Citation
Leadership behavior, teacher morale, and unexpected teacher absenteeism

Material Information

Title:
Leadership behavior, teacher morale, and unexpected teacher absenteeism in selected elementary schools in an urban school district
Creator:
Shaw, James Donald, 1940-
Publication Date:
Language:
English
Physical Description:
viii, 148 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Educational leadership -- Florida -- Orange County ( lcsh )
Absenteeism (Labor) -- Florida -- Orange County ( lcsh )
Teacher-principal relationships -- Florida -- Orange County ( lcsh )
Elementary school principals -- Florida -- Orange County ( lcsh )
Elementary school administration -- Florida -- Orange County ( lcsh )
Teacher morale -- Florida -- Orange County ( lcsh )
Teachers -- Leaves of absence -- Florida -- Orange County ( lcsh )
Elementary school teachers -- Florida -- Orange County ( lcsh )
Teacher morale ( lcsh )
Leadership ( lcsh )
Teachers -- Leaves of absence ( lcsh )
Dissertations, Academic -- Educational Administration and Supervision -- UF
Educational Administration and Supervision thesis Ed. D
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ed. D.)--University of Florida, 1980.
Bibliography:
Bibliography: leaves 138-147.
General Note:
Vita.
General Note:
Typescript.
Statement of Responsibility:
by James Donald Shaw.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
023446914 ( ALEPH )
07382258 ( OCLC )

Downloads

This item has the following downloads:


Full Text
LEADERSHIP BEHAVIOR, TEACHER MORALE, AND UNEXPECTED
TEACHER ABSENTEEISM IN SELECTED ELEMENTARY
SCHOOLS IN AN URBAN SCHOOL DISTRICT
BY
JAMES DONALD SHAW
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION
UNIVERSITY OF FLORIDA 1980




ACKNOWLEDGEMENTS
The writer wishes to express his sincere appreciation to the Superintendent of the Orange County Public School System, Dr. James L. Schott, and to the principals and teachers who made this study possible through their cooperation and participation. Gratitude also is expressed to Dr. Charles Walton and Dr. Wilfred Still from the district office who assisted the researcher with the technical requirements of this project.
A special expression of gratitude is extended to Dr. Michael Nunnery for his guidance and encouragement during the completion of the required course work and the writing of this dissertation. The writer is also grateful to the other members of his committee, Dr. James Hale and Dr. Joseph Shea, for their advice and assistance.
Special acknowledgement is also due to the writer's friend and colleague, Dr. Robert Cascaddan, who shared commuting and class time and who provided encouragement and understanding. Very
special thanks are expressed to the writer's wife, Susan, and his daughters, Betsy and Stacey, for their many sacrifices, faith, and patience during the completion of all the degree requirements.




TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS .................................... ii
LIST OF TABLES ......................................... iii
ABSTRACT .............................................. vi
CHAPTER
I INTRODUCTION ............................... 1
The Problem .................................... 5
Statement of the Problem ..................... 5
Delimitations ............................... 7
Limitations ...... I ........................ 8
Justification for the Study ................ 9
Definition of Terms ............................... 12
Procedures 17
Setting of t e Study ..................... 17
Identification of Participants ................ 18
Instrumentation and Data Collection ...... 18 Analysis of the Data ..................... 21
Organization of the Remainder of the Research
Report ......................................... 25
11 REVIEW OF RELATED LITERATURE ........... 26
Organizational Leadership Behavior ........... 26
The Influence of Perception on Behavior ........... 31
Employee Morale ............................... 34
Absenteeism as an Indicator of Organizational
Health ......................................... 43
Leadership Behavior, Morale, Job-Related
Variables, Personal Characteristic Variables,
and Employee Absenteeism ..................... 45
Leadership Behavior ..................... 46
Morale .................................... 48
Job-Related Variables ..................... 50
Personal Characteristic Variables ........... 51
III PRESENTATION AND ANALYSIS OF DATA ...... 61
iii




Descriptive Data by Schools Relative to TeacherRelated Variables ............................... 65
Comparisons Among Elementary Schools with
High, Average, and Low Unexpected Teacher
Absenteeism .................................... 77
Comparisons Among High and Low Morale Elementary
Schools Relative to Teachers' Perception of the
Leadership Style of the Principal ........... 84
Prediction of Unexpected Teacher Absenteeism
from Selected Variables ..................... 90
IV SUMMARY, CONCLUSIONS, AND DISCUSSION ..... 97
Summary ......................................... 97
Conclusions .................................... 106
Discussion. ................................... 108
Results of the Study in Relation to Its
Justification and Previous Research ...... 108
Practical Application of the Study Results .... 114
APPENDIX
A LEADER BEHAVIOR DESCRIPTION QUESTIONNAIREFORM XII ......................................... 117
B MEANS AND STANDARD DEVIATIONS OF SUBSCALE
LBDQ-XII SCORES FOR SELECTED LEADERS IN
LEADERSHIP POSITIONS .......................... 122
C RELIABILITY COEFFICIENTS OF SUBSCALES ON
THE LBDQ-XII .................................... 125
D PURDUE TEACHER OPINIONNAIRE ................ 127
E TEST-RETEST CORRELATIONS FOR PURDUE TEACHER
OPINIONNAIRE--FACTOR AND TOTAL SCORES ... 134
F CORRELATION MATRIX FOR INDEPENDENT AND
DEPENDENT VARIABLES .......................... 135
G RESULTS OF MULTIPLE REGRESSION OF
INDEPENDENT VARIABLES ON UNEXPECTED
TEACHER ABSENTEEISM .......................... 136
H RESULTS OF MULTIPLE REGRESSION OF
INDEPENDENT VARIABLES ON UNEXPECTED
TEACHER ABSENTEEISM WITH TEACHER
MORALES TREATED AS A COVARIATE ........... 137
REFERENCES .............................................. 138
BIOGRAPHICAL SKETCH .................................... 148
iv




LIST OF TABLES
Table Pg
1 Basic Information Relative to the 15 Elementary
Schools Selected for the Study ......................64
2 L8DQ-XII Dimension Means, Orientation Means,
and Standard Deviations for the 15 Elementary
Schools Included in the Study .......................66
3 PTO Dimension Means, Total Means, and Standard
Deviations for the 15 Elementary Schools Included
in the Study ...................................... 73
4 Number of Teacher Participants, Male-Female
Ratios, Distribution by Education Level,
Marital Status, and Experience in the School
District for the 15 Elementary Schools Included
in the Study ...................................... 78
5 LBDQ-XIl Dimension Means, F Values, and
Observed Alpha Level for Schools Classified as
High, Average, Low in Teacher Absenteeism .... 81
6 PTO Dimension Means, F Values, and Observed
Alpha Level for Schools Classified as High,
Average, Low in Teacher Absenteeism ...............83
7 F Values and LBDQ-XII Dimension, Orientation,
and Total Means for Schools Classified as High
and Low in Teacher Morale......................... 86
8 Results of Multiple Regression Analysis of
10 Independent Variables as Predictors of
Individual Teacher Unexpected Absenteeism ..........91
9 Results of Multiple Regression Analysis of
Nine Independent Variables as Predictors of Individual Teacher Unexpected Absenteeism
when the Teacher's Total Morale Score (PTO)
is Treated as a Covariate........................... 95
v




Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial
Fulfillment of the Requirements for the Degree of Doctor of Education
LEADERSHIP BEHAVIOR, TEACHER MORALE, AND UNEXPECTED
TEACHER ABSENTEEISM IN SELECTED ELEMENTARY
SCHOOLS IN AN URBAN SCHOOL DISTRICT by
James Donald Shaw
December 1980
Chairman: Michael Y. Nunnery
Major Department: Educational Administration and Supervision
Employee absenteeism costs reached enormous proportions in the United States during the 1970's. in the 1978-1979 school year, $854,000 was spent for substitute teachers in the public schools of Orange County, Florida. Excessive unexpected teacher absenteeism was determined to have a detrimental effect on the progress of the educational organization, since the teacher's absence directly affected a large number of pupils. Numerous studies in industry indicated a relationship between absenteeism and employee morale, and morale was found to be a function of the perceived leadership behavior of the manager. Given the limited research in educational institutions designed to assess the relationships among leadership behavior, teacher morale, and teacher absenteeism, the study was undertaken. The major focus was to determine the extent to which the unexpected absenteeism of teachers could be explained by teacher perceptions of the leadership behavior of principals, teacher morale, and selected teacher-related and school -related characteristics.
The Leader Behavior Description Questionnai re- Form XII and the Purdue Teacher Opinionnaire were administered to 423 teachers in 15
v i




elementary schools within the district (the 5 schools with the highest percentage of unexpected teacher absenteeism, the 5 schools with the lowest percentage of unexpected teacher absenteeism, and the 5 schools nearest the mean percentage of unexpected teacher absenteeism for all 67 elementary schools in the district). Also collected were demographic data concerning the teacher's level of education, sex, marital status, and years experience in the school district. Additionally, the schools studied were categorized as Title I or non-Title 1, large or small, and with high, average, or low unexpected teacher absenteeism. These constituted the independent variables employed in an effort to describe the dependent variable, unexpected teacher absenteeism.
Two basic statistical techniques wp e utilized to determine the contribution of the independent variables to the dependent variable. Differences among schools with high, average, and low unexpected teacher absenteeism in regard to perceived leadership behavior of the t.rincipal and teacher morale were analyzed with the one-way analysis of variance (F test). The teacher-related and school-related demographic variables, along with scores on the leader behavior and morale instruments, were analyzed with stepwise multiple regression to determine to what extent these independent variables contributed to teacher absenteeism. Standard descriptive statistics and a correlation matrix completed the analysis.
Based up9n an analysis of data, it was found that (a) teacher perceptions of the leadership style of principals were significantly different at the .05 level in schools with high unexpected teacher
vii




absenteeism when compared to schools with average and low unexpected teacher absenteeism; (b) there was a significant difference in teacher morale among schoois with high, average, and low unexpected teacher absenteeism; (c) significant differences existed in the teachers, perceptions of the leadership style of principals in high and low morale schools; (d) the independent variables of unemployment in a school with high or average unexpected teacher absenteeism, employment in a small school, one-five years experience in the district, employment in a Title I school, single marital status, and low morale score were significant predictors of individual teacher unexpected absenteeism. The independent variables collectively accounted for 17.72% of the variance in teacher absenteeism. Effective leadership behavior in the areas of representation, demand reconciliation, tolerance of uncertainty, initiation of structure, role assumption, and integration was identified as a significant contributor to low unexpected teacher absenteeism. The two dimensions of the Purdue Teacher Opinionnaire which had the greatest range in mean scores between high absenteeism and low absenteeism schools were teacher rapport with principal and rapport among teachers.
viii




CHAPTER I
INTRODUCT ION
In the context of the politically sensitive education arena of the 1970's, with commonplace power struggles for decision-making authority, the quality of human relations interactions between principals and teachers became a focal point of concern. Teacher morale, a somewhat nebulous term used in many contexts, had been recognized as a powerful force impacting education. According to Knezevich (1975),
morale is difficult to define and even more difficult to measure. It is a state of being more easily felt than described and verified. Morale is not necessarily an end in itself. It is a means of promoting a smoothly functioning
and productive institution. (p. 455)
Washington and Watson (1970) contended the effectiveness of a school may be directly dependent upon the morale of the teachers of that school, and "the school principal plays a key role in nurturing and maintaining positive teacher morale" (p. 4). His or her actions, decisions, and the staff's perceptions of his or her behavior have been found vitally important in determining the morale of a school in studies conducted by Halpir. and Croft (1963, p. 57) with the Organizational Climate Description Questionnaire. In research conducted in elementary schools they found the building of positive teacher morale is not entirely dependent upon the actions of the principal, but also upon the ways in which teachers perceive this behavior. Campbell, Dunnette, Lawler, and Weick (1970, p. 103) found that perceptions by others of leadership style are more reliable than self -descriptions.
1




21
Bruner and Tagiuri (1954, p. 83) reported it is the teacher's perception of a leader's behavior which can influence teacher behavior.
Teacher absenteeism has been noted as a possible result of low morale (Bamber, 1979, p. 18). An analysis of Department of Labor statistics indicated an increase in teacher absenteeism of 3% in the 1977-1978 school year as compared to the 1976-1977 year (United States Department of Labor Statistics, Division of Economic Studies, 1978, p. 51). On an average school day in the 1977-1978 school year 86,000 of the 2,400,000 public school classrooms in the United States were staffed by someone other than the regular teacher (p. 53). Indications from a 1977 poll of National Association of Elementary School Principals members showed that 15% considered teacher absenteeism a serious problem and another 59% considered it an increasing problem (Stryker, 1977, p. 24.). This increase in absenteeism has taken place when many diseases have been conquered and medications are available to provide relief from pain.
Personal factors such as sex, marital status, education level, and years of employment have been examined in relationship to absenteeism. Redmond (1978) found that gender was significantly related to the absence of professional personnel in the Fort Madison (Iowa) Community School District, i.e., women were absent more frequently than men. Sylwester (1979) reported absence rates for male teachers were significantly lower than for female teachers in his study of 335 Oregon elementary and secondary school teachers. Coller (1975) found that married teachers tended to have lower absence records than single teachers in the Livonia, Michigan, public schools. "Academic degree" was one of nine variables that Douglas (1976) found to be predictors of absenteeism when added in a stepwise




3
regression. Studies conducted to measure the impact of years of employment on the absence rates of educational personnel have produced conflicting results. Gibson and Lafornara (1972) reported that teachers with fewer than 10 years of service were absent more often than teachers with more than 10 years of service. However, Stall~ngs (1959) reported little difference in the use of sick leave by permanent (tenured) and probationary (non-tenured) teachers in 16 Southern California school systems. In his sample of Dade County (Miami), Florida, elementary school teachers, Manganiello (1972) reported no significant difference in the frequency of sick leave absences when the teachers were grouped in terms of length of service.
School-related characteristics such as the type of school and the work unit size have been related to absence rate. The Office of Education Performance Review of the State of New York (1974, p. 14) reported that the discretionary absence rate for New York City teachers in Title I elementary schools was 29% higher than in non-Title I elementary schools during the 1972-1973 school year. Bridges and Hallinan (1978) found that work unit size had direct impact on absenteeism of teachers in 57 California and Wisconsin elementary schools.
In 1977-1978, 56 school districts in the Philadelphia area participated in studies of employee absenteeism (Pennsylvania School Board Association, 1978, p. 53). increased absenteeism among teachers was associated with staff morale, salary scale, professional expectations and attitudes of teachers, administrative leadership, working conditions, emotional stress, physical weakness and chronic illness, and education programs. The monetary cost of providing substitute teachers in the Philadelphia area was $2,000,000 for the 1977-78 school year. Projected to all 504 school systems in the state, the




4
authors noted, the potential cost of filling all vacancies with substitutes would be more than $30 million. If the salaries for absent teachers were added to the cost of substitute teachers, they estimated that Pennsylvania school systems spent more than $88 million a year for total professional salaries directly attributable to teacher absence. In 1977-1978, this cost was 4.5% of that paid for teachers' salaries. To put this cost into perspective:
This $88 million cost of absenteeism accounts for a greater percent of the total budget than any one of the following budgetary line items: health services, food services, student activities, community services, or capital outlay.
(p. 36)
Teacher absenteeism has had an impact on school district budgets in terms of substitute teacher pay, administrative expenses, and record-keeping costs. In addition to financial considerations, instructional leaders have reported that the quality of the teaching -learning experience has deteriorated when the regular teacher has been absent from the classroom (Jacobson, Reavis, & Logsdon, 1963, p. 67).
Absenteeism resulting from professional leave, temporary duty assignments elsewhere, long-term leaves, or a personal leave approved as a result of a contractual agreement between the school board and the teacher has been viewed as less disruptive to the educational process because more time is available for the principal, the teacher, and a substitute teacher to formulate plans. It is unexpected absenteeism, therefore, which has been noted as a major concern to school districts and was the focus of the study reported herein.
Numerous investigations have been conducted to ascertain the relationships between leadership behavior and organizational climate. However, there have been relatively few studies designed to assess the relationships among leadership behavior, teacher morale, and




5
teacher absenteeism. Research is needed to identify those leadership behavior patterns which are crucial in dealing with teachers in the educational organization. A knowledge of effective leadership behavior could provide valuable insight for the managerial development of principals. In light of the foregoing, the study reported herein relative to teachers' perceptions of the principal's leadership behavior, teacher morale, teacher-related characteristics, school-related characteristics, and teacher absenteeism was undertaken.
The Problem
Statement of the Problem
The problem in the study was to determine the extent to which the unexpected absenteeism of teachers in selected elementary schools within the Orange County, Florida, Public School System and the unexpected absenteeism of individual teachers could be explained by teacher perceptions of the leadership style of principals, teacher morale, and selected teacher-related and school-related characteristics. Specifically, answers to the following questions were sought:
1. Were there differences in teacher perceptions of the leadership
style of the principals on each of the 12 dimensions of the Leader Behavior Description Questionnaire- Form XII (LBDQ-XII), developed by Stogdill (1963), among elementary schools with high,
average, and low unexpected teacher absenteeism?
2. Were there differences in the 10 dimensions of teacher morale as
measured by the Purdue Teacher Opinionnaire (PTO), developed by Rempel and Bentley (1967), among elementary schools with
high, average, and low unexpected teacher absenteeism?




6
3. Were there differences in the teachers' perceptions of the leadership style of principals in high and low morale elementary schools?
4. To what extent could individual teacher unexpected absenteeism
be predicted from
a. the teacher's perception of the systems orientation of the
principal,
b. the teacher's perception of the personal orientation of the
principal,
C. the teacher's total morale score on the PTO,
d. the teacher's level of education,
e. the teacher's sex,
f. the teacher's marital status,
g. the teacher's experience in the school district,
h. whether the teacher was employed in a Title I or a nonTitle I school,
i. whether the teacher was employed in a large or a small
elementary school, and
j. whether the teacher was employed in a school with high
unexpected teacher absenteeism or a school without high
unexpected teacher absenteeism?
S. When the teacher's total morale score was treated as a covariate,
to what extent could individual teacher unexpected absenteeism
be predicted from
a. the teacher's perception of the systems orientation of the
principal,
b. the teacher's perception of the personal orientation of the
principal,




7
C. the teacher's level of education,
d. the teacher's sex,
e. the teacher's maritW4 status,
f. the teacher's experience in the school district,
g. whether the teacher was employed in a Title I or non-Title I
school,
h. whether the teacher was employed in a large or a small
elementary school, and
i. whether the teacher was employed in a school with high
unexpected teacher absenteeism or a school without high
unexpected teacher absenteeism?
Delimitations
1. The study was confined to a single urban school district (the
Orange County Public School System) in central Florida.
2. The investigation of unexpected teacher absenteeism for a single
school year, 1978-1979, was confined to 15 elementary schools within the district--5 with the highest percentage of unexpected teacher absenteeism, 5 with the lowest percentage of unexpected teacher absenteeism, and 5 nearest the mean percentage of unexpected teacher absenteeism for all 67 elementary schools in
the district.
3. The measure of leadership behavior was confined to teachers'
responses to the 12 dimensions measured by the LBDQ-XII: representation, demand reconciliation, tolerance of uncertainty, persuasiveness, initiation of structure, tolerance of freedom, role assumptions, consideration, production emphasis, predictive
accuracy, integration, and superior orientation.




8
4. The measure of teacher morale was confined to teachers' responses
to the 10 dimensions measured by the PTO: teacher rapport with principal, satisfaction with teaching, rapport among teachers,
teacher salary, teacher load, curriculum issues, teacher status, community support of education, school facilities and services,
and community pressures.
S. Teacher-related and school-related characteristics were confined
to the teacher's level of education, sex (male or female), the teacher's marital status (single or married), the teacher's experience in the school district, whether the teacher was employed in a Title I or non-Title I school, whether the teacher was employed in a large or a small elementary school, and whether the teacher was employed in a school with high unexpected teacher absentee,z;m or a school without high unexpected teacher absenteeism.
Limitations
The descriptive study had certain limitations that must be identified and stated in order that the conclusions of the study may be regarded in the proper perspective. The following limitations were identified.
1. The design of the study was ex post facto. There are weaknesses inherent in the ex post facto design which restrict the interpretation of the findings to inferences about the relationship
of the variables rather than cause-effect.
2. The data source relating to teacher absenteeism was limited to
records within the district. To the extent that these data were
inaccurate or incomplete, the study lacked validity.




9
3. The accuracy of the teachers' responses to the instruments
administered was dependent upon the ability of the individual respondents to depict accurately their perceptions. Therefore, teachers' perceptions of leadership behavior and teacher morale were valid only to the extent that individuals accurately depicted
their perceptions.
4. The use of the variables included in the study did not preclude
that other variables may have influenced the dependent variable (teacher absenteeism) or what was treated on occasion as the intervening variable (morale). Therefore, the variables studied should not be considered the only influences on teacher
attendance.
5. The study was confined to a single school district, thus generalizations beyond the district must be regarded as suggestions.
Justification for the Study
Employee absenteeism costs reached enormous proportions during the 1970's. Robert Moore (1973), President of Searle Educational Systems, noted:
The cost to business and government of employee illness in wages alone has reached $15,000,000,000 annually. if the rate of employee absence due to illness and accidents--now 5.8 days per year--could be reduced by only one day, the Gross National Product could be increased by as much as $10,000,000,000 according to government reports. The figure would oe astronomical if we added absence due to laziness, employee discontent, and family problems. (p.
115)
Taylor (1979, p. 53), after analyzing data released by the United States Bureau of Labor Statistics, reported 1% of the total compensation paid by United States employers in 1976 was for paid sick leave. This amounted to six cents for every hour paid.




