I--.!--
a- Inn'u'r'i

TEACHER MORALE AND
EVALUATION OF TEACHERS

The“: §or the Degree o? E}. D.
MICHIGAN STATE UNWERSITY

C. Jarvis Wotring
1965

||\\\\\|\\|W\\|\|\\\\\\\ N

0307 87 58

|\;u\\\\1\u219\\g\\u\§\

IHESls-z

This is to certify that the
thesis entitled

TEACHER MORALE AND EVALUATION OF
TEACHERS

presented by

C. Jarvis WOtring

has been accepted towards fulfillment
of the requirements for

Ed-Do degree in Education

Datg‘lovemb

0-169

.5,
L I b’ 1»: A [<1 ‘2’
M is}; igan S La:

Umcmty

 
 

t .' -
14.633 i-u. m_ ..,

  

 

 

0“ s} N209 1 Ida

 

ABSTRACT

TEACHER MORALE AND EVALUATION OF TEACHERS

by C. Jarvis Wotring

Statement of the Problem

This research attempted to relate the effectiveness
of teaching to teacher morale. Its significance lay in the
continuing search for the things that contribute to the ef-

fectiveness of teachers.

The Sample

The hypotheses were tested in a population of 188
teachers in nine Michigan public secondary schools. Six
were schools having grades seven through twelve; one was a
ninth through twelfth grade school; one housed grades seven
through nine; and one had grades ten through twelve. The
first seven were in small rural communities under 1,669 in
population and the latter two were in suburban communities
of 75,000 or over. To the extent that 80 percent of Michigan
towns are smaller than 1700 in population, as given by the

1960 census,the schools are representative. To the extent

C. Jarvis Wotring

that no communities of 5,000 to 25,000 are represented and

the sample was small, it is not representative.
Instrumentation

The factors of morale were measured by the Teacher
Morale Form for which morale was defined operationally as:
teachers' satisfactions that their personal needs were being
met; teachers' satisfactions with administrative policies and
practices; and teachers' satisfactions with the tasks they
performed to reach the goals of the institution. It was a
40 item sentence completion test.

Teacher effectiveness was measured by the Evaluation
of Teachers tests in which teacher effectiveness was defined
as:

1. Knowledge the teacher has of the subject
2. The teacher's ability to explain clearly
3. The teacher's fairness

4. The teacher's discipline

5. The teacher's sympathy and understanding
6. The teacher's ability to be interesting
7. The teacher's efficiency

8. <The teacher's ability to get students to think for
themselves

9. .The teacher's ability to help students to know why
they learn certain things

10. The teacher's general teaching ability.

C. Jarvis Wotring

Both tests satisfied the standard criteria for validity and

reliability.

Procedures and Testing of
Hypotheses

 

This was an ex—post facto study. Therefore, cause
and effect relationships were not explored. The morale of
teachers was measured on teacher responses to 40 incomplete
sentence items of the TMF. Three teachers from each school
judged the teachers' statements and ranked them: 0--highly
positive, l--slightly positive, 2--neutral, 3--slightly nega—
tive, and 4--highly negative. A five percent sample check
showed agreement among judges' ratings to be significant at
the .01 level of COnfidence.

Evaluations of teachers were made by teachers them-
selves, one average class of their students, and their
principals. Estimates of teacher effectiveness on the ten
items were made on a five point rating scale. A mean score
was determined for each item of morale for each school and
for student evaluations of each teacher on each item.

The first null hypothesis that there would be no
difference in the teacher morale of the various schools was
done by comparing each school with every other school. The
median test was applied to the 40 items for each school in
the pair being compared. Significance of Chi—square was ac—

cepted at the .05 level of confidence.

C. Jarvis WOtring

The second null hypothesis that evaluations of
teachers do not differ with respect to high and low teacher
morale was tested by pairing each high morale school with
each low morale school and observing the number of high and
low evaluations in those schools. Using the median test,
Chi-square significance was accepted at the .05 level of
confidence.

To test hypothesis three: morale has more than one
factor, the mean morale scores for each item were converted
to standard scores and a factor analysis conducted.

Hypothesis four was also stated in the null form:
there will be no significant difference between high, medium,
and low divisions of each of the teacher morale factors and
high, medium, and low divisions of each of the items of
evaluation. Significance of Chi-square was accepted at the
.05 level of confidence.

Students were also asked to list the good qualities

of their teachers.

Findings

1. Teacher morale does differ significantly from school
to school. Three schools were significantly different from
six other schools with respect to teacher morale.

2. Only one high morale school showed principals' evalu-

ations of teachers to be related to teacher morale when

C. Jarvis Wotring

that school was compared with six low teacher morale
schools.

3. The factor analysis of the items of the TMF produced
three factors: teachers' satisfactions that their personal
needs were being met; teachers' satisfactions with adminis-
trative policies and practices; and teachers' satisfactions
with the tasks they performed to reach the goals of the
institution.

4. The most significant relationships were found between
teachers' satisfactions with their tasks and principals'

estimates of teacher effectiveness in the areas listed:

 

 

Item of Level of
Evaluation Confidence
1. Knowledge of subject .01
2. Ability to explain .01
3. Efficiency .005
4. Ability to get students to think .001
5. Ability to help students to know
whygthey are learning .02
6. General teaching ability .02

5. Students listed the five most important qualities of
a good teacher as: sympathy and understanding, sense of

humor, good personality, interestingness, and fairness.

C. Jarvis WOtring

Conclusions

Within the limitations of this study the following
conclusions are made:

1. Teacher morale does differ significantly among
schools. The difference lies in the way teachers feel their
personal needs are being met.

2. Dissatisfactions of teachers with faculty meetings,
the curriculum, and the Opportunity to help make policy indi-
cate the teachers' need for changes, and the blindness of ad-
ministrators to capitalize on this need to improve the con-
ditions of learning.

3. The function of administration, as it relates to com-
munication with teachers, to communication between teachers
and the community, to policy making, and to the satisfaction
of teachers' personal needs is the most important ingredient
of high teacher morale. This is described as the task-needs
integration function.

4. Students have a different concept of the "good"
teacher from that of either principals or teachers. While
students rank sympathy and understanding, a sense of humor,
fairness and interestingness highest on their list of quali—
ties of a "good" teacher, principals and teachers tend to

emphasize techniques and methods.

C. Jarvis Wotring

5. Principals by reinforcing the behavior of teachers
which emphasizes knowledge, ability to explain, efficiency,
ability to get students to think for themselves and to know
why they are learning certain things tend to perpetuate the
"system" and to continue making students' needs secondary to
techniques of teachers.

6. There is no satisfactory evaluating device to fit
all purposes. The California teachers reported in Teacher
Competence: Its Nature and Its Scope, that teachers wanted
to participate in evaluation. This study showed teachers to
be dissatisfied with evaluations, except for one high
teacher morale school. Since students have a different con—
cept of the "good" teacher than principals have, teachers
should seek evaluations by students and bring this infor-
mation to principals and supervisors for joint evaluation.
Administrators would have to set the climate for such action.

7. The factor analysis showed considerable inter—
relatedness of the teacher morale factors. It is not known
how other factors affect the task oriented morale items. To
study the effects of "classroom" teacher morale and its ef-
fects on students, a new form should be developed specifical-

ly for this purpose.

C. Jarvis Wotring
Specific Recommendations

1. The factor analysis, using the TMF, should be re-
peated with another set of schools at both the elementary
and secondary levels.

2. Search for good evaluative measures must go on.

Most promise may lie in guides for teachers and principals
to evaluate the teachers' work together. Certainly in light
of good human relations and tenure laws this procedure has
relevance.

3. There is more work to be done in the public schools
on student morale and student achievement and the part that
teacher morale plays in the learning environment.

4. Morale studies should be used by students, teachers,
and principals to explore their feelings about their jobs,
school, and community. Such studies would provide motivation
for studying the kinds of behavior that produce attitudes
which provide the most effective teaching and learning

conditions.
“Questions for Further Study

1. Are there more than three factors of teacher morale?
What are the factors? To what degree are the factors inde-
pendent? Is there a kind of teacher classroom morale that

is directly related to student behavior and achievement?

C. Jarvis Wotring

2. Do students of teachers who emphasize the importance
of sympathy and understanding in their teaching learn more
than students of teachers who emphasize techniques?

3. In how many schools do teachers participate in evalu-
ative procedures? What is the effect on teacher morale?

4. To what extent do teacher expectations determine the
behavior of principals?

5. HOW' does the community's concept of the "good"
teacher affect teachers' morale and their classroom behavior?
6. What is the effect of principals' evaluations upon

teacher morale and teacher effectiveness?

7. What effect does the social structure of a group of

teachers have on morale and classroom behavior of teachers?

TEACHER MORALE AND EVALUATION OF TEACHERS

BY
\

C?3Jarvis Wotring

A THESIS

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

DOCTOR OF EDUCATION

College of Education

1965

ACKNOWLEDGMENTS

The writer wishes to express gratitude to the many
peOple who have helped to make this study possible.

Special appreciation is expressed to Dr. John H.
Suehr who, as committee chairman, gave invaluable assistance
and guidance throughout the preparationtof this study.

The writer would also like to express his appreci-
ation to each of the other members of his committee for
their helpful assistance. The members of this committee
are: Dr. George Myers, Dr. Karl Hereford, and Dr. James
McKee. .Also,.thanks are due to Mr. Bruce Rogers for his in-
valuable assistance in analyzing the data.

Special thanks are due to the writer's wife, Violet,
and his daughters, Cessily and Rowena, for their patience
and understanding during the years this student devoted to

the doctoral program.

October, 1965 C. Jarvis WOtring

ii

TABLE OF CONTENTS

Chapter Page

I. INTRODUCTION . 1

Teacher Morale and Teacher Effectiveness 1

Significance 1

Purpose . 3

Need for the Study 3

Statement of the Problem 5

~Re1ated Assumptions 9

Hypotheses . . 11

Definition of Terms . 11

Limitations of the Study 16

Organization 19

Summary 20

II. RELATED IDEAS AND LITERATURE . . . . . . . . . 21

Teaching and Learning . . . . . . . . . . 21

Administration . . . . . . . . . . 23

Related Studies of Morale . . . . . . . . 25

Studies of Morale in Education . . . . . . 29

Summary of Morale Research . . . . . . . . 43

Evaluating Teachers . . . . . . . . 46

Summary of Teacher Evaluations . . . . . . 56

III., PROCEDURES USED . . . . . . . . . . . . . . . 58

The Sample . . . . . . . . . . . . . . . . 58

Instrumentation: Teacher Morale Form . . 59

Scoring the TMF» . . . . . . . . . . . . . 63
Instrumentation: Teacher Evaluation

Form . . . . . . . . . . 66

Scoring the Evaluation Forms . . . . . . . 70

Preparing for Computer Analysis~ . . . . . 71

Hypotheses . . . . . . . . . . . . . . . . 71

Summary . . . . . . . . . . . . . . . . . 78

iii

Chapter

IV. ANAL

YSIS OF RESULTS

Hypothesis One

Hypothesis Two

Hypothesis Three

Teacher Morale Factors
Hypothesis Four .
Student Comments on Evaluations
Summary of the Findings

V. SUMMARY, CONCLUSIONS, AND IMPLICATIONS

BIBLIOGRAPHY

APPENDICES

Summary

Conclusions

Implications .

Specific Recommendations
Questions for Further Study

iv

Page

80

81
85
88
93
96
104
105

108
108
108
111
115
115
117

123

Table

LIST OF TABLES

Percentage of persons in five occupational
groups expressing important personal
needs . . . . . . . .

This rank order correlation shows how well
teachers rated the teacher morale in their
schools compared with the total morale in
that school . . . . . . .

Teacher judges ranks of teacher responses to
TMF

Manner of determining item mean scores of
student evaluations

Matrix table showing schools compared on
teacher morale . . . . .

Median test to distinguish high from low
morale schools . . . . . . . . . . .

Comparison of evaluations in high and low
morale schools . . . . . . .

Numbers indicate each third of morale factors
and evaluations

Example of Chi-square test contingency table
for comparing Factor 1 with teacher self—
evaluation on knowledge . . . .

Schools different from each other and their
level of significance

Factors producing low morale in school 8:
the extreme low morale school

A comparison of the mean scores of evaluations
of teachers by themselVes, their Students~
and their principals,and mean morale
scores . . .

Page

28

62

65

70

72

73

73

77

77

81

84

86

Table

4.9

4.10

4.11

4.12

Principals' evaluations and their relation to
high and low teacher morale schools .

Means and standard deviations of forty items
of teacher morale arranged according to
factors . . . . . . . . . . . . . . . . .

Percentage of men and women in high and low
morale schools . . . .

Rotated factor loadings of morale items

Chi-square test of independence between
teacher morale Factor 1, teachers“ per—
ceived satisfactions of individual needs,
and items of evaluations by teachers them?
selves, their students and their prinicpals.
High, medium and low morale is compared with
high, medium and low evaluations. There are
four degrees of freedom . . . . .

Chi-square test of independence between
teacher morale Factor 2, teachers' per—
ceived satisfactions with administrative
policies and practices, and items of evalu-

. ation by teachers themselves, their stu-
dents, and their principals. High, medium
and low morale is compared with high,
medium and low evaluations. There are four
degrees of freedom . . . . . . . . .

Chi-Square test of independence between
teacher morale Factor 3, teachers' per-
ceived satisfactions with tasks, and evalu—
ations of teachers by themselves, their
students and their principals. High,
medium and low morale is compared with high,
medium and low evaluations. There are four
degrees of freedom . . . . . . . . . . .

Significance of Factor 3 and Principals'
evaluations .”. . . . . . . . . . . .

Comparison of teachers', students', and
principals'evaluations by mean scores on
all items . . . . . . . . .

vi

Page

87

9O

92

97

98

99

100

102

103

Table Page

4.13 Correlations between ranked mean scores of
evaluations of teachers by themselves,
students and principals . . . . . . . . . . 104
4.14 Qualities students liked about teachers . . . 105

vii

APPENDICES

Appendix
A Reliability of Judgements of TMF Responses
Five Percent Sample Check by Two
Judges . . . . . . . .

B Coefficient of Concordance Among Ranks of
Judgements of Teacher Morale Items (Five
Percent Sample)

C Instructions to Principals for Teacher
Evaluating Procedures

D Reliability of Student Evaluations

E Coefficient of Concordance' Among Rankings
of Three Factors of Teacher Morale

F Computation of the Chi;square Statistics
of the Median Test Applied to Teacher
Morale Items . . . . . . . .

G Computations of Important Median Tests of
Student Evaluations of Teachers in High and
Low Morale Schools

 

H Computations of Important Median Tests of
Principals' Evaluations of Teachers in
High and Low Morale Schools

 

I Mean Scores Ranked and Correlated for Items
of Teacher Self-Evaluations, Student Evalu-
ations of Teachers, and Principals' Evalu-
ations of Teachers . . . . . . . .

viii

Page

124

125

126

131

132

133

137

141

140

CHAPTER I
INTRODUCTION

Teacher Morale and Teacher
Effectiveness

The nature of teacher morale has not been dissected
and analyzed to the degree that morale of industrial workers
has been scrutinized and evaluated since 1914. Morale in
industry has dealt with workers' satisfactions with their
jobs and their relation to production. But, only during the
last decade have a few studies of morale in education at—
tempted to relate teacher morale to the effectiveness of
teaching.

This study on that subject is an "after-only" study
without controls as described by Barnes.1 The data are de—
scriptive in nature and only limited attempts are made to in—
fer causation among the variables of teacher morale and

teacher effectiveness.
Significance

The greatest significance to be derived from a study

of teacher morale and teacher effectiveness is the

 

1F. P. Barnes, Research for the Practitioner in Edu—

cation (Washington, D.C.: Department of Elementary School
Principals, N.E.A., 1964), p. 65.

1

identification of the conditions that make for optimal learn-
ing. There is considerable research that the personality of
the teacher and the conditions affecting his teaching be-
havior are determinants of a good learning environment for
students.

Studies of morale in education are not widespread.
Only one study was reported in 1964. Pryor studied 323
teachers in 19 administrative units of three districts to
investigate the relationship between teachers' perceptions
of administrative policies, procedures and practices, and
the morale status of teachers. He found the relationship to
be significant. He used Pearson's product moment coefficient
of correlation. Pryor reported that the strength and level
of the relationship varies among administrative units in the
district, that morale showeda sharper deviation in some in—
stances, that administrative policies affectedsteacher morale
according to the way the policies wereyperceived and that
there was no evident difference in morale that could be as-
sociated with the size of the district or salary. Dr. Pryor
concentrated on teacher perceptions of administrative

policy.2

 

2Guy Clark Pryor, "The Relationship Between Teachers'
Perceptions of Administrative Dimensions and the Morale
Status of Teachers in Certain Texas Schools" (unpublished
Doctoral dissertation, North Texas State University, 1964).

Purpose

The purpose of this study is t0 examine forty areas
of teachers' feelings about the tasks they perform, satis—
faction of their own personal needs, and the possible re-
lationship of morale to their effectiveness as perceived by

themselves, students and principals.

Need for the Study

The importance of research about teacher morale is
readily apparent.

There is a senseflof urgency to identify morale and
a concern agOut'discovering how to affect it
positively.

The importance of morale is such that effective
personnel administration will be constantly on the
alert to discover threats to high morale and to pro—
vide preventive or corrective measures necessary to
promote conditions of high morale.4

Research in industry has been long and conclusive in
this area. Such research shows that high morale is related
to high production.

Morale is directly and closely related to effiCiency.
EducatiOn might learn in this regard from industry,
which has been Spending large Sums on devices, tech—

niques, and benefits to improve morale and thereby in—
crease production.

 

3James A. Van Zwoll, School Personnel Administration

(New York: Appleton Century-Crofts, 1964) , p. 169.

 

41bid., p. 187.

5Percy E. Burrup, The Teachers and the Public School
System-(New York: Harper and Brothers, 1960), p. 298.

For, when there is a feeling of well being between
employees and management working together toward
common goals, there is a unique condition in which
the ambitions of all concerned are realized.6

Search for the effective teacher has been equally
long and more fervent than studies of morale in education.
Ellsworth Thompkins, Executive Secretary of the National

Association of Secondary School Principals, said:
The best evidence is that there is no one set of
practices that add up to good teaching——no one set
of competencies that every teacher must possess.
That conclusion has put some researchers on a new
track-—studying the teacher himself more than his
practices.
Teacher effectiveness may be essentially a relationship be—
tween teachers, pupils, and other persons concerned with the
educational undertaking.

Since 1945 the California Council on Teacher Edu-
cation, at first working alone, then with the National Com—
mission on Teacher Education and Professional Standards in
1953, and finally having its work reviewed by the American
:Association of Colleges of Teacher Education, had been de-
veloping a definition of the role of the teacher.‘ In 1957,
the report, Teacher Competence: Its Nature and Its Scope was
published by the National Commission on Teacher Education.

Between pages 32 and 42 various aspects of teacher competence

were delineated. Of the 116 definitions, 72 or 62 percent

 

6Carl E. Gregory, "What is Morale," Personnel Ad-
ministration, Vol. XXX (March, 1959), p. 33.

7Ellsworth Thompkins, “Evaluating Teachers," Spot—
light, No. 62 (March-April, 1964).

 

dealt with interpersonal relations with members of the com—

8 There is

munity, administration, colleagues and students.
little doubt that maturity and sensitivity to the needs of
others are the prime criteria in defining the role of the
teacher. Most people would agree with John Gustard's9 state-

ment that effectiveness is the most important trait or combi—

nation of traits of a teacher.
Statement of the Problem

The theoretical basis for the problem has been in-
ferred but should be clarified. It is not definitely known
what aspects of teacher morale affect the learning environ—
ment of pupils. Anderson found a significant relationship
between teacher morale and student achievement, but Redefer
refers to Anderson's study as showing need for further
research.10

'A study by Bills and McGhee showed that students

who have more positive attitudes toward psychology perform

8 ,
Teacher Competence: Its Nature and Its Scope,
(Burlingame, California:3 N.E.A} National Commission on

Teacher Education, 1957), pp..32-42.

9John Gustad, "On Improving College Teaching," N.E.A.

M (March, 1964), p. 38.

10Frederick L. Redefer, "Teacher Morale and Quality of
Education," Nation's Schools, LIX (February, 1957), p. 53.

 

rmare effectively in a learning experiment than those who have
negative attitudes.ll
The importance of feelings and attitudes of teachers
for their jobs can readily be inferred from French's dis-
cussion of behavioral goals. He said that the knowledge of
the behavioral sciences gives to modern culture the astonish-
ing power of the choice to set the conditions which mold "an
open, adaptive, free-thinking, self-respecting individual” or
its Opposite. It is the former person who will be able to
use the enormous powers of the behavioral sciences with in-
telligence and sensitivity to human values.12 Teachers must
be open, adaptive, free-thinking, self—respecting individuals
if they are to set the stage for students to learn this kind
of behavior. These feelings and attitudes are important to
high teacher morale and positive conditions of learning.
Wiles said:

Unless the emotional content of the situation is

right, the desired subject—matter learnings will

not occur. In any situation boys and girls are

striving to feel right about themselves, to feel

that they haVe worth, to feel that they are ac—

cepted. .Other learnings are secondary and will not

receive the full attention of students until these
first learnings receive satisfaction.l3

 

llR._E. Bills and C. R. McGhee, "The Effect of Atti-
tude Toward Psychology in a Learning Experiment," Journal of
Personality, Vol?’XXIII'[l955),‘pp. 499—500.

12Will French and Associates, Behavioral Goals of
General Education in High School (New York: Russel Sage
Foundation, 1957), p. 45.
l3Kimball Wiles, Teaching For Better Schools (Engle~
wood Cliffs: Prentice-Hall, Inc., 1959), p. 47.

 

Whert Wiles has said about students and learning might very
well have been said about teachers and teaching.

The problem under investigation was to find out if
teachers in schools where teacher morale was high would be
evaluated highly by students, by teachers themselves, and by
principals in those sdhools and vice versa. The hypotheses
were:

Null Hypothesis 1: Teacher morale will not differ
significantly from school to school.

