3 well as other

inntltu t ib n s, should attempt to keep their records
accurate, accessible,
feasible.

updated and as comprehensive as

30
Content Analysis
One useful,

but often overlooked,

form of

unobtrusive measures Is content analysis.

In recent

years content analysts have attempted to overcome
the reputation of their research technique as an
unsophisticated, descriptive,

counting procedure by

emphasizing that content analysis studies should
include not only information on content but also
information about the kinds of people who are commu­
nicating and their environments

[12:ix].

Several

authors [12:3,30: 23Y-<*239], agree that the purpose of
content analysis studies is to make inferences from
content to its antecedents or sources,

or from content

to its receivers or effects.
There is also agreement that content analysis is
most useful when it is used in conjunction with other
research methods and includes information on variables
other than those derived from c o n t e n t .

Webb et a l .

[3*1:1] object to the use of interviews and q u estion­
naires as the sole measurement of variables and recommend
that these research techniques be supplemented by
unobtrusive measures of the same variables.
[12: *1] point

Budd et a l .

out that because the analyst is concerned

with the -process and effects of communication,

he should

make use of additional information (attitude, personality,
demographic characteristics)

in order to make better

31
predictions about the relationships among the source,
the receiver and the content and adds that "in such
studies content analysis is considered a tool to be used
in combination with other techniques".
To summarize, the research technique of content
analysis can be useful in the following ways:

(1) to

describe occurrences or nonoccurrences of selected
variables,

(2) to predict relationships among source,

receiver and content, and (3) as an adjunctive tool to
supplement and cross-validate measures obtained by a
different method.
Content Analysis of Insti­
tutional Records
Institutional records of colleges and universities
are a rich source for content analysis studies because
they usually include:

(1) demographic and bibliographic

data of students and faculty,
aptitude,

etc.,

(2) measures of personality,

(3) measures of academic performance and

achievement, and (4) products of student and faculty work.
In addition,

institutional records could include measures

of aspirations,

motivations, morale or any measure

considered sufficiently Informative or valuable to warrant
the cost Involved in obtaining it and incorporating It
into the -records.

Dressel

[17:^10] notes that certain

data (e.g., sex, age, indices of ability, socio-economic
status,

etc.) should be collected continually on all

32
students.

Tf there is a generally recognized need to

include other variables,

cost would probably not be a

decisive factor.
Mot only can a variety of data be extracted from
institutional records but also treatment of the data
can vary.

Procedures range from the relatively simple

recording of information,

for which time and cost

factors are minimal, to the more complex task of a t­
tempting to measure psychological constructs through
analysis of content.

Although the latter procedure

is usually hig her in cost and consumption of time,

it

is more likely to reveal findings of greater import to
general knowledge, while the specific findings of the
former are likely to be of more interest to those
individuals involved in practical, day-to-day decision
making.
There is no attempt here to advocate either end
of this procedural continuum because each tends to serve
the different, but complementary,
above.

Osgood

functions mentioned

[25:37] stresses this compatibility of

approaches when he states

"just as the validation of

many specific inferences by practically oriented users
may provide insights into general relations,

so the

gradually accumulating generalities of the academician
may enrich the base for the practical content a n a l y s t ” .
He also mentions the ideal situation of the "tool

makers" and the

"bool. 110or a "

working in close assoc­

iation .
On the other hand, when Campbell

[13:267]

discusses the problems of extrapolation of findings
he emphasizes that we cannot wait for dependable
generalizations to be developed because need exists
now for cumulative processes where each situation is
evaluated.
Prompted by Campbell's position, and recognizing
that analysis of the loosely structured process of
doctoral study does not lend itself to "dependable
generalizations",
this study.

the cumulative approach is taken for

That is, in this study, attrition and

duration of doctoral candidates in a College of
Education, and factors which are posited to be related,
will be investigated in order to add to the cumulative
knowledge of these problems.

The methodology of this

research will include making use of Institutional
records in a content analysis framework.
Hopefully an additional outcome of the study might
be to demonstrate that:

(1) it is possible and practical

for decision makers to have empirical evidence— of
what ever kind they deem valuable— upon which to help base
decisions’ and (2) a variety of questions can be answered
from the relatively inexpensive research procedures which
make use of institutional records in a content analysis
framework.

34
Summary
1.

Duration of doctoral study,

in terms of the

B.A.-Ph.D. time lapse median of 13*8 years,

is higher

in Education than in any other field.
2.

The M.A.-Ph.D. time lapse median of 8.M years

in Education is far from the ideal time of three to
four y e a r s .
3.

The number of ABD*s is higher in Education than

in any other field.
4.

Lack of motivation is considered a major factor

in attrition of doctoral c a n d i d a t e s .
5.

Evidence about the relationship of academic

achievement to duration and attrition of doctoral
candidates is scanty and somewhat contradictory.
6.

Graduate faculty, recent graduates and ABD's

cite the research requirement for the dissertation as a
major cause of both extended duration and a t t r i t i o n .
7.

Graduate faculty, recent graduates and A B D ’s

express concern about the lack of research preparation
for the dissertation.
8.

Communication of research through the research

report is an Important aspect of the research process
and an expected research-related d u t y .
9.

"For this study, the amount of research

Information communicated in the dissertation abstract
is considered a measure of effective research
c om m un i ca t i on .

35
10.

The use of Institutional records in a content

analysis framework Is a practical and unobtrusive method
of obtaining empirical evidence.

CHAPTER III
THEORY AND OBJECTIVES

Amount of Research Preparation
The review of research indicates

that the disser­

tation process is considered a major cause of extended
duration of doctoral study and attrition of doctoral
candidates.

It Is inferred from this conclusion and

from the research findings discussed that the amount of
research preparation a doctoral candidate receives Is
associated with:

(1) duration of doctoral study,

(2) attrition at the ABD level and (3) effective r e ­
search communication of the doctoral candidate.
These relationships are more readily understood
if we examine some variables posited to be intervening
between the research preparation period and the per­
formance outcomes.
Intervening Variables
Task Familiarity
Most “research pr eparation courses require student
involvement in research tasks such as measurement and
analysis of data, or conducting, reporting or proposing
a research study.

Prom one learning theory viewpoint
36

37
reported

by Craig [16:21], an individual learns what

he does.

Therefore, the doctoral c a n d i d a t e ‘who has

learned to apply theoretical and methodological aspects
of research should have some degree of familiarity with
research ta3k s .

Since he is prepared for and familiar

with the tasks, he should perceive the research process
as less difficult.
Task Difficulty
Atkinson [7 ] reports that tasks of moderate d i f­
ficulty are more likely to evoke achievement orientation
and the desire to do well than are tasks which are
extremely difficult or extremely easy.

It is doubtful

that any doctoral candidate views the dissertation
process as an extremely easy task.

Thus, the doctoral

candidate who is well prepared for the dissertation task
should look upon the task as moderately difficult, rather
than extremely difficult, and consequently should desire
to do well at and be oriented to achieve the dissertation
task.
On the other hand, doctoral candidates who perceive
the dissertation task as extremely difficult would not be
as likely to achieve or do as well as doctoral candidates
for whom the task is moderately difficult.
Ta sk-Created Dissonance
Ac cording to theories o f cognitive consistency, an
individual will continually strive toward consistency or

38
balance.

That is, if there are inconsistencies among a

p e r s o n ’s cognitions, he will try to restore balance by
changing one cognition for another [2 2 :8 ].
Although it is generally agreed that most people
can tolerate some dissonance, it is the proposition of
balance theorists that the reduction of dissonance is
reinforcing to the individual.

Dissonance reduction

is viewed much the same as any other drive reduction
(e.g.,

it is reinforcing to reduce pain).
When dissonance is aroused,

there are several

hypothesized alternative responses to reduce the di s ­
sonance.

According to McGuire

[22:10],

these reduction

modes are not necessarily mutually exclusive.
the reduction modes are believed to

Some of

be [22:10-14]:

1 . repression (put out of mind);
2 . devaluating the task;
3.

submergence (submerging the inconsistency among a
larger body of consistencies);

4.

redefinition of the object (rather than changing
opinion about the object, the object itself is
redefined);

5 . changing opinion about

the object;

6 . changing the object.
In addition to reducing dissonance via some reduction
mode, another response to inconsistency is toleration
and continuance of the dissonance.
As McGuire [22:13] points out, these many
alternative reduction modes make it extremely difficult

39
for the researcher to predict which mode an individual
will use to reduce dissonance, particularly since it
is assumed that people have differing levels of d i s ­
sonance.

However, cognitive dissonance theory does

posit that the greater the amount of dissonance
aroused, the more likely it is that the individual will
use some mode of dissonance reduction.
Therefore, doctoral candidates who are inadequately
prepared for the dissertation task would be likely to
experience some level of task-created dissonance and
consequently, would be likely to either:

(1 ) use one

of the modes of dissonance reduction and complete the
dissertation,

in differing amounts o f time and at dif­

fering performance levels or (2 ) tolerate the
inconsistency and do little or nothing about completing
the dissertation.
The doctoral candidate who is b etter prepared for
the dissertation task could also either use one of the
modes of dissonance reduction or tolerate the
inconsistency and do nothing about the dissertation.
However, because it is assumed that the better prepared
candidate would experience a moderate rather than an
extreme amount of dissonance about the task, it seems
tenable that he would be more likely to:
the dissertation,
time span and

(1 ) complete

(2 ) complete it within an optimum

(3 ) pe rform well on it.

40
Task Motivation
One of the major conclusions of T u c k e r Ts study
[ 3 1 :2 9 2 ] was that motivation was an important factor
in attrition of doctoral candidates.
appears tenable when we consider that:

This conclusion
(1 ) the

dissertation is a major factor in both attrition and
extended duration,

(2 ) the dissertation undoubtedly

creates some level of dissonance in doctoral candidates
and

(3 ) the level of dissonance is related to the

degree of motivation.
There Is some disagreement among authors such as
Pestinger [19:31* Brehm and Cohen [11:228-231] and
Pepitone [26:273] about the complex interaction of
dissonance and motivation.

However, there is general

agreement that the arousal of dissonance and the
dissonance reduction process has motivational aspects.
Furthermore, Feldman [18:87] implies that the greater
the motivation,

the more permanent the effect on

evaluation of an element.
Thus,
candidate,

it is posited that the more-prepared
for whom the task is familiar and not d i f­

ficult, will experience a moderate, rather than an
extreme, amount of dissonance about the dissertation
task and will likely be motivated to achieve and
perform well on the task.

This position is also

supported by Craig [1 6 :5 6 ] who reports that motivation

41
Is aroused by slight to moderate variations of familiar
situations, but not by difficult situations.
A General Model for Studying
Doctoral Performance
In order that the relationships among the several
variables discussed can be more readily visualized, a
general model for studying doctoral performance is
presented in Figure 1.

When the variables discussed

previously are incorporated into the general model,

the

following relationships are assumed to hold:
1.

the greater the amount of research preparation, the
greater the degree of research familiarity:

2.

the greater the amount of research familiarity, the
less the dissertation task will be perceived as
di f f i c u l t ;

3 . the less difficult the dissertation task is perceived,
the less dissonance the dissertation task will create;
4.

the less dissonance produced by the dissertation task,
the higher the motivation to accomplish the task;

5 . the higher the motivation to accomplish the task, the
better will be the performance.
That is:
(1) the
more likely it is that the doctoral candidate will
complete the dissertation and consequently the degree,
(2 ) the shorter will be the elapsed time for com­
pletion of doctoral study and (3 ) the more effective
the research communication will be.
For the purposes of this study, the individual's
dispositions

(task familiarity, task difficulty, dissonance

and motivation) will be considered as intervening var*

iables.

T'hus, the model specific to this study is

presented in Figure 2.

Individual’s Dispositions
^Intervening Variables)

Antecedent
Conditions

Outcomes
\'

Amount of
Task
Preparation

Degree of
►
Task
Familiarity

Perception
• of Task Difficulty

Level of
►
Task
Dissonance

Degree of |
►
Task — «-► Performance
Motivation I
I

Fig. 1.— A general model for studying doctoral performance.

43
Antecedent Condition

Outcomes
Duration of Doctoral Program

Amount of Research
Preparation

^

Completion of Ph.D./Ed.D
Effective Research
Communication

Pig. 2.— A specific model for studying doctoral
performance.

Hypotheses
The hypotheses to be tested in this study are
derived from both the theoretical and established
relationships among relevant variables previously d is ­
cussed and illustrated in the models.
H-,:

H2:

The greater the amount of research preparation
completed b ^ doctoral graduates during their
doctoral program:
a.

the shorter will be the duration of the
doctoral program;

b.

the more effective will be the research
com m un i ca t io n .

For doctoral g r a d u a t e s , effective research co m ­
munication and duration of the doctoral program
are related to the amount of research preparation
completed during the doctoral program as follows:
Effective Research Communication

Elapsed Time
Admission - Ph.D.

Above
Average

Below
Average

Shorter (Below Average)

Group 1

Group 3

Extended (Above Average)

Group 2

Group 4

44
Group 1 > Group 2 > Group 3 > Group 4 on amount of
research preparation completed during -the doctoral
program.
H,:

Doctoral candidates who do not complete the disser­
tation and consequently the degree (ABD*S) will
have completed less research preparation dur ing
their doctoral program than doctoral g r a d u a t e s .

Groups 1, 2, 3» 4 > Group 5 on amount of research
preparation.
For hypotheses two and three, it is posited that
the doctoral candidate who is well prepared for the
dissertation task will experience moderate dissonance
about a familiar task and, therefore, will be motivated
not only to complete the task but to perform well.
Therefore,

the well prepared doctoral candidate would

be likely to be in:
Group 1:

Doctoral graduates who have completed the
doctoral program in a shorter amount of
elapsed time and who indicate more effective
research c o m m u ni c at io n .