10
Organizations that have an average employee absence rate of 4% actually carry 1 extra employee for every 25 to take care of average absence (Smardon, 1974, p. 12).
During the 1978-1979 school year, a total of $854,000 was spent for substitute teachers in the public schools of Orange County, Florida (OCPS Statistical Report, 1979). A total of 27,828 substitute days were reported, representing a 22% increase over the 1977-1978 school year. An average of 8.06 days of sick leave per teacher was reported for the 1978-1979 school year, as compared to an average of 7.47 days for the 1977-1978 school year. The percentage of absences reported on days preceding and following holidays was also significantly higher than those occurring at other times.
Excessive absenteeism can have a detrimental effect on the progress of an educational organization, since the teacher's absence directly affects a large number of pupils. Jacobson, Reavis, and Logsdon (1963, p. 67) asserted that the service of substitute teachers does not equal the service of regular teachers, because the substitute teachers cannot deal in an efficient manner with the individual differences of pupils in a relatively short period of time.
Teacher absenteeism may be a potential source of conflict between principals and parents, as Zimet (1973) explained in a book on decentralization in New York City Schools:
Principals have pointed out to their teaching staffs that absenteeism is highly visible to parents and is very difficult to defend. Although a principal can support a teacher's actions vis-b-vis classroom teaching techniques, he cannot defend persistent and unexplained absences. (p. 111)
Research studies designed to assess the relationship of teachers' perceptions of the principal's leadership behavior, teacher morale, and teacher absenteeism have been limited, both in number and in the




11
variables investigated. in one study Collar (1970) found teacher absenteeism related to teacher morale. The teachers who had low rates of absenteeism tended to have higher teacher morale scores than did teachers with high rates of absenteeism. The male teachers in the study tended to have lower absence records than did the female teachers, and married teachers had lower absence rates than did the unmarried teachers. Kokovich (1969) investigated leadership behavior of selected secondary school principals to determine if relationships existed between teachers' perceptions of leadership behavior and teacher morale. He found a strong re'itionship between the principal's leadership behavior and morale status, but determined leadership behavior is not the sole determinant of morale status. Miles (1967) noted the need for additional behavioral data, related to specific educational organizations, to study relationship among leadership behavior, teacher morale, and teacher absenteeism.
The study reported herein was conducted with the recognition of the need for increased knowledge in varied settings of the dimensions of leadership behavior and teacher morale which can influence teacher behavior to the extent that teacher absenteeism is affected. Such knowledge could be useful in training future administrators. At a more immediate and practical level, it was reasoned that the findings from such a study would be valuable to the leadership of the Orange County, Florida, Public School System as they struggled to reduce the rate of teacher absenteeism in that school district.




12
Definition of Terms
Elementary school. This term refers to a school established and organized to teach pupils on any grade level from kindergarten through grade six.
Elementary teacher. This term refers to a teacher employed to teach any grade level or provide music, art, physical education, media, and/or counseling services to pupils from kindergarten through., grade six.
Frequency of absence. This term refers to the number of instances of absences in a period of time (one school year in the present study).
High morale elementary school. This term refers to the five elementary schools in the study with the highest composite scores on the Purdue Teacher Opinionnaire.
Large elementary school. This term refers to those elementary schools in the study which had more than 33 certificated staff members, excluding administrators.
Leadership behavior. This term refers to the leader dimension measured by the LBDQ-XII (Stogdill, 1963, p. 3):
1. representation--speaks and acts as the representative of the
group;
2. demand reconciliation--reconciles conflicting demands and
reduces disorder to the system;
3. tolerance of uncertainty--is able to tolerate uncertainty and
postponement without anxiety or upset;
4. persuasiveness--uses persuasion and argument effectively;
exhibits strong convictions;




13
5. initiation of structure--clearly defines own role and lets
followers know what is expected;
6. tolerance of freedom--allows followers scope for initiative,
decision, and action;
7. role assumption--actively exercises the leadership role
rather than surrendering leadership to others;
8. consideration--regards the comfort, well-being, status, and
contribution of followers as significant;
9. production emphasis--applies pressure for production output; 10. predictive accuracy--exhibits foresight and ability to predict
outcomes accurately;
11. integration--maintains a closely knit organization, works to
resolve intermember conflicts; and
12. superior orientation--maintains cordial relations with superiors, has influence with them, and is striving for higher
status.
Long-term leave. This term refers to the absence of a teacher from an assigned work site for a period in excess of 10 consecutive days.
Low morale elementary school. This term refers to the five elementary schools in the study with the lowest composite scores on the Purdue Teacher Opinionnaire.
Org3nizational climate. This term refers to the "personality" of a school organization, viewed in terms of interpersonal relations (Halpin & Croft, 1963, p. 1).
Personal orientations of the principal. This term refers to those dimensions on the LBDQ-XII grouped by Brown (1967) as representing a leader's emphasis on the individual: representation, tolerance




14
of uncertainty, persuasiveness, role assumption, consideration, predictive accuracy, and tolerance of freedom.
Principal. This term refers to the administrator directly responsible for the management of the school and for carrying out the policies of the school board within that school.
Professional leave. This term refers to the extended absence of a teacher from an assigned work site for the purpose of continuing his or her formal education.
School district. This term refers to a school system serving all the pupils in a county within the State of Florida.
School year. This term refers to a 10-month contractual period for a teacher.
Small elementary school. This term refers to those elementary schools in the study which had 33 or fewer certificated staff members, exluding administrators.
Systems orientations of the principal. This term refers to those dimensions on the LBDQ-XII grouped by Brown (1967) as representing. a leader's emphasis on the institution: demand reconciliation, initiation of structure, production emphasis, integration, and superior orientation.
Teacher employment experience. This term refers to the number of years the teacher had been employed in the Orange County, Florida, Public School System. It was analyzed in four categories: 1-5 years, 6-10 years, 11-15 years, and more than 15 years.
Teacher level of education. This term refers to the highest academic degree held by the teacher: bachelor's, master's, specialist, or doctorate.




15
Teacher marital status. This term refers to whether the teacher is single or married.
Teacher morale. This term refers to factors as measured by the Purdue Teacher Opinionnaire (Rempel & Bentley, 1967, p. 4):
1. teacher rapport with principal --deals with the teachers'
feelings about the principal, his or her professional competency, his or her interest in teachers and their work, his or her ability to communicate, and his or her skill in human
relations;
2. satisfaction with teaching --pertains to teacher relationships
with students and feelings of satisfaction with teaching, the high morale teacher loves to teach, feels confident in the job, enjoys the students, and believes in the future of
teaching as an occupation;
3. rapport among teachers --focuses on a teacher's relationship
with other teachers by soliciting opinion regarding cooperation, preparation, eth ics, influence, interests, and
competency of peers;
4. teacher salary--pertains primarily to the teacher's feelings
about salaries and salary policies;
S. teacher load--deals with such matters as record-keeping,
clerical work, "red tape," community demands on teacher time, extra-curricular load, and keeping up-to-date professionally;
6. curriculum issues- -solicits teacher reactions to the adequacy
of the school program and meeting student needs, in providing for individual differences, and preparing students
for effective citizenship;




16
7. teacher status--pertains to feelings about the prestige,
security, and benefits afforded by teaching;
8. community support of education --deals with the extent to
which a community understands and is willing to support a
sound educational program;
9. school facilities and services --relates to the adequacy of
facilities, supplies and equipment, and the efficiency of the
procedures for obtaining materials and services; and
10. community pressures--gives special attention to community
expectations with respect to the teacher's personal standards, participation in outs ide-of -school activities, and the teacher's freedom to discuss controversial issues in the
classroom.
Temporary duty assignment. This term refers to the excused absence of a teacher from the classroom so that the teacher may work at another location, or attend a meeting, for a brief period of time (usually one day).
Title I school. This term refers to an elementary school with a sufficient number of economically-disadvantaged pupils to qualify for federal compensatory funding.
Unexpected absenteeism. This term refers to the degree to which teachers fail to report to work. It specifically excludes professional leave, long-term leave, temporary duty assignments, and prearranged personal leave which may be charged to sick leave according to the negotiated contractual agreement between the school district and the teachers' union.




17
Procedures
The reader is reminded that the problem in the investigation focused on teachers' perceptions of the principal's leadership behavior, teacher morale, selected teacher-related and school-related variables, and teacher absenteeism. Detailed in the following sections are the procedures which were used to conduct the investigation. First, the setting of the study is reviewed. Second, the procedures for identifying the participants are stated. Next, the data sources and collection procedures are explained. Finally, there is a description of the data analysis procedures.
Setting of the Study
The study was conducted in the Orange County Public School System, a large urban school district in central Florida. I n 1978, total population of the county was 438,558; 22% of the district population was of minority races, with the black race constituting the largest minority (Orange County Planning Department, 1979). In 1979, the economy of Orange County was based on tourism, manufacturing, agriculture, finance, and trade.
In the 1978-1979 school year, there were 81,453 pupils enrolled in kindergarten through twelfth grade. Seventy-five percent of those pupils were white, 22% black, 2% Hispanic, and the remaining 1% were Pacific Islanders and American Indians. Of the 40,959 elementary school pupils in the district 73% were white, 24% black, and the remaining 3% were Hispanic, Pacific Islanders, and American Indian (OCPS Statistical Report, 1979).




18
The 67 elementary schools in the district varied in both geographic and demographic features; they included urban, suburban, and rural schools. Student enrollment in the 67 elementary schools ranged from 248 to 957, with a mean enrollment of 600 (OCPS Statistical Report, 1979).
Identification of Participants
From the 67 elementary schools in the district, excluding exceptional education centers intended exclusively for pupils with special educational needs, those five with the highest percentage of unexpected teacher absenteeism, those five with the lowest percentage of unexpected teacher absenteeism, and those five nearest the mean percentage of unexpected teacher absenteeism rate for all elementary schools for the 1978-1979 school year participated in the study. All teachers in each of the 15 schrols (a total of 496 teachers) were asked to participate in the study. Records from the school district personnel and payroll departments were researched to select the 15 schools which met the criterion.
Instrumentation and Data Collection
Leadership behavior
The teachers' perceptions of the principal's leadership behavior for each of the 15 schools in the investigation were measured by the Leader Behavior Description Questionnaire-Form XII (LBDQ-XII) developed by Stogdill (1963). Each teacher was asked to complete this instrument which is composed of short descriptive statements of ways in which leaders are perceived as behaving (423 of the 496 teachers provided responses). Each respondent indicated the extent the behavior described by each item characterized the behavior of the principal of the school where he or she was employed. The LBDQ-XII




19
measures 12 dimensions of leader behavior (Appendix A). The 12 dimensions may be grouped into two general categories--personal behavior and systems behavior. Those grouped as personal behavior are representation, tolerance of uncertainty, persuasiveness, role assumption, consideration, predictive accuracy, and tolerance of freedom. The dimensions grouped as systems behavior are demand reconciliation, initiation of structure, production emphasis, integration, and superior orientation. Each of the 100 items on the LBDQ-XII has been related to one of the leader dimensions. The respondents described the leader behavior of the principal by indicating whether he or she always, often, occasionally, seldom, or never behaved as described by the item.
The LBDQ-XII was designed as a research device and has no norms (Stogdill, 1963, p. 8). However, the mean scores and standard deviations for several highly selected samples of commissioned and non-commissioned officers in an army combat division, the administrative officers in a state highway patrol headquarters office, the executives in an aircraft engineering staff, ministers of various denominations of an Ohio community, leaders in community development activities throughout the state of Ohio, presidents of successful corporations, presidents of labor unions, presidents of colleges and universities, and United States Senators were analyzed and reliability cr.fficiants for each of the LBDQ-XII subscales determined by using a modified Kuder-Richardson formula (Appendices B and C). Teacher morale
The teacher morale in the 15 schools selected for the study was measured by the Purdue Teacher Opinionnaire (PTO) developed by Rempel and Bentley (1967). Each of the teachers in the 15 schools




20
was asked to .;uiple' e the PTC (again 423 of the 496 teachers provided responses). The PTO contains 100 items and it provides a total morale score which is derived from the subscores on 10 dimensions of morale. The dimensions measured are teacher rapport with the principal, satisfaction with teaching, rapport among teachers, teacher salary, teacher load, curriculum issues, teacher status, community support of education, school facilities and services, and community pressures (Appendix D).
The reliability of the PTO was established by administering the instrument to school faculties of 20 or more teachers (Rempel & Bentley, 1967, p. 5). A stratified random sample of 60 Indiana schools and 16 Oregon schools was used, and test-retest data from 3,023 teachers were correlated. These correlations became the basis for the reliability data relative to each dimension score and the total score of the PTO (Appendix E).
Teacher-related and school -related variables
The researcher met with each of the 15 faculties and asked teachers to complete a researcher-developed personal data sheet. The researcher had previously coded the data sheet with the school number, school size (large-small), and a designation for Title I or nonTitle I school. Teachers were asked to provide information about sex, marital status, highest academic degree held, and years of employment experience in the school district. This data sheet was attached to a packet containing the LBDQ-XlI and the PTO Opinionnaires. All responses were coded to facilitate tabulation by computer. A self-addressed envelope was included and teachers were asked to return the completed forms to the researcher by means of the school




21
district courier system. Teachers were given assurance of confidentiality. Personal data sheets were received from 423 of the 496 teachers.
Teacher absenteeism
Records from the school district personnel and payroll departments were researched to determine teacher absenteeism in the 15 schools which participated in the study. These records were compared with data analyzed by the school district research department and cross-checked for accuracy. Excused teacher absences for temporary duty assignments, professional leaves, and long-term illness leaves for the 1978-1979 school year were subtracted from each school's total of teacher absences before selecting participating schools. Teacher absences were recorded in the district computer bank by teacher's social security number.
Analysis of the Data
Two basic statistical techniques were utilized in order to answer the questions posed in the statement of the problem. To answer questions 1, 2, and 3, which focused on differences among schools with high, average, and low absenteeism in regard to the 12 dimensions of the LBDQ-XII, the 10 dimensions of the PTO, and the leadership style of principals in high and low morale schools the basic technique utilized was the one-way analysis of variance (F test). In questions one and two, each of the dimensions was treated as a presumed dependent variable and the F test was applied to determine if there were significant differences among the three sets of schools when they were grouped according to degree of absenteeism. The same basic process was applied to question three which focused on leadership style of principals in high and low morale schools. The




22
schools were grouped according to the five highest morale schools and the five lowest morale schools and it was determined, using the oneway analysis of variance, if there were significant differences between these groups in teachers' perceptions of the leadership style of principals. Where a significant difference at the predetermined level (the .05 level of confidence) was found this was followed by the Tukey test to locate the difference within the set. It was recognized that the F test is designed to be used with interval data and some of the data in the study reported herein were ordinal (i.e., scores on the LBDQ-XII and the PTO); however, the impact on the F valLe has been considered by researchers to be relatively insignificant. To answer questions four and five the basic technique utilized was stepwise multiple regression. In the study, stepwise multiple regression was used to determine the extent to which teacher absenteeism could be explained by the presumed independent variables of leadership behavior, teacher morale, teacher's marital status, teacher's level of education, teacher's sex, teacher's experience in the school district, whether the teacher was employed in a Title I or a non-Title I school, whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absenteeism, and whether the teacher was employed in a large or a small elementary school. (Schools with high unexpected teacher absenteeism were those 5 schools classified as having the highest percentage of absenteeism in the district during 1978-79; the schools without high unexpected absenteeism were the other 10 schools included in the study.) In sum, stepwise multiple regression was applied for the total group of teachers participating in the study to determine to what extent teacher absenteeism could be explained by




23
these independent variables, with teacher morale treated as an independent variable in question four and ;,s a covariate in question five. This technique seemed appropriate since multiple regression can accommodate any number and type of independent variables, i.e., continuous, dichotomous, categorical. It is superior to frequency and percentage crossbreak tables as a means of analyzing non-experimental data. It yielded useful statistics for this stuoy; specifically, the overall variance accounted for (R2), the multiple correlation-coefficient
(R), regression coefficients, and F tests of significance. Though reliability is questionable in multiple regression due to the tendency for regression weights to change with different samples, with different numbers of independent variables, and with their order of entry in the equation, this can be enhanced through replication and sample size. Kerlinger and Pedhazur (1973, pp. 444-447) recommended at least 100 subjects for any analysis and as representative a sample as possible. Coupled with a fairly large sample size (423), the unreliability of regression weights did not constitute a problem because there were few independent variables (10 in question 4, and 9 in question 5) considered in the investigation, and the intercorrelations among these independent variables were low. The standard error of beta weights is .08 with samples of at least 100, and becomes .05 with samples of 200. With samples as large as 400 the standard error becomes .04. The standard error of beta weights can be as high as .25 with samples of 40. The higher the degree of intercorrelation, the more unstable the beta weights (Kerlinger, 1973, pp. 620-625).
The fundamental regression equation used was as follows: y (days absent) = b0 + b1 (systems orientation of the principal) + b2 (personal orientation of the principal) + b3 (PTO total score) + b4




24
(large school or small school) + b5 (unexpected absenteeism category of school high or not high) + b6 (Title I or non-Title I school) + b7 (sex) + b8 (marital status) + b9 (level of education) + bl0 (years of experience in the district) where b ... b 0 beta weights. Raw scores were used in the regression equation. Though the researcher acknowledged the fact that "the order which the independent variables are entered into the equation makes a great deal of difference in the amount of variance accounted for by each variable" (Kerlinger & Pedhazur, 1973, p. 95), order was less important when overall prediction was the main concern. The test variables were entered in the equation on a theoretical basis derived from the problem statement with systems orientation of the principal first, followed by personal orientation of the principal, and PTO total scores. The demographic variables--large or small school size, unexpected absenteeism category of the school, whether the school is a Title I or non-Title I school, sex, marital status, highest academic degree held, and years of experience in the district--were used as dummy variables to identify class membership.
The computer program Statistical Package for the Social Sciences (SPSS) was used to produce means and standard deviations for the dependent variables and each of the independent variables. In addition, to ascertain the degree of correlation among the variables and to detect multicollinearity an SPSS program which computed correlation coefficients was used. The importance of computing intercorrelations is related to the fact that if there are low intercorrelations among the variables, i.e., low multicollinearity, then the strength of the regression analysis is enhanced. On this point, Kerlinger and Pedhazur (1973) noted that independent variables are almost always correlated




25
to some extent in ex post facto studies; therefore, zero-order correlations were not to be expected.
Two SPSS regression programs were used. These programs included one with all 10 independent variables and one where the individual teacher's total PTO score was treated as a covariate. Using these programs, the multiple r, r square, adjusted r square, standard error, regression coefficient, residual, sum of squares, mean square, F values, B weights, and Beta weights were computed.
Organization of the Remainder of the Research Report
Chapter 11 contains a review of related literature. Chapter II I contains the presentation and analysis of the data. Chapter IV contains a summary of the findings, conclusions, and discussion.