Null Hypothesispg: Estimates of teacher effective—
ness in a school will not be significantly related to
teacher morale in that school.

Hypothesis 3: Teacher morale has more than one
factor.

Null Hypothesis 4: There will be no significant
difference between teacher morale factors and items of evalu-
ation of teachers.

Another aspect of the study was to discover which
factors of morale were most highly related with each of the
ten separate items of the evaluations of teachers. The
original question asked was: What feelings of teachers show
the strongest relationship to high and low morale and items
of teacher evaluations? Therefore it seemed advisable to

study the variables of morale. McNemarl4 had suggested that

 

l4Quinn McNemar, "Opinion Attitude MethOdOIOgy,n
Psychological Bulletdxn Vol. xLIII (July}”l946), pp. 365—367.

martain items of morale tend to form clusters. Bently and
Ihmnpl, seeking to validate a teacher morale measuring instru—
ment, divided 157 items into eight categories: (1) self
status, (2) relationship with students, (3) relationship
with other teachers, (4) factors relating to administration
and policies, (5) relationships with community, (6) curriculum
factors, (7) working conditions, and (8) economic factors.15
A factor analysis divided the items of the Teacher
Morale Form into three categories: (1) perceived job satis—
faction of individual needs of teachers; (2) satisfactions
with administrative policies and practices; and (3) satis—
factions teachers have with the tasks they perform. The
null hypothesis that there would be no relationship between
the factors of morale and selected teacher evaluation items
was tested. Chi-square was used to test the significance.
The high, medium, and low morale measures of factor one were
tested with high, medium, and low scores of each of the

items of the evaluations. The same test was applied in the

same way for factors two and three with the items of the

evaluations.

 

15Ralph R. Bently and Arno M. Rempl, "Peer-Selection

vs. Expert Judgement as a Means of Validating a Teacher
Morale Measuring Instrument," Journal of Experimental Edu—
cation, Vol. XXXI, No. 3 (March, 1963), p. 235-240.

Re 1 ated Assumptions

Several assumptions based on other research were
necessary for this study. One, free responses permit the
satisfactory assessment of morale. Suehrl6 compared the
word association method to the sentence completion tests and
found advantages in favor of the latter. In the SCT some
disguise is present and freedom of response is greater than
rating of items. For this reason this ambiguous type of
test is thought to be potentially more valid. (This point
will be developed more fully in Chapter IIIZ) The teacher
Morale Form used in this study is a sentence completion test.

The TMF has been found satisfactory for distinguish—
ing high teacher morale schools from schools of low teacher
morale. ~It is assumed that it will distinguish levels of
teacher morale for the schools in this study.

Two, morale can be effectively studied by the staff
itself.l7 Teacher judges have a high interscore reliability
when rating the responses on a 0 to 4 scale with 0 being
highly positive, 1 slightly positive, 2 neutral, 3 slightly
negative and 4 highly negative. The TMF was rated by
teachers in each school, and a five percent sample was

checked by two members of the Department of Administration

and Higher Education at Michigan State University.

 

16John N. Suehr, "A Study of Morale in Education,"
(unpublished Doctoral dissertation, University of Colorado,
1962), pp. 80-85.

17Ibid., pp. 80-85.

10

Three, students can evaluate satisfactorily teachers'
classroom behavior. McCall and Krause18 concluded from their
study of teacher merit that pupils are "far better" judges
of a teacher's merit than professionally trained adults.
Remmers, in one of his many comments on this subject, said
that pupil judgements of teacher effectiveness measure an im—
portant effect of teachers on students just as achievement

tests do.19

Four, sex, marital status, religion and race, in and
of themselves, seem to have little to do with morale of
teachers in junior and senior high schools. Redefer sug—
gests an exception to this. He says, " . . . when they form
the basis for relations within a faculty (they) may be
destructive of morale."20

Five, morale, as measured by the Teacher Morale Form,
will not vary during the school year. Suehr said:

that the results of the Incomplete Sentence

Form will not vary to any great extent during the
school year; some of the other instruments might.21

 

18William H. McCall and Gertrude Krause, "Measurement

of Teacher Merit for Salary Purposes," Journal of Educational
Research, Vol. LIII, No. 22 (October, 1959), p. 74.

19H. H. Remmers and N. L. Gage, Educational Measure—
ment and Evaluation (New York: Harper and Brothers, 1955),
p. 493.

20Redefer, 0p. cit., p. 55.

 

1Suehr, 0p. cit., (thesis), p. 193.

11

Hypotheses

There were four hypotheses to be tested. Three were
stated in the null form and tested by Chi-square measures.

1. Teacher morale will not differrsignificantly from
school to school.

2. There will be no significant difference between
evaluations in high teacher morale schools and low teacher
morale schools.

3. Teacher morale is composed of more than one factor.

4. There will be no significant difference between
teacher morale factors and items of teacher evaluation.

Interest in this problem was generated by discussions
of administrative purpose. Certainly the goal of adminis-
trators and teachers is quality education. But what con—
tributes to it? This researcher sought an answer in the
possible relationships between teacher effectiveness, as
judged by teachers themselves, their students, and their

principals.

Definition of Terms

Definitions of morale vary considerably. It has been
described as an emotion, a feeling, an attitude, and a group
of attitudes. It has been considered as a single entity and
a group of factors or clusters. To some, morale has inde-

pendent factors; to others the factors are quite dependent.

12

All definitions, however, are expressed in terms of job
satisfactions and interpersonnel relations.

Gregory thinks morale is an emotional reaction and
adjustment of peOple to their environment. People are
strongly influenced, pleasantly and unpleasantly, humanly
and physically, in their own unique way when they respond
emotionally with co—workers, supervisors, and management
policies. Constant pampering with new and varied programs is
necessary to maintain and raise morale.

Whitehill23 explains that morale must be explained
in a complex of reactions. The dualism of individual atti-
tudes or state of mind as opposed to a group phenomenon is
one of these complexities. An individual influences group
attitudes and in turn his own feelings are shpaed by the
"organizational state of mind." If the sources of satis—
faction outweigh those of dissatisfaction and unrest, morale
is high and vice versa.

D. E. Anthony sees morale as a total set of attitudes

toward one's job, management, and the company.24

 

22Carl E. Gregory, "What Is Morale?" Personnel, Vol.
XXXVI (March, 1959), p. 31.

23Arthur Whitehill, Jr., Personnel Relations (New
York: McGraw—Hill Book Company, Inc., 1955), p. 329.

24D. E. Anthony, "New Trends in Morale Development,"
Personnel Journal, vol. XLIII (May, 1964), pp. 265—266.

 

13

Quinn McNemar25 believes that the entities Of morale
are dependent and form clusters, which are conceivably inde—
pendent.

Bently and Rempel26 see morale as the "professional
interest and enthusiasm that a teacher displays toward the
achievement of individual and group goals in a given school
situation." Theirs was an operational definition in peer
selection versus expert judgement as a means of validating a
teacher morale measuring instrument.

Lonsdale suggests a way out of this definitional
problem.

By this View, morale is a feeling of participants
in an organization stemming from a combination of (a)
perceived productivity or progress toward the achieve-
ment of tasks of the organization, and (b) perceived
jOb satisfaction of individual needs through inter—
action Of the participant in his role within the work
group and the total organization. Further, high
morale is the participant's perception of a successful
task—needs integration. Since task-needs integration
is the ultimate purpose of administration, it fol-
lows that high morale is the participant's perception
of the consummatiOn Of administrative purpose. As
walker put it, the key questions about any new
practice or policy are these: "How will it affect
individual or collective efficiency?" and “How will
it affect the employees' enjoyment Of their jobs?"27

 

25Quinn McNemar, "Opinion Attitude MethodolOgy,"
Psychological Bulletin, Vol. XLIII (July, 1946), pp. 365-367.

26Ralph R. Bently and Arno M. Rempel, "Peer Selection
vs. Expert Judgement is a Means of Validating a Teacher
Morale Measuring Instrument," Journal of Experimental Edu—
cation, VOl. XXXI, NO. 5 (March, 1963), pp. 235—240.

27Richard C. Lonsdale, “Maintaining the Organization
in Dynamic Equilibrium," Chapter VII in Behavioral Science
and Educational Administration, the Sixty—third Yearbook of
the National Society for the Study of Education (Chicago:
University of Chicago Press, 1964), pp. 164-165.

 

 

l4

LonSdale's definition was chosen as one operational
definition for this study. The virtue of this conceptuali-
zation is that it not only concerns itself with teachers'
satisfactions with the tasks they perform and with their
feelings about their personal needs, but also with the ad-
ministration's responsibility for welding the two.

.Definitions of teacher effectiveness have been as
elusive as those of teacher morale. The California Teachers
Association and the N.E.A. Commission on Teacher Education
spent twelve years in defining teacher competence. Sixty-
two percent of the items mentioned dealt with interpersonal
relations.28 More specifically, Barr found teachers picked
200 items from a list of 6,939 in designating qualities that
are prerequisite to success in teaching. The first ten in
order of importance are:

.Instructional skill

Personal fitness for teaching
Scholarship and professional preparation
Ability to COOperate

Appearance

Classroom management

- Interest on work, pupils and subject
Considerateness

Leadership
Health29

OkocoxlmmprI—n

H

Ryans discovered the three most important areas to

be:

 

28Op.icit‘.
29A. S. Barr and Lester M. Emans, "What Qualities Are

Prerequisite to Success in Teaching," Nations Schools, Vol.
VI, No. 3 (September, 1930), pp. 60—64.

15

1. Understanding, friendly teacher behavior vs.
aloof egocentric behavior.

2. Systematic, responsible, businesslike behavior
vs. evading, unplanned slipshod behavior.

3. Stimulating, imaginative behavior vs. routine

dull behavior.
Nelson and others found disciplinary control, teacher—pupil
relations, and instructional excellence.31 Interpersonal re-
lationships and instructional excellence stand out in all
considerations of the effective teacher.

Teacher effectiveness for this study was defined
as the teacher's behavior exhibited in the following
areas:

Knowledge of subject
Ability to explain
Fairness

Discipline

Sympathetic understanding
Interestingness

Efficiency

Ability tO provoke thought
Ability to help students know why they
learn certain things

10. General teaching ability32

\OCDQO‘Ulwal-J

It was felt that the teachers who were evaluated high on
these qualities could be identified as good teachers.

To further the Operational definitions, the terms
high and low morale and high and low evaluations needed

attention.

 

30 3, Who's A Good Teacher (Washington, D.C.:
American Association of School Administrators, Department of
Classroom Teachers, National School Boards Association, 1961),
pp. 33—34.

31Ibid., pp. 30—34.

 

32 , Western Michigan UniVersity Continuous
Study of Student-Reaction Reports (Kalamazoo, Michigan:
Western Michigan University, 1964).

 

16

High Teacher Morale Schools and Low Teacher Morale

 

Schools were those determined to be significantly different
by the median test; Chi-square test of independence was ac—
cepted at the .05 level of confidence. High and low morale
was measured by the number of mean scores of the forty
Teacher Morale Form items which were above the median or at
or below the median.

When the morale factors were compared with the se-
lected items from the evaluation forms, high, medium and low
scores were arbitrarily determined by dividing them into

thirds.

Morale Factors were defined as the combination of
Teacher Morale Form items that were categorized by factor
analysis into three factors of morale: expressions of indi—
vidual needs satisfaction from teaching, satisfaction with
administrative policies and practices, and teacher satis—
factions with the tasks they perform in relation to the
goals of the institution.

All definitions were stated to be used Operationally

in this research.
Limitations of the Study

The limitations Of this study must be recognized.
Since it is an "ex-post facto" study,it is impossible to

cite cause and effect relationship. However, Barnes,33 said

 

330p. cit., pp. 65-67.

17

that studies of this type have a larger scope and can "range
more freely than experimental studies with control groups."
One of the main purposes is to suggest problem areas for con-
trolled studies.

The Teacher Morale Forms were administered at differ-
ent times between November and February. The author of the
TMF believes that measures of morale will vary little during
the year when this incomplete sentence form is used.34

The question, "Do all of the items contribute the
same amount to mean morale scores?" was raised. Kendall's
coefficient of concordance for the three factors of teacher
morale showed ME signifiCant by the F test at the .01
level. These will be discussed later in Chapter v. Thirty-
three of the forty items of the TMF discriminated at the .01
level, and six others at the .05 level indicating internal
consistency. Some interaction among the variables showed up
in the factor analysis and will be reported in Chapter IV.
One fact cannot really be reconciled. In a school of fifteen
teachers each teacher's responses contribute one-fifteenth
of the score and in a school of thirty-eight teachers re-
sponding each teacher contributes only one-thirty—eighth of

the score. However, size of school did not appear to be

 

34Suehr, Op. cit., p. 193.

18

related to morale. Dennerlein,35 Cohen,36 and Pryor37 be—

lieved from their work that size of the school, the district
or the socio-economic level of the community played only a
small part in the morale Of the school.

Ix) the TMF and the teacher evaluation instrument
measure different things or is there only a semantic differ-
ence? Because the interpersonal relations aspect of this
study is so strong, the question is a very real one. The re—
searcher's response to the question was that the instruments
did in fact measure teachers' feelings about their tasks and
satisfaction of their own needs. Principals, teachers and
students, on the other hand, were asked for their judgements
of the effectiveness of the teacher's performance of his
role.

It must be recognized that generalizations for all
Michigan schools cannot be made from this sample. The schools
tend to be small: three schools, twelve to fifteen teachers;
three schools, eighteen to twenty teachers; two are twenty-
eight and twenty-nine teachers; and one, thirty—eight

teachers. Because homogeniety could not be assumed,

 

35Gerald Dennerlein, "Factors Related to the Measure-
ment of Teacher Morale" (unpublished Doctoral dissertation,
University of Southern California, 1958).

6Gloria J. Cohen, "A Study of the Socio-Economic
Status of the School Community and the Morale of Teaching
Personnel in New York City (unpublished Doctoral dissertation,
New York University, 1959).

37 .
Pryor, Op. Cit.

19

non-parametric measures were used to study the over-all
sample of 188 teachers.
Limitations of this study do not keep the research

from being important to further studies.

Organization

To discuss the organization of this research study,
the reader must be reminded of the two aspects: teacher
morale and teacher effectiveness. In the final analysis
these will be discussed in the light of administrative
behavior.

Reviewing briefly, the reader will recall that the
morale of teachers was studied by the Teacher Morale Form be-
tween NOvember and February of the school year. Evaluations
of teachers by the teachers themselves, by students and by
principals was accomplished during the month of May. The
advantages of evaluations at this time of year are obvious.
Teachers could analyze their success and failures. Students
and principals had a full school year to look back at the
teacher's behavior and evaluate the skills and attitudes
which he brought to the teaching situation.

Of the eighteen principals who agreed to have their
junior and senior high schools participate, 10 submitted suf—
ficient data, and nine were used in the study. During July
the results were analyzed and are presented here in the fol-

lowing chapters:

20

II. Related Ideas and Literature
III. Procedures Used
IV. Analysis of Results

V. Summary, Conclusions and Implications

Summary

To this point the discussion has indicated the
direction of the study. It has indicated that there is a
need for studies of morale; that the problem under study is
based on the theory that high morale of teachers is conducive
to the aims of education; that the problem was to discover
whether or not a relationship exists between teacher morale
and teacher effectiveness, as perceived by teachers them-
selves, students and principals; and that there are limi—
tations to the study which prevent showing cause and effect
relations, but which permit theory develOpment and basis for
further research.

Related ideas and literature will be discussed in

Chapter II.

CHAPTER II

RELATED IDEAS AND LITERATURE

Because the findings of this research have certain
implications for administration, the first portion of this
chapter will be devoted to an underlying philosophy of teach-
ing and learning and administration upon which the impli—
cations rest. A discussion of recent studies in teacher
morale followed by references to important studies of teacher

effectiveness completes the chapter.

Teaching and Learning

 

The ferment in student activities, such as those at
Berkley, was overt testimony of the desire of youth to
actively participate in the building of American society.

It was not enough to make acquisition of knowledge identical
with education. Hart viewed that kind of education as dis—
integrating tO society and made a strong plea for knowledge
to be integrated with the human processes of living.38

Perhaps this generation will give real meaning to

education in a free society. Will French pointed out that

 

38Joseph K. Hart, Education in the Humane Community

(New York: Harper and Brothers, 1951), p. 29.

 

21

22

educators have an astonishing power of choice to mold youth
into submissive unsure individuals or to create the con-
ditions which will develOp Open, adaptive, independent, free-
thinking, self-respecting individuals.39
These are qualities necessary to living in a rapidly
changing society. The submissive and unsure prefer security.
About security and movement, Kelley and Rasey had this to say:
. when human beings are able to get control of
the process and movement, they will stOp seeking se—
curity in the unchanging and begin to live in a
changing world, where their security actually lies.4O
Kelley and Rasey believe that to achieve such a goal
the teacher's role should be that of the stage manager and
scene shifter who provides a suitable situation which facili—
tates growth in attitudes, habits, andknowledge.41 Getting
the teacher off his pedestal, which tends to erect barriers
to learning, is part of their prOposal to encourage learning
and increase communication.42 This philoSOphy recognizes
the individual's responsibility for his own learning and pro-

vides the freedom for him to pursue it. This philoSOphy

provides the learner with the Opportunity to creatively

 

39Will French and Associates, Behavioral Goals of
General Education in High School (New York: Russel Sage
Foundation, 1957), p. 45.

40Earl C. Kelley and Marie Rasey, Education and the
Nature of Man (New York: Harper and Brothers, 1952), p. 21.

41Ibid., pp. 75-77.

42Ibid., pp. 83-84.

23

participate in the learning process. He is not primarily re-
sponsible for regurgitation of facts, but he is responsible
for integrating his knowledge so that it is useful in solving
the problems of living in a complex society.
Kimball Wiles said:
When the first edition of Teaching for Better
Schools was published, I was convinced by evidence
from research in human relations, group development,
psychiatry and psychotherapy, counseling, communi-
cation, leadership, and the learning process that
the role of the teacher is to assist in the creation
of the environment in which learning occurs and to
relate to the learner in such a way that he will use
the help the teacher can give. I am still convinced.43
Evelyn Wenzel discussed the creative teacher in
terms Of one who had developed a sensitivity to students:
understanding their behavior as well as their words and
recognizing that learning can take place only if the learners

are members of an interacting group.44

Administration

What is true of the relationship between teachers
and pupils is also true of the relationships between ad-
ministrators and teachers. The human element is the basic
factor to be considered if the school is to achieve its

goals. Elsbree and Reutter, discussing morale, considered

 

43Kimball Wiles, Teaching for Better Schools (Engle-
wood, Cliffs, N.J., Prentice Hall, Inc., 19597, p. v}

 

4Evelyn Wenzel, "What Is A Creative Teacher,"
National Education Association Journal, Vol. LIII, NO. 6
(September, 1964), pp. 8-10.

24

the first task of school administrators as that of convincing
themselves that high morale " . . . is directly related to
certain well defined personnel policies and procedures that
have their roots in a philosophy of human relations."45

Eugene Kim saw the principle Of human relations as

basic in all types of human organization but especially so
in an educational institution:

First, the major process Operating in an edu-
cational organization is a human one, and the final
products of this "service oriented" organization are
human beings, not material goods, Second, the quality
of personnel who produce the human product constitute
the core which invariably influences the total edu-
cational achievement. These two factors bear more
directly upon personnel administration in education
than is generally true in other areas.

He discussed in this article five important aspects: recog-
nition of interpersonnel needs; scientific investigation on
interpersonnel relations; provision for social and psychic
space; executive leadership and a human relations skill; and
change in evaluating educational products. He noted that
lack of satisfaction of biological needs impairs the well
being of the organism and likens this to the interpersonnel

needs of an organization; that there is need for release from

arbitrary authority, for self-expression, for greater freedom,

 

45Willard S. Elsbree and E. Edmund Reutter, Jr., Staff
Personnel in the Public Schools (New York: Prentice Hall,
Inc., 1954), p. 265.

6Eugene C. Kim, "Five Realities Important to Faculty
Human Relations in School Administration," The Bulletin Of
the National Association of Secondary School Principals
(Washington, D.C., April, 1962), Vol. XLVI, NO. 273, pp. 121.

25

and for self-development; that further research is needed on
executive leadership; and that there is need for instruments
of evaluation that take into account the socio-psychological

aspects of performance.47

Related Studies of Morale

The current philOSOphy of educational administration
seems to be that the administrator's most important task is
"to release the creative capacities of the teachers on his
faculty." This thought expressed by Griffiths has also been
discussed in the preceding paragraphs, but Griffiths stated
further that administrators, for this reason, must pay more
and more attention to factors of morale.48

He made it clear that "groups must be worked with
and not on," and that "change comes from the inside."49
Such statements were made with the belief that high morale
promotes high production and low morale begets low production.

These beliefs have developed over a fifty year span
of time. Suehr points out in his comprehensive review of

the literature that morale studies in education have been

borrowed from industry and the military. The literature

 

47Ibid., pp. 121-124.

 

48Daniel E. Griffiths, Human Relations in School Ad-

ministration (New York: Appleton Century—Crofts, Inc., 1956),
p. 145.

49Ibid., p. 147.

26

increased in amount during war years. High three year
periods were: 1941-44; 1950-53 and 1957-1959.50

Two examples of important research, one from the
military and one from industry, discussed in the chapter:
"Research and Theory" of the book, Research Methods in Social
Relations, indicates the importance of research in these two
areas.

The first report is credited to "Stouffer (in Merton
and Lazersfeld, 1950)," The concept of "relative deprivation"
is described as a conclusion made by Stouffer from his
studies of the American soldier. He reports that Research
Branch I was stumped when it discovered:

. . Northern Negroes in Southern camps, in spite of
the fact that they said they wanted to be stationed
in the North and that they resented discrimination in
Southern police, showed as favorable or more favorable
responses to items reflecting personal adjustment in
the Army than did those in Northern camps. .