On the other hand, the doctoral candidate who is
inadequately prepared for the dissertation task should
experience more than moderate dissonance and would tend
to either use some mode of dissonance reduction or
tolerate and continue the dissonance.

Thus, the inad­

equately prepared doctoral candidate would likely belong
*
to one of the following groups:
»

45
Group 2:

Doctoral graduates who may have prolonged the
time period while acquiring more research
preparation, resulting in extended elapsed
time and more effective research communication.

Group

Doctoral graduates who may have proceeded with
the research despite inadequate preparation,
resulting in a shorter amount of elapsed time
and less effective research c o m m u n i c a t i o n .

3:

Group 4:

Doctoral graduates who may have delayed the
dissertation task unnecessarily long but did
not prepare further in research, resulting in
extended elapsed time and less effective
research communication.

Group 5:

Doctoral candidates (ABD *s ) who avoid the
dissertation task by leaving the institution
without the dissertation and consequently
the degree.

In addition to the hypotheses stated above,

the

following exploratory questions will be investigated.
Kxploratory Questions
1.

II.

To what degree are (a) duration of the doctoral
program, (b) effective research communication and
(c) completion of the doctoral program associated
with:
1.

sex;

2.

departmental affiliation;

3.

kind of degree

4.

assistantship held during the doctoral program;

5.

junior-senior undergraduate grade point average;

6.

doctoral

7.

age at admission to doctoral course work;

8.

score on the filler Analogies Test.

(Ph.D. or Ed.D.);

grade point average;

What is the average number of courses completed in
each kind of research preparation course (theory,
research methods, measurement, statistics)?

III.

IV.
V.

Does the amount of research preparation received
differ among cognate areas completed during the
doctoral program?
Is undergraduate major related to effective
research communication?
Does duration of the doctoral program vary by
major area within departments?
Operational Definitions

1.

Doctoral G r a d u a t e s :

All doctoral candidates who:

(1) completed their doctoral program in the College
of F.ducation, Michigan State University,

(2) trans­

ferred no more than nine credit hours from another
institution,

(3 ) completed comprehensive examinations

wit hin the two year period from Fall term 1963
through Summer term,

1965 and (*l) graduated by the

end of Summer term 1968.
2.

A B D 1s (All But Dissertations):
who:

All doctoral candidates

(1) took their doctoral program in the College

of Education, Michigan State University,

(2) trans­

ferred no more than nine credit hours from another
institution,

(3 ) completed comprehensive examinations

within the two year period from Fall term 1963
through Summer term 1965 and (4) did not graduate by
the end of Summer term 1968, thus exceeding the
university required time limit of completing the
*
Fh.D. within three years of completing comprehensive
examinations.

*17
3.

Effective Research C om m un i ca t io n :

Total score on

the Research Information Index (see Appendix I)
which Is used to measure the amount of research
Information communicated in the dissertation abstract.
The higher the score on the RII, the more effective
Is the research communication.
*1.

Duration of the Doctoral P r o g r a m :
number of terms

Elapsed time in

from admission to (first enrollment

in) doctoral course work to completion of the
doctoral degree.
5.

Amount of Research Prepara t io n :

Total number of

credit hours for all university courses identified
by faculty members as research preparation courses.
6.

Research Preparation Co u r s e s :

Courses which are:

(1 ) listed in the university catalogues,
identified by faculty members
ment or college and

(2 )

for their own d epart­

(3 ) offered primarily to prepare

candidates to apply theoretical and/or methodological
aspects of research
7.

(see Appendix II).

Kind of Research P r e p a ra t io n :

Four categories of

research preparation courses to which faculty members
assigned courses identified as primarily preparing
students to apply theoretical and/or methodological
aspects of research.
(1 ) theory,
methods,

The four categories are:

theory construction or logic,

(3 ) measurement or evaluation,

statistics or mathematics.

(2 ) research

and

(4)

48
8.

Kind of D e g r e e :

The degree, Doctor of Education

(Ed.D) or Doctor of Philosophy (Ph.D.) which:
(1 ) the candidate has received or expected to
receive and
9.

(2 ) is listed on the university record.

Departmental A f f i l ia t io n :

One of four departments

within the College of Education.
departments are:
Education,

The four

(1) Elementary and Special

(2) Secondary Education and Curriculum,

(3) Administration and Higher Education and (4)
Counseling,

Personnel Services and Educational

Psychology.
10.

Major A r e a :

The major within the department of the

College of E d u c a t i o n .

A list of major areas is in

Appendix III.
11.

Cognate A r e a :
of Education

The discipline outside of the field
(e.g., Psychology, Sociology, Co m­

munication) which has been selected by the candidate
and listed on the university records as the cpgnate
area.
12.

Doctoral G P A :

University recorded grade point

average for all courses completed during the doctoral
program.
13.

Junior-Senior G P A :

University recorded grade point

average for courses completed during the juniorsenior years of undergraduate school.

14.

Score on

the M A T :

Raw score on the Miller Analogies

Test taken by the doctoral candidate after c om ­
pleting undergraduate school.
15.

A ss l st a nt s hi p :

Position In the College of

Education held by the doctoral candidate while
completing his doctoral program.

This category

Includes the titles of graduate assistant and
assistant instructor.
16.

Undergraduate M a j o r :

The doctoral c a n d i d a t e 1s

recorded undergraduate major categorized into one
of three areas:
physical

(1 ) behavioral sciences,

(2 )

sciences and (3) other (see Appendix

IV).

CHAPTER IV
PROCEDURES

The Population
The population selected for this study included
all doctoral candidates who:
1.

took their doctoral program at the College of
Education, Michigan State University;

2.

transferred no more than nine credit hours from
another institution;

3.

were enrolled in one( o f the following four d ep a r t ­
ments in the College of Education:
(1) Elementary
and Special Education, (2) Secondary Education and
Curriculum, (3) Administration and Higher Education
and (4) Counseling, Personnel Services and
Educational Psychology;
completed comprehensive examinations in the two year
period from Pall term 1963 through Summer term 1965*
Candidates who met the above criteria and received

their degree by the end of Summer term 1968 are termed
Graduates.

Candidates who had not graduated by the end

of Summer term 1968 exceeded the university required
time limit of completion of the degree within three years
of completing comprehensive examinations, and thus are
considered -ABD's.
The selection criteria excluded doctoral candidates
who transferred to the College o f Education from another

50

51
institution during their doctoral program.

Because

of

their small number, doctoral candidates from the
department of Health, Physical Education and Recreation,
a fifth department within the College of Education, were
also excluded.
It should be noted that completion of comprehensive
examinations Indicates that the doctoral candidate has
completed:

(1 ) 80 per cent o f his course work and (2 )

the language requirement or the language substitute.
Size of the Population
Two hundred and thirty-nine doctoral candidates
completed comprehensive exams from Fall term 1963
through Summer term 1965*

Six candidates were excluded

because they were enrolled in the Department of HPER.
Twenty seven graduates and two ABD*s were excluded because
they had transferred more than 9 credit hours from another
institution.

The 204 doctoral candidates remaining in

the study included 180 graduates and 2_4 ABD *s .
Data and Instrumentation
Data were gathered from two kinds of institutional
records:

(1 ) university records of the doctoral

c a n d i d a t e ’s graduate work and (2 ) the dissertation
%
abstract.

52
Records of Graduate Work
Demographic and institutional data were collected
from university records for each subject included in the
research.

The following variables were recorded or

computed from these data:
1.

total number of research preparation courses completed
during the doctoral program;

2.

total number of credit hours in research preparation
courses;

3.

number and credit hours by kind of research prepa­
ration course (see Appendix II);
sex;

5*

departmental affiliation;

6.

major area within the department;

7.

cognate area;

8.

age at admission to doctoral course work;

9.

age at graduation;

10.

kind of degree (Ed.D. or Ph.D.);

11.

undergraduate major;

12.

total number of credits completed for the doctoral
program;

13.

doctoral GPA;

14.

junior-senior GPA;

15.

raw score on the Miller Analogies Test;

16.

assistantship held;

17.

elapsed time measures;

18.

extended duration Indices;

19.

ideal duration indices.

53
Elapsed Time Measures
In order to determine average duration for different
time periods of post-bachelor's education,

the following

elapsed time measures were computed:
1.
2.
3.

5.
6.
7.
8.
9.
10.
11.
12.

number of years from
number of years from
course work
number of terms from
course work
number of terms from
comprehensive exams.
number of terms from
of oral exams.
number of terms from

B.A.
M.A.

to M.A.
to admission to doctoral

number of years from
number of years from
number of terms from
number of terms from
to Ph.D.
number of terms from
exams to Ph.D.
number of terms from
comprehensive exams.

B.A. to Ph.D.
M.A. to Ph.D.
admission to Ph.D.
completion of course work

admission to completion of
course work to completion of
comprehensives to completion
orals to completion of the Ph.D.

completion of comprehensive
admission to completion of

Extended Duration Indices
Michigan State University has specified time criteria
for completion of different stages of the doctoral program.
These time criteria are:

(1) completion of comprehensive

exams within 5 years after first enrollment in courses
counted toward the doctorate,

(2) completion of the Ph.D.

within 3 years after completion of comprehensive exams and
(3) completion of the Ph.D. within 8 years after first
e n r o l l m e n t ’in doctoral course work.
In order to determine to what degree these criteria
are met, information was recorded on the number of
doctoral candidates who extended:

54
1.

admission to comprehensive exams over 5 years;

2.

comprehansive exams to Ph.D. over 3 years;

3.

a d m i s s i o n to Ph.D.

over 8 years.

Ideal Duration Indices
As mentioned previously, the three or four year
time lapse from B.A. to Ph.D.
from the actual time lapse.

considered Ideal is far
Thus,

in order to obtain

an ideal time lapse which is more realistic at this time,
medians of different elapsed time measures were used as
cut-off points to indicate ideal duration.

For doctoral

candidates In this study, median elapsed times were
approximately 4 years
exams,

from admission to comprehensive

1 year from comprehensive exams to Ph.D. and

5 years from admission to Ph.D.
Thus,

In order to explore correlates of ideal

duration Indices,

information was obtained on the number

o f doctoral candidates who completed:
1.

admission to comprehensive exams in 4 years or less;

2.

comprehensive exams to Ph.D.

3.

adm ission to Ph.D.

in 1 year or less;

In 5 years or less.

Kind o f Research Preparation Courses
In order to determine the kind of research prepa­
ration courses a candidate completes,

courses which

prepare candidates to apply theoretical and/or
methodological aspects of research were originally

55
assigned to six categories by faculty members from each
of four academic areas.

The six categories are:

theory, theory construction or logic,
methods,

(2 ) research

(3 ) measurement and/or evaluation,

and/or mathematics,

(1)

(4) statistics

(5 ) computer applications and (6 )

research design.
Because the computer applications and research
design categories contained only one course each, these
two categories were eliminated.

The course categorized

as computer applications was collapsed into research
methods and the course categorized as research design
was collapsed into statistics.

Thus, the categories used

for the study are:

theory construction or

logic,

(1) theory,

(?) research methods,

(3) measurement or

evaluation and (^) statistics or mathematics.
The four academic areas for which faculty members
categorized courses are:
(3)

Sociology and

(1) Education,

(4) Communication.

(2) Psychology,

Two faculty

members from each of these four areas were asked to
assign "research11 courses to the six categories.

Agreement

was not reached for six courses of the ninety three
categorized for Education and five courses of the forty
seven categorized for Psychology.

Therefore, a third

faculty member in each of these two areas was asked to
categorize those courses about which there was disagre­
ement.

The courses were then assigned to the category

selected by two of the three faculty members.

56

All courses completed under the departments of
Mathematics and Statistics were considered research
preparation courses and categorized under statistics/
mathematics.

Courses from departments other than the

four mentioned above were categorized by the writer
via course descriptions in the catalogues.

The

instructions concerning procedures followed in cate­
gorizing courses can be found in Appendix II.
The Dissertation Abstract
The dissertation abstract of each doctoral
graduate selected for the study was scored on the
Research Information Index in order to measure effective
research communication of the doctoral graduate.
The Research Information Index
The Research Information Index Is an Instrument
developed by the author for the purpose of measuring the
amount of research information communicated In a research
report.
The three main categories of the RII; The Research
Problem, The Research Procedures and The Research
Findings, were formed on the premise that Information in
the research report should usually include:
statement Of the problem,

(1 ) a brief

(2) a description of the

methods, techniques and data used and (3 ) the major
findings of the study.

Conclusions

(recommendations,

57
discussion,

implications),

usually Included in research

reports are omitted from the RII because of their sub­
jective nature.
An underlying assumption of the RII Is that the
bits of information listed above are basic to all
research communication, regardless of the type of research
conducted.

As Budd et a l . [12:46] emphasize, what Is not

included is often more important to understanding the
research than what is Included.
report is scored on:

Therefore, the research

(1) the presence of Information

which is considered essential to understanding the
research and (2 ) the objectivity and specificity of
the information.
As indicated in the explanation of the RII

(see

Appendix I), each item of Information considered
essential for effective research communication Is scored
one if present in the report and zero if absent from the
report.

For example, presence of a statement of the

purpose of the research would be scored one and absence
of such a statement would be scored zero.

In addition, a

score of one is given for Information which is objective
and specific and a score of zero given if it Is not
objective and specific.
to 16 points.