CHAPTER 11
REVIEW OF RELATED LITERATURE
In the review of literature major emphasis is placed upon
(1) leadership behaviors which have been identified as important in an organization, (2) the role of perception in influencing behavior,
(3) morale as a factor in employee behavior, (4) the role of employee absenteeism as an indicator of organizational health, and (5) the relationships between and among leadership behavior, morale, jobrelated variables, personal characteristics, and employee absenteeism.
Organize tonal Leadership Behavior
While researchers have concluded that effective leadership is a necessary component of a successful organization, there has been difficulty identifying appropriate leadership behavior. It has been found that variations in the situation may predicate the leadership behavior which is most effective in accomplishing organizational goals (Fiedler, 1967, p. 110). Hersey (1967) advanced the concept of adaptive leader behavior:
The more a manager adapts his style of leader behavior to meet the particular situation and the needs of his followers, the more effective he will tend to be in reaching personal
and organizational goals. (p. 15)
Two major categories of leadership behaviors have been identified as important in contributing to organizational effectiveness: taskoriented behavior and people-oriented behavior. While it has been demonstrated that organizations have been effective with leadership which has been mainly task-oriented or people-oriented, researchers 26




27
have concluded that organizations are more effective under the integrating approach which emphasizes both task orientation and people orientation (Blake & Mouton, 1964; Halpin, 1966; Huse & Bowditch, 1973; Stogdill, 1974.
Several dimensions of leadership behavior were identified in studies originating in 1945 by the Bureau of Business Research at Ohio State University (Stogdill & Coons, 1957). Leadership was defined as the behavior of an individual when he or she is directing the activities of a group toward a goal attainment. The Ohio State University study group identified two major dimensions of leader behavior, initiating structure and consideration. Halpin (1959) further defined initiating structure and consideration:
Initiating Structure refers to the leader's behavior in delineating the relationship between himself and members of the work-group, and in endeavoring to establish well-defined patterns for organization, channels of communication, and methods of procedure. Consideration refers to behavior indicative of friendship between the leader and members of
his staff. (p. 4)
Halpin (1966, p. 47) stressed that the dimensions of initiating structure and consideration are not traits of leadership, but rather the behavior of the leader. In the former, a fixed capacity is inferred; in the latter, changes can be induced through appropriate training. In studying aircraft commanders, Halpin observed that the successful leader is one who promotes both group maintenance and group achievement.
Blake and Mouton (1964, p. 17) applied the task-oriented and people-oriented concept to the development of management programs for industrial organizations. Five different types of leadership based on concern for production (task) and concern for people (relationships) were identified. Blake and Mouton attempted to train individuals




28
for "team management" behavior wich exhibits maximum concern for both production and people. They contended that production and people concerns are complementary, and the integration of the two optimizes both task orientation and people orientation. Slake and Mouton asserted that the "team manager"--high on production and high on people--will always be the most effective type of leader regardless of the situation. They contended that their training program could convert any willing participant into a team-oriented manager. According to Bernardin and Alvares (1976), the claims of Blake and Mouton have been subjected to minimal empirical testing.
Likert (1961), in his studies in industrial settings, noted the need for an integrative approach to management when he proposed his linking-pin concept which combines both structural and humanistic perspectives. In the linking-pin approach, each work group has a head who is a subordinate in the next higher group. According to Likert the feeling or support is the key to high motivation within the work group. Likert explained his principle of supportive relationships:
The leadership and other processes of the organization must be such as to ensure a maximum probability that in all interaction and all relationships with the organization each member will, in the light of his background, values, and expectations, view the experience as supportive and one which builds and maintains his sense of personal worth and
importance. (pp. 102-103)
Fiedler (1967), in his studies of surveying parties, military combat crews, steel companies, basketball teams, and members of boards of directors, found that appropriate management behavior depends on the situation. In Fiedler's Leadership Contingency Model three situational variables seemed to determine the favorableness of the situation for the manager: leader-member relations, task structure,




29
and the leader's position power. Fiedler defined favorableness of a situation as "the degree to which the situation enables the leader to exert his influence over his group" (p. 13). In testing his theory over a period of 16 years Fiedler concluded that both task-oriented leaders and human relations -oriented leaders are successful under some conditions. Taking a simHar stand, Stogdill (1974, p. 26), based on a survey of the leadership research, cautioned that leaders cannot rely upon their characteristic pattern of behavior to sustain them when the demands of the situation increase in severity.
In tracing the historical evolution of management thought, Davis (1968) described four models of organizational behavior: autocratic, custodial, supportive, and collegial. Although managers may employ all of these behaviors at one time or another, Davis contended that their predominant behavior can be classified under one of the models. The autocratic model, prevalent in the early days of the industrial revolution, was still in operation in the late 1960's. Davis explained that the custodial model, a form of paternalism under which employees become concerned with maintaining benefits rather than with production, was not being advocated in the 1960's due to advanced knowledge of human behavior. According to Davis, under the supportive model a manager furnishes psychological support to his or her employees at work. Under Davis' collegial model, the nanagerial orientation is toward teamwork with managers building a feeling of mutual contribution among participants in the organization. Davis concluded that there is no one best method of managerial behavior since it depends on what is known about human behavior and the situation in which it occurs. It is important for the manager to understand the different results brought about by different models




30
and to vary the behavior with the total human and task conditions surrounding the work.
Reddin (1970, p. 88) stressed the importance of the situation in determining appropriate leader behavior. When the leader's behavior is appropriate in a given situation, it is effective; when the leader's behavior is inappropirate in a given situation, it is ineffective. Reddin added a third dimension to leadership models--the effectiveness dimension.
Stogdill (1963), reacting to previous studies, developed tile Leader Behavior Description Questionnaire-Form XII (LBDQ-XII) to measure the presence of the following patterns of leader behavior: representation, demand reconciliation, tolerance of uncertainty, persuasiveness, initiation of structure, tolerance of freedom, role retention, consideration, production emphasis, predictive accuracy, integration, and influence with superiors. Stogdill noted that both theory and research suggest the above patterns of behavior are involved in leadership, though they are not equally important in all situations.
Brown (1967), utilizing the LBDQ-XII and grouping each of the 12 dimensions of leader behavior into systems orientation or persons orientation, found these two factors account for 76% of the total factor variance. Brown noted the similarity between systems orientation and persons orientation and Getzels' (1952) nomothetic (emphasis on the institution) and idiographic (emphasis on the individual) dimensions and Halpin's (1959) initiating structure and consideration. Stodgill (1974, p. 41) pointed out that since Form XII was developed in 1962, very little research using it has appeared in publication. After reviewing the studies which suggested that the subscales of LBDQ-XII




31
are differentially related to different dimensions or organizations, Stogdill urged that further studies be made to uncover some heretofore hidden complexities of leader behavior and leader influence.
The Influence of Perception on Behavior
The perception that people hold of others may or may not be consistent with the actual behavior. Bruner and Tagiuri (1954, p. 83) concluded from their research that people's perception of others may infl,,ence their behavior.
Research in perception has pointed to the need to incorporate a human relations approach to management. Four of the seven metapsychological assumptions upon which Carl Rogers' (1951) theory rests provide the skeletal structure for the human relations approach:
Proposition 1: "Every individual exists in a continually changing word of experience of which he is the center" (p. 483).
Proposition 2: "The organism reacts to the field as it is experienced and perceived. This perceptual field is, for the individual, 'reality"' (p. 484).
Proposition 4: "The organism has one basic tendency in striving--to actualize, maintain, and enhance the experiencing organism" (p. 487).
Proposition 7: "The best vantage point for understanding behavior is from the internal frame of reference of the individual himself" (p. 494).
Bruner (1973, p. 161) saw perception as the product of stimulus determinants and of experiential, motivational, personal, and social factors as well. Bruner advocated continued research on perception, particularly the investigation of environmental cues which aid in interpersonal adjustment.




32
Tagiuri (1969, p. 27) in his investigations into perception identified the following as the main elements in the process of perception: the characteristics or state of the perceiver, the characteristics or state of the one perceived, the cues or the manifestations of the one perceived available to the observer, the cues of the one perceived utilized by the observer, the cognitive process of the perceiver in utilizing cues, and the resulting judgment by the perceiver. These elements have been studied in relationship to each other to determine whether perceived manifestations yield reliable and valid cues, which cues are used by the observer, and whether or not there is a valid connection between the behavior of the observed and the perception of this behavior by the observer.
Writers in the field of management, both industrial and educational, have reiterated the importance of the human aspect in interpersonal relationships and the role that perception plays in determining behavior. Combs (1970) pointed to the need for administrators to become more sensitive to the subordinates' view of the world about them. He maintained administrators' awareness of how their workers come to know and regard them can make the difference in their managerial effectiveness. Likert (1961, p. 67) pointed out that the superior needs accurate information as to how his or her behavior is actually seen by subordinates. Likert noted the need for periodic measurements of perception. Rogers (1962, p. 70) noted that teachers' perceptions of the principal may affect the principal's success in securing change. Croft (1965) concluded that sensitivity to the perceptions of others is an important dimension of effective supervision. Leavitt (1972, p. 111) stressed that the perceived world is the world that determines behavior.




33
Stogdill (1974, p. 87) pointed out that characteristics of the followers helped to determine the kinds of leadership characteristics that will be acceptable to the group. Reviewing sociometric studies of friendship and leadership, Stogdill concluded that groups tend to choose leaders with attitudes and personalities similar to their own.
Marquit (1968) stressed that supervisory stimuli must be received by the teacher if the principal is to be credited with affecting change. A teacher must perceive the stimulus if he or she is to respond to it.
Bruner and Tagiuri (1954, p. 98) cautioned that the accuracy in judging others is often a consensual kind, basically in agreement with others regarding a person's characteristics. From studies in perception they concluded: (1) accuracy is aided by similarity between judge and judged; (2) accuracy is dependent on having cues to work on; therefore, traits with little behavioral manifestation are poorly judged; (3) certain systematic errors in judgment--halo effect, logical error, and the like--account for much of the error involved in judgment; (4) there are systematic relationships between various personality variables and judging ability; and (5) a global or intuitive approach seems to improve judgment.
Knowledge of the forces at work in the forming of impressions can be an aid in interpersonal behavior. Managers need to know which behaviors need additional exposure and which need to be curtailed. The adaptive managerial behavior may alter the perception formed by observers.
March and Simon (1958, p. 79) pointed out that because of perceptual biases, observers cannot be truly objective about anything. March and Simon explained that perception depends on both




34
internal states (including values, goals, beliefs, and perceived relations between actions and their outcomes) and external states (including upbringing, reading habits, and leisure-time activities).
Study after study of managerial behavior has shown that the accuracy of observers' judgments of others is questionable. The descriptions by others of a manager's behavior seem potentially more accurate than self -descriptions as predictors of overall effectiveness (Campbell, Dunnette, Lawler, & Weick, 1970, p. 78).
Likert (1961) concluded that "the higher the productivity, the greater the accuracy of perceptions" (p. 49). In a review of administrative research, Erickson (1967) found that although teacher perceptions of administrative behavior cannot be considered completely accurate, they are a vital subject of investigation in that they represent one of many criteria in determining overall administrative effectiveness.
Employee Morale
The nature of morale has been a controversial subject among scholars. There are basically two opposing points of view that have extended over a number of years: one has held that morale is an individual phenomenon (Silverman, 1957, p. 271) and another has treated morale as a group phenomenon (Maier, 1955, pp. 109-111). ;n the field of education there has been a decided tendency to use the term to apply to both individuals and groups. Maier (1955) expressed the relationship between individual and group morale:
In describing an individual, one can speak of his attitude, his motivation, and his adjustment. When one attempts to describe a group, one uses the term morale and it communicates all of these things, but with a group reference. This means that the relationship between individuals is also part
of the meaning of morale. (p. 111)




35
There is also a tendency in the field of education to use the term morale interchangeably with job satisfaction and job attitude (Herzberg, Mausner, Snyderman, 1969). In the review that follows, with full recognition of the controversy in the methodological problems, the rubric of teacher morale is treated both with what some authorities would call job satisfaction and others teacher morale as both an individual and as a group phenomenon.
The quality of education received by pupils depends in large part upon the teacher in the classroom. Teacher morale is one determinant of the behavior of the teacher. Ellenburg (1973) expressed the importance of morale:
It is one of the factors which may determine whether a school functions at its best, demanding and receiving the utmost from its students, or whether it is happy just to see the passing of another day. Call it what you will. It is easy to overlook, yet it can make a school stand ahead of
the rest. (p. 7)
Herzberg, Mausner, and Snyderman (1969, p. 37) theorized that worker job attitude is determined by two distinct classes of variables: job satisfiers, referred to as motivators or intrinsic factors, and job dissatisfiers, referred to as hygiene or extrinsic factors. According to Herzberg and his colleagues, the dimensions of work, responsibility, and advancement are major motivators, while company policy and administration, supervision, working conditions, and pay are hygiene factors. The motivators seem to be effective in motivating people to effective performance; the hygiene factors prevent dissatisfaction.
In the discussion of the Herzberg theory presented by Sergiovanni and Carver (1975), the motivation-hygiene or satisfierdissatisfier theory is deemed applicable to educational organizations for two reasons:




36
1 The theory Is consistent with the humanistic belief
pattern which forms one dimension of our applied
science of educational administration.
2. When the Herzberg model is tested using teachers and
other educators as respondents, results similar to those found for other groups are obtained. (p. 72)
Maslow (1970, p. 12) and McGregor (1960, p. 27) were interested in behavior and attitudes as psychological factors for human existence in work. Maslow noted that a person's needs may be different at different times and developed a hierarchy of needs to include low order needs such as physiological and safety needs, and high order needs including belonging and affection, ego satisfaction, and self -actualization.
McGregor (1960, p. 30) developed what he called "Theory X" and "Theory Y" in which opposite attitudes toward work were described. In "Theory X" McGregor assumed that man was essentially lazy and had a dislike for work and must be coerced into work performance. in "Theory Y" he assumed that man found work natural and would seek responsibility given proper conditions.
The work accomplished by Maslow, McGregor, and Herzberg described the factors necessary to meet the psychological needs of employees, and it was implied that it was management's responsibility to consider these factors when evaluating employee behavior. One of the basic needs of an employee was a continuity, a predictability, and a routine of relationships in his life. Fox and Scott (1943, p. 67) reported management's responsibility was to help provide for this need or to risk the consequences, one of which was absenteeism. Argyris (1957, explained that "need" referred to something that existed "in" the personality which required a particular behavior/ condition to occur. A need was said to be related to all other needs




37
and to initiate and guide behavior until the goal of the individual was reached. According to Argyris, "the needs to be integrated, to feel successful, to feel wanted and so on are felt by people in our culture, and come from within their personalities" (p. 31).
Using the Purdue Teacher Opinionnaire (PTO) developed by Rempel and Bentley (1967), Henderson (1976) found a positive relationship between participation in decision-making and job satisfaction. Teachers' feelings about the professional competency of their principal, the principal's interest in teachers and their work, and his or her skill in human relations were also found to have a positive effect upon morale. The PTO measures individual teacher morale, but scores can be used as an aggregate of the group studied.
Shealy (1975) studied readily accessible data about classroom teachers to determine the usefulness of such data in predicting teacher morale. The morale score and personal information were obtained from each teacher participating in the study. Shealy concluded that the data which are readily accessible about teachers were indeed useful in predicting teacher morale. A score for the National Teachers Examination, type degree held by the teacher, and the number of education courses taken by the teacher made contributions to the predictive efficiency of teacher morale. Shealy also found that the older, more experienced teachers with more education courses and advanced degrees tended to have higher morale scores. Males tended to score lower than did the females and the elementary teachers had higher morale scores than did the middle and secondary teachers.
The locale of the school is another aspect to be considered when studying teacher morale. Cook (1971) used the PTO to measure the morale of teachers in inner-city conventional school settings. He




38
reported that the mean scores of teachers in inner-city schools were significantly lower on three dimensions of the PTO. These dimensions were support among teachers, satisfaction with teaching, and community support of education. Teachers' perceptions of their principals' consideration affected their satisfaction with teaching.
In another study using the PTO, Herrmann (1971) found positive correlations between teacher morale and teacher involvement in school policy determination. He also discovered a positive correlation between faculty involvement and principal -teacher rapport.
When teacher morale and authoritarianism and the authoritarian level of administrators were examined by Gubser (1969), older teachers scored higher in morale factors and authoritarianism than did the younger teachers. This led Gubser to conclude
principals may have an indirect effect on the level of authoritarianism of their faculties if they discourage younger, more anti -authoritarian teachers from remaining on their
faculties. (p. 38)
Relationships between teacher morale and verbal teacher behavior were investigated by Greenwood and Soar (1973). They used the Reciprocal Category System, a modification of Flanders Interaction Analysis to determine teacher verbal behavior and the PTO to measure teacher morale. The product moment correlations developed by these researchers indicated:
1. Teacher morale was positively related to satisfaction
with teaching, teaching load, curriculum issues, and
support of education by the community.
2. Teacher morale was negatively related to teacher talk.
3. The amount of pupil talk followed by more pupil talk
was positively related to satisfaction with teacher
salary and teacher load.
4. The percentage of teacher acceptance behaviors was
positively related to satisfaction with teaching and
rapport among teachers. (p. 105)




39
Burgett (1977) studied the morale of 400 Montana teachers and found it was possible to predict the morale of those teachers by the use of the teachers' perceptions of the level of agreement between the school board and the superintendent on certain administrative practices. He also pointed out
there were three areas of morale that were most affected by the administrative practices: teacher rapport with the principal, curriculum issues, and community support of
education. (p. 54)
Several other studies which used the PTO dealt with the behavior of pri ncipals or supervisors. In one study examining selected principal behavior as seen by teachers and principals, Pennebaker (1970) suggested these generalizations:
1. Where teachers perceive their principal as setting a
good example by working hard, they will exhibit
higher morale.
2. Where teachers perceive their principal as being highly
considerate, they exhibit higher morale.
3. Principals feel that they are more considerate than
their teachers perceive them to be.
4. Teacher morale is not adversely affected when teachers
perceive their principal as being aloof and acting the
role of the "straw boss."
5. Teachers with high morale are more apt to keep abreast
of current professional problems.
6. The more aloof, close supervising principals had teachers who spent more time preparing for teaching.
(p. 36)
Kokovich (1969) investigated the leader behavior of some secondary school principals to determine whether any relationships existed between perception of leader behavior and teacher morale. He measured leader behavior with the LBDQ-XII and teacher morale with the PTO. He concluded that the leader behavior most highly associated with satisfaction is consideration. Kokovich said, "the behavior of




40
the principal does have a strong relationship to morale status, but it is not the sole determinant of morale status" (p. 24).
In another study using the PTO and the LBOQ-XII, Lambert (1969) found that teachers who perceived their principals as being above the mean on leader behavior perceived their own morale as being significantly higher than was the morale of those teachers who perceived leader behavior of their school principal as being below the mean. Perry (1977) used the PTO and the Executive Professional Leadership Scale (EPLS) to determine whether relationships existed between teacher morale and the principal's attempts to improve teacher performance. Perry found that each of the 10 morale factors of the PTO was significantly correlated with the principal's professional leadership as determined by the EPLS.
Blumberg and Weber (1968) studied the possible relationship that might exist between the manner in which a teacher perceives his or her supervisor's behavior and teacher morale. These authors found that the behavioral style of the supervisor as seen by the teacher was related to the morale of that teacher.
Numerous other studies dealing with teacher morale which did not employ the PTO were also found. In a study of 776 teachers, Schultz (1952) reported that the administrator is most often identified as the focal point of the teachers' satisfaction or dissatisfaction.
Gragg (1955) found that the most frequently mentioned contributor to teachers' high morale in his study was "confidence in the leadership of my principal and other administrators" (p. 494).
An 81-item questionnaire was administered by Silverman (1957) to a group of teachers with instructions to respond in relationship to the way they thought the item would affect morale. The teachers indicated




41
that 85% of the items dealing with a principal's character or behavior would have some effect on teacher morale. Silverman concluded that the principals represented the strongest morale-influencing factor and the principal is in a position to bring about change.
The principal's role in promoting teacher morale was also examined by Gezi (1962). Subjects were asked to respond to the openend question, "What do you think the principal should do in order to obtain and maintain his teachers' morale?" In discussion sessions, the teachers defined morale as
the group's attitude and feelings which are conducive to their working harmoniously and effectively with each other and with the administration so that by achieving the organization goals many of their individual goals may also be
achieved. (p. 88)
Three hundred and eighty suggestions were made for fostering teacher morale in response to the question posed. The findings indicated that in order for the respondents to be emotionally secure and in order for them to gain recognition as members of the school team, they must find in their principal the qualities which they perceived to be necessary for the achievement of their goals and those of the school.
Redefer (1962) provided insight concerning the selection of principals. As a result of his studies, he found that principals in high morale schools received their satisfaction from helping the faculty to grow professionally. The principals of low morale schools received satisfaction from the power of the position.
In a survey of 400 teachers, Prewett (1956) found that many of the barriers contributing to teacher inefficiency and low teacher morale are trivial and could be removed by an administrator who is people conscious and thinks in terms of personalities. Campbell




42
(1959) found that the highly satisfied teachers he studied consistently made references to certain qualities of their principals such as, "understanding, competent, courteous, makes teachers feel worthy, and guides teachers in such a way as to make it easy for them to teach" (p. 2). The dissatisfied teachers mentioned such factors pertaining to their work as, "lack of staff participation, too many clerical duties, shortage of supplies, and displeasure with teaching assignments" (p. 2).
Another study of factors affecting teacher morale was conducted by Suehr (1962), who identified some characteristics of high and low morale teachers. The high morale teachers had taught longer, felt they had fulfilled their parents' expectations for them, and most often came from upper or upper-middle class homes. Those teachers who felt they were not realizing their fullest potential were in the low morale group. This group indicated that an opposite-sexed parent had influenced them more, considered their self-confidence to be greater, and rated their level of ambition high.
Principals of open climate schools were researched by Tirpak (1971) and were found to be warmhearted, good-natured, sociable, and attentive to people. The open climate schools were also found to be high in esprit or morale and were staffed by faculties who characterized the principals by their high degree of emotional stability, frustration tolerance, and calm and realistic approach to life.
Perceptions of teachers of organizational climate and the reinforcing behavior of the principal were studied by Nelson (1972) using the Organizational Climate Description Questionnaire. He noted that teachers tended to perceive an open climate in schools led by principals whom the teachers perceived to reflect a high level of reinforcement behavior.




43
Absenteeism as an Indicator of Organizational Health
A certain number of legitimate absences from a job are due to illness, obligations, or emergencies. A high rate of absenteeism, however, may be an indicator of a problem within an organization. Herzberg -(1966, p. 113) and Vroom (1964, p. 79) concluded that absenteeism correlates negatively with indices of worker morale. Mann and Baumgartel (1953, p. 47) concluded that workers who had low absence rates were relatively more satisifed with their jobs, supervisors, and work associates.
Argyris (1964) found that absenteeism is an adaptive activity of the employee. Just as some employees learn to accept the work environment, some are "also capable of modifying their working world so that they can express some of their frustrations, decrease them, or partially avoid them" (p. 59). Kornhauser (1962, p. 26) found that people who have a high absenteeism rate tend to be frustrated on the job by poor supervision and non-challenging routine work.
Likert (1967, p. 85) observed that excessive absence tends to be highest when people are free to move under an exploitive-authoritative style of management. Abse.nteeism is likely to be moderately high under a benevolent-authoritative style of management, moderate under consultation-authoritative management, and low under a participative approach to management. Likert further noted than an unfavorable trend in worker attitude will, in turn, be manifested in increased absence. He observed that the factor "time" plays an integral part in the manisfestation of excessive absence.
When hostile attitudes, uncooperative motivation, and distrust
of management are widespread and deep-seated, it often requires years for even an extremely competent manager to
bring about any substantial improvement. (p. 88)




44
Ronan and Prien (1971, p. 67) found that excessive absenteeism indicates that an employee will eventually leave the company. Merritt (1974) found that management personnel believed that employee absenteeism was due to changing attitudes and the changing environment. Workers were no longer satisfied with
just an adequate income and work in a safe, acceptable working environment. They expect to be able to obtain some job satisfaction for the time and energy they contribute to the job. (p. 177)
Workers were no longer ingrained with the attitude that the job must be accomplished above all else.
Metzner and Mann (1953) found that frequency of absence proved to be more successful than the length of absence in yielding relationships between satisfaction and absenteeism for white-collar workers. They concluded that frequency of absence yielded more significant relationships with satisfaction because it minimized the effective 3bsenteeism due to prolonged illness. Metzner and Mann recommended the use of the frequency of absence measure for further research into the relationship between absenteeism and other concepts.
Hersey and Blanchard (1972, p. 74), in reviewing research, pointed out that absenteeism is one of the many end-result variables which may be indicators of deterioration of work attitudes. Merritt (1974) observed that government statistics demonstrated that the number of employed workers contributing to absenteeism has increased from 4.3% of the total employed workers in 1967 to 5% in 1972. In other words, 2,800,000 workers were absent at any one time in 1967; 3,500,000 were absent at any one time in 1972, an increase of 700,000 workers.