When, eventually, it was suggested that the Northern
Negro soldier in the South had very great advantages
over Negro civilians in the South and that advantages
over Negro civilians in the North were much less, a
clue to the paradox appeared.

Out of this paradox came a somewhat new theory which was

\
called "relative deprivation."

The concept Of "relative deprivation" has relevance

to this research in that it points up the importance of the

 

50John H. Suehr, "A Study of Morale in Education" (un—

published Doctoral dissertation, University of Colorado,
1961).

1 . . .
Cla1re Sell1tz, Marie Jahoda, Morton Deutsch, and
Stewart W. Cook, Research Methods in Social Relations (New
York: Henry Holt and Co., Inc., 1959), p. 496.

27

satisfaction of personal needs as a factor of teacher
morale.

Another germane example, this time from industry, may
be cited. The well known "Hawthorne Studies" by
Roethlisberger and Dickson in 1939 discovered the importance
of social and attitudinal factors to production. NO matter
whether physical conditions of work were improved or made
poorer, production increased. The workers in the experi—
mental group knew they were being set apart to study changes
in working conditions and a group cohesiveness develOped
among them. Someone had shown an interest in them. The re-
searchers concluded that output is influenced by social and
attitudinal factors within the work situation.52 Such re-
search reinforces the importance of satisfaction of personal
needs as a morale factor.

Boris Blai, Jr., searching for a job satisfaction
predictor, used six multiple choice questions in connection
with job satisfaction and needs satisfaction. His subjects
were 470 persons employed in Philadelphia by the federal
government in three very large and occupationally diverse
organizations. The questionnaire was designed to measure
two variables: (1) degrees of need satisfaction among four—
teen pre-selected psychological needs, and (2) the indi-
vidual's self—evaluated degree of job satisfaction. The

need statements were randomized among fourteen possible

 

521bid., p. 497.

28

forced choices in each of five Of the six questions. Re—
spondents selected from the fourteen needs, three they con—
sidered most desirable, satisfying or most important and
ranked them in order of importance.

In the sixth question, the subjects were asked to
indicate their satisfaction with their present jobs on a
seven point scale with one being negative and seven positive.
His hypothesis was: "Job satisfaction varies with the
strength of human needs satisfaction." The hypothesis was
true and he listed the strongest needs by occupations as

Table 2.1 illustrates.53

Table 2.1. Percentage Of persons in five occupational groups
expressing important personal needs.

 

 

 

Pro— Cleri- Trades—
fessional Managerial cal Service Manual
Self-
actualization 70 58 36
Interesting”
Duties 70 54 60 34
Advancement 39 33
Job Security 53 70 71
Independence 41
Esteem 36
Congeniality 34

 

 

53Boris Blai, Jr., "A Job Satisfaction Predictor,"
Personnel Journal, Vol. XLII, No. 9 (October, 1963), p. 453.

29

The foregoing examples serve the purpose Of showing
how the study of morale in education is related to research
in business and industry. For instance, it appears from the
studies of morale in education that all six of the items
mentioned by Blai in Table 2.1 are important to teacher
morale. Though the six names above, and the terms, "relative
deprivation" and "social and attitudinal factors" will not
appear, similarities can be noted in the literature concern-
ing studies of morale in education. Three factors: satis-
faction Of teachers' personal needs, teacher satisfaction
with administrative policies and practices, and teachers'
satisfactions with their tasks will receive considerable
attention.

Researchers of conditions of morale in education
only recently got on the band wagon. The studies reported
here are general in nature because few studies have attempted
to relate teacher morale to the evaluation of teachers. The
importance of administrative policies and practices to morale
in the educational organizations stands out just as inter—
personnel relationships have been reported important to

other organizations.
Studies of Morale in Education

Many things have been rediscovered from time to time

about morale in education. They have been reported as main

30

studies or tangentially to the issues being researched. In
1954 Elsbree and Reutter'wrote:

The quality of morale, like the efficiency of a

teacher, is somewhat elusive; but students of the

morale problem have now reached rather general

agreement as to what morale is and have identified a

number of common factors which seem to condition the

attitude of workers in every sphere of life.
Their chapter on "Morale" emphasizes the responsibility of
the administration for high morale of the staff and non-
certified employees. They relate high morale directly to
"well—defined personnel policies and procedures that have
their roots in a philos0phy of human relations."55 Personnel
policies in and of themselves, like the socio—economic nature
of the community, wages, tenure, and sick leave, are no
substitute for teachers' satisfactions with their tasks and
of their personal needs. Every teacher needs to feel that
he has worth and can make a contribution with his colleagues
and the administration toward the goals of the school organi—
zation. This emphasizes the need for good two-way
communications.

These and other research results will be presented

in chronological order in the following discussion. General—

ly speaking, the studies referred to will have been completed

during the last fifteen years.

 

54William S. Elsbree and E. Edmund Reutter, Jr.,
Staff Personnel in the Public Schools (New York: Prentice
Hall, Inc., 1954), p. 262.

55Ibid., p. 265.

31

.Anderson's attempts to relate student achievement to
teacher morale are referred to several times in the litera-
ture. He found that the achievement of students of low
morale teachers is less than that of students of high morale
teachers. -Anderson studied twenty Iowa schools selected on
the basis of the Iowa Tests of Educational Development. Ten
of the schools were above the 75th percentile and 10 were be-
low the 30th percentile. The teachers were interviewed and
tested on morale.56 Of this study, Suehr stated:

While more work needs to be done in this area in

order to determine how morale affects achievement

of students in relation to their ability, this study

alone should stimulate action.57
In 1957 Redefer said that Dr. Anderson had left many questions
unanswered and that studies suggested by Anderson had not
been forthcoming.58

Monford's thesis in 1956 on factors influencing the

work of teachers found the most important factors to be:
helpful, cooperative, understanding principals
and supervisors: freedom allowed teachers to plan
and teach as they think best and good human re-
lations among teachers within the school. Ele-

mentary teachers are hindered most by excessive
clerical work and lack of time away from pupils;

 

56Reported by Daniel E. Griffiths, Human Relations in
School Administration (New York: Appleton-Century-Crofts,
Inc., 1956), p. 145.

57John H. Suehr, "A Study of Morale in Education" (un—
published Doctoral dissertation, University of Colorado,
1961), p. 23.

8Frederick L. Redefer, "Teacher Morale and Quality
Education," Nations Schools, Vol. LIX (February, 1957), p. 53.

32

secondary teachers are hindered most by interruptions

during their teaching and by removal of pupils from

the classroom for other activities.
The theory of "relative deprivation" as applied in "Studies
of the American Soldier"60 appears operant here. Elementary
teachers compare themselves with secondary teachers most of
whom have a planning period and easily conclude that they
have less free time away from students. High school teachers
conclude that they have more extra curricular activities than
elementary teachers. The attitudes of both groups might be
classified as "relative deprivation" which affects their
morale.

Sometimes dissatisfactions are as prevalent in high
morale schools as in low morale schools. Making this con—
clusion, Roth also found dissatisfaction with working con-
ditions to be more prevalent among high school teachers. He
also noted that low morale teachers made more use of the
free response line available on his measuring instrument.61

It should be noted that freedom of response to the
Teacher Morale Form used in this study Of "Teacher Morale

and Evaluation of Teachers" is one of the attributes of the

TMF.

 

59Ida Bell Monford, "Factors Influencing the Work of

Teachers in Fairfax County Virginia" (unpublished Doctoral
dissertation, Ohio State University, 1956).

600p. cit.
61

James Lester Roth, "A Technique for Determining the
Sources of Job Dissatisfactions" (unpublished Doctoral
dissertation, Stanford University, 1956).

33

Two years later in the central New York area, O'Connor
examined sixteen general questions included in an extensive
questionnaire administered to 303 public school teachers.
Ninety—nine percent returned the questionnaire. His inquiry
attempted to (1) measure the overall level of satisfaction
or dissatisfaction with the conditions of work found in differ—
ent schools, and (2) to find out what factors in the school
situation might be associated with high or low morale. He
does say that no school could be identified as a high or low
morale school, and that independent judges asked to diagnose
morale were wrong two out of three times when designating
high and low morale schools. When the upper quartile,
seventy-seven high morale teachers was compared with the
lower quartile, he found:

1. Teachers perceived own morale to be higher than
their co-workers.

2. The most consistent relationship with the overall
level of morale was the way in which the teacher
felt about his administrator or supervisor.

3. Sex of supervisor, salary, size of class, and
extra duties did not appear as important as
number two.

4. High morale teachers estimated co-workers to be
COOperative and well disposed toward each other.

5. Low morale teachers had poor relations with super-
visors, fellow teachers, pupils and the community.

6. Happier teachers had more free time away from
children each day, were supervised more often,
and had more democratic relations with
supervisors.

34
7. High morale teachers were over 35, home owners,

and active church members.

Cohen determined that attitudes toward morale areas,
such as personnel practices, curriculum, and teacher—
supervisor relationships, exist independently of the socio-
economic nature of the community. -Comparing privileged and
underprivileged schools she reported that there were high
and low morale teachers in both, but that autocratic
practices of administrators contributed to low morale.63

‘Administrative behavior will undoubtedly come under
scrutiny again and again as one of the contributing factors
to high and low morale Of teachers and non—certified person—
nel. Wagoner made such a study in order to determine which
tests would discriminate between principals whose staffs have
high or low morale. Principals were divided into two groups
on the basis of total scores on the "Faculty Morale Rating
Scale" and on the basis Of the "Rating Scale of Hostile Be-
havior." He determined significant differences between high
and low morale groups by using the "t" test of the standard
error of the difference between the means of small inde-

pendent samples. He reported:

 

62William Francis O'Connor, Jr., "A Study of Some
Selected Factors Related to Teacher Morale" (unpublished
Doctoral dissertation, Cornell University, 1958).

63Reported by Clyde E. Bocker and Richard C.
Richardson in "Twenty-Five Years of Morale Research: A
Critical Review," Journal of Educational Sociology, Vol.
XXXVI, NO. 5 (January, 1963), p. 201.

35

1. With principals divided into high and low morale
groups by the RSNB the most effective tests in
discriminating between groups of low morale and
high morale were: the Public Opinion Questionnaire,
and the Edwards Personal Preference Schedule on
three variables: affiliation, heterosexuality, and
consistency.
2. With the principals divided into high and low morale
groups on the basis of the FMRS, . . . the Minnesota
Teacher Attitude Inventory and the EPPS discrimi-
nated between the means.
His conclusion was that the EPPS, the P00 and the MTAI may
be helpful in discriminating between principals whose staffs
have high or low morale.64

It seems to the writer that it would be much more
profitable to measure teacher morale directly. Studying
principals' morale undoubtedly has value, but Waggoner's con—
clusion hardly seems valid or efficient.

Ross used a battery of seven measuring instruments
to secure data from approximately 100 rural teachers in two
school systems in order to compare the morale status of rural
and suburban schools. The first instrument of 108 items
measured the high and low morale tendencies of teachers. The
next three instruments provided information by which to
interpret the morale and the last three were used to validate

the original test. One of Ross' findings was that rural

teachers of low morale tend to view their social status as

 

64Glen Hastings Waggoner, "Administrators' Scores on

Selected Standardized Tests and His Administrative Per-
formance as Reported by Classroom Teachers" (unpublished
Doctoral dissertation, Stanford University, 1959).

36

being lower than that of non—teachers, and to consider the
cultural level of the community as inferior.65 Cohen, on the
other hand, had reported that morale was independent of the
socio-economic nature of the community.66 Herein lie some
apparent differences. He also concluded that, in general,
the social origin of the teacher is not the determinant, per
se, of the morale status of the teacher.67 Suehr reported
the opposite. He found that high morale teachers come from
upper or upper middle class homes more often than low morale
teachers.68

There is general agreement with other conclusions
made by Ross.

1. Staff morale is a function of personnel policies
and practices which are characteristic, innately

of each staff group.

2. Married females tend to have higher morale than
married males.

3. The factor of desirability of community residence
is not related significantly to the morale status
of rural teachers.

 

65WalterE. Ross, "A Study of Personnel Factors Af—

fecting the Morale Status Of Teachers of Two Rural School
Systems in New York State and Including Comparisons of
Findings With Those of a Similar Study Completed for a New
Jersey suburban School System" (unpublished Doctoral disser-
tation, New York University, 1960).

66Blocker and Richardson, Op. cit., p. 201.

67Ibid., p. 201.

68John H. Suehr, "A Study of Morale in Education
Utilizing Incomplete Sentences," Journal Of Educational Re—
search, Vol. LVI, NO. 2 (October, 1962), p. 78.

37

4. Comparatively, rural and suburban teachers tend

to react similarly, with but minor variations to

personnel factors affecting morale status.

Suehr, interested in the personality factors that
might cause high or low morale in teachers, devised an incom-
plete sentence blank test which he called the Teacher Morale
Form. It was patterned after the semi-structured projective
technique used by Rotter and Rafferty.7O It is assumed, they
say, that the subject reflects his own wishes, desires,

fears and attitudes in the sentences he completes when the

first word or words are supplied. Its advantages are:

1. Freedom of response is provided.

2. Some disguise is present.

3. Group administration is relatively efficient.

4. NO special training is necessary for administration,
and interpretation can be done with a minimum of
clinical experience.

5. It lends itself easily to Objective scoring.

6. The time of administration tends to be shorter.

7. new sentence beginnings can be "tailor made”

to fit a variety of situations.
Its disadvantages are that it cannot be machine scored, not
as much disguise as in other projective techniques is present
and insufficient material is obtained in some cases.71

Using this information Suehr developed a forty item
form which was highly discriminating. It is described in de—

tail in Chapter III of this study. With it he measured the

 

69Op. cit.

7OJulian B. Rotter and Janet E. Rafferty, The Incom-
plete Sentences Blank: College Form (New York: The Psycho—
logical Corporation, 1950), p. 3.

71Ibid., pp. 3 and 4.

38

morale of teachers in the public schools Of Boulder, Colorado
and related psychological, sociological and biological
factors of teachers to their morale.

Suehr found twenty-six factors affected the person-
ality of the teacher and were significantly related to morale
at the five percent level. Quoted here are a few of those
items not generally reported in the literature:

1. Teachers indicating that an opposite sexed parent
had the most influence on them were more apt to be

low morale teachers.

2. Being the youngest in the childhood family was a
factor in low morale.

3. Low morale teachers feel that they are more stubborn
in their personality make-up.

4. High morale teachers are more apt to go to bed early
and get up early.

5. High morale teachers were more apt to rate them-
selves as bing slightly introverted, and low morale
teachers were more often found in the slightly
extroverted category.

6. High morale teachers more Often feel that their
childhood family was very close.

He also administered the Incomplete Sentence Blank
to principals to determine their personal adjustment and re-
late it to teacher morale. No principal was poorly adjusted
and no significant relationship was found between teacher
morale and the adjustment of principals.

A postcard questionnaire asked parents about their

degree of satisfaction with the job the school was doing for

 

72Op. cit.

39

their child. .A five percent sample showed that parent satis-
factions were less when teacher morale was low. Parent
satisfaction was significantly related to morale of teachers
but not to educational level or socio-economic factors of
parents.

Some Of Suehr's conclusions not reported by others
are:

l. The childhood background of teachers is highly

important to their morale as adult teachers.

High morale teachers have an advantage from the

beginning by more Often being born into higher

socio-economic status.

2. rAs among students, personality differences exist
among teachers.

3. Cause and effect relationships are evident in
the area of morale. Parents sending their
children to schools in which teacher morale is
low show a corresponding doubt of that school.

4. The incomplete sentences method is the best
method yet devised for determining teacher
morale.

5. Much is known about pupil motivation, but little
is known about what motivates teachers.

6. Teaching is not adequately recognized as a de—
sirable profession for a man; consequently the
type of man entering teaching might possibly be
different from the man entering medicine, law,
or business.
His other conclusions corroborate other research. They in—

clude the importance Of good communication, good human re—

lations, consideration of the whole personality of the

40

teacher, and good mental health when considering teacher

morale.73

Quinn McNemar, in 1946, had suggested that morale is
"not an entity," that there are many "morales,“ and that
they could be reliably measured by scale techniques. He

further hypothesized:

It seems more reasonable, however, to believe that
these "little things are not entirely independent,
either statistically or functionally, that certain
Of them tend to go together, or form clusters, and
that such clzsters are conceivably independent of

each other.7

The concept of clusters of morale was used by Bently
and Rempl of Purdue to study "Peer Selection vs. Expert
Judgement as a Means of Validating a Teacher Morale Measuring
Instrument." Of 169 items judged they retained 157 to
sample eight clusters:

self status

relationship with students

relationships with other teachers

factors relating to administration and policies
relationships with the community

curriculum factors

working conditions

economic factors

CDQONU'ItPOJNl-J

Their sample included 570 teachers in twenty-two Indiana
high schools. -Kuder-Richardson reliability coefficient for

total scores using the hold-out sample was .96. Coefficients

 

73John H. Suehr, "A Study of Morale in Education" (un-

published Doctoral dissertation, University Of Colorado,
1961).

74Quinn McNemar, "Opinion Attitude Methodology,"
Psychological Bulletin, Vol. XLIII (July, 1946), pp. 365—367.

41

of reliability using cluster scores ranged from .79 for eco-
nomic factors to .98 for factors relating to administration
and policies. They found that a large majority of the items
in the teacher morale measuring instrument showed low dis-
criminating power between "high" and "low" morale teachers
as identified by peers. They noted two dangers: (1) teachers
may be judged likeable or non-likeable, and (2) the marked
differences in the morale among schools could be explained as
only a relative difference since peers judged morale in terms
of a particular school, i.e., low morale teachers might be
at a high morlae level in another school. Using a scoring
key based on expert judgements, they found nearly all items
discriminated between teachers with "high" and “low" total
scores.75 Reporting the research by Bently and Rempl has
most relevance for this study in the use of clusters of
morale, as will be seen later in Chapters III and IV.

Three different pieces of research reviewed by
Blocker and Richardson have general relevance for this study.
They report: (1) Garrison using the McCluskey and Strayer
Test concluded that "nearly every aspect of the teacher's

environment is involved in adjustment to the job";76

 

75Ralph R. Bently and Arno M. Rempl, "Peer Selection
vs. Expert Judgement as a Means of Validating a Teacher
Morale Measuring Instrument," Journal Of Experimental Edu-
cation, Vol. XXXI, NO. 3 (March, 1963), pp. 235-240.

76Clyde E. Blocker and Richard C. Richardson, "Twenty—
five Years of Morale Research: A Critical Review," Journal
Of Educational Sociology, Vol. XXXVI, No. 5 (January, 1963),
p. 201.

42

(2) Knox using a modified McCluskey Test attempted to relate
sixty-five aspects of the teacher's environment to teaching

success. He discovered that a positive relationship existed
"between teacher efficiency and the sort of people that make

77 (3) Bidwell, they say, in a study that

up the community";
was well done reported:

. teachers who perceive administrative behavior

as being consistent with their expectations will tend

to be satisfied With the teaching situation; teachers

whose perceptions are not consistent with expectations

will tend to be dissatisfied, and the level of teach-

ing satisfaction depends on the expectation and

whether or not it is fulfilled. It does not depend

on the nature of the expectation.
Blocker and Richardson concluded that this research tended
to refute the position that democratic administration is
best for all situations.78 In summary, they felt that the
most important items of morale were those dealing with
personality and human relations, and that the principal may
be "the strongest morale producing factor."79

Pryor's research was about this strongest morale pro-

ducing factor. He studied 323 teachers in nineteen adminis—
trative units and three districts. The instrument, develOped
by the Industrial Relations Center of the University of
Chicago, dealt with teachers' perceptions of 105 different

descriptive items of policies and practices as they existed

 

 

 

 

43

in the administrative units. Teachers evaluated, by means
of a five point scale, how each of the 105 dimensions as it
existed in their school unit affected their morale status.
The responses were dichotomized into favorable and unfavorable
responses and the morale tendency scores. These were tested
by the Pearson Product Moment Coefficient of Correlation.
The two most important of eight conclusions were:
1. The relationship between teachers' perceptions
Of administrative policy, procedures and
practices, and morale status of teachers is
significant.
2. The administrative policies, procedures and
practices have their only important existence

in the way the teachers perceive them in Oper-
ation in their school system.

Summary of Morale Research

Two articles by Frederick L. Redefer will be borrowed
from extensively for the following summary of morale research
for two reasons:

1. Any discussion of morale in education without refer-
ence to this New York University professor would be remiss.

2. The findings of fifty graduate student researchers
under his direction over a period of seven years resulted
in twenty doctoral dissertations, the results of which rein—

force the research previously discussed here.

 

80Guy Clark Pryor, "The Relationship Between Teachers'

Perceptions Of Administrative Dimensions and the Morale
Status of Teachers in Certain Texas Schools" (unpublished
.Doctoral dissertation, NOrth Texas State University, 1964).

44

All the research in more than fifty school systems and in-
volving 10,000 teachers was conducted in the New York metro-
politan area.

In consulting industrial researchers, his students
found that school personnel differ from factory personnel,
but the principles were very much alike.

"Personnel policies and practices rather than salary
levels are the key to high and low teacher morale,"81 stated

Dr. Redefer: Studies by Monford,82 Cohen,83 O'Connor,84

Bidwell,85 and Pryor86 all reported the importance of ad—
ministrative policies and behavior to teacher morale.
"Several studies," reported the N.Y.U. professor, "noted
that commendation for good work done by teachers is frequent—
ly overlooked by administrators."87 The morale of teachers
in low_socio—economic areas was found to depend on many

factors. His students found that: "Democratic adminis—

tration by the school principal can counter-balance

 

81Frederick L. Redefer, "Studies of Teacher Morale,"
School and Society, Vol. XCII (February 22, 1964), pp. 63-64.

82Op. cit.

83Blocker and Richardson, Op. cit.
84Op. cit.

85Blocker and Richardson, Op. cit.
86Op. cit.