Total score can range from zero

The high er the score, the more research

information Is communicated.

58
In a small pilot study conducted by the author
(2 9 )» score on the RII discriminated among dissertations
abstracts of 12 doctoral graduates.

Results of the

analysis indicated that the higher the score on the RII,
the more likely it is that the doctoral graduate:

(1 )

completed a greater number of research courses and
(2 ) completed his doctoral program in a shorter amount
of elapsed time.

However, the results of this study

must be viewed with caution because:
convenient rather than random,

(1 ) the sample was

(2 ) the number of abstracts

scored was small and (3 ) there were no reliability or
validity indices.
Interrater Reliability
Prior to data collection in the major study, a
random sample of twenty doctoral candidates who were not
included in the population were selected and their
dissertation abstracts were scored on the RII by two
raters working independently.

The interrater reliability

coefficient, a Pearson Product-Moment correlation of
.9 2 , was sufficiently high to warrant use of the Instrument
on the basis of reliability.

The raw scores from which

the reliability index was computed are given in Appendix V.
In order to decrease personal bias, a rater who did
not personally know the doctoral candidates was selected
to score the abstracts for the study.

During the data

collection period, a random sample o f twenty doctoral

59
candidates included in the study was selected for a
check on the Interrater reliability.
Product-Moment correlation was
of interrater reliability.

The Pearson

.87* again a iiigh index

The raw scores from which

the reliability index was computed are shown in
Appendix V.
Validity
The jury method of validation, considered slightly
superior to logical validation [12:69 3* was used to
validate the RII.

Five experts in the theoretical and

methodological applications of research agreed that the
RII appeared to be a valid measure of the amount of
research information communicated in the dissertation
abstract.

CHAPTER V
FINDINGS

The hypotheses were tested and the data analyzed by
computing Pearson Product-Moment correlations, point
biserial correlations, analyses of variance and chi
squares.

The

.05 level of significance was preselected

as the criterion for rejecting the null hypotheses.
Hypotheses
Hypothesis la
Hypothesis la states that the greater the amount
of research preparation completed by doctoral graduates
during their doctoral program, the shorter will be the
duration of the doctoral program.

To test this hypothesis,

a Pearson Product-Moment correlation was computed
between the total number of credits in research pr e pa r a­
tion courses completed by doctoral graduates and the
number of terms from admission to graduation.

As Table 1

indicates, the correlation of -.21 is significant at less
than the

.0.5 level, thus supporting the hypothesis.

60

TABLE 1.— Correlations of total credits in research
preparation with number of terms admission to Ph.D. and
total score on RII.
No. Terms
Ad.-Ph.D.
Total Credits
Research Prep.

180

r

P

21

<.005

Total Score
RII

r

P

<.005

Hypothesis lb
Hypothesis lb states that the greater the amount of
research preparation completed by doctoral g r a d u a t e s . the
more effective the research communication will be.

To

test this hypothesis, a Pearson Product-Moment correla­
tion was computed between the total number of credits in
research courses completed by doctoral graduates and
the total score on the RII.

This hypothesis Is also

supported as the correlation of
less than the

.05 level

.2 *) is significant at

(see Table 1).

Hypothesis 2
For hypothesis 2, doctoral graduates were grouped
as follows:
Effective Research Communication
More
(g.12 on RII)
Elapsed

Less
(<12 on RII)

Shorter
(< 5 years)

Group 1

Group 3

Extended
(> 5 years)

Group 2

Group

62
The hypothesis stated that Group 1 > Group 2 > Group 3
> Group ^ on amount of research preparation.
Median elapsed time from admission to Ph.D. was
five years

(20 terms) and median score on the RII was 12.

Results of the analysis of variance, using total number
of credits completed in research preparation courses as
the score, supported hypothesis 2.

Average number of

credits completed by each group is shown in Table 2.
As indicated by the standard deviations given in
Table 2, Group One, with the greatest amount of research
preparation, was the least variable group in number of
credits in research p r e p a r a t i o n , score on the RII and
elapsed time from admission to c o m p r e h e n s l v e s .
Hypothesis 3
Hypothesis 3 states that A B D 's will have completed
less research preparation during their doctoral program
than doctoral g r a d u a t e s .

As the figures in Table 3

indicate, this hypothesis was not supported by result of
the analysis.

Average number of credits in research

preparation courses completed was 1 7 -1)2 for graduates
and 15.67 for ABD*s.

The standard deviations indicate

that the ABD's varied more than Graduates in the number
of credits completed in research preparation courses.
Upon '•examination of the data, it was found that
2956 of the ABD's completed only one or no research
preparation courses as compared with

of the Graduates.

TABLE 2.— Amount of research preparation by groups based on median elapsed time from
admission to Ph.D. and median score on the RII.

Score
on RII

Group

Terms
E l . Time
Ad.-Ph.D.

2.

3.

4.

More research comm.
shorter duration

13.46

1.16

Standard
Deviation

S.D.

S.D.
1.

N

Mean Number
of Credits
Res. PreD.

14.30

3-52

54

21.63

8.76
On
UJ

More research comm.
extended duration

13.77

1-31

31.38

9.12

39

16.41

10.42

Less research comm.
shorter duration

8.98

1.91

14.45

3.86

42

16.33

10.60

Less research comm.
extended duration

8.60

2.04

31.82

9.27

45

14.67

9.17

11.27

2.91

22.42

10.97

180

17.52

10.00

Overall

P 3,176 = 4 *93, p = *003

64

This post-hoc

finding indicates that future exploration of

the relationship between number of research preparation
courses completed and attrition at the dissertation stage
may be fruitful.

Distribution of number of research p r e ­

paration courses completed by graduates and ABD's is
reported in Table 4 .

TABLE 3.— Amount of research preparation completed by
graduates and A B D ’s.

Graduates
ABD's

N

Mean Number
of Credits
Res. Prep.

180

17.52

10.00

24

15.67

11.96

Standard
Deviation

P

.6956

N.S.

TABLE 4.— Distribution of number of research preparation
courses completed by graduates and ABD's.
Number of Research
Prep, Courses Completed

Graduates
N=l80

0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

4
10
20
19
16
24
18
23
15
11
8
8
2
0
1
1

ABD's
N=24
4
3
0
3
2
1
1
1
3
2
3
1

65

Exploratory Questions and Post-Hcc
Analysis
Although the relationships among the variables
reported in this section were not hypothesized, results
of the analyses are suggestive of their existence.
That is, if these relationships had been hypothesized,
all variables reported here as correlated or related
would have shown a significant relationship at less
than the

.05 level on a two tailed test of significance.

Because significance levels are not appropriate here,
values, means, or distributions are reported.

The

purpose of reporting information about these correlates
is to suggest that further study of the relationship
among these variables is warranted.
Correlates of Admission to
Ph.D. Duration
Average elapsed time from admission to P h .D .
for the doctoral graduates was 22.
mately 5 1/2 y e a r s .

terms or approxi­

In addition to the amount of

research preparation, other correlates

of elapsed time

from admission to Ph.D. are reported in Table 5.
Although the magnitude of the correlations indi­
cates that only a small amount of the variance in elapsed
time is accounted for by these variables, the correlations suggest that these relationships exist.

That is,

the shorter the duration from admission to Ph.D., the
more likely it is that the doctoral candidate:

66

TABLE 5.— Correlates of elapsed time from, admission to
Ph.D. for graduates.
No. Terms
Admission to
Ph.D. with:

N

Statistic

Credits Bes.

180

r

-.21

Jr. - Sr. GPA

121

r

-.19

Doctoral GPA

180

r

-.22

Total Program Credits

180

r

.15

Ph .D . / E d .D .

180

rpb

.38

No A s s i s t . / A s s i s t .

180

rpb

-.31

Undergraduate Major*

170

•Behavioral Science
Physical Science
Other

Value

N

X terms
Ad.-Ph.D.

Standard
Deviation

20
38
112

17.30
25.39
22.40

7.75
13.19
10.41

(1) completed more credits in research preparati on
courses,

(2) received a Ph.D. rather than an Ed. D.,

(3) held an a s s l s t a n t s h i p , (4) received higher grades
both during the junior-senior undergraduate years and
during the doctoral program,

(5) completed fewer total

credits for his doctoral program and

(6) had an under­

graduate major in the behavioral s c i e n c e s .

Age at

admission, sex and score on the Miller Analogies test
were not related to duration, as shown by the correla­
tions given in Table 22.

67
Departmental affiliation and major area within
department were also related to duration of doctoral
study.

Average number of terms from admission to Ph.D.

for each department and major area are reported in
Table 6.

In examining Table 6 and other tables reporting

figures by major area, it should be noted that means
based on a small number of subjects must be Interpreted
with caution.

TABLE 6.— Elapsed time frorn admission to Ph.D. by depart­
ment and major area for graduates.

Department

N

Mean
No. Terms
Ad.-Ph.D.

Elementary &
Special E d u c .

9

23.33

Elementary
Special

Secondary Ed.
& Curriculum

47

22.23

Administration
and Higher
Education

60

Counseling'
Per. Ser. and
Educ. Psych.

64

N

Mean
No. Terms
Ad.-Ph.D.

5
4

30.40
14.50

Secondary
Curriculum
Soc./Phil.
B u s ./ D i s t .
Industrial
Agriculture
Home E c .

2
24
8
5
2
4
2

32.50
21.79
19.75
20.80
33.00
19.00
26.50

25.07

Adult
Higher
Admin.

21
10
29

24.76
20.40
26.90

19.94

Couns./P.S.
Ed. Psych.
M e a s ./
Research

48
7

21.44
19.57

9

12.22

Maj or

68
Correlates of Effective
Research Communication
Average score on the RII was 11.27 of a possible
total score of 16 points.

In addition to amount of

research preparation, score on the RII was related to
departmental affiliation and major area within dep art ­
ments.

Average scores on the RII for each department

and each major area are shown in Table 7.

TABLE 7.— Score on the RII by department and major area
for g r a d uat es.
Mean
Score RII

Department

N

Elementary &
Special Educ

9

11.44

Elementary
Special

47

11.36

Secondary Ed.
& Curriculum

Major

Mean
N Score RII

5
4

11.60
11.25

Secondary
Curriculum'
S o c ./ P h i l .
B u s ./ D i s t .
Industrial
Agriculture
Home E c .

2
24
8
5
2
2

14 .00
11.67
9.88
12.60
13.00
10.50
8 .00

Administration
and Higher
Education

60

10.08

Adult
Higher
Admin.

21
10
29.

9.52
11.60
9.97

C o u n s . Per.
Ser. and
E d u c . Psych.

64

12.28

Couns./P.S.
Educ. Psych.
M e a s ./
Research

48
7

12.65
11.14

9

11.22

Score on the Miller Analogies Test, kind of degree
assistantship held, junior-senior undergraduate and
doctoral GPA, age at admission and sex were not related
to effective research communication.

The correlations

of research communication with these variables are shown
in Table 22.
Grouping by Effective Research
Communication and Duration
from Admission to Ph.D.
In order to synthesize the findings for admission
to Ph.D. duration and effective research communication,
means and distributions of variables related to grouping
on these two dimensions are reported in Tables 8 and 9.
Correlates of Attrition
Twenty-four of the 204 doctoral candidates in this
study who completed comprehensive examinations

from Fall

term 19&3 through Summer term 1965 did not complete the
doctoral program by the end of Summer term 1968.

The

attrition rate at the ABD stage, based on these figures,
is 12%.
Analysis indicated that ABD's completed more total
credits during their doctoral program than did graduates
Because ABD's have not completed their dissertations,
thesis cred-its

(36 credits) are not included in the

total number of program credits.

Table 10 gives the

average number of total program credits completed during

TABLE 8.— Means of variables related to groups based on median score RII and median
elapsed time from admission to Ph.D. for graduates.
Group 1

Group 2

More Res. Comm.
Shorter Duration
N=54
X

More Res. Comm.
Longer Duration
N=39

S.D.

X

21.63

8.76

16.41

3.65

.20

Total Program
Credits

69.70

Age at Ph.D.

34.70

Total Res.
Prep. Credits

Group 4

Group 3

S.D.

Less Res. Comm.
Shorter Duration
N=42

Less Res. Comm.
Longer Duration
N=45

X

S.D.

X

S.D.

10.42

16.33

10.60

14.67

9.17

3.55

.17

3.69

.19

3.56

.27

13.97

69.23

14.19

65.31

14.21

74.22

13.48

5.34

40.79

5.93

35.95

6.16

39.36

7.09

o
Doctoral GPA

TABLE 9.— Distributions of variables related to groups based on median score RII and
median elapsed time for admission to Ph.D. for graduates.*
Group 2

Group 1
*More Res . Comm.
Shorter Duration
%

N

More Res. Comm.
Longer Duration
%

Group 4

Group 3

N

Less Res . Comm.
Shorter Duration
%

N

Less Res. Comm.
Longer Duration
%

!J

N

Degree
Ph.D.
Ed.D.

36
19

(43)
(11)

17
30

(21)
(18)

28
14

(34)
( 8)

19
37

(23)
(22)

121
59

33
32
12
U5

( 3)
(15)
( 7)
(29)

22
23
23
19

( 2)
(11)
(14)
(12)

11
23
32
17

( 1)
(11)
(19)
(11)

33
21
33
19

( 3)
(10)
(20)
(12)

9
47
60
64

23
40

(23)
(31)

26
15

(27)
(12)

18
31

(18)
(24)

34
14

(34)
(11)

102
78

Department
El./Spec.
Sec,/Curr.
Admin/HE
CPS & EP

Assist.
No
Held

*
Read Table 9, 3&% of the Ph.D.'s and 19% of the Ed.D's were in Group 1.
Percentages may not add to 100 because of rounding.