45
Leadership Behavior, Morale, Job-Related Variables,
Personal Characteristic Variables, and Employee Absenteeism
In studying the management of organizations, Likert (1967) identified three sets of variables for use in research and operating purposes: causal variables, intervening variables, and end-result variables. The condition of the intervening variables was affected by the causal variables and, in turn, the intervening variables influenced the end-result variables.
Included in the causal variables were those independent variables which influenced developments within the organization and which could be altered or changed by the organization and its management. Among the causal variables, Likert listed managerial skills and managerial behaviors. Among the intervening variables, which reflected the internal state of the organization, were worker perceptions and attitudes. Among the end-result variables, the dependent variables, was absenteeism, a performance variable.
As managers learn more about the influence of the variables and adapt their behavior accordingly, there should be an improvement in the positive direction of the end-resuit variables (Likert, 1967, p. 121). Likert added that efforts to improve the end-result variables by altering the intervening variables will usually be less effective than the changing of causal variables.
When the effectiveness of an organization is evaluated, it is usually the end result or output alone which is considered (Hersey & Blanchard, 1972, p. 97). Reddin (1970, p. 41) stated that the effectiveness of a manager should be measured objectively by the




46
maximum output. It is necessary to measure all three sets of variables periodically in an organization to detect trends which may influence output, according to the research done by Likert (1961, p. 101).
Likert (1967, p. 38) cautioned that the causal, intervening, and end-result variables comprise a complex network with many interdependent relationships, sometimes making it difficult to classify organizational variables into categories. He added that the employment of managerial principles will prove to be causal in most situations.
While researchers concluded that no one causal factor alone could account for differences in employee morale and performance (Fiedler, 1967; Hersey & Blanchard, 1972; Likert, 1967), the identification of those conditions under which certain types of managerial behavior would be more likely to adversely affect worker behavior could provide additional insight.
Leadership Behavior
In describing a case study of management concern about the absenteeism of manufacturing plant workers, Covner (1950, p. 336) reported that when an attitude survey was conducted, the unfavorable attitude of workers toward management was the strongest predictor of absenteeism. Mann and Baumgartel (1953, p. 121) indicated that, among plant workers, low absenteeism was noted where workers reported that the foreman (a) created an atmosphere which contributed to free and easy discussion of work problems, (b) had time to talk to his men about personal problems, (c) held group discussions with his men, (d) could be counted to "go to bat" or "stand up" for his men.




- 47
Metzner and Mann (1953, p. 481) concluded that satisfaction with leader behavior was negatively related to the absence of white and blue-collar males studied, but had no effect on the absence of whitecollar female examined. Gerstenfeld (1969, p. 55) reported in a study of female production workers that there was a strong relationship between the worker's attitude toward her immediate supervisor and her absences. Those workers who felt that the boss was frequently unfair were generally the same workers with poor records of attendlanre. Johns (1978, p. 443) investigated three factors relating to the influence of the supervisor on the absenteeism of 208 operative employees in a manufacturing plant. Satisfaction with leader behavior was found to be negatively related to absence frequency, but unrelated to time lost. Consideration, one of the leadership style variables analyzed, was negatively related to both absence frequency and time lost. Initiating structure, the other leadership style factor studied, was negatively related to time lost, but unrelated to absence frequency.
Much of the research conducted among school employees contradicts the belief that satisfaction with leader behavior exerts an appreciable impact on absenteeism. Schroeder (1977, p. 83) reported no significant relationship between 12 patterns of the principal's managerial behavior and the absence rate of 96 teachers in eight schools randomly selected from a total population of 3,800 teachers in a school system located in metropolitan New Orleans. In addition, satisfaction with supervision, as measured by the Job Description Index, was found to have no relationship to teacher absence frequency. In a study of 1,500 Pennsylvania teachers, Slick (1974, p. 109) found that teacher absence frequently was not significantly related to the level




48
of rapport with the principal or the perceived level of aloofness, production emphasis, thrust, or consideration on the part of the principal.
Morale
There is a large body of existing literature dealing with the relationship of morale and employee absenteeism. I lgen and Hollenbach (1977) noted in their study of these two variables
although most of the reseach is correlational, the general model assumes that employees appro.'ch (attend) jobs perceived to lead to satisfaction and avoid (are absent from)
jobs perceived to lead to dissatisfaction. (p. 149)
The earliest known study in this area, according to Mochinsky (1977, p. 317) was conducted by Kornhauser and Sharp in 1932. No statistical analyses were included in their results, which focused on a sample of female factory workers, but they concluded that a slight relationship existed between job attitudes and lost time. Likewise, Noland (1945, p. 508) found a negative relationship existed between overall morale and the absence rate among a sample of industrial workers, although again no statistical procedures were presented. Kerr, Koppelmeier, and Sullivan (1951, p. 54) noted that the morale of employees in manufacturing departments was negatively related to unexcused absenteeism, but positively related with total absenteeism.
When frequency of absence was used in their study, Metzner and Mann (1953, p. 483) reported a negative correlation between overall morale and the absence of blue and white-collar workers. When absence was measured by a count of actual days lost, no relationship was found between these two factors. They contended that these findings would most likely appear when absence indices are used which increase the weighting factors of persons who have irregular attendance patterns and decrease the weighting factors of absences




49
caused by illness. Morale was negatively related to absenteeism in studies by van Zeist and Kerr in 1953 (for manufacturing workers), Fleishman, Harris, and Burtt in 1955 (for production workers), Harding and Bottenberg in 1961 (for airmen), Lundquist in 1958 (for Swedish factory workers), and Talacchi in 1960 (for office workers).
Bass (1965) wrote that certain work groups typically are absent more frequently than others. In particular, he noted that poor morale traditionally has been related to the absenteeism of "lower-skilled" employees, but not of more highly-skilled white-collar workers or women. A possible explanation for this finding, he explained, is that
high status white-collar employees have more freedom to use other forms of withdrawal when dissatisfied, like taking extra long coffee breaks or three-hour lunch periods.
(p. 107)
Ilgen and Hollenback (1977, p. 159) found that, for the most part, morale was unrelated to the absenteeism of a sample of 164 clerical workers at Purdue University. The results of a study by Garrison and Muchinsky (1977) involving 174 accounting department workers produced mixed results. Overall morale, as measured by the Job Description Index, was found to be a significant negative predictor of absenteeism without pay, but it had no relationship to absenteeism with pay.
In contrast to the wealth of research conducted on morale and absenteeism in business and industry, only three researchers have investigated this relationship as it affects educational personnel. Teacher morale was found to be significantly related to teacher absenteeism by Coller (1975, p. 131). He learned that the low absence teachers he studied tended to have higher teacher morale scores than did the high absence teachers. In a sample of 1,500 Pennsylvania teachers, Slick (1974, p. 90) found an inverse relationship between




50
teacher absence frequency and the perceived level of teacher morale. Douglas (1976, p. 83) reported that morale was found to be a significant predictor of absenteeism in a study of 154 teachers in Central Ohio.
Job-Related Variables
The job-related variables of work unit size and type of school were found to relate consistently with worker absenteeism. Covner (1950, p. 336), who studied 38 groups of plant and office workers, reported a positive relationship existed between work unit size and employee absenteeism. Argyle, Gardner, and Cioffi (1958, p. 36), who studied workers in 86 production departments, and Indik and Seashore (1961, p. 22), who studied 91 groups of factory workers, found similar positive correlations.
One of Gibson's (1968) conclusions in his study of staff absence in Boston, Massachusetts, area schools was that:
In small systems as compared with large systems, absence is more a function of the total social system while in large systems, absence is more associated with the characteristics
of the subsystem work group. (p. 7)
According to Bridges and Hollinan (1978, p. 41), subunit size and work system interdependence had direct, independent effects on the absenteeism of teachers in 57 California and Wisconsin elementary schools. Subunit size was positively associated with absenteeism and work system interdependence was negatively related to absenteeism.
In the Chicago, Illinois, public schools, staff absenteeism in elementary schools was lowest in the smallest schools in 1959-1960, but no other consistent relationship seemed to occur (Chicago Board of Education, 1960, p. 12). Data for February, 1960, indicated that the lowest average amount of sick leave taken was for elementary




51
schools with a staff of less than 10 (0.56 days). Elementary school teachers in eight larger staff size categories used between 0.89 and
0.72 sick leave days.
A relationship may be found between the type of school in which a teacher works and his or her absence rate. The State of New York Office of Education Performance Review (1974) reported that the absence rate for New York City teachers in Title I elementary schools was 29 percent higher than in non-Title I elementary schools during the 1972-1973 school year. As the report concluded
This leads to the inference that where the educational need is greatest, teacher absenteeism is highest, especially
absence over which the teacher has control. (p. 14)
Douglas (1976) attempted to relate various social -psychological factors, generally considered to be sources of teacher stress, to work attendance and the excessive use of sick leave. Twenty-seven predictor variables were analyzed by stepwise multiple regression using the Wherry program MULREG. He discovered that inner-city vs. suburban school was one of nine variables found to be predictors of absenteeism when added in stepwise regression. Personal Characteristic Variables
The impact of personal characteristic variables on employee absenteeism has been investigated by behavioral scientists. Literature related to an employee's sex, marital status, level of education, and employment experience as potential variables impacting absenteeism was reviewed and is reported in the paragraphs which follow.
The available research indicates that female employees have higher rates of absence than males, although men seem to be absent for longer periods of time than women. Hedges (1973, 1975) reported




52
that data contained in the Bureau of Labor Statistics Current Population Surveys indicated that the absence rate for females was approximately twice that for males, in part-week absences in 1972 and 1974. Females were also absent significantly more often than males for full-week absences in 1967, 1972, and 1974. Hedges (1977, p. 22) and Taylor (1979, p. 51) found that the incidence of absence and the proportion of available time lost was much higher for females than males in May 1976 and May 1978. But where time lost by absent
workers was measured, female workers lost a smaller proportion of their usual weekly hours than males.
The results from three studies were mixed. Kerr, Koppelmeir, and Sullivan (1951, p. 52) indicated that female production workers were absent more than males for total absences and certified (previously approved) absences, but that males were absent more than females for uncertified absences. Garrison and Muchinsky (1977, p. 228) found that female white-collar workers had significantly more paid absences than males, yet there was no correlation between sex and upaid absence. Johns (1978, p. 442) reported that women manufacturing. workers studied had a significantly higher frequency of absence than men, but that there was no relationsihp between sex and total time lost. However, he concluded from regression analysis that sex was the single best predictor of absence among job satisfaction, six personal characteristics variables, two leadership style ~variables, and six job content variables examined. In a study of workers at a light engineering factory, Martin (1971, p. 86) reported that the demographic factors of sex and age seemed more important than education or marital status when correlated with absenteeism. In all categories of absence, females were absent more often than males.




53
The Philadelphia Suburban School Study Council (1970, p. 15) conducted a survey on teacher absenteeism covering school years 19681969 and 1969-1970. Fifty-six Pennsylvania school systems in five study councils participated in the study. The researchers found that women were absent more than men in the "average" district. Female teachers averaged 6.95 days of leave, with pay and without pay in 1968-69 and male teachers, 7.54 days. Females also had a higher mean rate of absence--on the average, females were absent 3.73% of the total work days, compared to a rate of 1.92% for males.
Female teachers were absent significantly more than male teachers according to Marlin's (1976, p. 78) study of 425 teachers in a semirural school system. The Pennsylvania School Boards Association (1978, p. 23) found that female teachers were absent .97% more often than male teachers in school year 1977-1978. A Dade County, Florida, Public Schools Office of Management and Budget study (1978, p. 18) contained data indicating in school year 1977-1978 female teachers took approximately 70% of their sick and personal leave, where male teachers took approximately 30%.
Redmond (1978, p. 111) found that of 10 demographic variables studied, only gender was significantly related to the absence of professional personnel in the Fort Madison (Iowa) Community School District, i.e., women were absent more frequently than men. In conjunction with his study of the effects of stress on absence, Sylwester (1979, p. 20) reported absence rates by sex for a sample of 335 Oregon elementary and secondary school teachers and administrators. Male educators were absent an average of 3.6 days in school year 1977-1978, compared to 5.4 days for female educators. The average number of days absent for the total group was 4.7 days.




54
Although research findings appear to link sex and absenteeism, care should be taken in interpreting these findings. The United States Department of Labor Statistics Report warned that other factors may influence the sex-absenteeism relationship, such as age, marital status, and occupation. Occupation is especially critical to this relationship. Traditionally, more females have been new hires in the lower skilled, lower pay positions, two factors regularly associated with relatively high rates of absence. Similar groupings of employees should be considered when analyzing sex and absence. Fewer women are employed in high salary positions usually associated with lower absence. "Sex differences in absence rates narrow when comparisons are made within a particular occupation group," observed Hedges (1973), a Bureau of Labor Statistics economist, "even though within the group men tend to occupy the better paying jobs" (p. 28). Isambert-Jamati (1962) found that, even if they have several children to raise, highly-trained women in responsible positions are absent infrequently.
Researchers have found a relationship between absence and an employee's marital status. Jackson (1944, p. 293) reported that married men with several dependents had a steadier attendance record than either single or married men with no children. Shepherd and Walker (1957, p. 271) found that single males were absent the most among iron and steel workers studied. Hedges (1973, p. 28) found that married males had a lower part-week absence rate than single males, but married females had a higher rate than single females.
Taylor (1979, p. 51) reported from Bureau of Labor Statistics data that men who were never married had a higher absence rate than married men in May, 1978; the reverse was true for women during




55
this period. Married women also had a higher percent of time lost for both illness and injury and miscellaneous reasons than did single women. "Family duties may be one factor causing their higher absence rates," Taylor noted, "although absences due to child care activities and other family-related responsibilities could be expected to appear as miscellaneous reasons" (p. 51).
Sharpies (1973, p. 98), conducting research on classified civil service personnel in Ohio, found that both the high absence male and high absence female categories studied contained a higher percent of married employees than the low absence groups for males and females. Martin (1971, p. 87) reported that, of the light engineering workers he studied, single workers were absent more than married workers. Marital status was found to be a significant negative predictor of absence without pay, but not with pay, in a study of accounting workers by Garrison and Muchinsky (1977, p. 228).
Lee (1960, p. 23), in a study conducted in the Wichita, Kansas, public schools for school year 1959-1960, reported that 34.6% of the male teachers, 25.3% of the single female teachers, and 16.5% of the married female teachers had perfect attendance records. While more married females than single females were absent less than 5 days, more single females were absent 15 days or more. Coller (1975,
p. 93) found that married teachers tended to have lower absence records than single teachers in the Livonia, Michigan, public schools. However, Marlin (1976, p. 88) reported that the mean absence rate for married teachers was higher than for unmarried teachers in a semi-rural Mississippi school system. No relationship was found between marital status and teacher absence in Bridges and Hollinan's (1978) study in California and Wisconsin, Bundren's (1974) study in




56
Clark County, Nevada, Marchant's (1976) study in Richmond, Virginia, or Redmond's (1978) study in Iowa.
Education level, as a personal characteristic variable which may impact employee absenteeism, has been reported in numerous studies focusing on business and industrial employees with no consistent relationship emerging. Noland (1945, p. 508) reported that absenteeism among a sample of industrial workers was inversely related to the workers' education level. However, neither the absence measure used nor the range of education was described. Raouf (1973, p. 41) found that in the companies he surveyed in Windsor, Ontario, employees without high school diplomas generally had the highest rate of absence. Sharples (1973, p. 99) found that high absence females in his study of classified civil service personnel in Ohio had more formal education than low absence females. Martin (1971, p. 92) reported that light engineering workers with more than eight years of education appeared to have more absences of all types than workers with less education.
Citing data from the Current Population Survey of the Bureau of Labor Statistics, Taylor (1979, p. 52) reported that a negative relationship existed between education level and absenteeism in the United States work force in May 1978. The incidence rate (percentage of workers absent) was highest for workers who completed elementary school (8.3%), followed by high school graduates~ (6.6%), and college graduates (4.1%). The inactivity rate (percentage of aggregate time lost) for workers with an elementary school education was 4.9%; for high school graduates, 3.5%; and for college graduates, 2.1%.
Few studies focusing on the relationship of education level and teacher absenteeism have been reported. Lee (1960, p. 8), in a




57
study conducted in the Chicago, Illinois, public schools during February 1960, indicated that absenteeism decreased as the amount of education increased. Teachers with a bachelor's degree on the average took 0.78 days of sick leave; teachers with a master's degree, 0.73 days; teachers with a master's degree plus 30 hours, 0.67 days; and teachers with a doctor's degree, 0.12 days. The caution was offered that the type of school, age, and sex may have complicated these findings. "Academic degree" was one of nine variables that Douglas (1976, p. 87) found to be predictors of absenteeism when added in a stepwise multiple regression. However, there was no correlation between education level and absenteeism of professional personnel in the Fort Madison (Iowa) Community School District (Redmond, 1978, p. 113).
Conflicting results have been reported from studies focusing on the relationship of employment experience to absenteeism. A negative correlation existed between length of employment and absence of white-collar males in a study by Metzner and Mann (1953, p. 483), but there was no relationship between these variables for blue-collar males. Hill and Trist (1955, p. 148) reported that no relationship existed between employment experience and the absence rate of factory workers in a longitudinal study. Kahn (1957, p. 92) found that the longer the period of service in an organization, the lower the absence frequency, regardless of age.
Martin (1971, p. 89) noted that, for the British light engineering workers studied, a positive relationship between employment experience and absence was found for males; however, no significant correlation was noted for females. Raouf (1973, p. 41) noted that the highest absence rates began after the first six months of employment




58
and ended after the third year on the job. Little absence was found in the early days of employment in his sample of factory workers in Windsor, Ontario. Sharples (1973, p. 108) indicated that, in his study of classified civil service personnel employed at Ohio University, high absence females spent fewer years on the job than low absence females. Nicholson and Goodge (1976, p. 252) indicated that younger, short-service female food processing employees were most likely to have high levels of casual and unsanctioned absence. Two measures of sickness absence were found to increase with longer service among the oldest group and decrease with longer service among the youngest group.
After the effects of pay and age were partialled out, a negative relationship between employment experience and absence was found in Bernardin's (1976, p. 90) study of 109 male white-collar sales workers. Garrison and Muchinsky (1977, p. 227) found a significant positive relationship existed between length of employment and the paid absences of 195 accounting workers; however, a significant negative relationship was found between length of employment and unpaid absences.
Studies have also been conducted to measure the impact of the years of employment experience on the absence rates of educational personnel. Gibson and Lafornara (1972, p. 17) reported that in a 30-year longitudinal study of teacher absenteeism in a single school system from 1938-1939 to 1968-1969, "newcomers" (teachers with up to 10 years of service) were absent more often than "continuing" teachers (those with 10 or more years of service). Newcomers were seen as the bearers of "creeping legitimacy," whereby absence norms shifted from illness to other reasons as newcomers entered the school system and influenced the absence patterns of continuing personnel.




59
Stallings (1959, p. 62) reported little difference in the use of sick leave by permanent and probationary teachers in 16 Southern California school systems. For the 1955-1956 school year, permanent teachers used about a half a day or more sick leave than probationary teachers, on the average. An examination of sick leave days taken by teachers in Chicago, Illinois, in February, 1960, showed that the amount of sick leave taken generally increased with experience. Teachers with 1-2 years of experience averaged about .50 days of sick leave during this month; teachers with 4 years and those with 10-20 years of experience averaged .75 days; teachers with 21-35 years of experience averaged over .80 days; and teachers with 36 or more years of experience averaged 1 day of sick leave (Chicago Board of Education, 1960, p. 12).
In his sample of teachers in Livonia, Michigan, Coller (1975, p. 94) fnund that teacher absenteeism was significantly related to years of teaching experience in a curvilinear fashion. Teachers with 2-4 and 23-25 years of teaching experience tended to have low absence records. Marlin (1976, p. 87) reported that the mean rate of absenteeism for teachers with four or more years of experience was slightly higher than for teachers with less than four years of experience in a semirural Mississippi school system. Douglas (1976, p. 89) reported that "years of teaching experience" was one of nine variables that was found to be a predictor of teacher absenteeism when added in a stepwise regression, and one of five variables included in a "predicti-. profile" of likely high-absence teachers.
In 1977-1978, teacher absenteeism in Dade County, Florida, was influenced by employees' years of experience. Teachers with 1-3 and 7-9 years of experience used an average of 15% of their sick and




60
personal leave. Teachers with 4-7 years of experience used slightly more than 20% of this leave, and teachers with 11 or more years of experience, about 45% (Dade County, Florida, Public Schools Office of Management and Budget, 1978, p. 12).
In contrast, teacher absenteeism was not found to be significantly related to length of continuous employment in Clark County, Nevada, according to Bundren (1974, p. 121), nor to previous educational experience, according to Marchant (1976, p. 98). Redmond (1978, p. 114) concluded that neither the amount of teaching experience in the district nor total teaching experience was significantly related to the absenteeism of professional personnel in the Fort Madison (Iowa) Community School District.