87

Redefer, Op. cit.

 

uhlu

«\UP

s

\

45

unfavorable community factors."88 This was the work of Gloria

Cohen89 reported earlier.

-Another area was freedom to plan and teach: this in—
volves being able to choose books and discuss controversial
issues with students.90 Monford also reported that freedom
to plan and teach was essential to the teachers of Fairfax
County Virginia.91

Morale exists to the degree that teachers freely

and consciously released and focused the skills,
knowledge, and abilities they possess, to achieve
known and accepted educational Objectives which they
have actively participated in formulating.

Teachers reported, he said, that the principal's
most important job is to understand them as persons. When
principals complain that faculties don't read communications,
teachers may be blocking out the message because no one
listens to them. Redefer emphasized, "Only two way communi-
cation will solve the problem."93 The level on which

teachers communicate and receive communications has much to

do with the morale of individual teachers.

 

88Ibid.

 

89Gloria J. Cohen, "A Study of the Socio—Economic

Status of the School Community and the Morale of Teaching
Personnel in New York City" (unpublished Doctoral disser—
tation, New York University, 1959).

90Redefer, Op. cit.

9J'Monford, Op. cit.

92Frederick L. Redefer, "A Teacher Teaches Better,"
N.E.A. Journal, LIII (April, 1964), pp. 8-10.

93Ibid.

46

Commenting on the undesirability of reducing interpre-
tations of group morale to averages and standard deviations

he added:

Yet, since a faculty is a group of individual
teachers, it is possible in the study of schools to
Obtain insights into teachers as individual
personalities.

This was a basic assumption of most of the research reported
here and of this research on "Teacher Morale and Teacher
Evaluation."

Now that the recent research on morale and education

has been reviewed, the pertinent research on teacher evalu-

ation will be discussed in the second half of this chapter.

Evaluating Teachers

The amount of research devoted to identification of
good teachers has been long and varied without very much
agreement on the best methods of evaluating. Some of the
methods are: observing teacher behavior and noting methods
and pupil—teacher interaction; teacher self-ratings; pupils
ratings Of teachers; and supervisors' or administrators'
rating of teachers. All have their particular places and
purposes and studies of teacher effectiveness have been
numerous.

A few evaluations of teachers studies were completed

between 1896 and 1912; then there was a sharp rise to about

 

94mm.

 

I H.-

V:

b\

nK»

ea

-u\ hi. ‘4.»

47

sixty studies by 1917 followed by a leveling Off for the
next decade. Between 1927 and 1932 the number of studies
reached 250. ~A sharp decline followed to about 150 in 1937.
With the advent of Gestalt psychology the number of studies
based on objective data decreased.95
The search for the good teacher continues. A forty-
eight page book, Who's A Good Teacher, discusses the im-
portance of knowing what good teaching is; what is meant by
evaluation, and the purpose and effects of it. It considers
the three ways of measuring teacher effectiveness-—pupil
growth, teacher behavior and teacher effectiveness--and
problems connected with each one. The first method has
proved rather inconclusive but the other two show some
promise.96 Redefer's comment on the Anderson study casts
some doubt on the method of using student achievement alone
to measure teacher effectiveness.97
This would seem to infer that teacher effectiveness

should be measured by other means than the achievement of

their pupils.

 

95Who's A Good Teacher (Washington, D.C.: American
.Association of School Administrators, Department Of Classroom
Teachers of the N.E.A., and National School Boards Associ-
ation, 1961), p. 12.

96Robert B. Howsam, Who's A Good Teacher: »Problems
and Progress in Teacher Evaluation. Prepared for the Joint
Committee on Personnel Procedures of the California School
Boards Association and the California Teachers Association.

97Redefer, Nations Schools, Op. cit., p. 53.

48

The nature of this study lends itself to the more
subjective approach and the use of a rating scale. The
literature examined will be limited to discussions Of the use
of rating scales, teacher self-evaluations, student evalu-
ations of teaChers and principals' or supervisors' evalu—

ations of teachers.

Use of Rating Scales. H. H. Remmers, in his chapter,

 

"Rating Methods in Research on Teaching," says:

It is likely that no approach to the measurement of
variables in research on teaching has been used more
often than the rating method. . . . Since 1950 the

use of rating methods has not diminished in importance
or frequency.

He describes this type of research as widespread and basic
and notes that the measuring device is not the form but the
rater. The ratings he points out have certain limitations:
characteristics of the human rater--his in—
evitably selective perception, memory, and for-
getting, his lack of sensitivity to what may be
psychologically and socially important, his in—
accuracies of observation and in the case of self
ratings the well established tendency to put his
best foot forward, to perceive himself in a more
favorable perspective than others do.
In the following discussion the attempt will be made

to show how various researchers have met these problems.

Teacher Self Evaluations. In 1954, Elsbree and

Reutter said: "There are no valid self—rating scales by

 

98N. L. Gage (ed.), Handbook of Research on Teaching
(Chicago: Rand McNally and Co., 1963), p. 329.

99m...

 

49

which a teacher can rate his efficiency."100 Indeed, this

appears to be true, but six years later, A. S. Barr described
a self-rating scale based on the idea that teacher effective—
ness is determined, at least in part, by the teacher's
personality. In Parth, the teacher ranks himself, and ad-
ministrators and teachers he knows on fifteen qualities. In
Part B, the teacher compares himself with these adminis—
trators and teachers. In Part C, he notes the amount of
qualities he thinks they have and he has. Barr believes the
scale to be discriminating and Objective but that further
research is needed.101

Also in 1960, Simpson found the use of self-evaluation
instruments to be widespread and have many successful

102 .
users. However, there appears to be no conclus1ve re-

search on the subject.

Student Evaluations Of Teachers. The importance of
pupil ratings Of teachers is reported by Remmers who is re—
sponsible for the Purdue Rating Scales. Such measures are
valid and reliable. Reliability of ratings of teachers by

twenty-five or more students is as good or better than most

 

1000p. cit., p. 247.
101

A. S. Barr, "The Assessment of the Teacher's Person-
ality," School Review, Vol. LXVIII (Winter, 1960), pp. 400-
408.

102 . n .

Ray H. S1mpson, Use of Self Evaluative Tools for
the Improvement of Instruction," AACTE Bulletin, Vol. VIII
(American Association of Colleges of Teacher Education,
December, 1960), pp. 1—9.

 

50

educational or mental tests available. They are affected
only slightly or not at all by age, sex of students or
teachers, popularity, grades given by the teacher, or diffi-
culty of the courses.103

Gage's conclusions were reported in 1963. Earlier,
in 1937, Bryan, in a study of secondary school teachers,

stated that his purposes were:

1. To determine the effect of such factors as in-
telligence, school marks, and sex on ratings.

2. To determine the degree of correlation of
pupils' ratings of teachers.

3. To determine the degree of agreement between
pupils' and administrators' ratings of teachers.

4. To determine what items in the rating instru-

ment have the most weight in determining general

teaching ability.
He studied 900 eighth and ninth graders, 600 tenth and
eleventh graders, and administrators' ratings. (Each pupil
on unsigned forms rated four teachers. Chance half corre-
lations were .90 and internal consistency of the tests, shown
by comparing general teaching ability with the other ten
items, was .995 and .997 for the junior and senior high

groups, respectively. Significant correlations of .68 and

.69 were shown for ratings by students and administrators

 

103Gage, Op. cit., p. 368.

104Roy C. Bryan, Pupil Ratings of Secondary78chool
Teachers, New York: Teachers College, Columbia University,
1937). p. 7.

51

for junior and senior high groups, respectively.105 His

findings, arranged in the same order as the purposes, were:

1. There is no significant difference between ratings
of pupils and high or low intelligence.

2. There was some significant difference between
ratings of boys and girls of men and women
teachers. Nineteen percent of the total number
of differences were considered significantly
different.

3. There was a slight tendency for pupils with high
marks to rate teachers higher than pupils with
lower marks, but there were many exceptions.

4. The ratings of 40 pupils gave correlations Of
.90 or above.

5. Self consistency of pupils was nearly perfect.

6. Pupils ratings are highly reliable.106

Determining who are effective teachers increased in
importance with interest in merit pay. McCall and Krause
report:

Generally speaking, the means of evaluating
teachers for merit pay, which are commonly used such
as: hours of credit, activity in the community, pro-
fessional knowledge, 1.0., years of service, knowledge
of subject matter, were shown to be invalid measures
of teacher effectiveness.

Of all the measures used, the one which proved to
have the highest correlation with teacher merit was
the McCall-Herring Personality Measure, when used as
a rating-by-pupils device. The scale consists of
five items: is clean, has good manners, keeps temper,
is kind, and is a good citizen. Correlation ranged
from .22 to .39. A comparison Of these results with
the results Obtained from ratings of teachers by
peers, principals, and supervisors indicates quite

 

105Ibi

Q.»

 

1061101

OJ

 

52

clearly that a teacher's pupils are far better
judges of a teacher's merit than are professionally
trained adults.l
General agreement has even been reached on the criteria to
be used.
Investigators have been able to classify teacher be—
havior into a relatively small number of areas. Ryans

(1960) found three patterns:

1. Understanding, friendly teacher behavior vs.
aloof, egocentric behavior.

2. Systematic, responsible, businesslike behavior
vs. evading, unplanned, slipshod behavior.

3. Stimulating, imaginative behavior vs. routine
dull behavior.

Nelson and others (1956) found almost identical
areas, i.e., disciplinary control, teacher pupil relations,
and instructional excellence.109

Gupta discovered five factors when sixth grade

pupils' ratings of 130 teachers were analyzed. A 96 item

scale combining the Leeds forty-eight item My Teacher inven-

 

tory with a forty—eight item Teacher Descrippion Inventory

developed by Gage and Weitman was used. These factors were:

 

107William A. McCall and Gertrude R. Krause, "Measure-
ment Of Teacher Merit For Salary Purposes," Journal Of Edu—
cational Research, Vol. LIII, No. 2 (October, 1959), p. 74.

108Who's A Good Teacher (Washington, D.C.: American
Association of School Administrators, Department of Class-
room Teachers of the N.E.A., and National School Boards
Association, 1961), p. 12.

logIbid.

 

53

1. Affective Merit
2. Cognitive Merit in Motivating Learning

3. Cognitive Merit in Promoting Comprehension (Nega-
tive Items)

4. Cognitive Merit in Promoting Comprehension (Posi-
tive Items)

5. Disciplinary Abilityllo
..Factors prerequisite to success in teaching were dis—

cussed by Barr and Emans. In an attempt to improve rating

scales they selected from 6,939 items those most frequently

mentioned as important. The top ten items and their fre-

quencies were:

1. Instructional Skill 371
2. Personal Fitness for Teaching 369
3. Scholarship and Professional
Preparation 301
4. Ability to Cooperate with Others 235
5. Appearance 213
6. Classroom Management 205
7.: Interest in WOrk, Pupils and Subject 172
8. Considerateness 145
9. .Leadership 131
10. Health 106111

Bryan takes note of the fact that surveys concerning
traits of the ideal teacher list: sincerity, fairness, ap—
preciatiyeness, friendliness, industry, good judgement and

ability to explain clearly.112

 

llOPromila Gupta, "A Study of Cognitive Merit of
Teachers" (unpublished Doctoral dissertation, University of
Illinois, 1960), reported in Gage, Op. cit., p. 368.

111A. S. Barr and Lester M. Emans, "What Qualities are

Prerequisite to Success in Teaching?" Nations Schools, Vol.
VI, NO. 3 (September, 1930), pp. 60-64.

112Roy Coulter Bryan, Pupil Rating of Secondary School
Teaching (New York: Bureau of Publications, Teachers College,
Columbia, 1937), p. 2.

54

Western Michigan University's Student Reaction
Center develOped a "Student-Opinion Questionnaire" for evalu—
ating teachers. Teachers subscribe for the service and are
sent any number of forms requested. Usually the teacher has
someone else administer the tests for him so that the stu-
dents can respond freely. They do not sign the opinionnaires
which are mailed to WMU Student Reaction Center. The Center
analyzes the results and sends the teacher a composite re—
port. The reliability coefficients for the different scaled
questions using fifty teachers—-one class per teacher-—range
from .86 to .92. This test used in this research is dis-
cussed in Chapter III. However, it is mentioned here so
that the similarity of its items to those mentioned above
may be noted.
Knowledge of the subject
Ability to explain clearly
Fairness
Discipline
Sympathetic understanding
Ability to make classes lively and interesting
Efficiency
Skill in getting students to think for themselves
Ability to help students know why they are learn—

ing certain things 113
General teaching ability.

\OCDflmUlnthi-J

l—‘
O

H. H. Remmers' comments that student ratings are
valid in-as-much as there is no other way to know how stu-

dents feel about their teachers than to ask for their

 

113Western Michigan University Continuous Study of Stu—
dent Reaction Reports (Kalamazoo, Michigan: Western Michigan
University, 1962).

 

55

judgements. A fitting summary to the subject of student
evaluations is his statement that:
Just as achievement tests in the usual sense are
used to evaluate one group of effects of the teacher
on pupils, so pupils' ratings of teachers may be used
to evaluati another equally important group of
effects.11
Attention is called to the fact that the items used

in the Western Michigan University studies are the ones most

often named as qualities of an effective teacher.

Supervisors' Evaluations of Teachers. The reader
will recall that McCall and Krause reported that principals
and supervisors are poorer judges of teacher effectiveness

than students.115

Peck116 and Bryan117 discovered the opposite. At-
tempting to predict principals' ratings from independent
analysis of personality data of teachers, he found that
principals' judgements are stable and valid, at least where
they have a chance for first hand observation.

Can principals exclude their own personality charace

teristics when they rate teachers? Andrews and Brown asked

 

114H. H. Remmers and N. L. Gage, Educational Measure-
ment and Evaluation (New York: Harper and Brothers, 1955),
p. 493.

115Op. cit.
116

Robert F. Peck, "Principals' Ratings of Teacher Per-
formance from Personality Data," Journal Of Educational Psy—
chology, Vol. L (April, 1959), pp. 70—74.

117Bryan, Pupil Ratings of Secondary Teachers, Op. cit.

56

this question in a study designed to discover if similarity
in personality traits affected principals' ratings of
teachers. Six hundred and eight principals and teachers
were studied by means of the Edwards Personal Preference
Schedule, the Allport-Vernon-Lindzey "Study Of ValuepV and
the Minnesota Teacher Attitude Inventory. He found no sig—
nificant relationship.118
Nelson, Bicknell and Hedlund developed three
measures, one of disciplinary control, one on teacher-pupil
relations and one on instructional excellence. Their raters
were pupils, supervisors or pincipals, and trained observers
from college staffs. Inter-rater correlations were .55 for
Observers and .68 for pupils and there was greater agreement
between the supervisors and the pupils. They pointed out
that this did not imply lack of ability on the part of the
Observers but that the one-half day spent with each teacher

assigned to them was insufficient.119

Summary of Teacher Evaluations

It is very apparent that measuring teaching effective—

ness is extremely difficult and that no general device has

 

118John H. M. Andrews and Alan F. Brown. "Can Princi-

pals Exclude Their Own Personality Characteristics When They
Rate Their Teachers?" Educational Administration and Super—
vision, Vol. XLV (July, 1959), pp. 234—242.

119Kenneth G. Nelson, John E. Bicknell and Paul A.
Hedlund, Development and Refinement of Measures of Teaching
Effectiveness, First Report of the Cooperative Study to Pre-
dict Effectiveness in Secondary School Teaching (Albany:
The University of the State of New York and the State Edu—
cation Department, 1956), p. 17.

57

been develOped for doing it. Also it is clear that the
personality of the teacher, elusive as it is to measure, is
Of paramount importance. Ellsworth Thompkins says that this
has "put some researchers on a new track—-studying the teacher
himself more than his practices.120
The research generally reports: that pupils' ratings
of teachers are valid and reliable; that principals and
supervisors can evaluate objectively if they take the time

to Observe and can agree on the criteria; and that teachers

self-ratings are valuable but reliable measures have not

been established.

 

120Ellsworth Thompkins, "Evaluating Teachers," Spot—

light (March-April, 1964).

CHAPTER III

PROCEDURES USED

The Sample

 

The hypotheses were tested in a population of 188
teachers in nine Michigan public secondary schools. Six
were schools having grades seven through twelve; one was a
ninth through twelfth grade school; one housed grades seven
through nine; and one had grades ten through twelve. The
first seven were in small rural communities under 1,669 in
population and the latter two were in suburban communities
of 75,000 or over. To the extent that 80 percent of Michigan
towns are smaller than 1700 in population, as given by the
1960 census the schools are representative. To the extent
that no communities Of 5,000 - 25,000 are represented and
the sample was smalL it is not representative.

The rural schools were consolidated and within fifty
miles Of cities of 18,000 or more. They could be described
as lower—middle and middle class communities, respectively.
Seven schools housed grades seven through twelve; one had
nine to twelve; one was a junior high with grades seven
through nine;and one was a senior high with grades ten

through twelve. Scattered around the lower peninsula, the

58

59

schools can almost be said to be representative. The 1960
census shows 80 percent of Michigan cities to be under 1,700
persons. Cities in the 5,000 to 20,000 pOpulation category
were not represented.

Size of schools varied. Small schools ranged from
twelve teachers to twenty-nine teachers and from 210 students
to 663 students. The junior high had thirty-six teachers and
950 students. There were thirty—eight teachers and 1,350
students in the senior high.

All principals held master's degrees in administration
except two: one MA in biology and one MA in English. ‘All
were participants in.Michigan State University's Extern Pro—
gram for school administrators.

Instrumentation: Teacher
Morale Form

 

Developed by Dr. John H. Suehr of Michigan State
University, the Teacher Morale Form was the instrument used
to measure group teacher morale. The TMF is an incomplete
sentence blank of the type developed by Rotter and
Rafferty.121

Suehr selected 100 items from the most often repeated

thoughts of teachers on other morale forms and tested them

 

121Julian B. Rotter and Janet E. Rafferty, Manual for
the Rotter Incomplete Sentence Blank (New York: The Psycho—
logical Corporation, 1950).

 

on seventy teachers.-

he selected the forty most important items.

schools.

12
necessary.

60

122

These he gave a "dry run" in the Denver elementary

Sixty percent were returned and no changes were

Suehr as:

1.

A study of the Chapter on Analysis Of Data gives
evidence of maladjustment among the low morale
teachers.

The teachers were motivated strongly in their
acceptance of the test.

Evidence of internal consistency is shown by the
interscorer reliability figures in Chapter III
(given on page 111 as .97).

Psychological attributes which might account for
variance in test scores are included in the Cor-
relation Form. Most all of these are adequate
indicators of causal factors affecting morale.
Low morale teachers did miss more school than
high-morale teachers. This is Often referred to
as validating morale indicators used in industry.

:An administrator familiar with all the schools

in the system was able to select schools in the
extreme areas of morale.

An item analysis of the Incomplete Sentence Form
demonstrated its aptitude in discriminating be-
tween teachers.

Item four of the ISF gives a picture of how well
teachers can tell if morale is high or low in a
specific building. The validity coefficient as a
result of ranking schools by the ISF and teacher
opinion by item four was .532 which is considered
by most experts as being "substantial."

 

p. 99.
124

122

See Teacher Morale Form on the next page.

23Suehr, Op. cit. (unpublished Doctoral dissertation),

I_b_i_d_°: p' 177°

From the sixty-seven usable responses,

Criteria for construct validity were given by

61

TEACHER MORALE FORM

Complete the following in order to make each, one complete

sentence. Take your time and show your true feelingp about
your situation. Make complete sentences. Try to do every

one. All responses will be confidential.

 

Teaching school

 

My salary

 

My future in teaching

 

Morale of teachers in this school
Faculty meetings
The people of this community

 

My working environment
Teachers

My principal
10. Teamwork among teachers
11. Children today
12. Parents

KOCDQOAU'er-OJNH

 

13. Fringe benefits in teaching
14. Channels of communication
15. Clerical help

 

16. The school board
17. Administrative policies
18. Professional organizations

 

19. Evaluation of my work
20. .American education
21. Opportunity to help make policy
22. The PTA

23. Teaching materials and supplies
24. Custodians

25. Professional standards

26. My pupils
27. This community
28. The superintendent
29. Discipline

 

30. Teaching assignments
31. Personnel policies
32. School administrators
33. This school

 

34. My class size

 

 

35. The in—service program
36. Teacher welfare
37. The curriculum

 

38. Teacher opinions
39. My teaching ability
40. Teaching again

Copyright 1961. All rights reserved.
College of Education
Michigan State University
John H. Suehr, Author

 

 

 

 

1 e.H Ne.+H H+H wemHonemoo
. . IIII.H . u.lllll u H H pH 0
. mm. N.H IAINIIOIJIM XXHN ”7.70.. .Hm
HmmVOH HHumzvz
me 1 ohm . H I III I H u Hmmqm1.- H u [Immmwm.u H I m
oo.me o.e .
H o.H a me.H m mH.m HH
mm o.m e Hm.H m mH.m OH
e o.~ m eo.m a mmo.m m
2 o O OH mo.m OH mm.m m
6 mm. m. m.m oe.H e mo.m s
mm. m. m.m oe.H m so.~ e
.H e.H m m~.H e em.H m
mm. m. m mH.H m.m em.H e
mm.om m.m m oe.H m.m em.H m
o o H mm. em.H H
e e Hooeom mHeu eH mHmHoz mamuH HH< mo meme:
m Honomoa "v EmpH mo cmoz an poxcmm mHoonom
wen we emxemm mHoonom
.Hoonom umcu CH mHMHOE Hmuou mnu SHHB ponmmeoo maoonom Hflwnu CH OHmHOE
Hmzomou may pmumu muocomwu Hao3 3oz m3onw coflvmaonwoo Hopuo xcmn mane .H.m manmfi

 

63

Fisher and Yates' table on page 470 of Walker and
Lev's Statistical Inference gives r = .52 significant at .95
for Product Moment Correlation.