72

the doctoral program.

The standard deviations indicate

that ABD's are less variable than graduates in total
number of program credits completed.

TABLE 10.— Mean number of total program credits completed
for the doctoral program by graduates and ABD's.

Graduates

N

Mean Number
Total Credits

Standard
Deviation

180

69.70

14.18

2>\

75.83

12.55

ABD's

None of the other major variables were related to
attrition.

Means,

frequencies and correlations of de mo­

graphic and institutional variables

for Graduates and ABD's

are given in Tables 20, 21 and 22.
It is interesting to note that while ABD's extend
elapsed time after comprehensive exams

(the criterion by

which they are defined) they do not extend elapsed time
from admission to comprehensive exarns more than graduates
do.

As shown in Table 16, average elapsed time from

admission to comprehensive exams was 17.16 terms for
graduates and 16.88 terms for ABD's.
Kind of Research Preparation
For exploratory question II, the average number of
courses completed in each kind of research preparation
is shown in Table 11.

73
TABLE 11.— Mean number of courses in.each kind of research
preparation completed by graduates and ABD's.

Mean No. of Courses in:

Grads
N-180
X
S.D.

A B D ’s
X

S.D.

Overall
1

S.D.

Theory/Theory Con­
struction/Logic

2 .09

1.46

1.83

1.55

2.06

1.47

Statistics/Mathematics

1.63

1.61

1.71

1.46

1.64

1.59

Research Methods

.96

.85

.75

.90

.93

.86

Measurement/Evaluation

.85

1.14

.75

.94

.84

1.12

Number of courses completed in each kind of research
preparation did not differentiate A B D ’s from Graduates.
Number of credits completed in each kind of research
preparation varied by department and major area within
departments.

Average number of credits completed in each

kind of research preparation course for each department
and major area is reported in Table 12.
As indicated in Table 13» number of credits com­
pleted in each kind of research preparation also differed
by cognate area.

Doctoral candidates whose cognate area

was Psychology completed more credits in theory, me asure­
ment and statistics while those whose cognate was
Sociology completed more credits in research methods.
*
For exploratory question III, Table 13 indicates
that total amount of research preparation was greater
for doctoral candidates whose cognate was Psychology

TABLE 12.— Mean number of credits completed in each kind of research preparation course
by department and major area for graduates and ABD’s.

Mean No.
Credits Theory

Mean No. Credits
Research Methods

Mean No.
Credits Meas.

Mean No.
Credits Stat.

Maj or

N

Elementary
Special

6
6

6.00
6.00

3.50
3.00

5.67
2.50

4.00
7.17

Overall El/Spec

12

6.00

3.25

4.08

5.58

Secondary
Curriculum
Soc/Phil
Bus/Dist
Industrial
Agriculture
Home Economics

2
27
9
5
2
4
3

6.00
5.37
3.33
10.60
1.50
3.00
8.00

1.50
4.33
4.00
1.80
3.00
4.50
5.00

1.50
1.11
.33
.60
3.00
.75
2.00

3.50
4.48
2.67
7.00
3.50
6.00
6.67

Overall Sec/Curr

52

5.37

3.92

1.04

4.57

Adult
Higher
Administration

24
10
33

2.75
3.00
6.12

1.63
4.00
3.24

1.71
2.70
.55

.71
2.80
2.36

Overall Ad/H.E.

67

4.45

2.78

1.28

1.84

Couns/P.S.
Ed. Psych.
Meas./Research

55
9
9

8.67
7.00
9.33

2.18
3.67
2.6 7

4.73
3.67
7.00

8.65
7.89
14.78

Overall CPS & EP

73

8.55

2.42

4.88

9.32

TABLE 13.— Mean number of credits completed in each kind of research prep ration course
by cognate area for graduates and ABD’s.

Cognate

N

Number
Credits Theory
X

S.D.

No. Credits
Research Methods
X

S.D.

No. Credits
Measurement
I

S.D.

No.
Credits
Statistics
X

S.D.

Overall
X

S.D.

Sociology

97

5.52

4.13

3. 44

3.11

1.56

2.ill

4.08

4.2:

14.59

8.25

Psychology

53

8.79

4.28

2.34

2.20

5.98

4.69

9.45

4.4:

26.61

7.87

Other

54

5.04

4.32

2.69

2.66

1.43

1.96

3.91

6.1'

13.06

9.93

204

6.24

4.47

2.97

2.81

2.67

3.64

5.43

5.3^

Overall

76
than for doctoral candidates whose cognate was
Sociology or categorized as "other.”

In addition, the

overall standard deviations indicate that candidates
whose cognate was Psychology varied the least in amount
of research preparation.
Undergraduate Major
For exploratory question IV kind of undergraduate
major was not related to effective research communication.
Average scores on the RII by kind of undergraduate major
are reported in Table 14.

TABLE 14 .— Mean score on RII by kind of undergraduate
major for graduates.

Undergraduate Major

N

Mean Score
RII

Standard
Deviation

Behavioral Science

20

12.45

2.19

Physical Science

38

10.95

2.97

Other

112

11.04

3.01

Overall

170*

11.18

2.94

«

Undergraduate major was not available in the
records of 10 graduates.

It is interesting to note that undergraduate major
%
varied by departmental affiliation.
Distribution of kind
of undergraduate major by departments is reported in
Table 15.

77
TABLE 15.— Distribution of kind of undergraduate major
by departments for graduates and ABD's.

Department

Behavioral
Science
%
n

Physical
Science
%
n

%

Other
n

N

El/Spec

9

( 1)

9

( 1)

82

( 9)

11

Sec/Curr

8

( 4)

20

(10)

72

(36)

50

Ad/Higher

2

( 1)

2*4

(15)

75

(^7)

63

26

(1 8 )

21

(15)

53

(37)

70

C o u n s / P .S ,
Ed. P s y .
Overall

(2*4)

(*41)

(129)

194*

Undergraduate major was not available in the
records of 10 graduates.

Exploratory Question V
Duration of the doctoral program did vary by
departmental affiliation and by major area within the
department.

Average number of terms from admission to

Ph.D. for each department and major area is reported in
Table 6.
Elapsed Time Measures
Average amount of elapsed time for various time
periods of post-bachelor's education are given in Table
16.

All t;j.me lapse measures are from completion of one

stage through completion of the next stage.

Except for

extending the time after comprehensives past three years,

TABLE 16.— Mean elapsed time in terms* and years for various periods of post-bachelor’s
education for graduates and A B D ’s.
■■■

'

........ ....

y —

= ■

Elapsed Time

q

-

,

..

,

.

L.

-n

.

1

Grads
N=l30

In

,

' '

ABD’s
N=24

X

S.D.

X

Overall
N=2-4

S.D.

X

S.D.

B.A. - M . A .

years

4.96

4.27

4.33

2.62

4.89

4.10

M.A. - Ph.D. Admission

years

3.20

3.56

2.63

3.46

3.12

3.54

Admission - Coursework

terms

14.81

9.37

15.08

7.20

14.84

9.12

Coursework - Comprehensives

terms

2.35

5.39

1.79

6.41

2.28

5.51

Comprehensives - Orals

terms

4.98

3.79

—

—

—

—

Orals - Ph.D.

terms

.27

.56

—

—

—

—

B.A, - Ph.D.

years

13.54

5.97

—

—

__

—

M.A. - Ph.D

years

8.72

4.36

—

—

—

—

Admission - Ph.D.

terms

22.42

10.97

—

—

—

—

Coursework - Ph.D.

terms

7.60

6.04

—

—

—

—

Comprehensives - Ph.D.

terms

5.26

3.80

—

—

—

—

Admission - Comprehensives

terms

17.16

10.22

16.88

8.20

17.13

Number of terms reported includes summer term thus, 4 terms per year.

9.99

79
none of the elapsed time measures differentiated ABD's
from Graduates.
Extended Duration Indices
The number of doctoral candidates who extended
admission to comprehensive exams past 5 years, compre­
hensive exams to Ph.D. past 3 years and admission to
Ph.D. past 8 years is reported in Table 17.

Because the

time period selected for the study allowed up to 5 years
from completion of comprehensives to graduation,

8

graduates exceeded the 3 year period from comprehensives
to Ph.D.
Eighty-eight

(43$) of the 204 doctoral candidates

extended at least one of the duration criteria set by the
university.

Sixty-seven

(76%) of these 88 who did

extend the criteria extended the time from admission to
comprehensives past 5 years.
Ideal Duration Indices
Approximate median elapsed times of 4 years or less
from admission to comprehensives,

1 year or less from

comprehensives to Ph.D. and 5 years or less from a dmi s­
sion to Ph.D. were used as ideal time indices.
Correlates of ideal duration from admission to
comprehensives and admission to Ph.D. are reported in
Table 18.

It should be noted that these two indices are

overlapping and therefore include many of the same subjects

TABLE 17.— Number of graduates and ABD’s who extended duration from admission to
comprehensives past 5 years, comprehensives to Ph.D. past 3 years and admission
to Ph.D. past 8 years.
No. Graduates with
El. Time Ad.-Ph.D.
<g yps^
>8 yrs>

txtenaea.

116

None

26

Only Ad-Comp >5
Only Comp - Ph.D. >3

5
—

Both
Overall

147

Total Ad-Comp >5 years
Total Comp-Ph.D. >3 years
Total Ad-Ph.D.

>8 years

—

Total No.
Qrads

No.
AB D’s

Overall

% of 204

116

—

116

57

56

—

56

28

5

16*

21

10

3

3

8

11

5

33

180

24

204

59

8

67

33

8

24

32

15

12-24*

45-57*

22-28*

30
—

33

*
This figure includes 12 ABD’s who, at Summer Term 1968 had not yet extended
the elapsed time from admission past 8 years.

TABLE 18.--Correlates of ideal duration from admission to comprehensives 4 years
or less and admission to Ph.D. 5 years or less for graduates and ABD’s.

N

El. Time
Ad. - Ph.D.
<, 5 years - 0 (N—99)
> 5 years = 1(N=105)

El. Time
Ad. - Comps
<_ 4 years = 0(N=ll6)
> 4 years = 1(N=88)

.18

Degree (Ph.D.=l, Ed.D.=2)

204

.30

Assist. (None=0, Held=l)

204

-.30

-.37

Credits in Res. Prep.

204

-.19

-.25

Jr. - Sr. GPA

142

-.17

-.19

Doctoral GPA

204

-.25

-.23

Total Program Credits

204

.15

.19

ro

.15

CO

204

•

Sex (M=l, F=2)

32
(e.g.

81 of the 10^ doctoral candidates who extended

admission to Ph.D. past 5 years also extended admis­
sion to comprehensives past 4 years).
Cognate area, departmental affiliation and und er­
graduate major were related to one ideal duration index,
admission to comprehensives 4 years or less.

Distribu­

tions of these variables are shown in Table 19.

TABLE 19.— Distributions of cognate area, departmental
affiliation and undergraduate major by ideal duration of
admission to comprehensive exams 4 years or less for
graduates and ABD's.
Ad. - Comps
<4 years
n
%

Ad. - Comps
>4 years
n
%

52
70
54

48
30
46

(47)
(16)
(25)
(88)

97
53
54
204

N

Cognate
Sociology
Psychology
Other
Overall

(50)
(37)
(29)
(116)

Department
El/Spec
Sec/Curr
Admin/HE
Couns P.S./Ed Psych
Overall

50
58
^3
70

( 6)
(30)
(29)
(51)
(116)

50
42
57
30

( 6)
(22)
(38)
(22)
(88)

12
52
67
73
204

79
42
57

(19)
(17)
(74)
(110)

21
58
43

( 5)
(24)
(55)
(84)

24
41
129
194*

U.G. Major
Behavioral Science
Physical Science
Other
Overall,

Undergraduate major was not available in the
records of 10 graduates.

83
To summarize, doctoral candidates who completed
comprehensive exams in four years or less from admission
were more likely to:
1.

be male;

2.

have enrolled for a Ph.D. rather than Ed.D.;

3.

hold an assistantship;

. complete more credits in research preparation
courses;
5.

have higher grades both in the junior-senior
undergraduate years and the doctoral program;

6.

complete fewer total program credits;

7.

have Psychology as a cognate;

8.

be in the Department of Counseling, Personnel
Services and Educational Psychology;

9.

have an undergraduate major in the behavioral
sciences.

Eighty-five doctoral candidates completed the
doctorate in one year or less from completion of compre­
hensive exams.

None of the major variables selected for

this study were correlated with elapsed time from com­
prehensive exams to Ph.D. of one year or less.
Additional Tables
A summary of the means and frequencies of variables
selected for the study and discussed previously are
reported in Tables 20 and 21.

Correlations of major

84
variables are given in Table 22 and Table 25 (Appendix VI).
It is interesting to note that score on the Miller
Analogies Test, often used as a situation criterion for
admission to graduate school, is positively related to
amount of research preparation but is not associated with
duration, attrition or effective research communication.
i

Junior-senior undergraduate GPA, also often used as a
selection criterion, is associated with shorted duration
of the doctoral program and more research preparation but Is
not related to either effective research communication or
attrition at the ABD stage.
Summary
Results of the analyses Indicated support of
Hypotheses la, lb, and 2 but not Hypothesis 3*

Thus,

the greater the amount of research preparation a
doctoral candidate completes- during the doctoral program:
(1) the shorter is the duration of the doctoral program
and (2) the more effective Is the research communication.
The amount of research preparation was not related to
attrition at the ABD stage.