CHAPTER III
PRESENTATION AND ANALYSIS OF DATA
The purpose of the study was to determine the extent which the unexpected absenteeism of teachers in selected elementary schools within the Orange County, Florida, Public School System and the unexpected absenteeism of individual teachers could be explained by teachers' perceptions of the leadership style of principals, teacher morale, and selected teacher-related and school-related characteristics. There were five basic questions which gave direction to the study. The first three questions were preliminary and used single variant analysis to determine if there were differences among certain subsets of what were thought to be major presumed independent variables in regard to the dependent variable of teacher absenteeism. Specifically, question one was designed to determine if there were differences in teachers' perceptions of leadership style of the principals on each of the 12 dimensions of the LBDQ-Xlf among 15 selected elementary schools with high, average, and low unexpected teacher absenteeism. Question two was designed to determine if there were differences on each of 10 dimensions of teacher morale as measured by the PTO among the 3 groups of elementary schools. The intent in question three was to determine if there were differences in the teachers' perceptions of the leadership style of principals in the 5 schools with the highest morale scores and the 5 schools with the lowest morale scores from the 15 schools selected for the study. Questions four and five related to the extent to which the presumed independent variables 61




62
relating to the teacher's perception of the system's orientation of the principal, the personal orientation of the principal, the teacher's total morale score, and teacher-related and school-related demographic variables could be used to explain individual teacher's unexpected absenteeism. Questions four and five were different in that morale was considered an independent variable in question four and was held as a covariant in question five. As previously noted, two basic statistical techniques were utilized--one-way analysis of variance and stepwise multiple regression.
From the 67 elementary schools in the district, those five with the highest percentage of unexpected teacher absenteeism, those five with the lowest percentage of unexpected teacher absenteeism, and those five nearest the mean percentage of unexpected teacher absenteeism rate for all elementary schools for 1978-1979 school year were selected to participate in the study. The researcher solicited cooperation from the 496 teachers employed in the 15 schools and 423 teachers provided usable data.
The required materials for the research included two instruments--the Leadership Behavior Description Questionnaire-Form XII (LBDQ-XII) and the Purdue Teacher Opinionnaire (PTO). The
LBDOQ-XI was used to measure the independent variables related to teachers' perceptions of the leadership style of a principal, and the PTO was used to measure the independent variable of teacher morale. The school-related and teacher-related demographic variables were obtained from teachers by means of a researcher-developed personal data sheet.




63
The focus of the present chapter is a presentation and analysis of the data in regard to the aforementioned questions. However, before turning directly to the results of the statistical analysis, in order that the reader may have a perspective for understanding the analysis, descriptive data relative to the independent and dependent variables are presented. This is followed by comparisons in relation to the leadership style of the principals and teacher morale among elementary schools having high, average, and low unexpected teacher absenteeism. The third section is focused upon comparisons between high and low morale elementary schools relative to teachers' perceptions of the leadership style of the principals, and the fourth section deals with the prediction of individual unexpected teacher absenteeism f rom the presumed independent variables.
In Table 1 each school is described by number of certificated staff members in the school, the number of study participants, the mean unexpected absenteeism rate for the school, and whether it was a Title I or non-Title I school. As denoted in the table, the range by school was 20-47 certificated staff members %-Oth 8 of the schools classified as small and 7 as large schools. The number of voluntary participants at the 15 schools varied from 19 to 40, with the percentage of participants ranging from 70% in school 10 to 100% in school 12. Overall, 423 of the 496 teachers (85%) employed in the 15 schools participated. (To strengthen validity, a t test was utilized to determine if there was a significant difference between the mean absenteeism rate of respondents and non -respondents. No significant difference was found at the .05 level of confidence.) The mean unexpected teacher absenteeism ranged from a low of 3.61 days in school 3 to




64
Table 1
Basic Information Relative to the 15 Elementary Schools Selected for the Study
School Size Mean Unexpected
School a Certificated Study Absenteeism Rate Title I Non-Title I
Number aStaff Members Participants (in days) School School
1 47 (L) b 40 7.37 X
2 34 (L) 28 4.67 X
3 41 (L) 38 3.61 X
4 32 (S) 28 4.83 X
5 37 (L) 26 4.05 X
6 32 (5) 25 7.03 X
7 33 (S) 26 4.83 X
8 34 (L) 29 6.84 X
9 20 (S) 19 11.10 X
10 33 (S) 23 7.08 X
11 25 (S) 22 11.52 X
12 28 (5) 28 12.00 X
13 37 (L) 35 6.97 X
14 27 (S) 24 10.70 X
15 36 (L) 32 13.70 X
Total 496 423 11 4
a The school number is merely a designation used to protect the
anonymity of the school.
bThe L designation indicates the school was classified as Large for purposes of the study; the S indicates the school was classified as Small.




65
13.70 days in school 15. Eleven of the 15 schools were classified as Title I.
In order to evaluate properly the findings, it was necessary to collect descriptive data on the participants as well as compute standard descriptive statistics for the variables. The next section is focused on a descriptive analysis of the participants, the independent and dependent variables, and the schools.
Descriptive Data by Schools Relative to Teacher-Related Variables
To obtain descriptive statistics, the data were first aggregated by the subgroups--high absenteeism schools, average absenteeism schools, and low absenteeism schools. The means and standard deviations for each dimension of the LBDQ-XI I and PTO were computed. Also determined were distributions by teacher sex, education level, marital status, and experience in the school district. In the paragraphs that follow, these data are presented for high absenteeism schools, average absenteeism schools, low absenteeism schools, and for the total group of schools.
Presented in Table 2 are data regarding the independent variable, leadership style of the principal as perceived by the teachers. Data for each of the 12 dimensions of the LBDQ-XII and for the system and personal orientation categories are presented.
The reader is reminded that the systems orientation consisted of the following five dimension scores--demand reconciliation, initiation of structure, production emphasis, integration, and superior orientation.




Table 2
LBDQ-XII Dimension Means, Orientation Means, and Standard Deviations for the 15 Elementary Schools Included in the Study
LBDQ-XII Dimension Means and Standard Deviations
Demand Tolerance
Representation Reconciliation of Uncertainty Persuasiveness
School Standard Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation
High 9 13.68 3.98 12.89 1.85 25.37 3.35 33.21 4.14
Absenteeism 11 14.81 1.56 13.23 1.69 27.32 2.42 35.23 2.52
Schools 12 17.71 4.31 14.25 5.34 31.50 7.99 29.61 8.71
14 13.62 1.66 12.91 1.91 28.29 2.53 30.42 6.39
15 14.25 2.42 12.97 2.02 26.81 2.67 32.34 3.33
Average 1 15.45 3.67 13.43 3.81 27.98 6.58 33.00 6.68
Absenteeism 6 15.48 4.90 15.84 4.88 33.16 6.87 31.08 8.09
Schools 8 20.38 3.12 21.07 3.63 37.86 6.83 39.41 6.81
10 19.35 3.23 15.43 5.34 28.91 6.74 28.30 8.25
13 15.42 1.97 12.77 1.59 28.31 2.08 34.71 2.72
Low 2 17.75 2.88 16.29 4.79 30.71 7.29 35.89 6.34
Absenteeism 3 15.68 2.34 12.32 1.69 28.50 2.83 35.13 3.47
Schools 4 14.75 6.25 15.32 5.38 26.93 9.94 27.96 11.31
5 19.96 3.17 19.73 3.22 36.31 5.18 37.92 5.11
7 15.00 5.02 16.04 5.81 32.23 8.89 32.23 9.21
Total 16.22 4.08 14.87 4.52 29.98 6.82 32.22 7.24




Table 2
(continued)
LBDQ-XII Dimension Means and Standard Deviations
Initiation of Tolerance of Role
Structure Freedom Assumption Consideration
School Standard Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation
High 9 30.79 4.33 32.37 5.33 26.16 2.41 26.42 3.44
Absenteeism 11 32.95 2.84 33.77 3.53 26.64 1.87 26.73 2.99
Schools 12 32.18 9.69 36.68 8.73 28.18 10.34 34.96 8.43
14 26.75 5.10 31.29 5.54 25.83 2.35 25.08 3.03
15 30.06 3.44 30.84 5.54 27.78 3.32 27.38 3.03
Average 1 33.48 7.19 33.13 8.30 27.07 6.51 28.55 7.16
Absenteeism 6 33.48 8.38 38.72 4.24 31.16 10.04 31.60 8.86
Schools 8 40.86 4.72 42.83 4.68 41.69 6.45 41.24 5.60
10 32.83 8.73 33.43 7.36 30.87 8.16 31.57 7.29
13 33.54 2.75 32.89 5.47 25.63 2.49 28.60 2.49
Low 2 37.54 6.56 36.64 5.12 35.61 7.94 33.93 7.02
Absenteeism 3 33.97 3.36 34.68 5.00 26.79 2.55 26.45 2.65
Schools 4 32.18 10.16 28.29 12.57 28.14 12.78 28.25 11.82
5 41.00 4.51 38.46 7.08 40.92 6.63 38.96 5.57
7 33.81 8.44 30.38 8.88 32.12 6.52 26.58 10.72
Total 33.79 7.24 34.28 7.72 30.17 8.40 30.38 8.03




Table 2
(continued)
LBDQ-Xll Dimension Means and Standard Deviations
Production Predictive Superior
Emphasis Accuracy Integration Orientation
School Standard Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation
High 9 30.32 3.28 17.53 2.91 14.05 2.22 32.00 4.08
Absenteeism 11 29.23 3.56 17.64 2.01 14.27 1.55 34.36 3.09
Schools 12 28.57 7.75 15.11 4.31 15.07 4.89 26.89 11.35
14 25.42 4.46 16.88 2.56 13.25 2.36 30.25 4.73
15 29.81 2.83 19.06 2.03 14.28 1.94 32.03 3.06
Average 1 29.48 6.83 17.40 3.43 14.93 4.23 34.75 6.83
Absenteeism 6 30.36 7.62 16.32 4.91 15.56 5.32 29.12 10.50
Schools 8 32.45 4.59 19.86 2.81 21.10 4.43 34.59 9.79
10 30.35 7.73 13.96 3.80 13.83 4.46 30.35 11.80
13 31.31 3.49 19.03 2.02 15.17 1.64 33.43 3.16
Low 2 31.64 6.76 16.61 3.82 16.75 3.87 30.43 7.01
Absenteeism 3 30.68 3.17 18.76 2.19 '15.82 1.96 32.66 3.84
Schools 4 27.79 9.07 13.64 5.49 16.21 3.93 25.29 9.79
5 33.88 5.67 18.15 2.84 18.85 3.54 34.73 8.03
7 27.77 5.26 16.31 4.25 14.08 5.53 23.62 11.25
Total 29.99 6.01 17.22 3.80 15.61 4.12 31.15 8.35




Table 2
(continued)
LBDQ-XII Orientation Means and Standard Deviations
Systems Orientation Personal Orientation Total
School Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation
High 9 120.05 9.73 174.74 14.23 262.79 19.81
Absenteeism 11 124.05 8.43 182.14 7.45 271.81 12.50
Schools 12 116.96 35.36 193.75 47.76 283.82 73.37
14 108.58 14.39 171.42 15.82 249.75 24.64
15 119.16 8.45 178.47 12.09 265.59 15.77
Average 1 126.05 26.42 182.58 38.46 273.88 58.22
Absenteeism 6 124.36 31.02 197.52 38.89 292.76 62.53
Schools 8 150.07 20.32 243.28 29.25 358.76 40.45
10 122.78 32.50 186.39 38.76 278.83 61.58
13 126.23 8.29 184.60 10.33 277.40 14.66
Low 2 132.64 17.90 207.14 27.57 309.36 43.90
Absenteeism 3 125.45 9.77 186.00 11.91 278.79 17.01
Schools 4 116.79 33.60 167.96 66.78 259.46 92.98
5 148.19 16.45 230.69 25.42 344.15 34.53
7 115.32 30.94 184.84 47.48 276.54 68.59
Total 125.42 24.48 191.48 38.73 285.75 56.24




70
The personal orientation consisted of the remaining seven dimension scores- -representation, tolerance of uncertainty, persuasiveness, role assumption, consideration, predictive accuracy, and tolerance of f freedom.
It can be seen in Table 2 that mean scores for schools in the high absenteeism group for the representation dimension of leadership were, with the exception of one school, below the mean score for the total population surveyed. Scores for three of the five schools in the average and low absenteeism groups were also below the mean scores for this dimension. Mean scores for all schools in the high absenteeism group for the demand reconciliation dimension were below the mean for the total population, while scores for three of the average absenteeism schools and four of the low absenteeism schools exceeded the total population mean score. Four schools in the high absenteeism group were below the total population mean for the dimension tolerance of uncertainty, while two schools in the average absenteeism group and three schools in the low absenteeism group were above the total mean. Three schools in each group were above the total population mean on the persuasiveness dimension. All schools in the high absenteeism group and four schools in the average absenteeism group were below the total population mean on the initiation of structure dimension, while four schools in the low absenteeism group exceeded the total mean score. Four schools in the high absenteeism group and three schools in the average absenteeism group were below the total mean on the tolerance of freedom dimension, and three schools in the low absenteeism group were above the total mean. All schools in the




71
high absenteeism group were below the total mean on the dimension role assumption, while three schools in the average absenteeism group and three schools in the low absenteeism group were above the mean for the total population. Four schools in the high absenteeism catetory were below the total mean for the consideration dimension and three schools in the average absenteeism group and two schools in the low absenteeism group were above the total mean. Four schools in the high absenteeism group also were below the total mean on the production emphasis dimension while four schools in the average absenteeism group and three schools in the low absenteeism group exceeded the total mean score. However, for the predictive accuracy dimension, three of the high absenteeism schools and average absenteeism schools were above the total mean, and three of the low absenteeism schools were below the total mean. All schools in the high absenteeism group and three schools in the average absenteeism group were below the total mean on the integration dimension while four of the low absenteeism schools exceeded the total population mean. Three of the schools in the high absenteeism and the average absenteeism categories were above the total mean on the superior orientation dimension while three of the low absenteeism schools were below the total mean.
For the systems orientation of the principal dimension all of the schools in the high absenteeism group were below the total population mean. Three of the schools in the average absenteeism group and three in the low absenteeism group were above the total mean for this orientation. For the personal orientation of the principal, four schools




72
in the high absenteeism group and three schools in the average absenteeism group and three schools in the low absenteeism group were below the total population mean. Examining the LBDQ-XII total mean scores for each group it can be seen that all schools in the high absenteeism group were below the total population mean score, while one school in the average absenteeism group and two schools in the low absenteeism group were above the total population mean score. The mean leadership score for each dimension was low compared to the scores obtained by the groups upon whom the instrument was field tested.
Presented in Table 3 are data regarding each of the 10 dimensions of the PTO and a total morale score for each school is shown. It can be noted that one school in each category -school 12 in the high absenteeism category, school 8 in the average absenteeism category, and school 3 in the low absenteeism category- -consistently scored in a dissimilar manner from other schools in their categories. For the teacher rapport with principal dimension, four schools in the high absenteeism category were below the mean score for the total population, three schools in the average absenteeism category were above the total mean, and four schools in the low absenteeism category were above the total mean. Four of the high absenteeism category schools were below the total mean, and three of the average absenteeism category schools and four of the low absenteeism category schools were above the total mean for the satisfaction with teaching dimension. Four of the high absenteeism category schools were below the total mean on the rapport among teachers dimension, while three




Table 3
PTO Dimension Means, Total Means, and Standard Deviations for the 15 Elementary Schools Included in the Study
PTO Dimension Means and Standard Deviations
Teacher Rapport Satisfaction Rapport Teacher
with Principal with Teaching Among Teachers Salary
School Standard Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation
High 9 21.42 2.39 20.68 3.67 17.47 2.01 6.89 1.70
Absenteeism 11 20.05 3.23 23.59 2.84 17.32 2.17 6.73 2.00
Schools 12 58.90 11.36 66.18 9.17 39.71 6.05 17.46 3.97
14 21.88 3.84 22.29 2.39 15.08 2.87 7.83 1.27
15 22.81 2.28 21.34 3.24 17.16 3.34 7.50 1.14
Average 1 27.70 15.97 29.38 15.37 21.85 9.93 9.43 4.32
Absenteeism 6 42.24 21.08 48.40 22.32 33.56 15.39 13.60 5.79
Schools 8 69.28 10.90 70.41 6.91 50.41 5.34 17.66 2.83
10 50.22 12.78 63.17 9.17 37.30 7.18 16.70 3.78
13 22.60 3.36 23.51 3.04 17.86 2.61 7.46 1.20
Low 2 53.71 21.13 54.11 20.32 38.46 14.94 15.07 5.70
Absenteeism 3 26.24 2.77 24.84 2.40 19.39 1.72 8.63 1.28
Schools 4 55.29 11.72 62.43 12.30 45.04 5.33 19.29 3.85
5 65.50 11.18 64.92 11.16 44.81 7.15 17.38 4.43
7 45.69 19.63 66.58 14.41 39.85 11.08 16.73 4.80
Total 39.74 20.91 43.29 22.59 29.95 14.33 12.40 5.78




Table 3
(continued)
PTO Dimension Means and Standard Deviations
Curriculum Community Support
Teacher Load Issues Teacher Status of Education
School Standard Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation
High 9 23.89 3.20 9.89 2.11 9.42 1.80 8.42 2.65
Absenteeism 11 23.77 2.89 9.91 1.11 9.86 1.52 9.14 2.92
Schools 12 27.68 5.49 15.50 3.44 24.14 5.25 14.93 2.65
14 23.04 2.56 10.38 2.10 8.83 1.86 7.63 3.50
15 22.84 3.19 9.81 2.02 9.09 2.02 7.47 2.59
Average 1 24.08 5.63 10.65 3.13 11.58 5.01 9.60 2.37
Absenteeism 6 25.12 6.83 11.88 4.77 16.68 8.01 8.72 3. 06
Schools 8 31.97 4.92 18.10 1.84 22.90 5.21 15.07 2.43
10 26.87 7.46 14.61 3.31 20.43 6.60 13.26 2.90
13 23.51 3.11 9.49 1.63 10.31 1.55 9.06 2.15
Low 2 27.25 4.80 14.54 3.62 18.07 7.48 9.64 3.29
Absenteeism 3 23.53 2.39 9.74 1.46 10.32 1.47 9.21 1.74
Schools 4 29.21 7.04 16.07 3.23 23.32 7.07 15.64 2.59
5 28.15 5.86 15.35 3.19 20.81 6.63 13.77 3.36
7 30.77 7.43 14.54 4.65 23.38 7.99 13.58 4.01
Total 26.01 5.79 12.58 4.06 15.72 7.83 10.95 3.95
A.