Walker and Lev report: "How to deal with samples
larger than 8 and smaller than 25 is still a problem," on
page 282. They also say that rank order correlations must
be considerably higher than product moment. Since R = .62
is greater than r = .55, the student felt that R = .62 was
significant. The item analysis referred to in number seven
showed that only six items failed to discriminate at the one
percent level and only one of the six, "My Class Size"

failed to discriminate at the .05 level.125

Scoring the TMF

Instructions for scoring are given on the next page.
A panel of three teacher judges in each school scored the un-
signed morale forms. Anonymity was guaranteed to obtain as
true responses as possible. Each judge ranked each statement
as he evaluated it: 0-high1y positive; l-slightly positive;
2—neutral; 3-slight1y negative; and 4-high1y negative. The
mean of the judges' ranks was assigned to each item for that

school.. The same procedure was followed in all schools.

 

125John H. Suehr, “A Study of Morale in Education Uti—
lizing Incomplete Sentences," Journal of Educational Research,
Vol. LVI, No. 2 (October, 1962), p. 76.

 

64

INSTRUCTIONS FOR SCORING THE

TEACHER MORALE FORM

1. Read each statement over carefully.
2. Assign each statement a point value based on
the scale below.

0 - High positive statements--
denoting high morale

1 - Slightly positive statements
2 - Neutral statements
3 — Slightly negative statements

4 - Highly negative statements—-
denoting low morale

3. If a statement iS‘bOth positive and negative,
weigh the two, and decide if one is stronger
than the other. If one does not predominate,
score the statement as neutral.

4. Score blanks as neutral.126

 

126John Suehr, Op. cit., p. 76.

65

For example:

Table 3.2. Teacher judges' ranks Of teacher responses to

 

 

 

 

TMF.
Item Judge 1 Judge 2 Judge 3 Mean
1 0 0 1 .33
2 3 4 3 3.33
40 l 2 0 3.00

 

To test the reliability of such judgements, eighteen
items, a five percent sample of the responses from the TMF's
from various schools, were judged by two members of the
Michigan State University staff in the field of adminis-
tration. The product moment coefficient of correlation was
.80. When the "t" test was applied, "t" was significant at
.995. Computations are in Appendix A.

To test the agreement among all judges, the mean of
teacher judges scores for each item were ranked, the judge-
ments of judge X were ranked, and the determinations of
judge Y were ranked. That there was considerable agreement

among the ranks was shown by Kendall's coefficient of con-
127

 

cordance. The coefficient was .939; using the F test,
significance was .99. R, the correlation coefficient
127

Helen M. Walker and Joseph Lev, Statistical Inference
(New York: Henry Holt and Co., 1953), pp. 284—285.

66

determined from W, the coefficient of concordance, was .904,
also significant at .99. (See Appendix B for computations.)
Instrumentation: Teacher
Evaluation Forms
The evaluation forms served three purposes: teachers'
self-evaluations, student evaluations of teachers, and
principals' evaluations Of teachers. These forms were
adapted with the approval of the Student Reaction Center of
western Michigan University. The student evaluation of
teachers form is included on the next page. Instructions to
principals and other forms are Appendices C. The princi-
pal's evaluation of teachers form was identical except for
the heading and directions. Students and principals were
asked for their opinions regarding the ten items and teachers
were asked: How do you rate on the following:
1. Your knowledge of the subject taught.
(Have you a thorough knowledge and under-
standing Of your field?) 1 2 3 4 5
Wording the question in the second person was the adaptation.
The information which the Student Reaction Center of
Western Michigan University sends to teachers says:
if the responses of a chance-half of the stu—
dents in a class of 30 produces an average of 85
(3.5) on a given question, the average of the re-
sponses of the other 15 students in the same class
will usually be 85 or close to that number. The
fact that there is a difference of opinion within
chance-half groups does not mean that there is not
close agreement between the halves. One chance-

half group of fair size will usually contain about
the same number of dissenters from majority Opinion

67

as the other. The reliability coefficients for the
different scaled questions, using 50 teacE355—-One
class per teacher--range from .86 to .92.

 

STUDENT ESTIMATE OF TEACHING ABILITY

Please answer the following questions honestly and frankly.
DO not give your name. Neither your teacher nor anyone else
at your school will ever see your answers.

The person who is temporarily in charge of your class will,
during this period, collect all reports and seal them in an

envelOpe addressed to Michigan State University.

After completing this report, sit quietly or study until all
students have completed their reports. There should be no
talking.

Circle the appropriate number at the right of eadhof the ten
questions. 1 - Below average; 2- Average; 3 — Good; 4 - Very
good; 5 - The very best.

Teacher's Name School
WHAT IS YOUR OPINION CONCERNING:

l. The knowledge this teacher has of the sub-
ject taught? (Has he a thorough knowledge
and understanding Of his teaching field?) 1 2 3 4 5

2. The ability of this teacher to explain
clearly? (Are assignments and explan—
ations clear and definite?) 1 2 3 4 5

3.’ This teacher's fairness in dealing with
students? (Is he fair and impartial in
treatment of all students?) 1 2 3 4 5

4. The ability of this teacher to maintain good
discipline? (Does he keep good control of
the class without being harsh? Is he firm
but fair?) 1 2 3 4 5

 

——

12 . . . . .
8Western M1ch1qan Un1vers1ty Cont1nuous Study of

_§tudent—Reaction Reports (Kalamazoo, Michigan: Western
Michigan University, 1962).

68

5. The sympathetic understanding shown by this
teacher? (Is he patient, friendly, con-
siderate, and helpful?) 1 2 3 4 5

6. The ability this teacher has to make
classes lively and interesting? (Does he
show enthusiasm and a sense of humor? Does
he vary teaching procedures?) 1 2 3 4 5

7. The ability of this teacher to get things
done in an efficient and businesslike manner?
(Are plans well made? Is little time
wasted?) l 2 3 4 5

8. The skill this teacher has to get students
to think for themselves? (Are students'
ideas and Opinions worth something in this
class? DO they get at the real reasons why
certain things happen?) 1 2 3 4 5

9. The ability of this teacher to help students
know why they are learning certain things?
(Are there good reasons for studying the
topics in this course? DO you see the
value and importance of the things you
study?) 1 2 3 4 5

10. The General (All-Around) teaching ability
of this teacher? (All factors considered,

how close does this teacher come to your
ideal?) l 2 3 4 5

11. On the reverse side, name one or two things
you like about this teacher.

 

Since the items of this form are those generally con-
sidered by educational researchers to be important to the
evaluating of the effectiveness of teachers, they are con-
sidered to have face validity. Reliability coefficients for
this form when used in the WMU studies ranged from .86 to

.92.

69

Concerning the "halo" effect which might result from
the estimates of teacher effectiveness, Bryan reported that
it has "less influence on students' than administrators'
opinions of teachers" and that reactions of students in his
study were "reasonably free from the 'halo' effect."129
Whether there was the tendency of teachers, rating them—
selves, to put their best foot forward or be more self
critical is not known. One of the limitations of rating de-
vices is always the bias of the rater.

Reliability coefficients of student evaluations in
this study ranged from .40 to .95. These were determined by
choosing blindly one class from each school and comparing
chance-half groups. See Appendix D.

Principals distributed the evaluation forms for the
teacher self-evaluation and student evaluations in a faculty
meeting in their schools. Teachers filled out the self—
evaluation and selected one of their members to mail them to
the researcher. The principal evaluated each Of his teachers
and sent his entire set to the researcher. Each teacher
asked one average class of students to evaluate him, which
the students did while the teacher was absent from the room.
A responsible student or other teacher in charge Of the
class sealed all of the student evaluations of the teacher

into an envelope addressed to the researcher and took it to

 

lzgRoy c. Bryan, Reactions to Teachers by StudentS,

Parents, and Administrators (Kalamazoo: Western Michigan
University, 1963), pp. 18, 19, and 37.

 

 

70

the office for mailing. In this way, student comments could

not be identified by the teacher.

Scoring the Evaluation Forms

The ratings were simplified for the evaluators and
for scoring. All they were required to do was read the cate-
gory, such as: vKnowledge this teacher has of the subject,
and rate him: 1 — Below average, 2 — Average, 3 - Good,

4 — Very good, 5 — The very best, by circling one of the
five numbers at the right of the question to indicate his
opinion of that teacher on that particular item. ”Evaluators
rated teachers On ten items. .Teacher and principal evalu—
ations were single numbers circled for each item. Student
ratings used in this study were means of the judgements on

an item for the entire class, determined as follows:

Table 3.3. Manner of determining item mean scores of student

 

 

 

 

evaluations.
Item Ratings
1 . l 2 3 4 5
Number of
judgements 3 5 l6 3 1
3 10 48 12 5 = 78-28 = 2.789

 

Three students rated the teacher 1, five rated him 2, sixteen

rated him 3, three rated him 4 and one rated him 5. The

71

mean score was that teacher's rating on item 1. The same

procedure was followed for all items for all teachers.
geparincp for Computer Analysis

Coding the data was done in an arrangement to test
the hypotheses stated in the next section. For the morale
fOITn, forty mean scores of variables (items of morale) were
written down for each school. For the evaluations of
teachers, thirty items for each teacher were tabulated.
These constituted ten teacher self-evaluation scores, ten
student evaluations of teachers scores, and ten principals'
evaluations of teachers scores. Since the principal of
school 8 did not evaluate his teachers, these scores were

not available .
HYPOtheses

The study of high and low morale Of teachers and
high and low evaluations of teachers was approached with
four hypotheses. The first hypothesis was stated in null
form; teacher morale will not differ significantly from
school to school. The median test was applied to the items
Of morale and two schools compared at a time. The matrix,
Table 3.5, explains the comparisons. X's appear in the

cells of schools compared.

72

 

 

 

Table 3.4. Matrix Table showing schools compared on teacher
morale.
Schools by Number
School 4 5 6 7 8 9 10 ll
3 x x x x x x x x
‘4 x x x x x x x
5 x x x x x x
6 x x x x x
7 x x x x
8 x x x
9 x x
10 x

 

On the advice of a research consultant morale score
were converted to standard scores: mean of 50 and standard
deviation of 10. These were arranged in order from high
morale items to low morale items one column for each school
and the median test applied.*

The contingency table comparing the schools under
the null hypothesis: there will be no difference in the
teacher morale of schools,was as shown in Table 3.5.

The null hypothesis was rejected. The schools were
different. This procedure was followed for all schools.
-Any null hypOthesis was to be rejected at the .05 level of

confidence.

 

*
-Actually raw scores distinguished just as well as
standard scores.

S

(D

‘5’

(I)

'8

//

73

Table 3.5. Median test to distinguish high from low morale
schools.

 

 

 

School 3 School 6 Totals
Number of items above
the median 28 12 40
NUmber Of items at or
below the median 12 28 40
40 40 80

Chi—Square = 12.8
Chi-Square .995 with 1 df (degree of freedom = 10.8)

 

Similarly, the mean scores Of the evaluations of
teachers were arranged from high to low for each set of
evaluations: teacher-self, student, and principal, for each
school. Schools of high teacher morale were compared with
schools of low teacher morale under the second null hypothe—
sis: there will be no significant difference between evalu-
ations in high teacher morale schools and low teacher morale

schools. For example:

 

 

'Table 3.6. Comparison Of evaluations in high and low morale
schools.
High Morale Low Morale
School 3 School 6

 

bhnuber of Evaluations

 

above the median 8 6 14

thumber of Evaluations

.at or below the median 12 5 _i1_
20 11 31

Chi—Square = .56

Chi—Square .95 with 1 df = 3.8

 

74

The null hypothesis could not be rejected; there was
no significant difference between student evaluations of
teachers in high teacher morale school three and low teacher
morale school six. This same procedure was followed for stu—
dent evaluations of teachers, teachers' self-evaluations, and

principals' evaluations of teachers.

The Morale Factors. It was not originally hypothe—
sized that the morale of teachers was composed of more than
two factors or clusters defined as (l) the teacher's satis-
factions of personal needs and (2) satisfactions with the
tasks he performed related to goals of the organization
welded together by the administration. Under the hypothesis
that morale would have more than one factor, a factor analy—
sis was performed. This showed administrative policies and
practices to be a definite third cluster. Clusters were
identified for the study as: Factor 1: the teacher's satis-
faction that his personal needs were met; Factor 2: satis—
faction of the teacher with administrative policies and
practices, and Factor 3: satisfaction of the teacher with
the tasks he performed. Teacher feelings were thus cate-
gorized into three factors.

~Kerlinger says, "Factor analysis serves the cause of
scientific parsimony." It is expressed in two questions:
How many underlying variables or factors are there? What are

the factors? He adds:

 

b4

)-

‘5

T

75

The factors are presumed to be the underlying

unities behind test performances. They are re-

flected in correlation coefficients. If two or more

tests are substantially correlated, then the tests

share variance. They are measuring something in

common (communality).l3
He urgently points out that factor names are only tentative
hypotheses to be tested by further research. One of the re—
sults of factor analysis is a table of coefficients, called
a factor matrix, that expresses the relations between the
tests and the factors. Such coefficients are factor load-
ings.l3l Communalities are also discussed:

Communalities are the sums of squares of the factor

loadings. The communality 0; a test or variable is

its common factor variance.1 2
This is also expressed as the degree of association. For the
forty variables of the Teacher Morale Form responses only
one, children today, was .43. All others were above .56.

With communalities that high and factor loadings all

above .56 except professional standards .5003, the clusters
of items were considered to be a relatively true picture of
the factors of teacher morale in this study (See Chapter IV,
Table 4.6) .

To determine the number of factors, the factor load-

ings were rotated twice to obtain three factors which

 

130Fred N. Kerlinger, Foundations of Behavioral Re—
search (New York: Holt, Reinhart and Winston, Inc., 1964),
p. 650.

131Ibi

Q.

 

1321131

0..

 

 

 

76

appeared to be those already named. Rotating the factor
loadings a third time brought out only two items which might
be classed in another factor. These were teacher satis—
factions with teaching materials and custodians. Teaching
materials maintained a strong factor loading in Factor 2,
.56, while custodians came out distinctly .90. .All others
continued to maintain similar loadings to those of the
second rotation. The researcher was satisfied with three
factors.

Then the high, medium and low teacher morale of each
factor for each school was studied. This was done my multi-
plying the standard score of each item by the factor loading
for that item and finding the total Of all factor loaded
items for each school. Agreement among the ranks of the
factors was tested by the coefficient of concordance cor-
rected for continuity because N and m (number of factors)
were small. W1 was .559 significant at F.95. See Appendix
E for computations.

Chi-square was used to test the third null hypothesis
that there would be no difference among high, medium, and
low teacher morale factor values (Table 3.7) and high, medium,
and.1ow evaluations by items of evaluation. First, high,
xnedium,rand low teacher morale factor scores were divided
into thirds from high to low. Second, the evaluation scores
inere divided into high, medium, and low scores given in

Table 3.7.

 

 

77

Table 3.7. Numbers indicate each third of morale factors
and evaluations.

 

 

 

Morale Factors Low Third Middle Third High Third
1 635-612 611-538 537-513
2 572-499 498-414 413-345
3 546-444 593-401 400—378
Teacher and Principal
Evaluations 1-2 3'3 4'5
Student Evaluation 100-347 348-389 350-500

 

 

Table 3.8. Example of Chi-square test contingency table for
comparing factor 1 with teacher self—evaluation
on knowledge.

 

 

Teacher Self—Evaluation
Knowledge of subject

 

F-l Low Medium High
Low 5 12 50
Medium 1 17 42
High 2 13 46

II
p
p
N
m
Q.
Hi
II
p

Chi-square
Chi-square .75 = 5.4

 

 

The hypothesis of no differences cannot be rejected.
In.this case more teachers were evaluated high on knowledge
<5f subject matter in schools with low morale of teachers on
Factor 1.

This procedure was followed for all three factors of

'teacher morale and each item of teacher evaluations made by

78

teachers themselves, students and principals: thirty evalu-
ation variables and three factors. A total of ninety tests

were required.

Summary
In this chapter the sample, the measuring devices,
their validity and reliability, the hypotheses and the pro— ~1

cedures for testing them have been discussed. Since eighty
percent of the cities in Michigan are under 1,700 in pOpu-
lation, the schools in this study could be said to be almost
a representative sample. Seventy-eight percent or seven of
nine schools used in studying the relationship of teacher
morale to evaluation of teachers were schools in towns of
that size. However, no claim on representative sampling,

nor homogeneity of populations was made. The statistics used
were chosen with this limitation in mind.

The measures have reliability. The rank order corre—
lation of teacher satisfaction with the teacher morale in
their schools as related to the actual morale scores for the
schools was .62 with a reliability coefficient of .76°
Kendall's coefficient Of concordance showed agreement among
the judgements made of school morale, significant at F .999,
and among the three factors of morale, significant at F .95.
Evaluation of teachers by students' measurements showed sub-
stantial reliability ranging from .40 to .94. Both forms

Inet the standard criteria for validity.

79

Procedures for relating evaluations of teachers to
teacher morale were discussed. High and low teacher morale
schools were distinguished by applying the median test to
forty items of morale for each school. Comparing the number
Of high and low teacher evaluations by themselves, their stu-
dents and their principals, in the high and low teacher morale
schools by the median test, tested the null hypothesis that
there would be no difference in the number Of teachers evalu-
ated highly in high morale schools and low in low morale
schools.

It was shown here, too, that the relationship be-
tween items of morale would be studied by intercorrelations.
Three factors of teacher morale: 1. 'The satisfaction
teachers have of their personal needs; 2. Teacher satis-
factions with administrative policies and practices; and
3.’ Teacher satisfactions with the tasks they perform related
to the goals of the organization, were determined by factor
analysis.

Finally, it was pointed out that the null hypothesis:
there is no difference among high, medium and low evalu-
ations and high, medium, and low values of the teacher
Inorale factors would be tested by Chi-square.

Thus, the plan for analysis of the data is laid.

CHAPTER IV
ANALYSIS OF RESULTS

The hypotheses to be tested were:

Null Hypothesis 1: Teacher morale will not differ
significantly from school to school.

Null Hypothesis 2: Estimates of teacher effective-
ness in a school will not be significantly related to
teacher morale in that school.

HypOthesis 3: Teacher morale has more than one
factor.

Null Hypothesis 4: There will be no significant
difference between teacher morale factors and items of evalu—
ation of teachers.

Null hypotheses one and two were tested by the median test
and significance of Chi-square accepted at the .05 level of
confidence. Hypothesis three was tested by factor analysis.
Null hypothesis four was subjected to a Chi—square test with
factors and items of evaluation divided into high, medium,
and low categories.

The reader will recall that raw morale scores of
teachers were transformed to standard scores with a mean of

fifty and a standard deviation of ten. These standard

80

81

scores were arranged from high teacher morale scores to low
teacher morale scores for each school and the median test ap-
plied with acceptance of the Chi—square statistic at the .05
level of confidence. It is worth noting that raw morale
scores served the purpose just as well as standard scores

for this purpose in a previous trial run.

Hypothesis One

No. 1: Teacher morale does not differ from school to
school.

Table 4.1. Schools different from each other and their
level of significance.

 

Legend: NS = not significantly different

 

 

 

 

 

 

Y Yates correction applied
Significance was accepted at .05 level of confidence.
Schools Schools
6 7 8* 9 10 11
3 .001 .005 .001 .19 .001 .01
4 .005 .025 .001 .19 .005 .05Y
5 .005 .05Y .001 NS .01 .19
6 .19 .001 .05Y .19 .19
7 .001 NS NS NS
8 X .001 .001 .001
9 NS NS
10 NS
*

School 8's low teacher morale made it significantly
different from all other schools.

Schools 3, 4 and 5 were nearly alike and were signifi-
cantly different from all except 9.

 

82

Appendix F gives the computations.

Table 4.1 gives the pertinent data showing that the
null hypothesis was rejected. Conversely, schools do differ
in teacher morale. Schools three, four and five were high
teacher morale schools; their mean morale raw scores were:
1.86, 1.86 and 1.94 as judged by teacher judges. 'All other
mean scores Of teacher morale were above 2.00 on the scale 0
to 4 with 0 being positive and 4 negative.

Schools 6, 7, 9, 10 and 11 were nearly alike. Their
mean teacher morale scores were: 2.07, 2.08, 2.085, 2.14,
and 2.13, respectively. School 8 had a teacher morale score
of 2.58. Nothing this prevented one from making the judge—

ment that the nine schools above should be divided into high,

 

medium and low morale schools. Placed along a continuum
thus,

. l lLllll [J l

0 l 1.30 2 2.58’ ‘3 4

it appeared that all were medium teacher morale schools.
But the fact remained that schools 3, 4 and 5 had teachers
'whose morale as measured by responses to items of the
Teacher Morale Form were, statistically speaking, different
from the others, except for school 9.

Though school 9 had a teacher morale score of 2.085,
the reasons for its not being significantly different from
the other eight schools were interesting. Six of the 19

items on which the school was high were in Factor 1: the

 

83

teachers perceptions that his personal needs were being
satisfied. These were teacher satisfactions with their
future in teaching, faculty meetings, teacher evaluations,
professional standards, teaching assignments and school ad-
ministration. Teachers in this school were also well satis—
fied with their principal, the superintendent, the board of
education, the Opportunity to help make policy, and other
teachers, items in Factors two and three.

The feelings pulling down morale were in Factors 1
and 2. Those in Factor 1 were adverse feelings of satis-
faction with peOple of the community, the PTA, in—service edu—
cation and fringe benefits. These four indicated teachers
were dissatisfied with things affecting their individual
needs. The others fell in Factor 2: those items related to
administrative policies and practices. They were: my work—
ing environment, channels of communication, this school and
the curriculum. Ten of the nineteen items discussed in the
above two paragraphs were related to the personal needs
satisfaction for teachers; five were in the area of adminis—
trative policies and four in Factor 3; teacher satisfaction
with tasks they perform.

What was the other "maverick" school like? How did

it differ in teaching morale to the extreme?

84

Table 4.2. Factors producing low morale in school 8: the
extreme low morale school.