Analyses also indicated

that the shorter the duration from admission to Ph.D.,
the more likely it is that the doctoral graduate:
1.
2.

received a Ph.D. rather than an Ed.D;
*
held an assistantshlp;

TABLE JO.— Means of demographic and institutional variables for graduates and ABD rs.
Grads
M=l8C

ABD's
;i=24

S.D.
Age at Graduation

37.^9

Overall
N=204

S.D.

X

S.D.

6.57 (N=179)

Jr. - Sr. GPA

3.01

# Courses in:
Theory
Res. Meth.
Meas/Eval
Stat/Math

2.09
.96
.85
1.63

1.46
.85
1.14
1.61

1.33
.75
.75
1.71

1.55
.90
.94
1.46

2.06
•94
.84
1.64

1.47

TOTAL # Res. Courses

5.53

3.07

5.04

3.33

5.48

3.16

if Credits in:
Theory
Res. Meth.
Meas/Eval
Stat/Math

6.34
3.06
2.71
5.42

4.45
2.80
3.72
5.48

5.50
2.29
2.37
5.50

4.65
2.80
3.03
4.74

6.24
2.97
2.67
5.43

TOTAL # Credits Res. Prep.

17.52

10.00

15.67 11.96

17.31

Miller Analogies

52.39

15.04 (N=155)

55.20 12.58(N=20) 53.15

Score RII

11.27

2.91

Age at Admission

31.96

5.85 (N=179)

Total Program Credits

69.71

14.18

3.62

.22

Doctoral GPA

.46 (N=121)

.48(N=21)

2.90

-

-

32.46

-

-

2.99

-

-

6.35

32.02

75.83 12.55

70.43

3.62

.18

.86
1.12
1.59

3.62

.21

86

TABLE 21.— Frequencies of demographic and institutional
variables for graduates and A B D ’s
Grads

A B D ’s

Overall

Sex
M
K

163
17

20
4

183
21

121
59

18
6

139
65

9
47
60
64

3
5
7
9

12
52
67
73

102
78

15
9

117
87

87
47
46

10
6
8

97
53
54

20
38
112

4
3
17

24
41
129

Degree
Ph.D.
Ed.D.
Department
El/Spec
Sec/Curr
Admin/HE
Couns. P.S./Ed Psych
Assist.
None
Held
Cognate
Sociology
Psychology
Other
U.G. Major
Behavioral Sci.
Physical Sci.
Other

87
TABLE 22.— Correlations of major variables with duration,
attrition and effective research communication.
# Terms

Total Score

Ad-Ph.D.

RII

180

.2 4

180

-. 06

204

Jr. Sr. GPA

-.19

121

-.04

121

-.09

142

Doctoral GPA

-.22

180

.38

180

-.31

180

Total Program Credits

.15

Age at Admission

Miller Analogies
Male/Female

3.

-.05

204

.04

180

-.04

204

180

.00

180

.14

204

.08

179

-.05

179

.03

203

179

-.08

179

—

-.0*1

155

.02

155

.05

175

.09

180

.00

180

.08

204

•

Age at Graduation

180

«

No A s s i s t . / A s s i s t .

204

•

P h .D ./ E d .D .

180

•

Prep

N

.t

-.21

# Credits Res.

r

o
o

N

1— 1
o

r

o
o

N

CO

r

_
. .
G r a d ./
ABD

received higher grades during the Juniorsenior undergraduate years and during the
doctoral program;

4.

completed fewer program credits;

5.

had an undergraduate major in the behavioral
sciences.

Average duration by departmental affiliation and
major area within departments is reported in Table 6.

—

88

The relationships listed above for duration from
admission to Ph.D. also hold for an ideal duration from
admission to comprehensives of 4 years or less.

In

addition, doctoral candidates who take 4 years or less
from admission to comprehensive^

are more likely to

be male and have Psychology as a cognate.
Effective research communication varied by depart­
mental affiliation and major area within departments as
reported in Table 7.

A B D ’s completed more total pro­

gram credits than did graduates.
Mean elapsed time for various duration periods and
number of doctoral candidates who exceed the university
time criteria are reported In Tables 16 and 17.

Means

and frequencies of demographic and institutional
variables are shown in Tables 20 and 21.

Correlations

of major variables with duration, attrition and effec­
tive research communication are shown in Table 22.
A correlation matrix of major variables for
graduates and ABD's is given in Table 25, Appendix VI.
i

CHAPTER VI
CONCLUSIONS AND RECOMMENDATIONS

A synthesis of the findings reported in Chapter V,
conclusions and recommendations concerning duration,
attrition, research communication and research p repara­
tion are presented in this chapter.
Duration of the Doctoral Program
B.A.-Ph.D.

Elapsed Time

Elapsed time from B.A. to Ph.D. ranged from 3 years
to 38 years.

The average elapsed time from B.A. to Ph.D.

of 1 3 .b years for the doctoral graduates In this study
is very close to the median of 13.8 years reported by
the National Academy of Sciences
graduates in Education.

[23] for doctoral

The means of 8.9 and 9.2 years

from B.A. to Ph.D. reported by Tucker [31] and Wilson
[35] indicate that doctoral graduates in four fields
outside of Education take approximately four years less
from B.A.-Ph.D. than doctoral graduates in Education.
Thus, the findings in this study correspond with
%

findings from other research that B.A.-Ph.D.

time lapse Is:

(1) approximately 13 1/2 years for doctoral candidates

89

In Education, and

(?) longer In Education than other

studies have shown for other* fields.
M.A.-Ph.P.

Elapsed Time

Elapsed time from M.A. to Ph.D.
years to 30 years.
to Ph.D.

ranged from 1 3/^

The average elapsed time from M.A.

of 8.7 years for the doctoral graduates in

this study is also similar to the median of 8.4 years
reported by the N.A.S.
Education.

[23] for doctoral graduates in

The means of 4.8 and 5.9 reported by Tucker

[31] and Wilson [35] indicate that doctoral graduates
in four fields outside of Education take 3 to 4 years
less from M.A.

to Ph.D.

than doctoral graduates

in

Education.
Again, the findings in this study correspond
with findings from other studies that M.A.-Ph.D.
lapse is:

time

(1) approximately 8 1/2 years for doctoral

graduates in Education, and

(2) longer in Education

than other studies have shown for other fields.
Admission to Ph.D. Elapsed Time
Elapsed time from admission to Ph.D.
7 terms

(1 3/4 years)

to 59 terms

ranged from

(14 3/4 years).

Average elapsed time from admission to Ph.D. was 22.42
terms or approximately 5 1/2 y e a r s .

The extended d u r a ­

tion indices given in Table 17 indicate that from 22%28% of the doctoral candidates in this study extended the

91
university criterion of 8 years from admission to Ph.D.
In addition,

Ideal duration indices given in Table 18

Indicate that 10t> or approximately half of the doctoral
candidates extended elapsed time from admission to Ph.D.
past 5 y e a r s .
The findings that approximateley one-fourth of the
doctoral candidates extended admission to Ph.D.
years and approximately one-half past 5 years,
that admission to Ph.D. time lapse is:

past 8
indicate

(1) long for

many doctoral candidates in Education, and (2) far greater
than the time required to complete the doctoral program.
Other Klapsed Time Measures
The elapsed time Indices in Table 16 indicate
which periods of post-bachelor's education might be con­
sidered as extended.

Approximate means in number of

years elapsed time for each of these periods is:

i
i

•
<m

1.

M. A. , 5 years;

2.

M.A.

AD.,

3.

AD.

Course work,

n.

Coursework —

C o m p s , 1/2 year;

5.

Comps

P h .D . , 1 1/4 years

—

3 years;
3 3/4

The time periods which seem to be extended for
these doctoral candidates are:
M.A.

to admission, and

(1) B.A. to M.A.,

(2)

(3) admission to coursework.

92
Correlates of Duration
The finding that the greater the amount of research
preparation a doctoral candidate completes during his
doctoral program, the shorter is the elapsed time from
admission to Ph.D. does support the relationship hy poth­
esized in the theoretical model.

That is, the more prepared

an Individual is, the better will be the per formance— in
this case, shorter duration.

However, until all the

relationships in the model are tested, the full model
must remain tentative.

In addition, the correlates of

shorter duration suggested by the analysis may modify
interpretation of the relationship between research
preparation and duration.
For example, a greater amount of research prep ara­
tion and a behavioral science undergraduate major were
associated with shorter duration from admission to Ph.D.
In addition, a cognate in Psychology was associated with
an ideal duration of 4 years or less from admission to
c om p r eh e ns i ve s .
Counseling,

Affiliation with the department of

Personnel Services and Educational Psychology

not only was related to shorter duration of the doctoral
program but also with the three variables mentioned
above.

That is, doctoral candidates in the CPS & EP

department:

(1) completed more credits in research

preparation courses, and

(2) had cognates in Psychology

and undergraduate majors in the behavioral sciences more

93
often than did doctoral candidates in other departments.
However, there were no departmental differences in the
other correlates of duration:
(2)

(1) proportion of^ P h . D . 1s ,

proportion of a s si s ta n ts h ip s , (3) .junior-senior GPA

and doctoral G P A . (4) total number of program c r e d i t s .
and

(5) proportion of m a l e s .

Thus, although departmental

affiliation is associated with some correlates of shorter
duration, it cannot be suggested as the major factor in
shorter duration.
Upon examining the correlates of duration, one might
speculate that motivation is an important factor in
duration of doctoral study.

For example, although score

on the Miller Analogies Test, considered one of the
better Indicators of academic potential, was not related
to duration, higher grade point average for doctoral course­
work was associated with shorter duration.

It is possible

that those candidates with higher GPA's are motiv ated to
perform well both in coursework and in duration.
When graduate assistantships are considered, again
motivation seems a plausible factor.

It requires a

highly motivated individual to accept a graduate assistantship, often a less prestigious position than previous ones
and usually accompanied by a lower income.
Durfng the time of this study, the requirements for
the Ph.D. in the College of Education included at least
one language examination.

Since the language requirement

94
is often considered one of the more difficult r e quire­
ments, it is probable that candidates who enroll for the
Ph.D. are more highly motivated to attain the degree p e r­
ceived as more prestigious.
In addition, many doctoral candidates in Education
hold responsible, full time positions which leave little
time for doctoral work.

Often, receiving the doctorate

makes no difference either in salary or position for
these Individuals.

For these Individuals to complete

the degree quickly, perhaps at some sacrifice, would seem
to require strong motivation.
This discussion on motivation is, of course, highly
speculative.

In order to discover if motivation is a

plausible factor in duration of doctoral study, further
research is required.
Recommendations
If shorter duration from B.A. to Ph.D.

is considered

desirable, attention might be focused on the three time
periods considered extended; B.A. to M . A . , M.A. to Admission
and Admission to Coursework.

The apparent need is to:

(1) attract candidates into the M.A. and Ph.D. programs
earlier than at present and

(2) after the candidate begins

the doctoral program, find some means of motivating hi m
to complete the coursework period at a faster rate.
An average of almost four years from admission to
work completion does appear excessive.

course

Since the doctoral

95

candidates completed an average of 70 credits, or
approximately 17-23 courses, it is probable that many of
them averaged less than 2 courses per term.

The extended

duration indices indicated that 76% of those who did
extend the university time criteria, extended the time
from admission to comprehensives past 5 years.

As this

period included admission to coursework for 164 of the
doctoral candidates, and since average time from c o urse­
work to comprehensives is one-half year, it seems clear that
elapsed time from admission to coursework completion is one
time period of the doctoral program on which to con cen­
trate efforts toward shorter duration.
Although the amount of research preparation com­
pleted by doctoral candidates did not account for much of
the variance In duration, the correlation between the two
variables suggests that a relationship exists.

In addition,

the standard deviations given in Tables 16 and 20 indicate
that these doctoral candidates are highly variable both
in the length of the doctoral pr ogram and in the amount
of research preparation completed.

Considering that

doctoral candidates in Education are a heterogeneous group
in respect to research preparation and duration of the
doctoral program, and that amount of research preparation
completed is related to duration of the doctoral program,
it is recommended that research preparation be considered
as one focus of the doctoral program for all candidates.

96
Attrition at the ABD
The attrition rate of 1256 for

Stage
the ABD's in this

study is higher than the 8% attrition rate reported by
Tucker [31] for A B D ’s.

However, Tucker selected doctoral

candidates from four fields outside of Education.

Since

attrition rate of ABD's is reported to be higher in Educa
tion [15],

it is possible that this difference in

tion rates

reflects a difference in fields.

attri­

In addition, Tucker defined ABD's as individuals
who were enrolled for doctoral work between 1950 and 1953
and had completed all requirements but the dissertation
by 1962.

In this study, ABD's were defined as doctoral

candidates who completed comprehensive exams

(and there­

fore had completed any language requirements and at least
8056 of the coursework) from Fall term 1963 through Summer
term 1965 and had not graduated by the end of Summer term
1968.

The different time criteria in this study may

partially account for the higher attrition rate.
It should be pointed out that attrition rate at the
ABD stage is probably much lower than attrition at
earlier stages of the doctoral program.

Tucker [31]

reported that attrition of post-master students from 24
universities ranged from 14 per cent to 67 per cent with
an average'of 38 per cent.

Thus, the problem of attri­

tion might be considered twofold:

(1) attrition between

admission to the doctoral program and completion of

97
comprehensives— or all requirements except the di sserta­
tion, and

(2) attrition at the ABD stage.