Table 3
(continued)
PTO Dimension Means and Standard Deviations
School Facilities Community Total
and Services Pressures Mean Morale
School Standard Standard Standard
Category Number Mean Deviation Mean Deviation Mean Deviation
High 9 5.47 0.84 8.21 2.10 131.79 12.63
Absenteeism 11 5.50 0.96 8.00 1.89 lt, 3.86 10.01
Schools 12 14.61 2.73 14.43 1.48 293.54 36.88
14 5.46 0.66 8.08 2.12 130.50 12.28
15 5.56 1.32 8.00 1.81 131.59 12.63
Average 1 6.55 4.27 9.33 2.44 160.13 63.00
Absenteeism 6 10.16 4.57 12.97 3.98 223.32 86.01
Schools 8 17.14 2.46 15.76 2.23 328.69 27.64
10 14.65 2.53 14.83 2.52 272.04 46.73
13 5.66 1.06 7.94 1.75 137.40 11.70
Low 2 13.04 4.89 13.64 4.05 257.54 80.37
Absenteeism 3 5.58 0.72 7.50 2.00 144.97 7.39
Schools 4 14.21 5.60 13.68 3.61 294.18 49.97
5 14.12 3.59 15.00 2.83 299.81 44.36
7 13.46 4.43 15.54 3.48 280.12 60.61
Total 9.89 5.35 11.35 4.16 211.90 86.71
Ln




76
schools in the average absenteeism category and four schools in the low absenteeism category were above the total population mean. For the teacher salary dimension, four of the high absenteeism category schools were below the total mean, while three of the average absenteeism category schools and four of the low absenteeism category schools were above the total mean. Four of the high absenteeism category schools and three of the average absenteeism category schools were below the total mean on the teacher load dimension, while four of the low absenteeism category schools were above the total population mean. For the curriculum issues dimension, four of the high absenteeism category schools and three of the average absenteeism category schools were below the total mean, and four of the low absenteeism category schools were above the total mean. Four of the high absenteeism category schools were below the total mean, while three of the average absenteeism category schools and four of the low absenteeism category schools were above the total population mean. For the community support of education dimension, four of the high absenteeism category schools and three of the average absenteeism category schools were below the total mean, while three of the low absenteeism category schools were above the total mean. Four of the high absenteeism schools were below the total mean for the school facilities and services dimension and for the community pressures dimension, and three of the average absenteeism schools and four of the low absenteeism schools were above the total mean. The low
absenteeism category schools exhibited, as a group, the highest total morale scores as measured by the PTO. Average absenteeism category schools had the second highest group morale scores, with high




77
absenteeism category schools attaining the lowest morale scores. if the aforementioned schools--school 12, school 8, and school 3--were treated as outliers, the within and between group comparisons would become even more distinct.
Shown in Table 4 are data regarding each of the four teacherrelated independent variables--sex, education level, marital status, and experience in the school district. As one reviews the table, it can be seen that 44 of the teacher participants in the study were male and 379 female. Two hundred and seventy had bachelor's degrees, 151 had master's degrees, 1 had a specialist's degree, and 1 had a doctorate. One hundred and twenty were single and 303 were married. One hundred and thirty-two teachers had 1-5 years experience in the school district, 123 had 6-10 years experience, 74 had 11-15 years experience, and 94 had more than 15 years experience in the district. In terms of the biographical characteristics of the teachers, the three categories of schools were similar, and within group differences were more pronounced than differences among groups. In terms of teacher characteristics the high absenteeism category schools were most different from average absenteeism and low absenteeism category schools in the percentage of teachers with 15+ years in the district. Twelve percent of the teachers in the high absenteeism schools had 15+ years of experience, compared to 24% of the teachers in the average absenteeism schools, and 29% in the low absenteeism schools.
Comparisons Among Elementary Schoois with
High, Average, and Low Unexpected Teacher Absenteeism
As has been noted, one-way analysis of variance (F test) was used to analyze the extent of the differences in principals' leadership




Table 4
Number of Teacher Participants, Male-Female Ratios, Distribution by Education Level, Marital Status, and Experience in the School District for the 15 Elementary Schools Included in the Study
Teachers'
Number of Teachers Education Level of Teachers Marital Status
School
Category Number Male Female Total Bachelors Masters Specialist Doctorate Single Married
High 9 2 17 19 15 4 0 0 3 16
Absenteeism 11 1 21 22 14 8 0 0 5 17
Schools 12 0 28 28 18 10 0 0 6 22
14 2 22 24 18 6 0 0 11 13
15 4 28 32 24 7 0 1 5 27
Average 1 5 35 40 17 23 0 0 10 30
Absenteeism 6 4 21 25 15 10 0 0 7 18
Schools 8 0 29 29 15 14 0 0 6 23
10 5 18 23 18 5 0 0 8 15
13 6 29 35 25 10 0 0 9 26
Low 2 4 24 28 21 7 0 0 15 13
Absenteeism 3 4 34 38 27 11 0 0 10 28
Schools 4 2 26 28 17 11 0 0 7 21
5 2 24 26 10 15 1 0 9 17
7 3 23 26 16 10 0 0 9 17
Total 44 379 423 270 151 1 1 120 303




Table 4
(continued)
Teachers' Experience in School District
School
Category Number 1-5 years 6-10 years 11-15 years 15+ years
High 9 7 8 3 1
Absenteeism 11 4 5 10 3
Schools 12 5 9 7 7
14 14 5 1 4
15 16 11 5 0
Average 1 10 11 10 9
Absenteeism 6 8 5 4 8
Schools 8 3 8 4 14
10 12 9 0 2
13 9 14 9 3
Low 2 9 12 4 3
Absenteeism 3 16 11 5 6
Schools 4 4 3 3 18
5 7 6 7 6
7 8 6 2 10
Total 132 123 74 94
-.4




80
styie and teacher morale, among the three groups of schools. The focus herein is on those comparisons relative to mean scores on the LBDQ-XII and the PTO for schools grouped by high, average, and low unexpected teacher absenteeism.
Table 5 displays indicators of the teachers' perceptions of the principals' leadership style among the three groups of schools. Statistical comparisons among the groups, with F values and level of significance, are shown for each dimension of the LBDQ-XII instrument. It can be seen that a significant statistical difference, at the .05 level of confidence, was found for 10 of the 12 dimensions when groups of schools were compared. As noted previously, when a significant difference was found among the groups, the Tukey test was used to locate the differences. Two dimensions--persuasiveness and predictive accuracy--were not statistically different among groups at the .05 level.
For the representation dimension, the high absenteeism schools were significantly different at the .05 level from schools with average or low absenteeism. There was no significant difference between average and low absenteeism schools on this dimension. The same pattern existed among the three groups for the dimensions of demand reconciliation, tolerance of uncertainty, initiation of structure, role assumption, and integration. The difference between high absenteeism schools and low absenteeism schools was not significant at the .05 level for the tolerance of freedom dimension. Schools with average teacher absenteeism had a higher mean score on this dimension which was significantly different from both high absenteeism and low absenteeism schools. For the consideration and production emphasis




Table 5
LBDQ-XII Dimension Means, F Values, and Observed Alpha Level for Schools Classified as High, Average, Low in Teacher Absenteeism
Mean Schools Mean Schools Mean Schools Observed
LBDQ-XII with High with Average with Low F Alpha
Dimensions Absenteeism Absenteeism Absenteeism Value Level
Representation 14.92 16.98 16.54 9.83* .0001
Demand Reconciliation 13.28 15.43 15.63 11.56* .0000
Tolerance of Uncertainty 28.01 30.93 30.67 7.67* .0005
Persuasiveness 32.16 33.30 33.83 2.60 .0749
Initiation of Structure 30.52 34.80 35.53 20.16* .0000
Tolerance of Freedom 32.98 35.88 33.73 5.53* .0042
Role Assumption 27.04 30.77 32.20 14.15* .0000
Consideration 28.37 31.94 30.47 6.96* .0011
Production Emphasis 28.66 30.74 30.36 4.59* .0106
Predictive Accuracy 17.24 17.31 16.69 1.37 .2539
Integration 14.22 16.09 16.30 10.65* .0000
Superior Orientation 30.94 32.82 29.57 5.81* .0032
*2 <.05




82
dimensions there was no significant difference between high absenteeism schools and low absenteeism schools, nor was there a significant difference between average absenteeism schools and low absenteeism schools. However, for both of these dimensions there was a significant difference at the .05 level between average absenteeism schools and high absenteeism schools. There was no significant difference between high and low absenteeism schools on the superior orientation dimension, but there was significant difference at the .05 level between average and low absenteeism schools.
Shown in Table 6 are the results of a one-way analysis of variance which provided comparative statistics among the three groups of schools on the basis of unexpected teacher absenteeism category and teachers' scores on the PTO. The reader is reminded that the PTO instrument was used to measure teacher morale. Mean scores, F
values, and significance levels are shown for each dimension of the instrument. The differences among groups for each of the 10 dimensions were found to be significant at the .05 level of confidence. Again, where significant differences were found the Tukey test was used to locate the differences.
For the dimensions, teacher rapport with principal, satisfaction with teaching, rapport among teachers, teacher salary, curriculum issues, teacher status, community support of education, and school facilities and services significant differences at the .05 level were found between the following pairs: high absenteeism schools--low absenteeism schools, high absenteeism schools- -average absenteeism schools, and low absenteeism schools --average absenteeism schools. For the dimensions teacher load and community pressures, significant




Table 6
PTO Dimension Means, F Values, and Observed Alpha Level for Schools Classified as High, Average, Low in Teacher Absenteeism
Mean Schools Mean Schools Mean Schools with High with Average with Low Observed
PTO Teacher Teacher Teacher F Alpha
Dimensions Absenteeism Absenteeism Absenteeism Value Level
Teacher Rapport with 30.01 40.25 47.53 26.59* .0000
Principal
Satisfaction with Teaching 31.86 44.09 52.23 31.51* .0000
Rapport among Teachers 21.88 30.64 36.13 39.71* .0000
Teacher Salary 9.56 12.32 14.91 33.16* .0000
Teacher Load 24.28 26.04 27.44 10.46* .0000
Curriculum Issues 11.22 12.60 13.72 13.58* .0000
Teacher Status 12.60 15.62 18.49 20.89* .0000
Community Support of 9.60 10.92 12.11 14.41* .0000
Education
School Facilities and 7.54 10.18 11.58 21.49* .0000
Services
Community Pressures 9.48 11.66 12.63 21.79* .0000
*p <.05




84
differences were found between high absenteeism schools and low absenteeism schools and high absenteeism schools and average absenteeism schools. No significant difference was found between low absenteeism schools and average absenteeism schools for these two dimensions.
Mean scores in low absenteeism schools were higher on each of the 10 dimensions than were the means in the average and high absenteeism schools. The average absenteeism schools also had higher means on every dimension than did the high absenteeism schools. The greatest difference in mean scores occurred between high absenteeism schools and low absenteeism schools for the dimensions of teacher rapport with principal (30.01--47.53), satisfaction with teaching (31.86--52.23), rapport among teachers (21.88--36.13), and teacher salary (9.56--14.91).
Comparisons Among High and Low Morale Elementary Schools
Relative to Teachers' Perception of
the Leadership Style of the Principal
As has been noted, one-way analysis of variance (F test) was used to analyze the differences in the teachers' perceptions of the leadership style of principals in high and low morale elementary schools. The schools were grouped according to the five highest morale schools and the five lowest morale schools. (Four of the five low morale schools were also four of the five schools with the highest unexpected teacher absenteeism.) Where a significant difference at the predetermined level (the .05 level of confidence) was found, the Tukey test was employed to locate the difference within the set.




85
Displayed in Table 7 are F values and dimension means, orientation means, and total mean scores for the two groups of schools--high morale and low morale--on the LBDQ-Xll. It may be observed that four of the high morale schools were above the total population mean (based on data from all 15 schools) while all of the low morale schools were below the total mean on the representation and demand reconciliation dimensions. Four of the high morale schools and three of the low morale schools were above the total population mean on the tolerance of uncertainty dimension. However, on the persuasiveness dimension three of the high morale schools were below the total mean, while four of the low morale schools were above the total mean. Three of the high morale schools and all of the low morale schools were below the total mean on the initiation of structure dimension. For the tolerance of freedom and role assumption dimensions, three of the high morale schools were above the total population mean, while all of the low morale schools were below the total mean. Four of the high morale schools were above the total mean on the consideration dimension and all of the low morale schools were below. Three of the high morale schools were above and three of the low morale schools were below the total population mean on the production emphasis dimension. The high morale schools were below the low morale schools on predictive accuracy, with three high morale schools being below the total mean score and four of the low morale schools being above. Three high morale schools were above the total mean while all of the low morale schools were below the total mean on the integration dimension. On the final dimension, superior orientation, two of the high morale schools and four of the low morale schools were above the total




Table 7
F Values and LBDQ-XII Dimension, Orientation, and Total Means for Schools Classified as High and Low in Teacher Morale
Dimension Means
School Demand Tolerance of
Category Number Representation Reconciliation Uncertainty Persuasiveness
High 4 14.75 15.32 26.93 27.96
Morale 5 19.96 19.73 36.31 37.92
Schools 8 20.38 21.07 37.86 39.41
10 19.35 15.43 28.91 28.30
12 17.71 14.25 31.50 29.61
Low 9 13.68 12.89 25.37 33.21
Morale 11 14.81 13.23 27.32 35.23
Schools 13 15.42 12.77 28.31 34.71
14 13.62 12.91 28.29 30.42
15 14.25 12.97 26.81 32.34
Total Populationa 16.22 14.87 26.98 32.22
F Valueb 9.83* 11.56* 7.67* 2.61
Observed Alpha Level .0001 .0000 .0005 .0749
*p <.05




Table 7
(continued)
Dimension Means
School Initiation of Tolerance of Role
Category Number Structure Freedom Assumption Consideration
High 4 32.18 28.29 28.14 28.25
Morale 5 41.00 38.46 40.92 38.96
Schools 8 40.86 42.83 41.69 41.24
10 32.83 33.43 30.87 31.57
12 32.18 36.68 28.18 34.96
Low 9 30.79 32.37 26.16 26.42
Morale 11 32.95 33.77 26.64 26.73
Schools 13 33.54 32.89 25.63 28.60
14 26.75 31.29 25.83 25.08
15 30.06 30.84 27.78 27.38
Total Population 33.79 34.28 30.17 30.38
F Value 20.16* 5.53* 14.16* 6.97*
Observed Alpha Level .0000 .0042 .0000 .0011
*p <.05




Table 7
(continued)
Dimension Means
School Production Predictive Superior
Category Number Emphasis Accuracy Integration Orientation
High 4 27.79 13.64 16.21 25.29
Morale 5 33.88 18.15 18.85 34.73
Schools 8 32.45 19.86 21.10 34.59
10 30.35 13.96 13.83 30.35
12 28.57 15.11 15.07 26.89
Low 9 30.32 17.53 14.05 32.00
Morale 11 29.23 17.64 14.27 34.36
Schools 13 31.31 19.03 15.17 33.43
14 25.42 16.88 13.25 30.25
15 29.81 19.06 14.28 32.03
Total Population 29.99 17.22 15.61 31.15
F Value 4.60* 1.38 10.66* 5.82*
Observed Alpha Level .0106 *.2539 .0000 .0032
*2 <.05




Table 7
(continued)
Orientation Means Total Means
School
Category Number Systems Personal
High 4 116.79 167.96 259.46
Morale 5 148.19 230.69 344.15
Schools 8 150.07 243.28 358.76
10 122.78 186.39 278.83
12 116.96 193.75 283.82
Low 9 120.05 174.74 262.79
Morale 11 124.05 182.14 271.81
Schools 13 126.23 184.60 277.40
14 108.58 171.42 249.75
15 119.16 178.47 265.59
Total Population 125.42 191.48 285.75
F Value 14.02* 30.94* 33.97*
Observed Alpha Level .0002 .0000 .0000
a Refers to the mean computed for all 15 schools in the study b Refers to the computed F value when the high and low morale schools were compared
*2 <.05
OD




90
population mean. Two of the high morale schools and one of the low morale schools were above the total population mean on the systems orientation section of the LBDQ-XII. Three high morale schaels were above the total mean on the personal orientation section while all of the low morale schools were below this mean. Two of the high morale schools were above the total population mean and all of the low morale schools were below the mean on the total instrument score.
The high morale schools were significantly different at the .05 level from the low morale schools on 10 of 12 dimensions of the LBDQ-XII--representation (F value--9.83), demand reconciliation (F value--11.56), tolerance of uncertainty (F value--7.67), initiation of structure (F value--20.16), tolerance of freedom (F value--5.53), role assumption (F value--14.16), consideration (F value--6.97), production emphasis (F value--4.60), integration (F value--10.66), and superior orientation (F value--5.82). In each instance the means in the high morale schools were higher. No significant difference at the .05 level between the groups was found for the persuasiveness (F value--2.61) and predictive accuracy (F value--1.38) dimensions. The scores in the high morale schools on the systems (F value--14.02) and personal (F value--30.94) orientations were significantly higher at the .05 level than the scores in the low morale schools. The total LBDQ-XII mean scores were also significantly higher in the high morale schools than in the low morale schools (F value--33.97).
Prediction of Unexpected Teacher Absenteeism From Selected Variables
Presented in Table 8 are the results of the regression analysis using all 10 independent variables to predict the extent to which




Table 8
Results of Multiple Regression Analysis of 10 Independent Variables as Predictors of Individual Teacher Unexpected Absenteeism
Standard Error
Regression of Regression Beta F
Variable Coefficient Coefficient Weight Value
Systems Orientation of Principal -0.108 0.020 -0.095 0.96
Personal Orientation of Principal -0.088 0.014 0.045 0.18
Teacher PTO Total Score -0.039 0.004 0.110 2.63
Teacher Education Level
Bachelor -0.034
Master 0.037 0.507 0.024 0.25
Specialist -0.032
Doctorate 0.010 4.816 -0.021 0.21
Teacher Sex Female -0.029 0.766 -0.042 0.87
Male 0.029
Teacher Marital Status
Single -0.174 0.537 -0.090 3.66
Married 0.174
Teacher Experience
in the District 1-5 years -0.173 0.622 -0.153 7.42
6-10 years 0.060
11-15 years 0.037 0.713 -0.015 0.08
15+ years 0.094 0.691 0.063 1.28
Title I School -0.108 0.629 0.149 7.12
Non-Title I School 0.108 to




Table 8
(continued)
Standard Error Regression of Regression Beta F
Variable Coefficient Coefficient Weight Value
School Size Small 0.157 0.10 0.197 6.31
Large -0.157
School Absentee- Low -0.222
ism Category Average -0.033 0.581 0.131 5.83
High 0.266 0.829 0.396 28.90
Notes:
Multiple R =.42 R2 =18
F with 16, 406 df 1.69*
*e (15




Full Text

PAGE 1

LEADERSHIP BEHAVIOR, TEACHER MORALE, AND UNEXPECTED TEACHER ABSENTEEISM IN SELECTED ELEMENTARY SCHOOLS IN AN URBAN SCHOOL DISTRICT BY JAMES DONALD SHAW A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 1980

PAGE 2

ACKNOWLEDGEMENTS The writer wishes to express his sincere appreciation to the Superintendent of the Orange County Public School System, Dr. James L. Schott, and to the principals and teachers who made this study possible through their cooperation and participation. Gratitude also is expressed to Dr. Charles Walton and Dr. Wilfred Still from the district office who assisted the researcher with the techn i cal requirements of this project. A special express i on of gratitude is extended to Dr. Michael Nunnery for his guidance and encouragement during the completion of the required course work and the writing of this dissertation. The writer is also grateful to the other members of his committee, Dr. James Hale and Dr Joseph Shea, for their advice and assistance. Special acknowledgement is also due to the writer's friend and colleague, Dr. Robert Cascaddan, who shared commuting and class time and who provided encouragement and understanding Very special than ks are expressed to the writer s wife, Susan and his daughters, Betsy and Stacey, for their many sacrifices, faith, and patience dur i ng the completion of all the degree requirements i i

PAGE 3

TABLE OF CONTENTS Page ACKNOWLEDGMENTS LIST OF TABLES ABSTRACT CHAPTER ii i ii vi II Ill INTRODUCTION The Problem . . . . . . . . 5 Statement of the Problem . . . . . 5 Delimitations . . . . . . . 7 Limitations . . . . . . . 8 Justification for the Study . . . . 9 Definition of Te r ms . . . . . . . 12 Procedures . . . . . . . . . 17 Setting of the Study . . . . . 17 Ident i fication of Participants . . . . 18 Instrumentation and Data Collection 18 Analysis of the Data . . . . . 21 Organization of the Remainder of the Research Report . . . . . . . . . . 25 REVIEW OF RELATED LITERATURE 26 Organizational Leadership Behavior . . 26 The Influence of Perception on Behavior.... ... .. 31 Employee Mo r ale . . . . . . . 34 Absenteeism as an Indicator of Organizationa l Health . . . . . . . . . . 43 Leadership Behavior, Morale, Job-Related Var i ables Personal Character i stic Variables, and Employee Absenteeism . . . . . 45 Leadership Behavior . . . . . 46 Morale . . . . . . . . 48 Job-Related Variables . . . . . 50 Personal Characteristic Var i ables . . 51 PRESENTATION AND ANALYSIS OF DATA 61 iii

PAGE 4

IV APPENDIX Descript i ve Data by Schools Relat i ve to Teacher Related Variables . . . . . . . 65 Comparisons Among Elementary Schools with High, Average and Low Unexpected Teacher Absenteeism . . . . . . . 77 Comparisons Among High and Low Morale Elementary Schools Relative to Teachers' Perception of the Leadership Style of the Principal . . 84 Pred i ction of Unexpected Teacher Absenteeism from Selected Var i ables . . . . . 90 SUMMARY, CONCLUSIONS, AND DISCUSSION ... .. 97 Summary . . . . . . . . . . 97 Conclusions . . . . . . . . . 106 Discuss i on . . . . . . . . 108 Results of the Study i n Relat i on to Its Justification and Prev i ous Research 108 Practical Applicat i on of the Study Results . 114 A LEADER BEHAVIOR DESCRIPTION QUEST I ONNA I REFORM XII .. . . . . . . . .. 117 B MEANS AND STANDARD DEVIATIONS OF SUBSCALE LBDQ-XI I SCORES FOR SELECTED LEADERS IN LEADERSHIP POSITIONS . . . . . . 122 C RELIABILITY COEFFICIENTS OF SUBSCALES ON THE LBDQ-XII . . .. . . .. . .. 125 D PURDUE TEACHER OPINIONNAIRE . . . . 127 E TEST-RETEST CORRELATIONS FOR PURDUE TEACHER OPINIONNAIRE--FACTOR AND TOTAL SCORES 134 F G H CORRELATION MATRIX FOR INDEPENDENT AND DEPENDENT VARIABLES .. .. ..... .. ....... .. .. RESULTS OF MULTIPLE REGRESSION OF INDEPENDENT VARIABLES ON UNEXPECTED TEACHER ABSENTEEISM . ........ ....... ..... RESULTS OF MULTIPLE REGRESSION OF INDEPENDENT VARIABLES ON UNEXPECTED TEACHER ABSENTEEISM WITH TEACHER MORALES TREATED AS A COVARIATE ....... .. REFERENCES BIOGRAPH I CA L SKETCH i v 135 136 137 138 148