 

——

 

 

Teacher Dissatisfactions With:

 

Factor 1 Factor 2 Factor 3
Personal Needs Administrative Related to Tasks
Practices
My salary Morale of teachers School board
Faculty meetings Teamwork among Opportunity to
People of this teachers help make policy
community Channels of Com— .Superintendent
Parents munication Teacher Opinions
PTA My principal Discipline
Fringe benefits Administrative Pupils
In-service education policies
Teacher welfare Curriculum

 

Except for morale of teachers, teamwork among
teachers, their prinicpal and the superintendent, teachers
of other schools were also dissatisfied, as school 8 was.
However, school 8 was extreme. The point is that those four
items were related to leadership and it was very apparent
from teachers' comments about the principal that he was lack—

ing in that quality. Their comments ran: My principal is a

nice guy, . . . a good poker player, . . . fun to be with,
. lacking in leadership, . . . should have stayed in the
classroom, . . . is a poor disciplinarian, . . . is weak.

Later, a more detailed discussion will treat teacher
morale factors and the relationships Of items of morale to
each other and to evaluation in answering question three.

Hypothesis two is next.

 

85
Hypothesis Two

No. 2: Evaluations of teachers do not differ with re—
spect to high and low teacher morale.

For clarity's sake, the converse of this null hy-
pothesis was that evaluations of teachers would vary with re—
spect to morale of teachers. It was expected that: when
teacher morale was high, evaluations of teachers would also
be high, and in schools where teacher morale was low, evalu—
ations of teachers would also be low.

A quick look at Table 4.3 might lead one to believe
that the null hypothesis was false. In fact, it was true for
student evaluations of teachers and teacher self-evaluations.
In no high morale school was a larger prOportion of teachers
evaluated high by these two groups than in the schools of
low teacher morale. Two pairs of schools, 3 and 10 and 5
and 10, were actually reversed and the Chi—squares were sig-
nificant at the .05 level (computations are in Appendix G).
A larger prOportion of teachers were evaluated high in the
low teacher morale schools than in the high morale schools.

With the principals' evaluations Of teachers it was
a slightly different story. In the comparisons of high
teacher morale schools 3 and 5 with the low teacher morale
schools 6, 7, 9, 10 and 11 the same pattern prevailed: the
null hypothesis Of no difference had to be accepted. But,
the comparisons of principal's evaluations of teachers in

high teacher morale school 4 and the low teacher morale

86

Table 4.3. .A comparison of the mean scores of evaluations of
teachers by themselves, their students and their
principals and mean morale scores.

 

 

 

      

Evaluation Mean Morale
Ratings Scores
High High
4.0
3.9 1.60
3.8 1.70
3.7 1.80
3.6 1.90
3.5 f 2.00
. \ \. , .
3.4 i. \ //.-0 (.63500000 2°10
/' “'7" ‘2. .5“ ~ \ .‘
3 3 l. ‘ '. 0 06 l \ ' 2020
I v (' 0 a n \ .0
3.2 I ‘ I o 0: ‘ \ 2.30
O I o O ‘. '\
3.1 a." 0 £0 0 .: \\ 2.40
. o ,1 ,
3 o (3.," ' 2 50
o‘ '

2 9 2.60
Low 3 4 5 6 7 8 9 10 11 Low
Schools

.Iegend: Students' Evaluations

—--—Principals' Evaluations
....Teachers' Self-Evaluations
ooooMorale Scores

 

87

schools were significantly different in every case. Compu-
tations may be found in Appendix H, but Table 4.4 summarizes

the results.

Table 4.4. Principals' evaluations and their relation to
high and low teacher morale schools.

 

Level of Rejection of
Null Hypothesis of NO

 

Schools Chi-Square Difference
4 and 6 4.196 .05

4 and 7 7.5 .01

4 and 9 3.99 .05

4 and 10 7.30 .01

4 and 11 7.91 .005

Principal of school 8 returned no evaluations.

 

Principals' evaluations of teachers in schools of
high teacher morale were high when evaluations were low in
low morale schools.

One asks the question, why? It couldn't be at-
tributed to size. Schools 4, 5 and 6 were about the same
with 13, 12 and 11 teachers, respectively. Principals'
evaluations in schools 5 and 6 were not significantly differ—
ent. Schools 3 and 11 were the same size, 20 and 19 teachers
respectively and the evaluations were not significantly
different. No principal had reason to think his evaluations
‘would be reported to teachers. SO)the assumption that princi—

pals would be honest seemed valid. Did this principal in

 

In

(I)

88

school 4 know his teachers better? What kind of interpersonal
relations existed in this school? Responses to the TMF indi-
cate good interpersonal relations.

The means of morale form items indicate that teachers
were more satisfied with items related to administrative
policies and practices. Their scores were the best of any
school for the items: morale of teachers in this school, my
working environment, teachers, my principal, clerical help,
teaching materials and supplies, personnel policies, school
administrators, this school, teacher welfare and the cur—
riculum. Eight of the ten items were in morale Factor 2:
those practices for which the administration was responsible.

This study would agree withPryor's.133 School
morale is related to teachers perceptions of administrative
policies and practices.

However, this is not the place for conclusions, morale
factors and their relationship to evaluation must be dis-

cussed first.

Hypothesis Three

This portion Of the study required a study of morale
items and a hypothesis about the factors of morale. This
portion will be discussed by looking first at the means and
standard deviations for all items for all schools, and

second, by studying the factors of morale. Though the

 

133 .
Pryor. M~

89

factors of teacher morale have been mentioned several times,
the discussion of them has been reserved for the last of
this chapter. SO that the reader will be ready for the dis—
cussion of factors, Table 4.5 will be organized by factors.
The means are really the means of nine schools in
which varying numbers of teachers-—e1even to forty-six—-
responded to the morale form. Therefore, the standard
deviations are not as meaningful as they would have been if
all schools were the same size or each individual teacher's
scores had been available. On the other hand, Pryor found
that teachers' perceptions of administrative policies and
practices and their relationship to teacher morale were not
associated with the size of the school.134 Ross concluded
that rural and suburban teachers tend to react similarly,
with but minor variations, to personnel factors affecting

morale status.135 Dennerlein reported that size of school

played a small part.136

 

134 .
Pryor, Op. c1t.

135Walter E. Ross, "A Study of Personnel Factors Af—
fecting the Morale Status of Teachers of Two Rural School
Systems in New York and Including Comparisons and Findings
with those of a Similar Study Completed for a New Jersey
suburban School System" (unpublished Doctoral dissertation,
New York University, 1960).

136Gerald Dennerlein, "Factors Related to the Measure—
ment of Teacher Morale" (unpublished Doctoral dissertation,
University of Southern California, 1958).

 

90

Table 4.5. Means and standard deviations of forty items of
teacher morale arranged according to factors.

 

 

Standard
Item Mean Deviation

 

Factor 1: Teachers' Perceptions of Satisfactions of Personal
Needs

.65 .46
.66 .22
.40 .62
.22 .58
.59 .58
.74 .45
.32 .33

2. My salary

3. My future in teaching

5. Faculty meetings

6. PeOple of this community
12. Parents
13. Fringe benefits
18. Professional organizations

NNI—‘l-‘NNNNl-‘NNNNNl—‘N
\O
0‘

19. Evaluation of my work .23
20. American education .22 .29
22. The PTA .94 .63
23. Professional standards .06 .34
27. This community .16 .53
30. Teaching assignments .96 .53
32. School administrators .85 .53
35. In-service education .76 .49
36. Teacher welfare .28 .43

NOte: All but 4 Of these mean scores indicate contributions
to poor morale.

*In-service education had the same factor loading for

Factor 2. Arbitrarily it was included in Factor 1.

Factor 2: Teachers' Satisfactions with Administrative
Policies and Practices '

4. Morale of teachers in this school 1.57 .60

7. My work environment 1.40 .53

9. 'My principal 1.35 .63
10. Teamwork among teachers 1.66 .47
14. Channels of communication 2.46 .54
15. Clerical help 1.82 .43
17. Administrative policies 2.24 .42
23. Teaching materials 1.86 .58
31. Personnel policies 2.09 .35
33. This school 1.84 .62
34. My class size 1.81 .28
37. The curriculum 2.46 .46

Note: Only four of these contribute to poor morale.

rn

. .
.J‘

di

91

Table 4.5. (continued)

_ w

 

 

Standard
Item Mean Deviation

 

Factor 3: Teachers' Satisfactions with Tasks Performed to
Attain the Goals of the Organization

 

1. Teaching school 1.03 .35
8. Teachers 1.78 .42
11. Children today 2.13 .39 .13
16. The school board 2.13 .63 '
21. Opportunity to help make policy 2.27 .55
24. Custodians 1.48 .61
26. My pupils 1.85 .37
28. The superintendent 1.94 .61
29. Discipline 1.85 .52
38. Teacher Opinions 2.04 .53
39. My teaching ability 1.62 .24
40. Teaching again 1.59 .28

Note: Only four of these contribute to poor morale.

 

Was the number of men and women teachers in the
schools associated with high and low morale different? It
does not appear so. The percentage of men in high teacher
morale schools was about the same as that of the low morale
schools.

Table 4.6 gives the percentages of men and women in
high and low teacher morale schools.

Noted here but discussed later is the point that
dissatisfactions lie primarily in Factor 1: satisfactions
of personal needs.

Teacher morale appeared to be highest in satisfaction
'with teaching schoOl. Most dissatisfaction was found with the

PTA.

92

 

 

Table 4.6. Percentage of men and women in high and low morale
schools.
Men Percentage WOmen Percentage
High Morale
3 14 70 6 30
4 5 38 8 62
5 7 58 5 42
Average percent 55.3 47.6
Low Morale
6 7 64 4 36
7 17 58 12 42
8 27 71 ll 29
9 19 65 9 35
10 8 44 10 56
ll 6 31 13 69
Average percent 55.5 44.5

 

Areas of satisfaction and dissatisfaction were clear.

Teachers were satisfied with tasks and task oriented things,

namely:

Teaching school

My future in teaching
My teaching ability
Teaching again
Evaluation of my work
Teaching assignments
My work environment
Teaching materials

Though teachers were satisfied with their pupils,

generally dissatisfied with children today.

My class size

This school

Teachers

Teamwork among teachers
Discipline

Custodians

Clerical help

Their pupils

they were

Another contrast

presented itself in that teachers felt good about school ad-

ministrators,

felt adversely toward channels of communication,

opinions, personnel policies,

school board,

administrative policies,

and the opportunity to help make policy.

their principals and the superintendent, but

teacher
the

They

 

93

were discontented with salary, fringe benefits and teacher
welfare. Faculty meetings, in—service education, profession-
al organizations and standards, American education and the
curriculum felt the lash. Parents, the community, and the
PTA were criticized.

Ross indicated that rural teachers of low morale
tend to consider the community cultural level to be inferior
to their own.137 Only two schools of nine expressed satis-
faction with the community and none felt good about parents
and the PTA. Seven of the nine were rural in nature.

In brief summary five points stood out: 1. Teachers
were dissatisfied with the community and the PTA. 2. They
were dissatisfied with policies. 3. They showed the most
dissatisfaction in that their personal needs were not being
met. 4. For the most part, they were satisfied with ad—
ministrators. 5. Teachers were satisfied with their tasks

and task oriented things.

Teacher Morale Factors

H 3: Teacher morale is composed of several factors; it
is not a single entity.

 

137
Ross, Op. cit.

94

By Lonsdale's definition, morale was expected to
have at least two factors:

(a) perceived job satisfaction or the satisfaction of
individual needs through the interaction of the
participant in his role within the work group and
the total organization.

(b) perceived productivity or progress toward the
achievement of the tasks of the organization.
Further, high morale is the participant's per-
ception of a successful task—needs integpation.
Since task-needs integration is the ultimate
purpose of administration, it follows that high
morale is the participant's perception of the
consumation of administrative purposes.

As was seen in Table 4.5 the first two factors could be
labled as (a) and (b) Of the definition, and the third
factor turned out to be administrative policies and practices.
To use Lonsdale's terms, it was the teachers' perceptions of
the "task-needs integration" provided by the administration.
Turning to Table 4.5 the reader might raise a series
Of questions. 1. Why did teacher satisfactions place the
school board, the superintendent and Opportunity to help
make policy in Factor 3, task oriented satisfactions, rather
than in Factor 2, administrative policies and practices? It
must be remembered that teacher perceptions were being
measured. The schools were small and in most instances the
superintendent and board of education were probably close to
the base of operations. Their interaction with teachers was

probably good and teachers felt that their tasks depended on

relations with the superintendent and board.

 

138Lonsdale, op. cit.

95

Why did the item, teacher satisfactions with school
administrators, place it in Factor 1, perceptions of satis-
faction of the individual needs of teachers? It is a rather
well accepted principle that the purpose of school adminis—
trators is to serve the needs of teachers. Why did satis—
factions with working environment, morale Of teachers and
teamwork among teachers appear in Factor 2, administrative
policies and practices? These are things perceived by
teachers to be responsibilities of the administration.

But, this is rationalization and conjecture, no matter
how much the Operational definition supports the logic. In
another set of schools the picture could be different.

At this point, one might ask: Was there a signifi—
cant relation within the factors? The answer is that this is
the basis Of factor analysis: the variables share variance.
However, there were three notable exceptions. In nine Of
fifteen correlations in Factor 1, the item, teaching assign-
ments was not significant at the 95th percentile. Four of
eleven times satisfactions with teaching materials in Factor
2 failed to be significant. Attitudes toward custodians
were negatively correlated ten of eleven times in Factor 3.

Did these items belong with these factors? In
Chapter III it was pointed out that the third rotation of
:fiactor loadings kept all but satisfactions with custodians

twithin the original factors.

2.

96

Another consideration, it will be remembered from
Chapter III, was whether the factors contributed equally to
the total morale. The coefficient of concordance, Appendix
C, showed Wl significant by the F test at .95. The question
was answered affirmatively.

Earlier some unsatisfactory rationalizations were made
for maintaining variables within certain factors. Some de-
gree of association was found. A look at factor loadings
above .40 indicates common factor variance. Table 4.7 shows
the shared variance. Rarely did it occur in more than two
factors.

Referring to Table 4.7 again, some interrelations ap-
peared plausible. For instance, My future in teaching, has
an important factor loading in Factor 3, items related to
tasks as well as the satisfaction of personal needs, Factor
1. Other examples can be more readily examined in the chart
than explained here in words.

Parsimony of teacher morale into factors made it
possible to examine the factors and their relationship to
items of evaluations of teachers. The null hypothesis was:

Ho 4: There will be no difference between high, medium,
and low divisions of each of the teacher morale
factors and high, medium and low designations of
each of the items of evaluations by teachers them—
selves, by students and by principals.

Tables 4.8, 4.9 and 4.10 record the Chi-squares and

the acceptance or rejection of the null hypothesis at the .05

level of confidence.

97

 

 

 

Table 4.7. Rotated factor loadings of morale items.*

Morale Item F2 F2 F3
1. Teaching school —.8377
2. My salary .8122
3. My future in teaching .5832 —.4689
4. Morale of teachers —.7302 —.6136
5. Faculty meetings .7140
6. People of the community .5636
7. My work environment -.9425
8. Teachers —.6468
9. My principal -.6l45 -.4510

10. Teamwork among teachers —.6724 -.4693

11. Children today -.6499

12. Parents .8672

13. Fringe benefits .6977 —.4181

14. Channels of communication -.7531

15. Clerical help -.7493

16. The school board -.7999

17. Administrative policy -.5863 —.5462

18. Professional organizations .7943

19. Evaluation of my work .7579

20. American education .6359

21. Opportunity to help make

policy .5205 -.7675

22. The PTA .8725

23. Teaching materials .5068 —.7183

24. Custodiansf* .4567 -.5744

25. Professional standards .5003 -.4958

26. My pupils .5010 -.6986

27. This community .7309 -.5728

28. The superintendent —.7781

29. Discipline —.4991 —.6376

30. Teaching assignments .7416

31. Personnel policies .5133 —.7732

32. School administrators .6882 -.4202 -.5458

33. This school —.9406

34. My class size .5998 —.6303

35. In-service education .5771 —.5771

36. Teacher welfare .6230 —.5043 -.4683

37. The curriculum .5899 -.6717

38. Teacher opinions, —.6688

39. My teaching ability .5963

40. Teaching again - 5491 -.6207

:Factor loadings under .40 are not listed.
In the next rotation custodians appeared a fourth

factor: .9043.

Note: PrOportions of variance: F1: .2759; F2: .2433;

F3: .2249."
variance.

These account fOr 17441 percent of the

 

Table 4.8.

98

Chi-square test of independence between teacher

morale Factor 1, teachers' perceived satis—

factions of individual needs,
ations by teachers themselves,

and their principals.
morale is compared with.high, medium and low

evaluations.

and items of evalu—
their students
High, medium and low

There are four degrees of freedom.

 

 

TSE SE PE
Evaluation Item Chi—Square Chi—Square Chi-Square
1. Knowledge 4.425 2.029 11.372
HO rejected NO No .025
2. Ability to explain 1.368 .934 3.795
H0 rejected NO NO No
3. Fairness 1.826 3.214 2.968
HO rejected No No No
4. Discipline 2.760 3.506 2.993
H0 rejected No No No
5. Sympathy and Under—
standing 5.530 4.246 2.806
Ho rejected No No No
6. Interestingness 5.835 2.932 7.186
Ho rejected No No No
7. Efficiency 3.400 4.358 4.607
Ho rejected NO No No
8. Gets students to think 3.862 3.055 3.202
Ho rejected No NO No
9. Helps students to know
why 9.622 2.371 7.426
Ho rejected .05 No No
10. General ability 2.029 3.331 9.049
Ho. rejected .No No (9.5:.05) No
Legend: TSE - Teachers' Self-Evaluations
SE - Students' Evaluations of Teachers
PE — Principals' Evaluations of Teachers
Ho - Null hypothesis level of rejection = .05

Table 4.9 .

99

Chi-square test of independence between teacher
morale Factor 2, teachers' perceived satisfactions
with administrative policies and practices, and
items of evaluation by teachers themselves, their
students, and their principals. High, medium and
low morale is compared with high, medium and low

 

 

 

 

 

evaluations. There are four degrees of freedom.
m =—
TSE SE PE
Evaluation Item Chi-Square Chi-Square Chi—Square
1. Knowledge 2.062 2.111 8.720
Ho rejected No No No
2. Ability to explain 6.387 6.847 4.431
H0 rejected No No No
3. Fairness 1.610 7.967 1.651
Ho rejected NO NO No
4. Discipline 3.187 5.975 8.503
HO rejected NO NO No*
5. Sympathy and under-
standing .999 6.967 4.915
HO rejected No NO No
6. Interestingness 8.169 2.223 6.794
Ho rejected No* No No
7. Efficiency 4.100 3.686 7.413
H0 rejected No NO No
8. Gets students to think 2.313 5.961 3.072
' Ho rejected No No No
9. Helps students to know
why 8.049* 4.222 4.569
HO rejected No No No
10. General ability 1.623 13.298 6.161
HO rejected NO .01 No
*Chi-square 7.8 significant at level .10
Chi—square 9.5 significant at level .05
Legend: TSE - Teachers' Self—Evaluations
SE - Students' Evaluations of Teachers
PE - Principals' evaluations of Teachers
Ho — Null hypothesis level of rejection = .05

Table 4.10.

morale Factor 3,

100

factions with tasks,

by themselves

Chi-square test of independence between teacher
teachers' perCeived satis-
and evaluations of teachers
, their students and their princi-

 

 

 

 

 

pals. High, medium and low morale is compared
with high, medium and low evaluations. There
are four degrees of freedom.
T SE SE PE
Evaluation Item Chi-Square Chi-Square Chi-Square
1. Knowledge .705 1.763 13.539
HO rejected No NO .01
2. Ability to explain 3.102 5.293 13.555
Ho rejected No No .01
3. Fairness 1.380 3.313 5.614
HO rejected No NO No
4. Discipline 2.646 4.680 5.092
H0 rejected No NO No
5. Sympathy and under-
standing 7.781 4.581 3.730
HO rejected No No No
6. Interestingness .603 4.990 8.861
H0 rejected No No No
7. Efficiency 2.544 2.911 16.846
HO rejected No No .005
8. Gets students to think 3.915 10.483 22.681
Ho rejected No .05 .001
9. Helps students to know
why 2.136 15.700 12.491
Ho rejected NO .005 .02
10. General ability 1.949 3.014 12.323
Ho rejected NO No .02
Legend: TSE - Teachers' Self-Evaluation
SE - Students' Evaluations of Teachers
PE - Principals' Evaluations of Teacher
Ho - Null hypothesis level of rejection = .05

101

.Accepting the null hypothesis means that high teacher
perceived satisfactions with individual needs was not general—
ly related to high evaluations of teachers. Only twice was
a Chi-square significant: principals' perceptions Of
teachers' knowledge of subject matter was significantly re—
lated to teachers satisfactions Of their personal needs, and
teachers' self-evaluations of helping students to know why
they were learning certain things was also significantly re-
lated to teacher morale Factor 1. Principals' evaluation of
general teaching ability and teacher morale Factor 1 was sig-
nificant at .075, not enough to reject the null hypothesis.

About the same picture was presented by Factor 2:
perceived teacher satisfactions with administrative policies
and practices. Relationships between teacher morale Factor
2 and the evaluations were almost nil. The relation between
students' evaluations of teachers general ability and Factor
2 was significant at the .01 level. Deserving mention were:
principals' evaluations of knowledge — .07 level and of
discipline — .08 level; and teacher self evaluations of
interestingness — .09 level, and helping students to know
why they were learning - .10 level.

In eight of the thirty tests of the null hypothesis
that there were no differences between Factor 3, teachers'
perceived satisfactions with the tasks they perform and
evaluations of teachers, it was rejected. A significant re-

lationship existed between student judgements of teachers'

 

102

ability to get students to think, and to know why they were
learning with teacher mora‘Ie Factor3 at the .05 and .005
levels, respectively. Most agreement of any of the factors
was found with Factor 3 and principals' evaluations. When
teachers felt good or bad about the tasks they performed,
principals evaluated them correspondingly on six of ten
items. A seventh almost reached significance. See Table

4.11.