While attri­

tion during the first time period is un doubtedly larger,
attrition at the ABD stage is probably more serious in
terms of more immediate doctorate production.
Correlates of Attrition
The results of the analyses suggested no correlates
of attrition at the ABD stage except for total number of
program credits.

Because the analyses indicated that the

amount of research preparation was not related to attrition,
hypothesis 3 was not supported.

When hypotheses are not

supported it is customary to examine the selection of the
subjects, the measures used and the theoretical approach
to better understand the results.
For this study the subjects selected were, with the
exceptions noted previously, the total population who co m­
pleted comprehensive exams over a two year period, and had
graduated or not graduated within the following 3 years.
Thus, is appears that selection of subjects followed
acceptable research procedures.
The measure used to define ABD's was non-completion
of the Ph.D. over a 3-5 year period after completion of
comprehensive exams.

This seems to be a logical d efini­

tion of an "all but dissertation" doctoral candidate.
The measure used to define amount of research p r e ­
paration was the number of credits completed by doctoral

98
candidates in courses categorized as research courses.
While this definition could be criticized from several
aspects, such as the point that every different instructor
of a particular course varies the kind of course coverage,
the measure did correlate with both duration and effective
research communication as hypothesized.
Therefore, if the selection of subjects and the
measures used are considered appropriate, and until further
research is conducted, neither the hypothesis nor the
theoretical model is considered tenable concerning the
relationship of attrition at the ABD stage and amount of
research preparation.
The finding that the total number of program credits
is greater for ABD's should probably be interpreted with
caution.

For example, it is possible that doctoral candi­

dates who have not graduated during the time period
expected of them by others, or who have long time lapses
in attendance, may be required to take additional course
work.
The significant outcome of this portion of the
study may be that while ABD's do differ from graduates in
amount of elapsed time after comprehensive exams, they do
not differ from graduates in the amount of elapsed time
from admis'sion to comprehensive exams or other elapsed time
measures.

That is, these ABD's are apparently not "hard

core" laggards since they do not take longer from B.A. to

99
M.A., from M.A. to admission,

from admission to com­

pletion of course work or from course wor k to comprehen­
sive exams than graduates do.

In this study, other than

program credits, A B D ’s seem to differ from graduates only
in amount of elapsed time past comprehensive exams.
Speculating with the same rationale given in the
previous section concerning duration, one might concur
with Tucker [31] who concluded that lack of motivation
(not measured by Tucker or in this study) is a major
factor in attrition of doctoral candidates.
Recommendat ions
In view of the results of this study, it is
recommended that further research be conducted to explore
factors which may differentiate ABD's from graduates.
If it is decided to further explore research p re ­
paration of ABD's, the finding that 2 9 % of the ABD's
completed only one or no courses in research preparation
may provide a productive basis for exploration.

That is,

is there a possibility of a critical or minimum amount
of research preparation below which doctoral candidates
tend to be ABD's and beyond which they tend to be
graduates?

In addition, future studies might include

motivational factors thought to be associated with attri­
tion at the ABD stage.
Because attrition is a problem throughout the doctoral
program, It may also' be productive to examine attrition at

100
different stages of the doctoral programs to see if
similar factors are related to attrition at these d i f ­
ferent stages.
The findings
reported in this study do not rule out
/
the possibility that ABD's experience an extreme amount of
dissonance about the dissertation task.

Because motiva­

tion is aroused by slight to moderate variations of
familiar situations, but not by difficult situations
[16:56], and a moderate amount of dissonance is considered
as motivating, it may be worthwhile to prepare doctoral
candidates in research processes before they attempt the
dissertation.

In this way, dissonance created by the

dissertation task might be kept at a moderate and m o ti ­
vating level rather than at an extreme and non-motivating
level.
Research Communication
Average score on the Research Information Index,
the instrument used to measure effective research communi­
cation, was 11.27 of a possible total score of 16 points.
The findings indicated that the greater the amount
of research preparation completed during the doctoral p r o ­
gram, the more effective is the research communication.
This finding supports the relationship hypothesized in the
theoretical model.

That is, the more prepared an individual

is, the better will be the performance— in this case more
effective research communication.

1

101
Effective research communication was also associated
with departmental affiliation.

The highest average score

on the RII was made by doctoral graduates
ment of Counseling,
Psychology.

in the depart­

Personnel Services and Educational

This finding is commensurate with the career

functions and expectations of that department.

That is,

because of the nature of their current and prospective
careers, both faculty members and doctoral candidates
in the CPS & PIP department are more involved in research
related duties than is usual in other departments.
The amount of research preparation completed during
the doctoral program, a correlate of effective research
communication,

is greater for doctoral candidates in the

CPS & EP department.

This finding follows naturally from

the expectation of career functions of individuals
associated with the department.
Attention might be warranted by the overall average
score of 11.27 on the RII.

Because the instrument con­

sists of 16 points considered essential for effective
research communication, this average suggests that many
doctoral candidates are not aware of the kinds of informa­
tion to be Included in a research report.

While it can be

argued that the dissertation abstract may be written
quickly and carelessly, and indeed it may, it is quite
natural for the individual experienced in conducting and
reporting research to include essential bits of information

102
In the report, even though the report may be short.

It

Is expected that such items as the research problem, p r o ­
cedures and findings would be included in the more lengthy
dissertation.

However, as mentioned previously, a short

research report usually requires more sophistication of
the writer In order to Include the ma ximum amount of
essential information in limited space.

Therefore,

the

abstract, a short report, would be more likely to d is crimi­
nate among doctoral candidates than the longer dissertat ion.
Recommendations
If doctoral candidates in Education, regardless of
departmental affiliation, are expected to communicate
research problems,

procedures and findings effectively,

the findings in this study suggest that these candidates
can be partially prepared for research communication
through coursework.
However,

in view of the average score on the RII,

even doctoral candidates who complete a greater amount of
research preparation coursework do not communicate
research as effectively as might be desired.
it is suggested that,

Therefore,

in addition to coursework, tasks

such as conducting, analyzing and reporting research be
made available during the doctoral program.

Thus, doctoral

candidates will become more familiar with all of these

103
tasks and more doctoral candidates can be associated with
research before they begin the dissertation process.
Research

Preparation

The doctoral candidates in this study completed an
average of 5 1/2 courses characterized as research p r e ­
paration.

Average number of research preparation courses

varies by department from approximately

3 courses for

the department of Administration and Higher Education to
approximately 8 courses for the CPS & EP department.
That is, research preparation courses would vary from
approximately one-eighth to one-third of a 24 course
doctoral program.

The overall average of 5 1/2 research

preparation courses would constitute less than one-fourth
of a 24 course doctoral program.
If research is considered a major function of higher
education, this appears to be a relatively small number
of courses in which to prepare doctoral candidates for
the research function.
The average number of courses in each kind of
research preparation course were approximately:

(1) two

courses in t h e o r y , theory construction or logic,

(2) one

course in research m e t h o d s , (3) one course in me asure­
ment or evaluation and (4) one and a half courses in
statistics or mathematics.
Number of credits completed in each kind o f research
preparation course varied by departmental affiliation.

104
As shown in Table 12, doctoral candidates in the CPS & EP
department completed more credits in theory, measurement
or evaluation, and statistics or mathematics.

Doctoral

candidates from the department of Secondary Education
.and Curriculum completed more credits in research methods.
Candidates

from the department of Administration and

Higher Education completed fewer credits in theory and
statistics than any of the other three departments.
Cognate area was also associated with kind of
research preparation.

Doctoral candidates whose cognate

area was Psychology completed more credits in theory,
measurement

and statistics while those whose cognate was

Sociology completed more credits in research methods.
Thus, candidates in the CPS & EP department and
candidates whose cognate was Psychology completed more
credits in theory, measurement and statistics.

Candi­

dates in the Secondary Education and Curriculum d e pa r t­
ment and candidates whose cognate was Sociology completed
more credits in research methods.

Because cognate area

completed is associated with departmental affiliation,
this finding is not unexpected.

That is, many candidates

from the department of CPS & EP have a cognate in
Psychology and many candidates from the Secondary Educa*
tion and Curriculum department have a cognate in
Sociology.

105
Recommendations
Ultimately, the decision concerning the importance
of research preparation for doctoral candidates lies with
the separate departments and the candidates'

guidance

committees, each with somewhat differing goals and
expectations for the individuals involved.
However, to
or no familiarity

produce doctoral graduates with little
with research in an

age when research,

development and di ssemination are highly desirable and
often well supported functions is to do both the doctoral
candidates and society as a whole a disservice.

It is

evident that society today needs individuals who can make
rational

judgments.

Judgments

made after consideration

of evidence, appropriately gathered and analyzed, are
almost always better than Judgments
tion or erroneous

information.

It is

of individuals in

higher education to

based on no informa­
one of the functions
see that the

appropriate information is available.
Therefore, it would seem almost obligatory to prepare
doctoral candidates in research so they might better
understand, conduct and disseminate research.
It is not always necessary to require separate
courses in order to better prepare doctoral candidates in
various aspects of research through coursework.
the better methods of learning is by doing.

One of

Thus, an

individual who is required to conduct a study associated

106

with the content of a particular course can gain much
Information about the research process.
Therefore, It Is recommended that doctoral candi­
dates be prepared for research related duties by:
1.

completing coursework in research-related courses;

2.

including research-related tasks in courses not
considered as primarily research courses;

3.

providing opportunities for doctoral candidates to
be associated with research-related tasks outside
of coursework.
These three related experiences

should be provided

prior to the beginning of the dissertation process so
that the doctoral candidate can approach the dissertation
task with a greater degree of preparation and familiarity
which should, in turn, produce less dissonance and greater
motivation.

The prepared doctoral candidate is more

likely to accomplish the dissertation process with
realistic expectations and appropriate research p r o ­
cedures .
The dissertation, like any other research project,
is hard work and usually requires several months.

However,

the dissertation process should not be so traumatic as
to be a possible factor in duration and attrition of
*
doctoral candidates nor should it be so unfamiliar a task
as to lead doctoral candidates to unrealistic time expec­
tations, unrealistic task expectations or inappropriate
research procedures.

107
The Theoretical Model
The premise of the model given in Chapter III was
that the more prepared an individual is, the better he will
perforin.

The finding:’ supported two of the three

hypotheses derived from this premise.

That is-, the more

research preparation a doctoral candidate completes, the
more likely it is that he:

(1) completes the doctoral,

program in a shorter amount of elapsed time, and
communicates research more effectively.

(2)

However, the

premise did not hold for attrition at the ABD stage.
The model also included intervening variables betweeh
the antecedent task preparation and the performance out­
comes.

It was posited that:

of task preparation,
familiarity,
difficult,

(1) the greater the amount

the greater the degree of task

(2) the less the task will be perceived as

(3) the less dissonance the task will create,

(4) the higher will be the motivation to accomplish the
task, and (5) the better will be the performance.
Given that:

(1) shorter duration was related to a

higher grade point average, a Ph.D. rather than an Ed.D.,
and assistantship held, and

(2) that it appears to require

greater motivation to attain these three goals, motivation
may be a plausible factor to consider in duration of
4

doctoral study.
In addition, both effective research communication
and attrition at the ABD stage undoubtedly involve

108
motivational aspects.

Therefore, the Intervening variable

of motivation posited in the model is considered a factor
worthy of future research.

Future exploration of research

pre paration of doctoral candidates in Education might also
include measures of research familiarity, candidate p e r ­
ception of the difficulty of the dissertation task and the
amount of dissonance created by the dissertation task.
It seems apparent from this study that modifications
are needed in order to Improve prolonged duration of the
doctoral program, attrition rate at the ABD stage and
research communication of doctoral candidates in Education.
i
While more research preparation is one factor to consider,
it is not advocated as a panacea.

What is needed is an

en compassing p ro gram which includes processes and p r o ­
cedures associated with improvement in these problems.
The findings and conclusions from this study and other
research previously cited suggest that future investigations
and current attention might be focused on:

(1) the role

of factors such as a s s i s t a n t s h i p s , grade point averages
and research preparation in better performance in these
pr oblem areas and (2) the development of motivation for
better performance in these problem areas.

However, because

of the unstructured nature, of the doctoral program, efforts
toward Improvement of these problems will probably have
to be generated from wi thin the institution rather than
from the doctoral candidates.

109
Summary of the Findings and
Conclusions
Duration
1.

The B.A.

to l'h.D. average time lapse of 13 1/2 years

and the M.A.

to ih.IJ. average time

lapse of 8 1/2

years for doctoral graduates in Education in this
study is longer than elapsed times reported in other
studies for doctoral graduates

in fields outside

Education.
2.

The admission to Ph.D. average time lapse of 5 1/2
years indicates that duration of the doctoral program
is not only long but also far greater than the time
required to complete the doctoral program.

3.

Forty-three per cent [88] of the 20^ doctoral candi­
dates extended one or both of the university time
criteria ofi

(1) completion of comprehensive exams

5 years after admission, or (2) graduation 3 years
after comprehensive examinations.
4.

Seventy-six per cent

[6 7 ] of the 88 candidates who

extended the time criteria, extended admission to
comprehensive exams past 5 years.
5.

Three time periods are considered extended:
B.A. to M.A.,

(2) M.A. to admission, and

(1)

(3)

admission to completion of coursework.
b.

Although the magnitude of the correlations account
for little variance in duration, shorter duration of

110
the doctoral program from admission to Ph.D.
associated with:
preparation,
an Ed.D.,

is

(1) a greater amount of research

(2) enrolling for a Ph.D. rather than

(3) holding an a s s i s t a n t s h i p , (4) higher

junior-senior undergraduate and doctoral GPA,

(5)

fewer number of total program credits completed, and
(6) a behavioral science undergraduate major.
7.