PAGE 5

Table 2 3 4 5 6 7 8 9 LIST OF TABLES Basic Information Relative to the 15 Elementary Schools Selected for the Study .. ... .... .. .... LBDQ-XI I Dimension Means, Orientation Means, and Standard Deviat i ons for the 15 Elementary Schools Included i n the Study ...... .... ... ... PTO Dimension Means Total Means, and Standard Deviat i ons for the 15 Elementary Schools Included in the Study .. ..... .... ...... ....... .. Number of Teacher Participants Male-Fema l e Ratios, Distribut i on by Education Level, Marital Status, and Experience i n the School District for the 15 Elementary Schools Included in the Study .. .. .. .. ..... ................. LBDQ-XII Dimens i on Means, F Values, and Observed Alpha Level for Schools Classified as High, Average Low in Teacher Absenteeism ..... PTO D i mension Means, F Values, and Observed Alpha Level for Schools Classified as High, Average, Low in Teacher Absenteeism .... ...... F Values and LBDQ XI I Dimension, Orientation, and Total Means for Schools Classified as High and Low in Teacher Morale .. .. .. .. .. .... .. .. Results of Mult i ple Regression Analysis of 10 Independent Variables as Predictors of Individual Teacher Unexpected Absenteeism Results of Multiple Regression Analysis of Nine Independent Variables as Predictors of Individual Teacher Unexpected Absentee i sm when the Teacher's Total Morale Score (PTO) is Treated as a Covariate ... ..... .. ... .. .... V 64 66 73 78 81 83 86 91 95

PAGE 6

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education LEADERSHIP BEHAVIOR, TEACHER MORALE, AND UNEXPECTED TEACHER ABSENTEEISM IN SELECTED ELEMENTARY SCHOOLS IN AN URBAN SCHOOL DISTRICT by James Donald Shaw December 1980 Chairman: Michael Y. Nunnery Major Department: Educational Administration and Supervis i on Employee absenteeism costs reached enormous proport i ons i n the United States during the 1970's. In the 1978-1979 school year, $854,000 was spent for substitute teachers in the public schools of Orange County, Florida. Excessive unexpected teacher absenteeism wa!' determined to have a detrimental effect on the progress of the educational organization, since the teacher's absence directly affected a large number of pupils. Numerous studies in industry i ndicated a relationship between absenteeism and employee morale, and morale was found to be a function of the perceived leadership behavior of the manager Given the limited research in educational institutions designed to assess the relationships among leadership behavior, teacher morale, and teacher absenteeism, the study was undertaken The major focus was to determine the extent to which the unexpected absenteeism of teachers could be explained by teacher perceptions of the leadership behavior of principals, teacher morale, and selected teacher-related and school-related characteristics. The Leader Behavior Description Questionnaire-Form XI I and the Purdue Teacher Opinionnaire were administered to 423 teachers in 15 '.ti

PAGE 7

elementary schools within the district (th:. 5 schools with the highest percentage of unexpected teacher absenteeism, the 5 schools with the lowest percentage of unexpected teacher absenteeism, and the 5 schools nearest the mean perc'.!ntage of unexpected teacher absentee ism for all 67 elementary schools in the district). Also collected were demographic data concerning the teacher's level of education, sex, marital status, and years experience in the school district. Addi tionally, the schools studied were categorized as Title I or non-Title I, large or small, and with high, average, 0r low unexpected teacher absenteeism. These constituted the independent variables employed in an effort to describe the dependent variable, unexpected teacher absenteeism. Two basic statistical techniques we"e l!tilized to determine the contribution of the independent variables i.:o the dependent variable. Differences among schools with high, average, and low unexpected teacher absenteeism in regard to perceived leadership behavior of the 1-"incipal and teacher morale were analyzed with the one-way analysis of variance ( F test) The teacher-related and school-related demo graphic variables, along with scores on the leader behavior and morale instruments, were analyzed with stepwise multiple regression to determine to what extent these independent variables contributed to teacher absenteeism. Stancard descriptive statistics and a correlation matrix completed the analysis Based up9n an analysis of data, it was found that (a) teacher perceptions of the leadership style of principals were significantly different at the .OS level in schools with high unexpected teacher vii

PAGE 8

absenteeism when compared to schools with average and low unexpected teacher absenteeism; (b) there was a significant difference in teacher morale among schoois with high, average, and low unexpected teacher absenteeism; (c) significant differences existed in the teachers' per ceptions of the leadership style of principals in high and low morale schools; (d) the independent variables of unemployment in a school with high or average unexpected teacher absenteeism, employment in a small school, one-five years experience in the district, employment in a Title I school, single marital status, and low morale score were significant predictors of individual teacher unexpected absenteeism. The independent variables collectively accounted for 17. 72% of the variance in teacher absenteeism. Effective leadership behavior in the areas of representation, demand reconciliation, tolerance of uncertainty, initiation of structure, rote assumption, and integration was identified as a significant contributor to !ow unexpected teacher absenteeism The two dimensions of the Purdue Teacher Opinionnaire which had the greatest range in mean scores between high absenteeism and low absenteeism schools were teacher rapport with principal and rapport among teachers viii

PAGE 9

CHAPTER I INTRODUCTION In the context of the politically sensitive education arena of the 1970's, with commonplace power struggles for decision-making authority, the quality of human relations interactions between principals and teachers became a focal point of concern. Teacher morale, a some what nebulous term used in many contexts, had been recognized as a powerful force impacting education. According to Knezevich (1975), morale is difficult to define and even more difficult to measure. It is a state of being more easily felt than described and verified. Morale is not necessarily an end in itself. It is a means of promoting a smoothly functioning and productive institution. (p. 455) Washington and Watson (1970) contended the effectiveness of a school may be directly dependent upon the morale of the teachers of that school and "the school principal plays a key role in nurturing and maintaining positive teacher morale" (p. 4). His or her actions, decisions, and the staff's perceptions of his or her behavior have been found vitally important in determining the morale of a school in studies conducted by Halpi~ and Croft (1963, p. 57) with the Organ izational Climate Description Questionnaire. In research conducted in elementary schools they found the building of positive teacher morale is not entirely dependent upon the actions of the principal, but also upon the ways in which teachers perceive this behavior. Campbell, Dunnette, Lawler, and Weick (1970, p. 103) found that perceptions by others of leadership style are more reliable than self-descriptions.

PAGE 10

2 Bruner and Tagiuri (1954, p. 83) reported it is the teacher's percep tion of a leader's behavior which can influence teacher behavior. Teacher absenteeism has been noted as a possible result of low morale (Bamber, 1979, p. 18). An analysis of Department of Labor statistics indicated an increase in teacher absenteeism of 3% in the 1977-1978 school year as compared to the 1976-1977 year (United States Department of Labor Statistics, Division of Economic Studies, 1978, p. 51). On an average school day in the 1977-1978 school year 86,000 of the 2,400,000 public school classrooms in the United States were staffed by someone other than the regular teacher ( p. 53). Indications from a 1977 poll of National Association of Elementary School Principals members showed that 15% considered teacher absen teeism a serious problem and another 59% considered it an increasing problem (Stryker, 1977, p. 24.). This increase in absenteeism has taken place when many diseases have been conquered and medications are available to provide relief from pain Personal factors such as sex, marital status, education level, and years of employment have been examined in relationship to absen teeism. Redmond (1978) found that gender was significantly related to the absence of professional personnel in the Fort Madison (Iowa) Community School District, i.e., women were absent more frequently than men Sylwester (1979) reported absence rates for male teachers were significantly lower than for female teachers in his study of 335 Oregon elementary and secondary school teachers Coller (1975) found that married teachers tended to have lower absence records than single teachers in the Livonia, Michigan public schools. "Aca demic degree" was one of nine variables that Douglas (1976) found to be predictors of absenteeism when added in a stepwise

PAGE 11

3 regression. Studies conducted to measure the impact of years of employment on the absence rates of educational personnel have pro duced conflicting results. Gibson and Lafornara (1972) reported that teachers with fewer than 10 years of service were absent more often than teachers with more than 10 years of service. However, Stall:ngs (1959) reported little difference in the use of sick leave by permanent (tenured) and probationary (non-tenured) teachers in 16 Southern California school systems. In his sample of Dade County (Miami), Florida, elementary school teachers, Manganiello (1972) reported no significant difference in the frequency of sick leave absences when the teachers were grouped in terms of length of service. School-related characteristics such as the type of school and the work unit size have been related to absence rate. The Office of Education Performance Review of the State of New York (1974, p. 14) reported that the discretionary absence rate for New York City teach ers in Title I elementary schools was 29% higher than in non-Title I elementary schools during the 1972-1973 school year Bridges and Hallinan (1978) found that work unit size had direct impact on absen teeism of teachers in 57 California and Wisconsin elementary schools. In 1977-1978, 56 school districts in the Philadelphia area partici pated in studies of employee absenteeism (Pennsylvania School Board Association, 1978, p. 53). Increased absenteeism among teachers was associated with staff morale, salary scale, professional expectations and attitudes of teachers, administrative leadership, working condi tions, emotional stress, physical weakness and chronic illness, and education programs. The monetary cost of providing substitute teachers in the Philadelphia area was $2,000,000 for the 1977-78 school year. Projected to all 504 school systems in the state, the

PAGE 12

4 authors noted, the potential cost of filling all vacancies with sub stitutes would be more than $30 million. If the salaries for absent teachers were added to the cost of substitute teachers, they estimated that Pennsylvania school systems spent more than $88 million a year for total professional salaries directly attributable to teacher absence. In 1977-1978, this cost was 4.5% of that paid for teachers' salaries. To put this cost into perspective: This $88 millicn cost of absenteeism accounts for a greater percent of the total budget than any one of the foilowi;ig budgetary line items: health services, food services, student activities community services, or capital outlay (p 36) Teacher absentee i sm has had an impact on school district bud gets in terms of substitute teacher pay, administrative expenses, and record-keep i ng costs. In addition to financial considerations, instruc tional leaders have reported that the quality of the teaching-learning experience has deteriorated when the regular teacher has been absent from the classroom (Jacobson, Reavis, & Logsdon 1963, p 67) Absenteeism resulting from professional leave, temporary duty assignments elsewhere, long-term leaves, or a personal leave approved as a result of a contractual agreement between the school board and the teacher has been viewed as less disruptive to the educational process because more time is available for the principal, the teacher, and a substitute teacher to formulate plans. It is unexpected absen teeism, therefore, which h
PAGE 13

5 teacher absenteeism. Research is needed to identify those leadership behavior patterns which are crucial in dealing with teachers in the educational organization. A knowledge of effective leadership behavior could provide valuable insight for the managerial development of principals. In light of the foregoing, the study reported herein relative to teachers' perceptions of the principal's leadership behavior, teacher morale, teacher-related characteristics, school-related charac teristics, and teacho:!r absenteeism was undertaken. The Problem Statement of the Problem The problem in the study wa~ to determine the extent to which the unexpected absenteeism of teachers in selected elementary schools within the Orange County, Florida, Public School System and the unexpected absenteeism of individual teachers could be explained by teacher perceptions of the leadership style of principals, teacher morale, and selected teacher-related and school-related characteris:ics. Specifically, answers to the following questions were sought : 1 Were there differences in teacher perceptions of the leadership style of the principals on each of the 12 dimensions of the Leader Behavior Description Questionnaire-Form XII (LBDQ-XII), devel oped by Stogdill (1963), among elementary schools with high, average, and low unexpected teacher absenteeism? 2. Were there differences in the 10 dimensions of teacher morale as measured by the Purdue Teacher Opinionnaire (PTO), developed by Rempel and Bentley (1967), among elementary schools with high, average, and low unexpected teacher absenteeism?

PAGE 14

6 3. Were there differences in the teachers' perceptions of the leader ship style of principals in high and low morale elementary schools? 4. To what extent could individual teacher unexpected absenteeism be predicted from a. the teacher's perception of the systems orientation of the principal, b. the teacher's perception of the personal orientation of the principal, c the teacher's total morale score on the PTO, d. the teacher's level of education, e. the teacher's sex, f. the teacher's marital status, g. the teacher's experience in the school district, h. whether the teacher was employed in a Tit l e I or a non Title I school, i. whether the teacher was employed in a large or a small elementary school, and j whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absenteeism ? 5. When the teacher's total morale score was treated as a covariate, to what extent could individual teacher unexpected absentee i sm be predicted from a the teacher's perception of the systems orientation of the principal, b t he teacher s perception of the personal orientation of the pr i ncipal

PAGE 15

7 c. the teacher's level of education, d. the teacher's sex, e. the teacher's marit~ status, f the teacher's experience in the school district, g. whether the teacher was employed in a Title I or non-Title I school, h. whether the teacher was employed in a large or a small elementary school, and i. whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absentee i sm? Delimitations 1. The study was confined to a single urban school district (the Orange County Public School System) in central Florida. 2. The investigation of unexpected teacher absenteeism for a s i ngle school year, 1978-1979, was confined to 15 elementary schools within the district--5 with the highest percentage of unexpected teacher absentaeism, 5 with the lowest percentage of unexpected teacher absenteeism, and 5 nearest the mean percentage of ur.expected teacher absenteeism fo r al! 67 elementary schools in the district. 3. The measure of leadership behavior was confined to teachers' responses to the 12 dimensions measured by the LBDQ-XI I: representation, demand reconciliation, tolerance of uncertainty, persuasiveness, initiation of structure, tolerance of freedom role assumptions, consideration, production emphasis, predictive accuracy, integration, and superior orientation.

PAGE 16

8 4. The measure of teacher morale was confined to teachers' responses to the 10 dimensions measured by the PTO: teacher rapport with principal, satisfaction with teaching, rapport among teachers, teacher salary, teacher load, curriculum issues, teacher status, community support of education, school facilities and services, and community pressures 5 Teacher-related and school-related characteristics were confined to the teacher's level of educat i on, sex (male or female), the teacher's mar i tal status (single or married), the teacher's exper i ence in the school district, whether the teacher was employed in a Title I or non-Title I school, whether the teacher was employed in a large or a small elementary school, and whether the teacher was employed in a school with high unexpected teacher absentee ::;m or a school without high unexpected teacher absentee i sm Limitations The descriptive study had certain l i mitations that must be identi fied and stated in order that the conclusions of the study may be regarded in the proper perspective The following limitations were identified. 1 The design of the study was post facto. There are weak nesses inherent in the post facto design which restrict the interpretation of the findings to inferences about the relationship of the variables rather than cause-effect. 2. The data source relating to teacher absenteeism was limited to records within the district. To the extent that these data were inaccurate or incomplete the study lacked validity.

PAGE 17

9 3. The accuracy of the teachers' responses to the instruments administered was dependent upon the ability of the individual respondents to depict accurately their perceptions. Therefore, teachers' perceptions of leadership behavior and teacher morale were valid only to the extent that individuals accurately depicted their perceptions. 4. The use of the variables included in the study did not preclude that other variables may have influenced the dependent variable (teacher absenteeism) or what was treated on occasion as the intervening variable (morale). Therefore, the variables studied should not be considered the only influences on teacher attendance. 5. The study was confined to a single school district, thus general izations beyond the district must be regarded as suggestions. Justification for the Study Employee absenteeism costs reached enormous proportions during the 1970's Robert Moore ( 1973), President of Searle Educational Systems, noted : The cost to business and government of employee illness in wages alone has reached $15,000,000,000 annually. If the rate of employee absence due to illness and accidents--now 5. 8 days per year--could be reduced by only one day, the Gross National Product could be increased by as much as $10,000,000,000 according to government reports. The figure would oe astronomical if we added absence due to laziness, employee discontent, and family problems. (p. 115) Taylor (1979, p. 53), after analyzing data released by the United States Bureau of Labor Statistics, reported 1% of the total compensation paid by United States employers in 1976 was for paid sick leave. This amounted to six cents for every hour paid.

PAGE 18

10 Organizations that have an average employee absence rate of 4% actually carry 1 extra employee for every 25 to take care of average absence (Smardon, 1974, p. 12). During the 1978 school year, a total of $854,000 was spent for substitute teachers in the public schools of Orange County, Florida (OCPS Statistical Report, 1979). A total of 27,828 substitute days were reported, representing a 22% increase over the 1977 school year An average of 8. 06 days of sick leave per teacher was reported for the 1978-1979 school year, as compared to an average of 7 47 days for the 1977-1978 school year. The percentage of absences reported on days preceding and following holidays was also signifi cantly higher than those occurring at other times. Excessive absenteeism can have a detrimental effect on the progress of an educational organization, since the teacher's absence directly affects a large number of pupils. Jacobson, Reavis, and Logsdon (1963, p. 67) asserted that the service of substitute teachers does not equal the service of regular teachers, because t he substitute teachers cannot deal in an efficient manner w i th the individual differ ences of pupils in a relatively short per i od of time Teacher absenteeism may be a potential source of conflict between principals and parents, as Zimet (1973) explained in a book on decen tralization in New York City Schools: Pr i ncipals have pointed out to their teaching staffs that absenteeism is highly visible to parents and is very difficult to defend. Although a principal can support a teacher's actions visa-vis classroom teaching techniques, he cannot defend persistent and unexplained absences. (p. 111) Research studies designed to assess the relationship of teachers' perceptions of the principal's leadership behavior, teacher morale, and teacher absentee i sm have been limited both in number and in the

PAGE 19

11 variables investigated. In one study Coller (1970) found teacher absenteeism related to teacher morale. The teachers who had low rates of absenteeism tended to have higher teacher morale scores than did teachers with high rates of absenteeism. The male teachers in the study tended to have lower absence ret:ords than did the female teachers, and married teachers had lower absence rates than did the unmarried teachers Kokovich (1969) investigated leadership behavior of selected secondary school princ i pals to determine if relationships existed between teachers' perceptions of leadership behavior and teacher morale. He found a strong re' 1tionship between the princi pal's leadership behavior and morale status, but determined leadership behavior is not the sole determinant of morale status. Miles (1967) noted the need for additional behavioral data, related to specific educational c!"'ganizations, to study relationship among leadership behavior, teacher morale, and teacher absenteeism. The study reported herein was conducted with the recognition of the need for increased knowledge in varied settings of the dimensions of leadership behavior and teacher morale which can influence teacher behavior to the extent that teacher absenteeism is affected. Such knowledge could be useful in training future administrators At a more immediate and practical level, it was reasoned that the findings from such a study would be valuable to the leadership of the Orange County, Florida, Public School System as they struggled to reduce the rate of teacher absenteeism in that school district.