Table 4.11. Significance Of Factor 3 and principals'

 

evaluations.
Knowledge, .01 level
Ability to explain, .01 level
Efficiency, .005 level
Getting students to think, .001 level
Helping students know why, .02 level
General ability, and .02 level
Interestingness, .07 level

not enough to reject the null hypothesis.

The null hypothesis at the .05 level of confidence was
rejected.

 

Absent from this list were: fairness, discipline,
and sympathy and understanding. When students were asked to
name two things they liked best about their teachers during
the evaluation period, sympathy and understanding ranked
first, fairness ranked fifth and discipline twelfth of
thirteen important qualities mentioned. There was a dis—
crepancy among the perceptions of students and principals
and teachers about what makes a good teacher. But, more of

this later.

 

103

The graph of Table 4.12 shows the manner in which
teachers rated themselves, students and principals rated
teachers and all ten items;_ When these means were ranked,
they were not highly correlated. (See Appendix I for

computations.)

Table 4.12. Comparison of teachers', students', and princi-
pals evaluations by mean scores on all items.

 

 

_ __ -_—

 

 

 

 

 

401 1
400 U

l
3.9 1
3.8 " ,

\. " u ’ , Students'
3'7...¥§§9.?§.\§9§?@F§..}1?T.M???§w.., ..... Evaluation
3 6 ‘npA‘ ,1 ,oa., 1~‘fl' Of Teachers

0 \ /'*" O o o o o o
/ \( / ‘

3.5 Means of achers) Ite Means Teachers'
§sl§:§yslsssisee

3. 7 Principals'
Evaluations

3 of Teachers'

3.

 

K E F D SU I EF T W GA

 

Knowledge

Ability to explain

Fairness

Discipline

Sympathy and understanding
Interestingness

<Efficiency‘

Getting students to think
Helping students to know why
General ability

OSHPJHUJU'UDJX
?I 9 II II

?

 

104

Table 4.13. Correlations between ranked mean scores of
evaluations of teachers by themselves, students
and principals.

 

Teachers and Students R = .28
Teachers and Principals R = .45
Students and Principals R = .09

None are significant

 

Student Comments on Evaluations

Student comments about the qualities they like in
their teachers were mentioned in anticipation Of this dis-
cussion. They do have significance from the standpoint of
evaluations of teachers. More than 4,000 students were
asked to evaluate 188 teachers. At the end of each evalu—
ation each student was asked to write two qualities he liked
about his teacher. Though negative comments weren't
solicited, some were written. No negative comments were re-
corded because they were opposites Of the good qualities.

In summary form Table 4.14 presents the results.

Among the students in grades seven through twelve
the order of importance of the qualities was about the same.
Students from suburban junior and senior high schools held
the same Opinions as students from rural areas.

Judging from the comments and the appearance of the
graphs of Table 4.12, there is considerable disagreement

about the qualities of a good teacher.

105

Table 4.14. Qualities students liked about teachers.

L J
l

 

 

 

 

 

 

Qualities Number of Times
Mentioned

1. Sympathy and understanding 668
2. Sense of humor 642
3. Good personality 540
4. Interestingness 538
5. Fairness 517
6. Is helpful and kind 513
7. Explains well 504 ,
8. Knows the subject well 431 5
9. Good teacher 308
10. Good methods 271
ll. Friendly (not generally related to

classes) 248
12 Is efficient and businesslike 215
13. vHas good discipline 210
14. Keeps temper-doesn't speak harshly 80
15. Has patience 61

Miscellaneous: has enthusiasm, respects students, is
well dressed, has a good voice, and is
honest and frank.

 

Summary of the Findings

,At the outset three questions were asked: (1) Could
high teacher morale schools be distinguished from low
teacher morale schools? (2) WOuld evaluations of teachers
be high when teacher morale was high and low when teacher
morale was low? and (3) What items and factors of teacher
morale would be strongly related to other aspects of teacher
morale and items of evaluations? The null hypotheses for
these questions were tested by the Chi-square statistic with
rejection of the null hypothesis at the "confidence ”level

of .05. The median test was used.

106

The null hypothesis that there would be no difference
in teacher morale of the various schools was rejected. Three
schools were significantly different from five others.
Neither size of school nor community seemed to have much ef-
fect on teacher morale. It was decided that because the
schools were significantly different the results would be
valid.

Null hypothesis two was: evaluations of teachers do
not differ with respect to high and low teacher morale. For
the most part this hypothesis had to be accepted. Student
evaluations of teachers, and teachers self-evaluations did
not vary with teacher morale. In the case of principals'
evaluations of teachers only one high teacher morale school
had a larger percentage of high evaluations when the low
morale schools had high percentages of low evaluations. The
median test was used.

A factor analysis of teacher morale showed three im—
portant factors of morale as measured by the Teacher Morale
Form. Factor 1 was designated: teachers' perceptions of
satisfactions of personal needs. Factor 2 was labeled:
teachers' satisfactions with administrative policies and
practices. Factor 3 was titled: teachers' satisfactions
‘with tasks performed to attain the goals Of the organization.

Dissatisfactions predominated in Factor 1, but Factors
2 and 3 tended toward high morale. Teachers were satisfied

‘with administrators and teaching but not with administrative

 

 

107

policies, personnel policies, and the opportunity to help
make policy. Channels of communication were criticized.
Dissatisfaction with items: the PTA, in-service education,
fringe benefits, salary, parents, faculty meetings, pro-
fessional organizations, people of this community, this com-
munity, American education, and professional standards
brought teacher morale scores down in Factor 1.

Except for dissatisfactions with channels of communi—
cation, the curriculum,eadministrative policies and personnel
policies in Factor 2, feelings of satisfactions about the
task-needs function of the administration produced high
morale. Factor 3 was similar. Except for feelings of dis-
satisfaction with the school board, opportunity to help make
policy, and teacher Opinions, satisfaction with teaching
tasks resulted in high morale.

When the null hypothesis: there will be no signifi-
cant difference between evaluations was tested, little agree-
ment was found. Most significant relationships existed be-
tween principals' evaluations and teacher morale. When
teacher morale was high, medium or low on Factor 3, teacher
satisfactions with tasks, principals rated teachers corre-
spondingly high, medium, and low on knowledge of the subject,
ability to explain, efficiency, ability to get students to
think, ability to help students know why they were learning
and general teaching ability at .02 level of confidence or

better. Also, on Factor 3, students correspondingly rated

 

108

teachers high, medium or low on ability to get students to
think and ability to help students know why they were learn-
ing, at levels of .confidencer .05 and .005 respectively.

Three other relationships were important. Teachers'
self-evaluations corresponded to teacher morale in Factor 1:
Satisfaction of personal needs, with the item helping students
to know why, at the .05 level. ~Again in relation to Factor
1, principals evaluations of teachers' knowledge showed a
significant relationship at the .025 level. Factor 2,
teacher satisfactions with administrative policies and
practices, were high, medium and low when student evaluations
of general ability were high in high teacher morale schools
and low in low teacher morale schools.

Out of ninety possible relationships tested, ten
were significant.

In the final analysis it appeared from student com—
ments that students may have a different concept of the "good"

teacher than either principals or teachers.

CHAPTER V

SUMMARY, CONCLUSIONS, AND IMPLICATIONS

Summary

In the beginning, it was hypothesized that teacher
effectiveness as evaluated by teachers themselves, their stu-
dents and their principals would be related to teacher
morale. The findings showed principals' estimates of teacher
effectiveness in the areas of teachers' knowledge Of sub-
ject matter, ability to explain, efficiency, ability to get
students to think for themselves, ability to help students
knowfwhy they were learning certain things and general teach-
ing ability were significantly related to teachers' satis-
factions with their tasks. Teachers' self evaluations and
student evaluations of teachers were not generally related

to the factors of teacher morale.

Conclusions

Within the limitations of this study the following
conclusions are made:
1. Teacher morale does differ significantly among
schools. The difference lies in the way teachers

feel their personal needs are being met.

109

 

110

Dissatisfactions of teaChers with faculty meetings,

 

the curriculum, and the opportunity to help make
policy indicate the teachers' need for changes, and
the blindness of administrators to capitalize on
this need to improve the conditions of learning.

The function of administration, as it relates to com-
munication with teachers, to communication between
teachers and the community, to policy making, and to
the satisfaction of teachers' personal needs is the
most important ingredient of high teacher morale.
This is described as the task-needs integration
function.

Students have a different concept of the "good"
teacher from that of either principals or teachers.
While students rank sympathy and understanding, a
sense of honor, fairness and interestingness highest
on their list of qualities of a "good" teacher,
principals and teachers tend to emphasize techniques
and methods.

Principals by reinforcing the behavior of teachers
which emphasizes knowledge, ability to explain, ef—
ficiency, ability to get students to think for them-
selves and to know why they are learning certain
things tend to perpetuate the "system" and to con—
tinue making a students' needs secondary to tech-

niques of teachers.

111

6. There is no satisfactory evaluating device to fit
all purposes. The California teachers reported in
Teacher Competence: *Ips Nature and Its Scope, that
teachers wanted to participate in evaluation. This
study showed teachers to be dissatisfied with evalu—
ations, except for one high teacher morale school.
Since students have a different concept of the "good"
teacher than principals have, teachers should seek

evaluations by students and bring this information

 

to principals and supervisors for joint evaluation.
Administrators would have to set the climate for
such action.

7. The factor analysis showed considerable inter—
relatedness of the teacher morale factors. It is
not known how other factors affect the task oriented
morale items. To study the effects of "classroom"
teacher morale and its effects on students, a new
form should be developed specifically for this

purpose.

Implications

1. Teachers and principals who ignore the importance of
the conditions of learning as perceived by students
are contributing to the disintegration of the learn-
ing environment. Teachers faced with the dilemma of

the necessity to behave in ways to be evaluated

112

highly by the principal and giving attention to the
needs of students are likely to choose the former be-
havior. Principals and supervisors should reevaluate
their administrative roles and reinforce that kind

of teacher behavior which is more consistent with

the current knowledge of the science of human
behavior. I _
General dissatisfaction of teachers with faculty

meetings, in-service education, American education,

 

curriculum, policies, and the Opportunity to help
make policy indicates that teachers feel the need
for changes. Sensitive administrative leadership is
needed to listen to what teachers are saying and use
their dissatisfactions as motivation for teachers to
participate in the processes of improvement.

Another related aspect is suspected at this point,
and it is noticed in teachers' dissatisfactions with
the community and the peOple of the community, items
related to satisfaction of the personal needs Of
teachers. During the past century the secularization
of cities has taken place at a very rapid pace.
Teachers may be aware of the passing of the double
standard of the American way of life. Today there
is a rejection of the old standards in which the

Puritan ethic held sway in favor of a freedom to

make one's own decisions. This is ineVitable in an

113

educated society. Society is disrupted by this
cataclysm of change and middle class teachers are
especially uncomfortable in it. A slogan like "free-
dom to teach," the emphasis on dealing with contro—
versial issues, and the inability Of many middle
class teachers to understand disadvantaged children,
are examples of the discomfort teachers are under-
going in this process of change. Colleges of edu-
cation and administrators in the schools have an im—
portant responsibility to help future teachers and
teachers become free and responsible, if they are
going to foster this kind of behavior in youth.
General dissatisfaction with the community, parents,
and the PTA may indicate that teachers in these com-
munities feel superior to the rest of the community.
The question, "Why?" is important. Is this feeling
a kind of compensating satisfaction for the feeling
of dissatisfaction with salaries and fringe benefits
and the cultural level of the community? Or is this
an indication that the communication lines are down
between school and community? In either case, satis-
faction Of these needs Of teachers to be more a part
of the community is important to high teacher morale.
If poor morale is a quality of the personality of

the teacher, then this compounds the job of the ad-

ministrators who hOpe to effect change. It implies

114

that teacher personalities provoking poor morale
must be worked with by administrators competent in
psychology and human understanding. The human re-
lations aspect reaches paramount proportions when
changes in people and interpersonnel relations are
sought.

6. For the kind of changes required, a democratic ad—
ministration is a necessity. PeOple become more im—
portant than routine directives. They take on added
importance in the functioning of the organization:
There is a greater chance for free and open dis-
cussion; communication flows in two directions; de—
mocracy is taken Off a verbal level and put to work.
This is the atmosphere in which teachers must work
with colleagues if they are to practice open, free,
and responsible behavior in the classroom. Con-
ditions Of learning for students will be improved by
such practices.

That high teacher morale is important to the satis-
factory functioning Of the school organization is undeniable.
It deserves further research, especially with designs having
control of as many variables as possible.

Perhaps the most significant part of this research
was the identification of the three factors of teacher morale

and the suggestion that there might be such a thing as a

 

115

teacher's classroom morale, which directly affects the con-

ditions of learning.
Specific Recommendations

1. The factor analysis, using the TMF, should be
repeated with another set of schools at both the elementary
and secondary levels.

2. Search for good evaluative measures must go on.
Most promise may lie in guides for teachers and principals
to evaluate the teachers work together. Certainly in light
of good human relations and tenure laws this procedure has
relevance.

3. There is more work to be done in the public
schools on student morale and student achievement and the
part that teacher morale plays in the learning environment.

4. Morale studies should be used by students,
teachers, and principals to explore their feelings about
their jobs, school and community. Such studies would pro—
vide motivation for studying the kinds of behavior that pro-
duce attitudes which provide the most effective teaching

and learning conditions.
_Questions for Further Study

1. Are there more than three factors of teacher

morale? What are the factors? To what degree are the

116

factors independent? Is there a kind of teacher classroom
morale that is directly related to student behavior and
achievement?

2. Do students of teachers who emphasize the im-
portance of sympathy and understanding in their teaching
learn more than students of teachers who emphasize
techniques?

3. In how many schools do teachers participate in
evaluative procedures? What is the effect on teacher morale?

4. To what extent do teacher expectations determine
the behavior of principals?

5. How does the community's concept of the "good"
teacher affect teachers' morale and their classroom behavior?

6. What is the effect of principals' evaluations
upon teacher morale and teacher effectiveness?

7. What effect does the social structure of a
group of teachers have on morale and classroom behavior of

teachers?

BIBLIOGRAPHY

BIBLIOGRAPHY

Anastasi, Anne. Principles of Psycholggical Testing. New
York: (MacMillan Company, 1954.

Andrews, John H. M. and Brown, Alan F. "Can Principals Ex-
clude Their Own Personality Characteristics When
They Rate Their Teachers?" Educational Adminis-
tration and Supervision, Vol. LIV (July, 1959).

Barnes, Fred P. Research for the Practitioner in Education.
Washington, D.C.: Department Of Elementary Princi-
pals, National Education Association, 1964.

Barr, A. S. "The Assessment of the Teacher's Personality,"
School Review, Vol. LXVIII (Winter, 1960).

Barr, A. S. and Emans, Lester M. "What Qualities are Pre—
requisite To Success in Teaching?" Nations Schools,
Vol, VI, No. 3 (September, 1930).

Bently, Ralph R. and Rempl, Arno M. "Peer Selection vs. Ex—
pert Judgement as a Means of Validating a Teacher
Morale Measuring Instrument," Journal of Exppri-
mental Education, Vol. XXXI, NO. 3 (March, 1963).

Best, John W. Research in Education. Englewood Cliffs:
Prentice Hall, Inc., 1959.

Blocker, Clyde E. and Richardson, Richard C. "Twenty—five
Years of Morale Research: A Critical Review,"
Journal of Educational Sociology, Vol. XXXVI, No. 5
(January, 1963).

Boris, Blai J. "A Job Satisfaction Predictor," Personnel
Journal, Vol. XLII, NO. 9 (October, 1963).

Brookover, Wilbur B. A Sociology of Education. New York:
American Book Company, 1955.

Brown, Frank B. "Education in the Age of Space," The North

Central Association Quarterl , Vol. XXXIII, No. 3
(Winter, 1964).

118

 

119

Bruce, William F. and Holden, John. The Teacher's Personal
Develgpment. New York: Henry Holt and Co., 1957.

 

 

Bryan, Roy C. Pupil Ratings of Secondary School Teachers.
New York: Teachers College, Columbia University,
1937.

Cohen, Gloria J. "A Study of Socio—Economic Status of the
School Community and the Morale of Teaching Personnel
in New York City:" Unpublished Doctoral dissertation,
New York University, 1959.

Dennerlein, Gerald. "Factors Related to the Measurement of
Teacher Morale." Unpublished Doctoral dissertation,
University of Southern California, 1958.

Elsbree, Willard S. and Reutter, Edmund E., Jr. Staff
Personnel in the Public Schools. New York:
Prentice Hall, Inc., 1954.

French, Will and Associates. .gehavioral Goals of General
Education in High School. New York: Russel Sage
Foundation, 1957.

Gage, Nathaniel L. (ed.). Handbook of Research on Teaching.
Chicago: Rand McNally and Company, 1963.

Griffiths, Daniel E. Human Relations in School Adminis—
tration. New York: Appleton-Century-Crofts, Inc.,
1956.

Hart, Joseph K. Education in the Humane Community. New
York: Harper and Brothers, 1951.

Hilgard, Ernest R. Theories of Learning. New York:
Appleton-Century-Crofts, Inc., 1956.

Howsam, RObert B. Who's a Good Teacher: Problems and Pro—
gress in Teacher Evaluation. Prepared for the Joint
Committee on Personnel Procedures of the California
School Boards Association and the California Teachers
Association.

Kelley, Earl C. and Rasey, Marie I. Education and the Nature
of Man. New York: Harper and Brothers, 1952.

Kerlinger, Fred N. Foundations of Behavioral Research. New
York: Holt, Reinhart and Winston, Inc., 1964.

Learning and the Teacher. Washington 6, D.C.: Association
for Supervision and Curriculum Development, 1959.

 

120

Nelson, Kenneth G., Bicknell, John E., and Hedlund, Paul A.
Development and Refinement of Measures of Teacher Ef-
fectiveness. -Albany: The University of the State
of New York and the State Education Department, 1956.

McCall, William A. and Krause, Gertrude R. “Measurement of
Teacher Merit for Salary Purposes," Journal of Edu-
cational Research, Vol. LIII, NO. 2 (October, 1959).

 

McNemar, Quinn. "Opinion Attitude Methodology," Psychological
Bulletin, Vol. XLIII (July, 1946).

Monford, Ida Bell. "Factors Influencing the WOrk of Teachers
in Fairfax County Virginia." Unpublished Doctoral
dissertation, Ohio State University, 1956.

O'Connor, William Francis, Jr. "A Study of Some Selected
Factors Related to Teacher Morale." Unpublished
Doctoral dissertation, Cornell University, 1958.

Peck, Robert F. "Principals' Ratings of Teacher Performance
From Personality Data," Journal of Educational Ppy-
chology, Vol. L (April, 1959).

Pryor, Guy Clark. "The Relationship Between Teachers' Per—
ceptions of Administrative Dimensions, and the Morale
Status of Teachers in Certain Texas Schools." Un-
published Doctoral dissertation, North Texas State
University, 1964.

Rasmussen, Glen R. "Perceived Value Discrepancies of
Teachers and Principals-—A Threat to Creative Teach-
ing," The Bulletin of the National Association of
Secondary School Principals, Vol. XLVI, No. 272
(March, 1962).

 

Redefer, Frederick L. "A Teacher Teaches Better," N.E.A.
Journal, Vol. LIII (April, 1964).

Redefer, Frederick L. "Teacher Morale and Quality of Edu-
cation," Nations Schools, Vol. LIX (February,
1957).

Redefer, Frederick L. "Studies of Teacher Morale," School

and Sociepy, Vol. XCII (February 22, 1964).

 

Remmers, H. N. and Gage, H. L. Educational Measuremenp'and
Evaluation. New York: Harper and Brothers, 1955.

121

Ross, Walter E. "A Study of Personnel Factors Affecting the
Morale Status of Teachers of Two Rural School Systems
in New York and Including Comparisons and Findings
with Those of a Similar Study Completed for a New
Jersey Suburban School System." Unpublished Doctoral
dissertation, New York University, 1960.

Roth, James Lester. "A Technique for Determining the Sources
of Job Dissatisfactions." Unpublished Doctoral
dissertation, Stanford University, 1956.

Rotter, Julian B. and ROfferty, Janet E. Manual for the
Rotter Incomplete Sentences Blank. New York: The
Psychological Corporation, 1950.

Sellitz, Claire; Jahoda, Marie; Deutsch, Morton and Cook,
Stewart W. Research Methods in Social Relations.
New York: Henry Holt and Co., Inc., 1959.

Simpson, Ray H. "Use of Self-Evaluative Tools for Improve-
ment of Instruction," AACTE Bulletin, Vol. VIII,
American Association of Colleges of Teacher Edu-
cation, December, 1960.

Suehr, John H. "A Study of Morale in Education," Un-
published Doctoral dissertation, University of
Colorado, 1961.

Suehr, John H. "A Study of Morale in Education Utilizing In—
complete Sentences," Journal of Educational Research,
Vol. LVI, NO. 2 (October, 1962).

Symonds, Percival M. "Learning is Reacting," What Education
Has to Learn from Psychology. New York: Teachers
College, Columbia University, March, 1960.

Thompkins, Ellsworth. "Evaluating Teachers," Spotlight,
NO. 62 (March—April, 1964).

Waggoner, Glen Hastings. "Administrator's Scores on Selected
Standardized Tests and His Administrative Performance
as Reported by Classroom Teachers." Unpublished
Doctoral dissertation, Stanford University, 1959.

Walker, Helen M. and Lev, Joseph. Statistical Inference.
New York: Henry Holt and Co., 1953.

_Wpstern Michigan University Continuous Study of Student Re—
action Rpports. Kalamazoo: Western Michigan Uni—
versity, 1962.

 

 

122

Wiles, Kimball. Teaching forpgetter Schools. Englewood
Cliffs: Prentice Hall, Inc., 1959.