Duration from admission to comprehensives of four
years or less is similarly

related to the variables

listed above in addition to:
in Psychology, and
8.

(1) having a cognate

(2) being male.

Affiliation with the CPG & El' department was asso­
ciated with:
to Ph.D.,

(1) shorter duration from admission

(2) a greater* amount of research preparation,

(3) a behavioral science undergraduate major, and
(
9.

a cognate in Psychology.

There were no departmental differences in:
proportion of Ph.D.'s,
ships,
GPA,

(1)

(2) proportion of assistant-

(3) junior-senior undergraduate and doctoral

(4) total number of prog ram credits, and

(5)

proportion of males.
10.

Although affiliation with the CPU & EP department is
associated with correlates of duration, departmental
%
affiliation is not suggested as the major factor in
duration.

Ill

11.

Motivation is probably an important factor in
shorter duration of the doctoral program.
«

Attrition
1.

It is probable that the attrition rate of 12% for
ABD's is higher than the 8% reported for other fields
because of:

(1) the difference between Education

and other fields, and

(2) the differing definitions

of ABD's.
2.

Attrition rate at the ABD stage is undoubtedly lower
than the attrition rate at earlier stages of the
doctoral program.

3.

Since amount of research preparation was not
associated with attrition and selection of the
subjects and measures used is considered appropriate,
the relationship hypothesized In the theoretical model
is not considered tenable at this time.

4.

Because various interpretations could be made of
the finding that ABD's completed more total program
credits than Graduates, this finding should be
Interpreted with caution.

3.

Perhaps a significant finding is that ABD's do not
differ from Graduates

in any other elapsed time

measures except extending the time after comprehensive^
past three years.
core" laggards.

Thus, these ABD's are not "hard

112
6.

Lack of motivation id probably a major factor in
attrition.

Research Communications
1.

More effective research communication is
associated with a greater amount of research
proparat i o n .

2.

That more effective research communication and a
greater amount of research preparation are
associated with affiliation with the CPS & EP
department

is commensurate with the career functions

and expectations within that department.
3.

Average score on the RII suggests that many doctoral
candidates are not aware of the kinds of Information
to be included in a research r e p o r t .

Research Preparation
1.

The average of

1/2 research preparation courses,

which ranged from approximately

3 to 8 courses over

departments and constituted from approximately oneeighth to one-third of total program courses,
appears to be a relatively few number of courses
devoted to preparation for a major function of

2.

higher education.
*
Since many candidates in the CPS & EP department
have a cognate

in Psychology and many candidates in

the Secondary Education and Curriculum department have

113
3.

cognate In Sociology,

it is not unexpected that:

(1) candidates from CPG & EP and candidates with a
cognate in Psychology have a greater amount of
research preparation in theory, measurement and
statistics, while

(2) candidates from the Secondary

Education and Curriculum department and candidates
with a cognate in Sociology have a greater amount of
research preparation in research methods.
Summary of Recommendations
1.

If shorter duration from B.A. to Ph.D.

is desirable,

attention should be given to attracting candidates
into the M.A.

and Ph.D. programs earlier than at

present.
2.

After candidates begin the doctoral program,
efforts should be concentrated on finding some
means of motivating them to complete the admission
to completion of coursework period at a faster rate.

3.

Because a greater amount of research preparation
is associated with shorter duration of the doctoral
program, research preparation might be considered
one focus of the doctoral program for all candidates.

.

To avoid having candidates who prolong the doctoral
program or who drop out at the AliD stage,

it may

be worthwhile to prepare candidates in research so
that the dissertation process is perceived as
familiar and not overly difficult.

Thus, dissonance

about the dissertation task might be kept at a
moderate and motivating level rather than at an
extreme and non-motivating level.
If additional study is concerned with research
preparation of ABD's, exploration of a possible
minimum number of research courses may be fruitful.
Future research on attrition might be more pro­
ductive If motivational measures are included.
Exploration of attrition rate and factors associated
with attrition at different stages of the doctoral
program would be valuable.
To better prepare doctoral candidates in research
the College might arrange for more doctoral candi­
dates to be involved with:
coursework,

(1) research-related

(2) research-related tasks in non­

research courses, and

(3) research-related tasks

outside of coursework.
The three related experiences listed above should
be provided to doctoral candidates prior to the
beginning of the dissertation process.
Future exploration of research preparation of
doctoral candidates in Education might include
measures of:

(1) motivation,

(2) research

%

familiarity,

(3) candidate perception of the d i f­

ficulty of the dissertation task, and

(*1) the amount

of dissonance created by the dissertation task.

REFERENCE

REFERENCES

American Association of Colleges for Teacher
Education.
The Doctorate in Education:
An
Inquiry Into Conditions Affecting Pursuit of the
Doctoral Degree in the Field of Education— Vol. I :
The G r a d u a t e s . Washington, D.C.:
The Author, I960.
_______ . The Doctorate in Education— Vol. II;
The
I ns t it u ti o ns . Wa s h i n g t o n , D .C .: The Author, I 9 6 0 .
_______ . The Doctorate in Education— Vol. I l l :
Conference Report.
Washington, D.C.:
The Author,

1961 .
_______ . The Doctorate in Education— Vol. I V :
Follow-Up S t u d y . Washington, D.C.:
The Author,
1964.
American Institute for Research.
Development of
a Record Form for Evaluating Research Through the
Report I Pittsburgh :
The Author, 195*1.
_______ . A Procedure for Evaluating Graduate
Research on the Basis of the T h e s i s . Pittsburgh:
The Author, 1955.
Atkinson, J. W.
Princeton, N.J.:

An Introduction to M o t i v a t i o n .
Van Nostrand, 1964.

Beach, Leonard B.
"The Graduate Student."
Graduate Education T o d a y . Edited by Everett Walters
Washington, D.C.:
American Council on Education,
1965.
Berelson, Bernard.
States.
New York:

Graduate Education in the United
McGraw-Hill Book Company, Inc.,

I960.
_______ . "Graduate Education Today" in American
Association of Colleges for Teacher Education, The
Doctorate in Education— Vol. Ill:
Conference Report
W a s h i n g t o n , D .C .: The Author, 1 9 6 l .

117

11.

Brehrn, J. W . , anil Cohen, A. R.
Explorations In
Cognitive 1>1 ssonance . IJew York:
Wiley, 1962 ,

I.’.

Rudd,

Richard

W .; Thorp,

Robert

K.;

and

Donohew,

br-wlu . (‘onl’Tit A n a l y s i s of* Communlcat. Iona .
Y<>r*k : 'J'iTe MucM 1 LJart C o m p a n y , I'i6 '(.

Mew

13.

Campbell, Donald T.
"Adminlutrat 1.ve Experimentation,
institutional Records, and Nonreactive Measures."
Improving Experimental Design and statistical
A n a l y s i s . Edited by Julian Stanley.
Chicago:
Rand McNally and Company, 1967.

IN.

Cartter, Allan M.
"The Decades Ahead:
Trends and
Problems."
Graduate Education T o d a y . Edited by
Everett Walters.
W a s h i n g t o n , C.D.:
American Council
on Education, 196 5.

19.

Chase, John L.
Doctoral S t u d y . U.S. Department of
Health, Education, and Welfare, Office of Education,
Circular Mo. 6N6.
Washington, D.C.:
Superintendent
of Documents, Government Printing Office, 1963.

16.

Craig, Robert D.
The Psychology of Learning in the
Cl ass r o o m . New York:
The MacMillan Company, 1966.

17.

Dressel, Paul L. and Associates.
Evaluation in
Higher Education.
Boston:
Houghton Mifflin Company,
1961.

18.

Feldman, S.
"Motivational Aspects of Attltudinal
Elements and Their Place in Cognitive Interaction."
Cognitive C o n s i s t e n c y . Edited by Shel Feldman.
Mew York:
Academic Press, 1966.

19.

B’estinger,
Stanford:

20.

Fusak, John A. and Cobb, T. Clinton.
"Memorandum
to Faculty," College of Education, Michigan State
University, October, 1967.
(Mimeographed.)

01.

GrLgg, Charles M.
Graduate E d u c a t i o n . New York:
The Center for Applied Research in Education, Inc.,
ll,o 9 .

2J.

McGuire, William J.
"The Current Status of Cognitive
Consistency Theories."
Cognitive C o n s i st e nc y .
Edited by Shel Feldman.
Mew York:
Academic Press,

L.
A Theory of Cognitive D i s s o n a n c e .
Stanford University Press, 1957.

%

1966

.

118
23.

National Academy of Sciences.
Doctorate Recipients
from United States Universities 1958-1966.
Publication 1489.
Washington, D.C.:
The Author, 1 9 6 7 .

24.

Office of Advanced Graduate Studies.
Guide to a
Graduate D e g r e e . East Lansing:
Michigan State
University,1985.

25.

Osgood, Charles E.
"The Kepresentational Model and
Relevant Researcli Methods."
Trends in Content
A n a l y s i s . Edited by Ithiel de Sola Pool.
Urbana:
University of Illinois Press, 1959.

26.

Fepitone, Albert.
"Some Conceptual and Empirical
Problems of Consistency Models."
Cognitive Con­
sistency . Edited by Shel Feldman.
New York:
Academic Press, 1 9 6 6 .

27.

Sanford, Nevitt.
The American C o l l e g e .
John Wiley and Sons^ I n c . , 1966.

28.

Sorenson, Garth.
Evaluation C o m m e n t . Los Angeles:
Center for the Study of Evaluation of Instructional
Programs, January, 1968, Vol. 1, No. 1, p. 3.

29.

Sproull, Natalie,
"Nonreactive Research and Evaluation
in Higher Education:
An Example Using Institutional
Records and Content A na l y s i s . 11 Occasional Paper No. 2,
Office of Research Consultation, School for Advanced
Studies, College of Education, Michigan State Univer­
sity, June, 1 9 6 8 . (Mimeographed.)

30.

Stone, Philip J.; D u n p h y , Dexter C.; Smith, Marshall
S.; and Ogilvie, Daniel M.
The General I n q u i r e r .
Cambridge:
The M.I.T. Press^ 1966.

31.

Tucker, Allan, Gottlieb, David, and Pease, John.
Factors Related to Attrition Among Doctoral S t u d e n t s .
East- Lansing, Michigan:
Michigan State University,
1964, Cooperative Research Project No. 1146.

32.

Tyler, R. W.
"Modern Aspects of Evaluation."
California Journal of Secondary Education, 1954, 29,
4l0-4l2.

33.

Walters, Everett, e d . Graduate Education T o d a y .
Washington, D.C.:
American Council on Education,
1965.

New York:

119
3*1.

Webb, Eugene J.; Campbell, Donald T. ; Schwartz,
Richard D. ; and Sechrest, Dee.
Unobtrusive M e a s u r e s :
Nonreactlve Research In the Social S c i e n c e s .
C h i c a g o : Rand McNally and C o m p a n y , 1966.

3*3.

Wilson, Kenneth M.
Of Time and the D o c t o r a t e :
Report of an Inquiry Into the Duration of Doctoral
S t u d y . SREB Research Monograph No. 9.
Atlanta:
Southern Regional Education Board, 1965.

APPENDICES

APPENDIX I

The Research Information Index

121

THE RESEARCH INFORMATION INDEX

The Research

Information Index

(RI1) was developed

for the purpose of measuring the amount of’ research
information communicated in a research report.

Research

information is considered to be communication which helps
the reader to understand the research problem, procedures
and findings.
An underlying assumption of the RII is that informa­
tion in the research report should usually
a brief statement of the problem,

include:

(1)

(2) a description of

the methods, techniques and data used, and

(3) the major

findings of the study; and that this information is basic
to all research communication, regardless of the type of
research conducted.
The whole research report is the unit to be analyzed.
The report

is scored on the presence of information

commonly considered essential in research communication
as well as the objectivity or specificity of the informa­
tion.

Thus, the total score reflects the assumption that

objectIve% and specific research communication which
Includes essential

information is more informative than

exclusive and less precise communication.

Total score

on the Rll can range from zero to sixteen points.
122

The

123
higher the acore on the RII, the greater the amount of
research Information communicated by the research report.
While almost all reports include items such as the
purpose or objectives, these items are retained so that
the Rll is more inclusive of essential information.
Therefore, if desired, the RII may also be used as a
check list of the kind of information to be included in
research reports.

If used in this manner, it should be

noted that conclusions

(recommendations, discussion,

implications), omitted from the RII because of their
subjective nature, are usually included in the research
report.
The examples given are from existing research
reports and are not intended to illustrate "good" or
"bad" useage but rather:

(1) how to apply the RII to

the kinds of research communication likely to be found
and (2) that one form of research reporting communicates
more relevant research information than does another
form.

THE RESEARCH PROBLEM
Score
I.

STATEMENT OF THE PROBLEM
Statement <>l i!»«• Prohl i‘in: Re iercince 10 IIn: existing problem or
concerns which prompted initiation o[ the research...................

1

Examples:
a.

Special class placement has not improved academic performance
of Educable Mentally Retarded students.

b.

Thereis general agreement that Negro history is presented in a
biased manner in children's history textbooks.

c.

Inadequate attention has been paid the conceptual and empirical
separation of the normative expectations directed toward a
given social position.

d.

Elementary school buildings are inadequate for present day
instructional modes and media.

e.

The conventional analysis of covariance is not appropriate
when the covariate contains errors of measurement.

Absence of Statement of Problem........

2.

STATEMENT OF THE PURPOSE

OR OBJECTIVES

Statement of the P urpose: Reference to the primary objectives, or
hypotheses, or questions or purpose of the research...................
Examples:
a.

The purpose of this study was to investigate the relationship
between motivation and academic achievement for males in...

b.

The primary objective of this research was to explore changes in
number and kind of reference to Negroes in history books from...

c.

This study was designed to test hypotheses relating perceptual
classifications of principals and teachers to frequency witb
which they interact.

d.

The purpose of this study was to identify characteostics of
great teachers...

Absence of Statement of the Purpose.

12*1

1

125
Score
3.

SPECIFICITY OF THE

PURPOSE OK OBJECTIVES

Specific: Clearly specified objectives, hypotheses, or questionis
or purpose of the research...................... ..................
Example:
...to investigate change in academic self-concept during the
second year of a special class...
Non-Spec ific: Vague reference to the purpose of the research....
Example:
...to make a survey of prison inmates.

THE RESEARCH PROCEDURES
4.

PRIMARY IDENTIFICATION OF THE SAMPLE OR POPULATION UNITS
Primary Identification: Major identification of the sample or population
of Individuals, Communications, or Objects, or Events.................. 1
Examples:
a-

Individuals
Educable Mentally Retarded Siudents
Principals of Secondary Schools
Prison Inmates
An Influential Educator

b.

Communicat ions
Philosophies of Self-Actualization
Children's Picture Books
Diaries of Sino-Soviet Decision-Makers

c.

Objects
Elementary Schools
Religious-Affiliated Colleges

d.

Events
Dice throws
The sampling distribution of the test statistic...

Absence of Primary Identification.....................................

.0

. SECONDARY IDENTIFICATION OF THE SAMPLE OR POPULATION UNITS
SECONDARY Identif Leat ion; Any information in addition to primary
identification, which further identifies the kind of research
units selected for the sample or population..........................
Examples:
a.

Ind ividuals
Educable Mentally Retarded students ranging in age from
eight to sixteen...

b.

Comrounicat ions
Childrens Picture Books which use color illustrations...

c.

Objects
Elementary schools which offer first through sixth grades...

d.

Events
The sampling distribution of the test statistic arising from
analysis of covariance procedures...

Absence of

Secondary Identification..................................

SETTING. EXTENT OR RANGE OF THE STUDY
Setting: Site, or range, cr area,.or time period covered by the study
Where or when or under what circumstances............................
Examples:
a.

Individuals
...students selected from six Ohio public secondary scuools...

b.

Communications
Children's Picture Books published in Russia from 1915-1965...

c.

Objects
...elementary schools in four Southern states...
*

d.

—

—

EvenLs
...the sampling distribution of :;.e analysis of covariance when
the covariate is estimated true scores rather than observed
fallible scores.

Absence of Setting

127
Score
7.

NUMBER OF SAMPLE OR POPULATION UNITS
Sample Size; Reference to the number of sample or population
units for which data were collected...................................

1

Examples:
a-

Individuals
...65 Educable Mentally Retarded students...

b . Communicat•ons
Fifteen Children's Picture B > o k s ...
c . Objects
...six elementary schools which...
d . Events
"

'

i*

...1,000 sampling distributions

Absence of Sample S i z e .................................................

8.

0

INSTRUMENTATION
Instrumentation: Reference to the instruments or. measures used to
define the
variables.................................................

1

Examples:
a.

Students were asked to respond

b.

The amount of prejudice communicated was measured by counting
the number of references to...

c.

Schools were judged to be open or closed by...

d.

The rapidity with which convergence occurred was shown...

Absence of Instrumentation

%

to the Academic Aspiration Scale...

0

128
Score
9.

DATA COLLECTION METHODS
Data Collection:

Reference

to how datawore collected................

1

Examples:
a.

Student behavior was observed by...

b.

Prisoners were interviewed...

c.

Questionnaires were mailed to...

d.

The XYZ Intelligence Test was given...

e.

...by randomly generated theoretical distributions...

f.

The A Projective Test was given to theclient...

g.

Data were recorded from the records of the institution...

h.

Each paragraph was examined by the raters...

Absence of Data Collection.............................................

10.

0

DATA ANALYSIS*
Data Anlysis: Reference to how

data were analyzed....................

1

Examples:
a.

...was exhibited by a significant linear trend...

b.

The factor analyses resulted in...

c.

Percentages were computed for...

d.

It was determined by chi square techniques that...

e.

Significant associations were found...

f.

Differences in the various philosophies were indicated by
differences in the number of references to...

Absence of Data Analysis...............................................

*

0

Reference to data analysis sometimes occurs in the findings of the research report.

129
THE RESEARCH FINDINGS*
Score
11.

STATEMENT OF THE FINDINGS
Statement of the Findings; Reference to positive or negative
results or findings of the research..........................

1

Examples:
a.

There were no significant differences between the two groups
on number of extracurricular activities...

b.

Thirty seven (63%) believed that the supervisor...

Absence of Statement of the Findings........................

12.

0

OBJECTIVITY OF THE FINDINGS
Objective: Statement of positive or negative findings indicating
support by data analysis.............

t

Examples:
a.

Academic aspirations did not change significantly...

b.

92% stated that...

Subjective: Statement in the findings with no referent or which
is unsupported by data analysis...........................................
Examples:
a.

b.

*

It was not understood that this substitution takes place
at times in middle-class speech.
There is a sensitivity, generally, to the position in which
other members of their group find themselves.

It should be noted that items 11-16 apply only to the research findings.
Findings refer to results of the research and are separate from conclusions,
recommendations, explanations or discussion which may be included in the
research report.

0

i

130

Score

13.

VERBS AND VERB CONNECTORS
Specific:Abaencc of weak verb phrases, or presence of unqualified
verbs, or presence of weak verb phrases accompanied by an
explanat ion....................... ........................................
Examples:
a.

Neither academic performance nor academic expectations changed...

b.

Frequency of disciplinary action was negatively related to...

c.

Although no significant differences were found, there appears
to be a positive relationship between...

Non-specific:
Verbs which are carried by connectors which imply only possible
or hypothetical relationships rather than relationships indicated by
the desalts (e.g., appears to be, may indicate, seems, suggests)....
Examples:

14.

a.

Frequency of occurrence teems related to...

b.

Differences appear to be...

ADVERBS AND ADJECTIVES

S p e d fic:

Absence of non-specific adverbs or adjectives, or presence
of non-specific adverbs and adjectives accompanied by an explanation
or modifier.................................................................
Examples:
a. ...were considerably higher at 42.25...
b. ...were slightly, but not significantly, higher...
Non-Specific: Adverbs or adjectives that tend to modify, restrict
or tone down (slightly, occasionally, somewhat) or tend to exaggerate
(definitely, considerably, extremely)....................................
Examples:
a.

...were considerably higher...

b.

Teachers were extremely...

15.

EVALUATION
Objective: Absence of unsupported evaluations or value
statements, or presence of evaluations supported by d a t a.............
Examples:
a.

All of the respondents felt that children should be

b.

More than half of the group considered it a

allowed to...

good practice...

Subjective:
Value statements or statements which contain
evaluative terms (good, bad, should, ought) not supported
by data...................................................................
Examples:

16.

a.

All groups should be allowed to...

b.

It must be understood that...

QUANTITY
Spec ific: Absence of non-specific quantity reference or presence
of specific quantity reference..........................................
Examples!
a.

57% of the prisoners...

b.

There was a significant increase...

Non-Specific: Vague, ill-defined quantity reference (most, a great
majority, a large amount, some).........................................
Examples:
a.

The great majority feel...

b.

The greatest number had...

TOTAL SCORE FOR AMOUNT OF RESEARCH INFORMATION COMMUNICATED CAN RANGE FROM

*

ZERO TO SIXTEEN POINTS.

APPENDIX II

Categories For Research Preparation
Courses

132

CATEGORIES FOR RESEARCH PREPARATION COURSES
Categories one to six are intended only for courses which,
in your judgement,

are offered by your department or

college for the purpose of preparing students to apply
theoretical and/or methodological aspects of research.
Category zero in for courses which do not meet this
criterion.

Courses which primarily cover reviews of

research rather than application should not be included as
research preparation courses.
PLEASE ANSWER QUESTION ONE FOR EACH 800 AND 900 LEVEL COURS
LISTED UNDER YOUR DEPARTMENT OR COLLEGE IN THE CATALOGUE.
1.

To which one of the following categories would you
assign this course?
0.

This course does not meet the criterion of pr e ­
paring students to apply theoretical and/or
methodological aspects of research.
This course primarily

2.

covers:

1.

theory,

theory construction or logic.

2.

research methods.

3.

measurement and/or evaluation.

4.

statistics and/or mathematics.

5.

computer applications.

6.

research design

Would you please list and place into categories one to
six all 400 level courses in your department or college
which-are offered for the purpose of preparing students
to apply theoretical and/or methodological aspects of
research.
Course Number

Course Title
133

Category Number

APPENDIX III
Departments and Major Fields of Emphasis
Within Departments, The College of
Education Michigan
State University

134

DEPARTMENTS AND MAJOR FIELDS OF EMPHASIS WITHIN DEPARTMENTS
THE COLLEGE OF EDUCATION, MICHIGAN STATE UNIVERSITY
1.

2.

3.

4.

b.

Elementary and Special Education
1.

Elementary Education

2.

Special Education

Secondary Education and Curriculum
1.
2.

Agriculture Education

3.
H.

Comparative and International Education

5.
6.

History and Philosophy of Education

7.
8.

Industrial Education
Secondary Education

9.

Vocational and Technical Education

Business and Distributive Education
Curriculum
Home Economics

AdministratIon and Higher Education
1.

Adult Education

2.

Higher Education

3.

Educational Administration

Counseling,
Psychology

Personnel Services and Educational

1.

Educational Psychology

2.

Counseling and Personnel Services

3.

Measurement, Evaluation and Research Design

Health,

Physical Education and Recreation

1.

Health and Physical Educaton

2.

Recreation

135

APPENDIX IV

Categories for* Undergraduate Majors

136

Categories

for Un dergraduate Majors
Ot her Majors
( c o n t .)

Agriculture Econ omics

Physical Sciences
( c o n t .)
M ic robiology

Anthropology

Nurs ing

Re ligion

Communication

Ph ysics-Astronomy

Speech

Economics

Ph ysiologyPh armocology

Textiles

Behavioral Sciences

Geography
Police Admini st r at i on
Political Science

Poultry-Soil
Science
Public Health

Psychology

Statistics

Social Work

S urgery-Medicine

Sociology
Urban Planning

Ve terinary Pathology
Wild Life
Zoology

Physical Sciences
Anatomy

Other Majors

Anim al Husbandry

Accoun t in g

Aud lology

Ad vertising

Biological Sciences

Art

Ch emistry

Business

Agriculture Chem.

Personnel

Dairy

Education

E ng i ne e ri n g

Foreign Studies

Chemical Eng.

History

Civil Eng.

Home Management

E lectrical Eng.

Hotel Management

M echanical Eng.

Industrial Arts

M et a ll u rg i ca l Eng.

Journalism

Entomology

L an g uage-Literature

Foods

Library Science

F orestry-Ho r ti c ul t ur e

M ar k eting

Geology

Music

Mathematics

Philosophy
137

R adio-T e le v is i on

APPENDIX V
Tables

of S co r es

for

Reliability

138

Interrater

139

TABLE 2 3 .— Total scores on the Research Information
Index is ed for interrater reliability prior to the study.
,, .
IJumber

Abstract

Total Score
Rater #1

Total Score
Rater 02

1

12

10

2

7

5

3

12

12

i\

in

15

3

in

m

6

9

9

1

8

8

3

6

8

9

8

8

10

13

m

11

in

15

1.2

13

15

13

12

12

L'l

19

15

19

in

15

16

12

15

17

15

15

18

8

10

19

11

11

12

13

20
i*

*

=

.92

140

"

TABLE 24.— Total scores on the Re search Information Index
used for interrater reliability for the study.
«u
m
u
Abstract Number

Total Score
Rater #1

Total Score
Rater §2

1

5

3

2

10

8

3

7

2

4

11

12

5

12

14

6

7

9

7

8

12

8

15

14

9

8

8

10

8

11

11

8

9

12

13

14

13

12

15

14

14

15

15

15

15

16

15

14

17

15

15

18

12

11

19

12

13

16

16

20

"

r - .87

A P P E N D I X VI
Correlation Table

141

TABLE 25.— Correlation matrix of major variables for graduates and ABD’s.
# Credits
Res. Prep.
(N=204)

Jr. Sr. Doctoral Degree Assist.
GPA
GPA
Ph.D.=l None=0
(11-142) (N=204) Ed.D.=2 Held=l
(N=204) (N=204)

Total
Program
Credits
(N=204)

Miller
Age at
Analogies
Admission
Test
(N=203)
(N=175)

Sex
Male=l
Female=2
(N=204)

# Credits
Res. Prep.
(N*204)

Jr. Sr. GPA
(N=l42)

.20
-.04

.35

Degree Ph.D. = 1
Ed.D. = 2
(N=204)

-.05

-.10

- .1 1

Assist.
None « 0
Held « 1
(N=204)

.16

-.06

.02

-.14

Total Program
Credits
(N*204)

.28

-.05

-.21

-.04

.01

Age at Admission
(N«203)

-.24

-.06

-.01

.01

-.23

-.09

Miller Analogies
Test
(N-175)

.15

.08

.13

-.10

.13

.09

.08

Sex
Male « 1
Female ■ 2
(N-204)

.04

.17

.13

-.02

-.16

.02

.25

142

Doctoral GPA
(N=204)

.10