Who's A Good Teacher. Washington, D.C.: American Association

 

of School Administrators, Department of Classroom
Teachers of the N.E.A. and National School Boards As—
sociation, 1961.

~ APPENDICES

Scorer Y

1

APPENDIX A

Reliability of Judgements Of TMF Responses
Five Percent Sample Check
by Two Judges

 

 

 

 

 

 

 

 

 

 

 

Scorer X

0 l 2 3 4 Totals

4 0 0 0 13 31 44
3 1 2 8 87 55 153
2 3 12 32 19 4 70
1 11 42 13 12 2 80
0 27 14 2 2 1 46
42 70 551133 93 = 393

H

fx x fo fx(Xl)2 2Y1 Xlz‘Yl fy fyl nyl fy(Yl)2 zXl YlZXl

 

 

 

 

 

42 0 0 0 20 0 44 4 176 704 163 652
70 1 70 70 72 72 153 3 459 1377 499 1497
55 2 110 720 102 202 70 2 140 280 149 298
133 3 399 1197 363 1089 80 1 80 80 112 112
93 4 372 1488 299 1196 46 0 0 0 28 0
393 951 2975 855 2559 393 55 2441 951 2559
c E A D A B c D
NEXY = (ND-AC) = 192,582
NZXZ = (NE-C2 ) = 254,774 192, 582 =
NEYZ = (NB-A2 ) = 228,288* F =V (254, 774) (228, 288) .798
t = VNez ,= 80 V391 _ .80 (19.92) = 15.936 = 16 48
Ni]... 2 All—.064 0967 .967
_ _ **
df _ 391 t 995 _ 3.291

*Walker and Lev, Statistical Inference, p. 238.

**Ibid.,

 

p. 241.

124

APPENDIX B

Coefficient of Concordance* Among Ranks of Judgements
of Teacher Morale Items
(Five Percent Sample)

 

 

 

 

 

Sum of
Ranks Sum of
School Judge Judge (Read L Ranks
Items Judges X Y to R) Squared
1 1 1“ 1 3 9
2 18 18 18 54 2916
3 6 6 2 14 196
4 2 7 4 13 169
5 3 3 6 12 144
6 16 16 14 46 2116
7 4 2 3 9 81
8 5 5 5 15 225
9 15 ll 8 34 1156
10 12 13 13 38 1444
11 9 12 12 33 1089
12 14 15 17 46 2116
13 17 17 16 50 2500
14 10 10 10 30 900
15 7 4 9 20 400
16 13 14 15 42 1764
17 8 8 7 23 529
18 11 9 ll _31 961
Sum of Ranks 5 3 18,715
m = Number of judges
R = Ranks 263 169
W’= Coefficient of Concordance S = 18,715 - _—18___ = 4094.5
R = Correlation Coefficient
2 _ 12 (4094.51,_ 49134 _
s = 282 - 3%L W ‘ 3.18 (323) ’ “ 52,326 ‘ '939
w = 128 I». = 3 (339)-1 = ié—B—l-Z = .904
m2 N(N2-1,
F = 43—1) .939 _ 1.878 _ 26 45
§=IILW_'_1_ F 1.1111211 1-.939 ‘ .071 ‘ '
m-l l—W
df = nl = N-l-i Significance of F = .01 level
df = n = (m—l) (N=1.3)
2 m

*Walker and Lev, Statistical Inference, pp. 284-285.
125

 

APPENDIX C

Instructions to Principals for
Teacher Evaluating Procedures

Each principal should have the following materials:

One "Principal's Estimate of Teaching Ability" for
EACH OF HIS TEACHERS.

One "Teacher Self-evaluation for EACH TEACHER.

Enough copies of "Student Estimate of Teaching
Ability" so that EACH teacher may be evaluated by ONE AVERAGE
CLASS. If there are 20 teachers, this would be approximately
600 forms.

In a faculty meeting, after explaining the purposes
of evaluation, ask the faculty to fill out the "Teacher Self—
evaluation." It should take about ten minutes.

Ask one highly respected staff member to collect the
evaluations, put them into an envelope provided, and have the
office secretary mail it.

Then ask each teacher to have one average class evalu—
ate him. .Another teacher or responsible student may be put in
charge of the class so that students can respond freely. Seal
the responses in an envelOpe and mail or take to Higgins Lake,
whichever is permissable to your staff. -A report of the esti-
mates made by the students will be given to each teacher who
requests it. No principal will be given a report of the stu—
dent evaluations.

Each principal is asked to evaluate his staff. Put
all the evaluations into one envelope and mail or take to
Higgins Lake in May.

The results of these evaluations will be used in a
doctoral disSertation, so your help is greatly appreciated.
Your staff has two things to gain: 1. Self—evaluation is an
important part of the teacher's role, and the items on these
forms are the ones generally agreed upon as being important to
good teaching. 2. The majority of teachers who have had re-
ports on the student evaluations have improved their teaching.

Thank you very much for your interest and assistance.

Jarvis WOtring

126

127

PRINCIPAL'S ESTIMATE OF TEACHING ABILITY

Please answer the following questions honestly and frankly.

Your replies will be compared with those of the teachers'

appraisal on the same questions.

Circle the appropriate number at the right of each of the

self-

questions. 1 - Below average; 2 — Average; 3 - Good; 4 — Very
good: 5 — The very best.

Teacher's Name School

1.

10.

 

The knowledge this teacher has of the subject
taught? (Has he a thorough knowledge and under-l 2
standing of his teaching field?)

The ability of this teacher to explain clearly?
Are assignments and explanations clear and 1 2
definite?

This teacher's fairness in dealing with students?
(Is he fair and impartial in treatment of all 1 2
students?)

The ability of this teacher to maintain good
discipline? (Does he keep control of the class 1 2
without being harsh? .Is he firm but fair?)

The sympathetic understanding shown by this
teacher? (13 he patient, friendly, considerate l 2
and helpful?)

The ability this teacher has to make classes

lively and interesting? (Does he show enthusi- l 2
asm and a sense of humor? Does he vary teach-

ing procedures?)

The ability of this teacher to get things done
in an efficient and businesslike manner? (Are
plans well made? Is little time wasted?)

|._a
N

The skill this teacher has to get students to

think for themselves? (Are students' ideas and 1 2
opinions worth something in this class? Do they
get at the real reasons why certain things happen?)

The ability of this teacher to help students know
why they are learning certain things? (Are . l 2
there good reasons for studying the tOpics in

this course? Do students see the value and im-
portance of things'they study?)

The general (all-round) teaching ability of this
teacher? (All factors considered, how close 1 2
does this teacher come to flflur ideal?)

Name

Please answer the following questions as objectively as possi—

ble. Circle the appropriate number at the right of each of

128

TEACHER SELF-EVALUATION

School

the ten areas. 1 — Below average; 2 — Average; 3 - Good;
4 — Very good; 5 - The very best.

HOW DO YOU RATE ON THE FOLLOWING:

l.

10.

Your knowledge of the subject taught. (Have
you a thorough knowledge and understanding of

your field?) 1 2
Your ability to explain clearly. (Are assign—
ments and explanations clear and definite?) 1 2
Your fairness in dealing with students. (Are

you fair and impartial in treatment of all 1 2
students?)

Your ability to maintain good discipline. (Do
you keep good control of the class without be- 1 2
ing harsh? Are you firm but fair?)

Your sympathy and understanding. (Are you
patient, friendly and considerate and helpful?)

|—'
N

Your ability to make classes lively and
interesting. (Do you show enthusiasm and a 1 2
sense of humor? Do you vary teaching procedures?)

Your ability to get things done in an efficient
and businesslike manner. (Are plans well made? 1 2.
Is little time wasted?)

Your skill to get students to think for them-
selves. (Are students' ideas and opinions l 2
worth something in this class? Do they get at

the real reasons why certain things happen?)

.Your ability to help students know why they are

learning certain things. (Do your students 1 2
understand the value and importance of things
studied?)

Your general teaching ability. (All factors
considered how close do you come to being the l 2
best teacher you know how to be?)

129

STUDENT ESTIMATE OF TEACHING ABILITY

Please answer the following questions honestly and frankly.
Do not give your name. Neither your teacher nor anyone else
at your school will ever see your answers.

The person who is temporarily in charge of your class will,
during this period, collect all reports and seal them in an
envelope addressed to Michigan State university.

After completing this report, sit quietly or study until all
students have completed their reports. There should be no
talking.

Circle the appropriate number at the right of each Of the ten

questions. 1 — Below average; 2 - Average; 3 - Good; 4 - Very

good; 5 - The very best.
Teacher's Name School
WHAT IS YOUR OPINION CONCERNING:

l. The knowledge this teacher has of the subject
taught? (Has he a thorough knowledge and under 1 2 3 4
standing of his teaching field?)

2. The ability of this teacher to explain clearly?

(Are assignments and explanations clear and l 2 3 4
definite?)

3. This teacher‘s fairness in dealing with students?
(Is he fair and impartial in treatment of all 1 2 3 4
students?)

4. The ability of this teacher to maintain good
discipline? (Does he keep good control of the 1 2 3 4
class without being harsh? Is he firm but fair?)

5. The sympathetic understanding shown by this
teacher? (Is he patient, friendly, considerate, l 2 3 4
and helpful?)

6. The ability this teacher has to make classes

lively and interesting? (Does he show enthusi- l 2 3 4
asm and a sense of humor? Does he vary teaching
procedures?)

7. The ability of this teacher to get things done in
an efficient and businesslike manner? (Are plans 1 2 3 4
well made? Is little time wasted?)

8. The skill this teacher has to get students to
think for themselves? (Are students' ideas and l 2 3 4
.opinions worth something in this class? Do they
get at the real reasons why certain things happen?)

9. The ability of this teacher to help students know
why they are learning certain things? (Are there 1 2 3 4
good reasons for studying the~tppics in this
course? no you see the value and importance of
the things you study?)

10.

11.

130

The general (All-round) teaching ability of this
teacher? (All factors considered, how close ' l 2 3 4 5
does this teacher come to your ideal?)

On the reverse side,
this teacher.

name one or two things you like about

APPENDIX D

Reliability of Student Evaluations

Split—half correlations of student evaluations drawn blindly -
one class from each school.

School Rank Order Correlation Reliability Coefficient
3 .83 .90
4 .89 .94
5 .54 .73
6 .35 .52
7 .24 .40
8 .72 .84
9 .90 .95
10 .85 .92
ll .39 .57

r
_._24J§1
Formula for reliability — zrxx - l + rxx

Quinn McNemar, Psychological Statistics (New York:
John Wiley and Sons, Inc., 1955), p. 157.

131

APPENDIX E

Coefficient of Concordance
Among Rankings of Three Factors
of Teacher Morale

 

 

(Wl - corrected for continuity because N and m are small.)

 

 

 

W1 = 12(s—1) *
m2 N(N2+1)+24 S = Sum of ranks squared
Schools Ranked (L to R)
3 .4 5 ~ 6 7 8 . 9 10 11
Fl 1 2 4 9 6 8 3 7 5
F2 3 l 2 7 4 9 8 6 5
F3 .5. _1. _1V ._2 ._§ ._2 ._2 ._2 ._2
8 10 7 l9 18 26 l3 19 15 = 135
Mean 0f 12. .12.. 12 .12 .12 .12 .12 .12 .12
Sum of Ranks
62 72 = 52 + 82 + 42 + 32 + 112 + 22 + 42 + 0 = 304
12(303 _ 3636 _
2 6504 ‘ '559

3 9 (81-1) +24

F = 3wl 1.677 = 3.8
1-w .441

N41-2/3 = 9—1—2/3 = 7 1/3 = 7

14 2/3 = 15

m—l (N41-2/m) = 2 (8~2/3)

F.95 = 2.70 F = 4.14 W1 significant F .95

99

Helen M. walker and Joseph Lev, Statistical Inference
(New York: Henry Holt and Co., 1953), pp. 285-287.

132

APPENDIX F

Computation of the Chi-square Statistics
of the Median Test Applied to Teacher Morale Items

Legend: Md — median; 0 — observed frequency; E - expected fre-
quency; df - degrees of freedom is one for all
computations

40-E) 2
E

Rejecting the null hypothesis at .95 indicates that teacher
morale is significantly different.

Formula: Chi—square = sum of

2 Null hypothesis

 

 

 

 

« _ _' 2_Q;§ accepted or
O E O E (0 E) E rejected
School 3 School 4
Above Md 20 20 20 20 0 0 0
At or below .
Md 20 20 20 20 0 0 0
20 20 0 0 0
*Equally non-independent20 20 0 0 .9
are schools 7 and 10 0 accepted
School 3 School 5
Above Md 22 18 22 20 2 4 .2
At or below
Md 18 22 18 20 2 4 .2
18 20 2 4 .2
*Equally non-independent22 20 2 4 .;2
are schools 7 and 9; 9 .8 accepted
and 10; 10 and 11
School 3 School 6
Above Md 28 12 28 20 8 64 3.2
At or below
Md 12 28 12 20 8 64 3.2
12 20 8 64 3.2
28 20 8. 64 342
12 8 rejected

Chi-square percentile .999 = 10.8

*Equally independent are schools
3“and 10

133

134
Null hypothesis

 

 

 

 

 

 

2
_ _ 2 9;; accepted or
O E O E (O E) E rejected
School 3 School 7
Above Md 27 13 27 20 7 49 -2.45
At or below
Md 13 27 13 20 7 49 2.45
13 20 7 49 2.45
20 20 7 49 2.45
9.80 rejected
Chi-square percentile .995 = 7.9
*Equally independent are schools
4 and 6; 5 and 6; 4 and 10
School 3 School 8
Above Md 34- 6 34 20 14 196 9.2
.At or below
Md 6 34 6 20 14 196 9.2
6 20 14 196 9 2
34 20 14 196 9.2
36.8 rejected
Chi-square percentile .999 = 10.8
School 3 School 9
Above Md 23 17 23 20 3 9 .45
At or below
Md 17 23 17 20 3 9 .45
17 20 3 9 .45
23 20 3 9 .45
1.80 accepted
Chi-square percentile .90 = 2.7
*Equally non-independent are schools
4 and 9; 5 and 11; 6 and 7; 6 and 10;
6 and 11
School 4 School 7
Above Md 25 15 25 20 5 25 1.25
At or below
Md 15 25 15 20 5 25 1.25
15 20 5 25 1.25
25 20 5 25 1.25
5.0 rejected
Chi-square percentile .975 = 5.0
Sohool 4 School 8
Above Md 33 7 33 20 13 169 8.45
At or below
Md 7 33 7 20 13 169 8.45
7 20 13 169 8.45
33 20 13 169 8.45
33.80 rejected

Chi-square percentile .999 = 10.8
Equally independent are schools
5 and 8

135

2 Null hypothesis
_ _ 2 0*E accepted or
O E O E (O E) E rejected

School 5 School 7'

 

Above Md 24 16 24.5 20 4.5 20.25 1.0125
At or below
Md 16 24 15.5 20 4.5 20.25 1.0125
15.5 20 4.5 20.25 1.0125
24.5 20 4.5 20.25 1.0125
Chi—square percentile .95 = 3.8 4.0500 rejected

Equally independent are schools 4 and 11;

6 and 9

*Walker and Lev, p. 106: If the significance level obtained
from Chi-square is smaller than the predetermined significance
level, the Yates correction should be computed.

School 5 School 9

 

Above Md 21 21 21 20 1 l .05
At or below
Md 19 19 19 20 1 l .05
19 20 l 1 .05
21 20 l l .05
.20 accepted
Chi-square percentile .50 = .46

Equally lacking independence are schools
7 and 11; 9 and 11

School 5 School 10

 

 

 

Above Md 26 14. 26 20 6 36 1.8
At or below
.Md 14 26 14 20 6 36 1.8
14 20 6 36 1 8
26 20 6 36 .148
Chi—square percentile .99 = 6.6 7'2 rejected
Equally independent are schools 3 and 11
.School 6 School 8
Above Md 29 ll 29 20 9 81 4.05
At or below
Md 11 29 11 20 9 81 4.05
11 20 9 81 4.05
29 20 9 81 4.05
16.20 rejected

Chi-square percentile .999 = 10.8

136
2 Null hypothesis

 

 

' 2 9:; accepted or
0 E O-E”(O-E1 E rejected
School 7 School 8
Above Md 32 8 32 20 12 144 7.2
At or below
Md 8 32 8 20 12 144 7.2
8 20 12 144 7.2
32 20 12 144 7.2
28.8 rejected

Chi—square percentile .999 = 10.8

*Equally independent are schools 9 and 8;
10 and 8; 11 and 8. Nine, 10 and 11 are
the higher morale schools.

Schools 3, 4, 5 are clearly distinguishable with higher
teacher morale than 6, 7, 8, 10 and 11. School 8 is clearly
lower morale than all others.

APPENDIX G

Computations of_;mportant Median Tests
of Student Evaluations of Teachers
in High and Low Morale Schools

Numbers of high teacher morale schools: 3, 4, 5
Numbers of low teacher morale schools: 6, 7, 8, 9, 10, ll

 

 

0 3 0-1:: (O-E)2 O-E 2
E
School 3 School 10
Above Md 6 13 14 10 4 16 1.6
At or below Md 14 5 6 10 4 16 1.6
13 9 4 16 1.78
5 9 4 16 1.78
Chi—square .99 = 6.6 df - l Chi-square = 6'76
The null hypothesis of no difference could be rejected, but -
in reverse. More low evaluations were found in high morale
school 3 and more high evaluations in low morale school 10.
This strange phenomenon occurred between schools 5 and 10.
2 (O—E)2
0 E O-E (O-E) E
School 5 School 10
Above Md 3 12 3 6 3 9 1.5
At or below Md 9 6 9 6 3 9 1.5
12 9 3 9 1.0
6 9 3 9 1 0
Chi-square = 4.5

Chi-square .95 = 3.8 df - 1

Null hypothesis could be rejected, .05 level, but, high
evaluations here found in low morale schools and vice versa.

137

APPENDIX H

Computations of Important Median Tests
of Principals' Evaluations of Teachers in
High and Low Morale Schools

 

Numbers of high teacher morale schools: 3, 4, 5
Numbers of low teacher morale schools: 6, 7, 8, 9, 10, 11
2 (0-822
0 E O-E (O-E) E
School 4 School 6
Above Md 9 3 9 6.5 2.5 6.25 .962
At or below Md 4 8 4 6.5 2.5 6.25 .962
3 5.5 2.5 6.25 1.136
8 5.5 2.5 6.25 1.136
Chi-square =s4.196

Chi—square .95 = 3.8
Null hypothesis of no difference was rejected, .05 level.

I§chool 4 §chool 7

 

Above Md 9 7 9 5 4 16 3.20
At or below Md 4 22 4 8 4 16 2.00
7 11 4 16 1.45

22 18 4 16 .85

Chi-square = 7.50

Chi—square .99 = 6.6

Null hypothesis of no difference was rejected, .01 level.

School 4 School 9

Above Md 9 10 9. 6.02 2.98 8.8824 1.48
At or below Md 4 18 4. 6.98 2.98 8.8824 1.27
10 13:98 2.98 8.8824 .64

18 15.02 2.98 8.8824 .59

Chi-square = 3.98

Chi-square .95 = 3.8

Null hypothesis of no difference was rejected. Principals'
evaluations of the teachers in these schools were significantly
different at the .05 level and yet school 9 was different from
4 at the .19 level of significance.

138.

.139
2
0 E O-E (O-E)2 g’E

School 4 School 10

 

Above Md 10 5 10 6.29 3.71 13.764 2.05
At or below Md 3 13 3 6.71 3.71 13 764 2.19
5 9.29 3.71 13.764 1.48

13 8

.71 3.71 13.764 1.58
Chi-square = 7.30
Chi-square .99 = 6.6

Null hypothesis was rejected at .01 level.

School 4 School 11

 

Above Md 10 5 10 6.1 3.9 15.21 2.49
At or below Md 3 l4 3 6.9 3.9 15.21 2.20
5 8.9 3.9 15.21 1.71

14 10.1 3.9 15.21 1.51

Chi-square = 7.91

Chi-square .995 = 7.9

Null hypothesis was rejected at .005 level.

APPENDIX I

Mean Scores Ranked and Correlated for Items of Teacher Self-
Evaluations, Student Evaluations of Teachers, and Principals
Evaluations of Teachers

 

 

 

 

TSE's j . . SE's 1. PE's

Mean SD Rank Mean SD Rank Mean SE Rank
K 3.81 .742 13 4.06 .502 1 3.81 .795 l
E 3.45 .786 5 3.65 .623 5 3.49 .763 2
F 3.61 .891 2 3.589 .666 10 3.41 .933 3
D 3.41 1.020 6 3.592 .745 9 3.41 1.109 5
SV 3.57 .895 3 3.74 .614 2 3.395 .952 7
S 3.47 .883 4 3.643 .758 6 3.22 .945 10
EF 3.30 .942 9 3.637 .607 7 3.41 1.069 4
T 3.31 .835 8 3.68 .515 4‘ 3.25 .884 8
W 3.27 .817 10 3.62 .550 8 3.23 .834 9
GA 3.38 .794 7 3.72 .624 3 3.40 .797 6
R Of TSE & SE R Of TSE'S & PE R Of SE'S & PE'S

d d2 d d2 d d2

0 0 O 0 0 0

0 '0 3 9 3 9

8 64 1 1 7 49

3 9 1 1 4 16

1 1 4 l6 5 25

1 1 6 36 4 16

2 4 5 25 3 9

4 16 0 0 4 16

2 ..4 l -.l 1 ..1

4 16 1 __1 3 ____9_

118 90 150
708 "' ' '
R = l 540 900
10 (107-1) R — 1 — 90 R — 1 — 990

R = l - .72 R = 1 - .545 R = 1 - .909
R = .28 R = .45 R = -09

140

HICHIGRN STATE UNIV. LIBRARIES
m1|”WINWNWWllWIMHIHIWIWWI
31293103078758