PAGE 20

12 Definition of Terms Elementary school. This term refers to a school established and organized to teach pupils on any grade level from kindergarten through grade six. Elementary teacher. This term refers to a teacher employed to teach any grade level or provide music, art, physical education, media, and/or counseling services to pupi l s from kindergarten through grade six. Frequency of absence This term refers to the number of in stances of absences in a period of time (one school year in the present study) H i gh mor::lc elementary school. This term refers to the five elementary schools in the study with the highest composite scores on the Purdue Teacher Opinionnaire. Large elementary school This term refers to those elementary schools in the study which had more than 33 certificated staff mem bers, excluding administrators. Leadership behavior This term refers to the leader dimension measured by the LBDQ-XII (Stogdill, 1963, p 3) : 1. reoresentation--speaks and acts as the representative of the group; 2 demand reconciliation--reconciles confl i cting demands and reduces disorder to the system; 3 tolerance of uncertainty--is able to tolerate uncertainty and postponement without anxiety or upset; 4. persuasiveness--uses persuasion and argument effectively; exhibits strong convictions;

PAGE 21

13 5. in:tic1tion of structure--ctearly defines own role and lets followers know what is expected; 6. tolerance of freedom--allows followers scope for initiative, decision, and action; 7. role assumption--actively exercises the leadership role rather than surrendering leadership to others; 8. consideration--regards the comfort, well-being, status, and contribution of followers as significant; 9. production emphasis--applies pressure for production output; 10. predictive accuracy--exhibits foresight and ability to predict outcomes accurately; 11. integration--maintains a closely knit organization, works to resolve intermember conflicts; and 12. superior orientation--maintains cordial relations with super iors, has influence with them, and is striv i ng for higher status. Long-term leave. This term refers to the absence of a teacher from an assigned work site for a period in excess of 10 consecutive days. Low morale elementary school. This term refers to the five elementary schools in the study with the lowest composite scores on the Purdue Teacher Opinionnaire. Org : mizational climate. This term refers to the "personality" of a school organization, viewed in terms of interpersonal relations (Halpin & Croft, 1963, p. 1). Per!>onal orientations of the principal. This term refers to those dimensions on the LBDQ-XI I grouped by Brown (1967) as represent ing a leader's emphasis on the individual: representation, tolerance

PAGE 22

14 of uncertainty, persuasiveness, role assumption, consideration, pre dictive accuracy, and tolerance of freedom. Principal. This term refers to the administrator directly respon sible for the management of the school and for carrying out the policies of the school board within that school. Professional leave. This term refers to the extended absence of a teacher from an assigned work site for the purpose of continuing his or her formal education. School district. This term refers to a school system serving all the pupils in a county within the State of Florida. School year. This term refers to a 10-month contractual per i od for a tear.her. Small elementary school. This term refers to those elementary schools in the study which had 33 or fewer certificated staff mem bers, exluding administrators. Systems orientations of the principal This term refers to those dimensions on the LBDQ-XI I grouped by Brown (1967) as represent ing a leader' s emphasis on the institution: demand reconciliation, initiation of structure, production emphasis, i ntegration and superior orientation Teacher employment experience. This term refers to the number of years the teacher had been employed in the Orange County, Florida, Public School System. It was analyzed in four categories: 1-5 years, 6-10 years, 11-15 years, and more than 15 years. Teacher level of education. This term refers to the highest academic degree held by the teacher : bachelor's, master's, specialist, or doctorate

PAGE 23

15 Teacher marita: status. This term refers to whether the teacher is single or married. Teacher morale This term refers to factors as measured by the Purdue Teacher Opinionnaire (Rempel & Bentley, 1967, p 4): 1. teacher rapport with principal--deals with the teachers' feelings about the principal, his or her professional compe tency, his or her interest in teachers and their work, his or her ability to communicate, and his or her skill in human relations; 2. satisfaction with teaching--pertains to teacher relationships with students and feelings of satisfaction with teaching, the high morale teacher loves to teach, feels confident in the job, enjoys the students, and believes in the future of teaching a::. an occupation; 3. rapport among teachers--focuses on a teacher's relationship with other teachers by soliciting opinion regarding coop eration, preparation, ethics, influence, interests, and competency of peers; 4. teacher salary--pertains primarily to the teacher's feelings about salaries and salary policies; 5. teacher load--deals with such matters as record-keeping, clerical work, "red tape," community demands on teacher time, extra-curricular load, and keeping up-to-date pro fessionally; 6. curriculum issues--solicits teacher reactions to the adequacy of the school program and meeting student need:;, in pro viding for individual differences, and preparing students for effective citizenship;

PAGE 24

16 7. teacher status--pertains to feelings about the prestige, security, and benefits afforded by teaching; 8. community support of education--deals with the extent to which a community understands and is willing to support a sound educational program; 9 school facilities and services--rela t es to the adequacy of facilities, supplies and equipment, and the efficiency of the procedures for obtaining materials and services; and 10. community pressures--gives special attention to community expectations with respect to the t eacher's personal stan dards, participat i on in outside-of-school act i v i ties and the teacher's freedom to discuss controversial issues i n the classroom Temporary duty assignment. This term refers to the e x cused absence of a teacher from the classroom so that the teacher may work at another locat i on, or attend a meeting, for a brief period of t i me (usually one day). Title I school. This term refers to an elementary school with a sufficient number of economically-disadvantaged pupils to qualify for federal compensatory fund i ng Unexpected absentee i sm. This term refers to the degree to which teachers fail to report to work It specifically excludes profes sional leave, long-term leave, temporary duty ass i gnments and pre arranged personal leave which may be charged to sick leave according to the negotiated contractual agreement between the school district and the teachers' union

PAGE 25

17 Procedures The reader is reminded that the problem in the investigation focused on teachers' perceptions of the principal's leadership behavior, teacher morale, selected teacher-related and school-related variables, and teacher absenteeism. Detailed in the following sections are the procedures which were used to conduct the investigation. First, the setting of the study is reviewed. Second, the procedures for identi fying the participants are stated. Next, the data sources and collec tion procedures are explained. Finally, there is a description of the data analysis procedures. Setting of the Study The study was conducted in the Orange County Public School System, a large urban school district in central Florida. In 1978, total population of the county was 438,558; 22% of the district popula tion was of minority races, with the black race constituting the largest minority (Orange County Planning Department, 1979). In 1979, the economy of Orange County was based on tourism, manufacturing, agriculture, finance, and trade. In the 1978-1979 school year, there were 81 453 pupils enrolled in kindergarten through twelfth grade Seventy-five percent of those pupils were white, 22% black, 2% Hispanic, and the remaining 1% were Pacific I slanders and American Indians. Of the 40,959 elementary school pup i ls in the district 73% were white, 24% black, and the remaining 3% were Hispanic, Pacific Islanders, and American Indian (OCPS Statistical Report, 1979).

PAGE 26

18 The 67 elementary schools in the district varied in both geo graphic and demographic features; they included urban, suburban, and rural schools. Student enrollment in the 67 elementary schools ranged from 248 to 957, with a mean enrollment of 600 (OCPS Statis tical Report, 1979). Identification of Participants From the 67 elementary schools in the district, excluding excep tional education centers intended exclusively for pupils with special educational needs, those five with the highest percentage of unex pected teacher absenteeism, those five with the lowest percentage of unexpected teacher absenteeism, and those five nearest the mean percentage of unexpected teacher absenteeism rate for all elementary schools for the 1978-1979 school year participated in the study. All teachers in each of the 15 scrnols (a total of 496 teachers) were asked to participate in the study. Records from the school district personnel and payroll departments were researched to select the 15 schools which met the criterion. Instrumentation and Data Coilection Leadership behavior The teachers' perceptions of the principal's leadership behav i or for each of the 15 schools in the investigation were measured by the Leader Behavior Description Questionnaire-Form XII ( LBDQ-XII) developed by Stogdill (1963). Each teacher was asked to complete this instrument which is composed of short descriptive statements of ways in which leaders are perceived as behaving ( 423 of the 496 teachers provided responses). Each respondent indicated the extent the behavior described by each item characterized the behavior of the principal of the school where he or she was employed. The LBDQ -X II

PAGE 27

19 measures 12 dimensions of leader behavior (Appendix A). The 12 dimensions may be grouped into two general categories--personal behavior and systems behavior. Those grouped as personal behavior are representation, tolerance of uncertainty, persuasiveness, role assumption, consideration, predictive accuracy, and tolerance of freedom. The dimensions grouped as systems behavior are demand reconciliation, initiation of structure, production emphasis, integra tion, and superior orentation. Each of the 100 items on the LBDQ-XII has been related to one of the leader dimensions. The respondents described the leader behavior of the principal by indicating whether he or she always, often, occasionally, seldom, or never behaved as described by the item. The LBDQ-XI I was designed as a research device and has no norms (Stogdill, 1963, p 8). However the mean scores and stan dard deviations for several highly selected samples of commissioned and non-commissioned officers in an army combat division, the admin istrative officers in a state highway patrol headquarters office, the executives in an aircraft engineering staff, ministers of various denominations of an Ohio community, leaders in community develop ment activities throughout the state of Ohio, presidents of successful corporations, presidents of labor unions, presidents of colleges and universities, and United States Senators were analyzed and reliability Cr.<:!t'fici<1nts for each of the LBDQ-XI I subscales determined by using a modified Kuder-Richardson formula (Appendices B and C). Teacher morale The teacher morale in the 15 schools selected for the study was measured by the Purdue Teacher Opinionnaire (PTO) developed by Rempel and Bentley (1967). Each of the teachers in the 15 schools

PAGE 28

20 was asked to <;o,.;ple-::e the PTC (again 423 of the 496 teachers p:-o vided responses). The PTO contains 100 items and it provides a total morale score which is derived from the subscores on 10 dimensions of morale. The dimensions measured are teacher rapport with the prin cipal, satisfaction with teaching, rapport among teachers, teacher salary, teacher load, curriculum issues, teacher status, community support of education, school facilities and services, and community pressures (Appendix D). The reliability of the PTO was established by administering the instrument to school faculties of 20 or more teachers (Rempel & Bentley, 1967, p 5). A stratified random sample of 60 Indiana schools and 16 Oregon schools was used, and test-retest data from 3,023 teachers were correlated. These correlations became the basis for the reliability data relative to each dimension score and the total score of the PTO (Appendix E). Teacher-related and school-related variables The researcher met with each of the 15 faculties and asked teachers to complete a researcher-developed personal data sheet The researcher had previously coded the data sheet with the school num ber, school size (large-small), and a designation for Title I or non Title I school. Teachers were asked to provide information about sex, marital status, highest academic degree held, and years of employment experience in the school district. This data sheet was attached to a packet containing the LBDQ-XI I and the PTO Opinion naires. All responses were coded to facilitcite tabulation by computer. A self-addressed envelope was included and teachers were asked to return the completed forms to the researcher by means of the school

PAGE 29

21 district courier system. Teachers were given assurance of confiden tiality. Personal data sheets were received from 423 of the 496 teachers Teacher absenteeism Records from the school district personnel and payroll depart ments were researched to determine teacher absenteeism in the 15 schools which participated in the study. These records were com pared with data analyzed by the school district research department and cross-checked for accuracy Excused teacher absences for temporary duty assignments, professional leaves, and long-term illness leaves for the 1978-1979 school year were subtracted from each school's total of teacher absences before selecting participating schools. Teacher absences were recorded in the district computer bank by teacher's social security number Analysis of the Data Two basic statistical techniques were utilized in order to answer the questions posed i n the statement of the problem. To answer questions 1, 2, and 3, which focused on differences among schools with high, average, and low absenteeism in regard to the 12 dimen sions of the LBDQ-X II, the 10 dimensions of the PTO, and the leader ship style of principals in high and low morale schools the basic technique utilized was the one-way analysis of variance (F test). In questions one and two, each of the dimensions was treated as a presumed dependent variable and the F test was applied to determine if there were significant differences among the three sets of schools when they were grouped according to degree of absenteeism. The same basic process was applied to question three which focused on leadership style of principals in high and low morale schools. The

PAGE 30

22 schools were grouped according to the five highest morale schools and the five lowest morale schools and it was determined, using the one way analysis of variance, if there were significant differences between these groups in teachers' perceptions of the leadership style of prin cipals. Where a significant difference at the predetermined level (the 05 level of confidence) was found this was followed by the Tu key test to locate the difference within the set. It was recognized that the F test is designed to be used with interval data and some of the data in the study reported herein were ordinal (i e., scores on the LBDQ-XII and the PTO); however, the impact on the F value has been considered by researchers to be relatively insignificant. To answer questions four and five the basic technique utilized was step wise multiple regression In the study, stepwise multiple regression was used to determine the extent to which teacher absenteeism could be explained by the presumed independent variables of leadership behavior, teacher morale, teacher's marital status, teacher's level of education, teacher's sex, teacher's experience in the school district, whether the teacher was employed in a Title I or a non-Title I school, whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absenteeism, and whether the teacher was employed in a large or a small elementary school. (Schools with high unexpected teacher absenteeism were those 5 schools classified as ha .1ing the highest percentage of absenteeism in the district during 1978-79; the schools without high unexpected absenteeism were the other 10 schools included in the study.) In sum, stepwise multiple regression was applied for the total group of teachers participating in the study to determine to what extent teacher absenteeism could be explained by

PAGE 31

23 these independent variables, with teacher morale treated as an inde pendent variable in question four and ;,s a covariate i n question five. This technique seemed appropriate si:ice multiple rE>gression can accommodate any number and type of ind.:-pendent variables, i.e., continuous, dichotomous, categorical. It is sup
PAGE 32

24 (large school er small school) + b 5 (unexpected absenteeism category of school high or not high) + b 6 (Title I or non-Title I school) + b 7 (sex) + b 8 (marital status) + b 9 (level of education) + b 10 (years of experience in the district) where b 1 ... b 10 = beta weights. Raw scores were used in the regression equation. Though the researcher acknowledged the fact that "the order which the independent vari ab l es are entered into the equation makes a great deal of difference in the amount of variance accounted for by each variable" ( Kerlinger & Pe,:ihazur, 1973 p 95), order was less important when overall pre diction was the main concern The test variables were entered in the equation on a theoretical basis derived from the problem statement with systems orientation of the principal first, followed by personal orientation of the principal, and PTO total scores. The demographic variables--large or small school size, unexpected absenteeism category of the school, whether the school is a Title I or non-Title I school, sex, marital status, highest academic degree held, and years of experience in the d i strict--were used as dummy variables to identify class membership The computer program Statistical Package for the Socia l Sciences (SPSS) was used to produce means and standard deviations for the dependent variables and each of the independent variables. In addi tion, to ascertain the degree of correlation among the variables and to detect multicollinear i ty an SPSS program which computed correlation co.effic i ents was used The importance of computing intercorrelations i s related to the fact that if there are low intercorrelations among the variables, i.e., low multicollinearity, then the strength of the regres sion analysis i s enhanced On this point, Kerlinger and Pedhazur (1973) noted that independent variables are almost always correlated

PAGE 33

25 to some extent in post facto studies; therefore, zero-order correla tions were not to be expected. Two SPSS regression programs were used. These programs included one with all 10 independent variables and one where the individual teacher's total PTO score was treated as a covariate. Using these programs, the multiple r, r square, adjusted r square, standard error, regression coefficient, residual, sum of squares, mean square, F values, B weights, and Beta weights were computed. Organ i zation of the Remainder of the Research Report Chapter 11 contains a review of relat e d literature. Chapter 111 contains the presentation and analysis of the data. Chapter IV contains a summary of the findings, conclusions, and discussion.

PAGE 34

CHAPTER II REVIEW OF RELATED LITERATURE In the review of literature major emphasis is placed upon (1) leadership behaviors which have been identified as important in an organization, (2) the role of perception in influencing behavior, (3) morale as a factor i n employee behavior, (4) the role of employee absenteeism as an indicator of organizational health, and (5) the relationships between and among leadership behavior, mora l e job related variables, personal characteristics, and employee absenteeism Organizational Leadership Behavior While researchers have concluded that effective leadership i s a necessary component of a successful organization, there has been difficulty identifying appropriate leadership behavior It has been found that variations in the situation may predicate the leadership behavior wh i ch is most effect i ve in accomplishing organizational goals (Fiedler, 1967, p. 110) adaptive leader behavior: Hersey (1967) advanced the concept of The more a manager adapts his style of leader behavior to meet the particular situation and the needs of his followers, the more effective he will tend to be in reaching personal and organizational goals. (p. 15) Two major categories of leadership behaviors have been i dentified as important in contributing to organizational affectiveness: task oriented behavior and people-oriented behavior While it has been demonstrated that organizations have been effective with leadership which has been mainly task-oriented or people-oriented, researchers 26

PAGE 35

27 have concluded that organizations are more effective under the inte grating approach which emphasizes both task orientation and people orientation (Blake & Mouton, 1964; Halpin, 1966; Huse & Bowditch, 1973; Stogdill, 1974. Several dimensions of leadership behavior were identified in studies originating in 1945 by the Bureau of Business Research at Ohio State University (Stogdill & Coons, 1957). Leadership was defined as the behavior of an individual when he or she is directing the activities of a group toward a goal attainment. The Ohio State University study group identified two major dimensions of leader behavior, initiating structure and consideration. Halpin (1959) fur ther defined initiating structure and consideration: Initiating Structure refers to the leader's behavior in delin eating the relationship between himself and members of the work-group, and in endeavoring to establish well-defined patterns for organization, channels of communication, and methods of procedure. Consideration refers to behavior indicative of friendship between the leader and members of his staff. (p. 4) Halpin (1966, p. 47) stressed that the dimensions of initiating structure and consideration are not traits of leadership, but rather the behavior of the leader. In the former, a fixed capacity is in ferred; in the latter, changes can be induced through appropriate training. In studying aircraft commanders, Halpin observed that the successful leader is one who promotes both group maintenance and group achievement. Blake and Mouton (1964, p. 17) applied the task-oriented and people-oriented concept to the development of management programs for industrial organizations. Five different types of leadership based on concern for produc~ion (task) and concern for people ( relationships) were identified. Blake and Mouton attempted to train individuals

PAGE 36

28 for "team management" behavior ,vhich exhibits maximum concern for both production and people. They contendf!d that production and people concerns are complementary, and the integration of the two optimizes both task orientation and people orientation. Blake and Mouton asserted that the "team manager--high on production and high on people--will always be the most effective type of leader regardless of the situation. They contended that their training program could convert any willing participant into a team-oriented manager. According to Bernardin and Alvares (1976), the claims of Blake and Mouton have been subjected to minimal empirical testing. Likert (1961), in his studies in industrial settings, noted the need for an integrative approach to management when he proposed his linking-pin concept which combines both structural and humanistic perspectives. In the linking-pin approach, each work group has a head who is a subordinate in the next higher group. According to Likert the feeling of support is the key to high motivation within the work group. ships: Li kert explained his principle of supportive relationThe leadership and other processes of the organization must be such as to ensure a maximum probability that in all interaction and all relationships with the organization each member will, in the light of his background, values, and expectations, view the experience as supportive and one which builds and maintains his sense of personal worth and importance. (pp. 102-103) Fiedler (1967), in his studies of surveying parties, military combat crews, steel companies, basketball teams, and members of boards of directors, found that appropriate management behavior depends on the situation. In Fiedler's Leadership Contingency Model three situational variables seemed to determine the favorableness of the situation for the manager: leader-member relations, task structure,

PAGE 37

29 and the leader's position power Fiedler defined favorableness of a situation as "the degree to which the situation enables the leader to exert his influence over his group" (p. 13). In testing his theory over a period of 16 years Fiedler concluded that both task-oriented leaders and human relations-oriented leaders are successful under !ome conditions. Taking a simi:ar stand, Stogdill (1974, p. 2o), based on a survey of the leadership research, cautioned that leaders cannot rely upon their characteristic pattern of behavior to sustain them when the demands of the situation increase in severity. In tracing the historical evolution of management thought, Davis (1968) described four models of organizational behavior: autocratic, custodial, supportiv~, and collegial. Although managers may employ all of these behaviors at one time or another, Davis contended that their predominant behavior can be classified under one of the models The autocratic model, prevalent in the early days of the industrial revolution, was still in operation in the late 1960's. Davis explained that the custodial model, a form of paternalism under which employees become concerned with maintaining benefits rather than with produc tion, was not being advocated in the 1960's due to advanced knowl edge of human behavior. According to Davis, under the supportive model a manager furnishes psychological support to his or her employees at work Under Davis' collegial model, the nanagerial orientation is toward teamwork with managers building a feeling of mutual contribution among participants in the organization Davis concluded that there is no one best method of managerial behavior since it depends on what is known about human behavior and the situation in which it occurs. It is important for the manager to understand the different results brought about by different models

PAGE 38

30 and to vary the behavior with the total human and task conditions surrounding the work. Reddin (1970, p. 88) stressed the importance of the situation in determining appropriate leader behavior. When the leader's behavior is appropriate in a given situation, it is effective; when the leader's behavior is inappropirate in a given situation, it is ineffective. Reddin added a third dimension to leadership models--the effective ness dimension. Stogdill (1963), reacting to previous studies, developed tile Leader Behavior Description Questionnaire-Form XI I ( LBDQ-Xl I) to measure the presence of the following patterns of leader behavior: representation, demand reconciliation, tolerance of uncertainty, per suasiveness, initiation of structure, tolerance of freedom, role reten tion, consideration, production emphasis, predictive accuracy, inte gration, and influence with superiors Stogdill noted that both theory and research suggest the above patterns of behavior are involved in leadership, though they are not equally important in all situations Brown (1967), utilizing the LBDQ-XII and grouping each of the 12 dimensions of leader behavior into systems orientation or persons orientation, found these two factors account for 76% of the total factor variance Brown noted the similarity between systems orientation and persons orientation and Getzels' (1952) nomothetic (emphasis on the institution) and idiographic (emphasis on the individual) dimens i ons and Halpin's ( 1959) initiating structure and consideration. Stodgill (1974, p. 41) pointed out that since Form XI I was developed in 1962, very little research using it has appeared in publication After reviewing the studies which suggested that the subscales of LBDQ-Xl I

PAGE 39

31 are differentially related to different dimensions or organizations, Stogdill urged that further studies be made to uncover some hereto fore hidden complexities of leader behavior and leader influence. The Influence of Perception on Behavior The perception that people hold of others may or may not be consistent with the actual behavior. Bruner and Tagiuri (1954, p. 83) concluded from their research that people's perception of others may i r1fl i ence their behavior. Research i n percept i on has po i r ted to the need to incorporate a human relations approach to management Four of the seven meta psychological assumptions upon which Carl Rogers' (1951) theory rests provide the skeletal structure for the human relations approach : Proposition 1: "Every individual exists in a continually changing worid of experience of which he is the center" (p 483). Proposition 2: "The organism reacts to the field as it is exper i enced and perceived. This perceptual field is, for the individual, 'reality'" (p. 484) Proposition 4: "The organism has one basic tendency in striving--to actualize, maintain, and enhance the experiencing organ i sm" (p. 487). Proposition 7: "The best vantage point for understanding behavior is from the internal frame of reference of the individual himself" (p. 494) Bruner (1973, p. 161) saw perception as the product of stimulus determinants and of experiential, motivational, personal, and social factors as well. Bruner advocated continued research on perception, particularly the investigation of environmental cues which aid in interpersonal adjustment.

PAGE 40

32 Tagiuri (1969, p. 27) in his investigations into perception i denti fied the following as the main elements in the process of perception : the characteristics or state of the perceiver, the characteristics or state of the one perceived, the cues or the manifestations of the one perceived available to the observer, the cues of the one perceived utilized by the observer, the cognitive process of the perceiver in utilizing cues, and the resulting judgment by the perceiver. These elements have been studied in relationship to each other to determine whether perceived manifestations yield reliable and valid cues, which cues are used by the observer, and whether or not there i s a valid connection between the behavior of the observed and the perception of this behavior by the observer Writers i n the field of management, both industrial and educa tional, have reiterated the importance of the human aspect in i nter personal relationships and the role that perception plays in deter m i ning behav i or Combs ( 1970) pointed to the need for admin i strators to become more sensitive to the subordinates' v i ew of the world about them. He maintained administrators' awareness of how their workers come to know and regard them can make the difference in their man agerial effectiveness. Likert (1961, p 67) pointed out that the superior needs accurate i nformation as to how his or her behav i or is actually seen by subordinates. Likert noted the need for periodic measurements of perception Rogers (1962, p 70) noted that teach ers' perceptions of the principal may affect the principal's success in securing change Croft (1965) concluded that sensitivity to the percept i ons of others i s an important d i mension of effective super vision Leavitt ( 1972 p 111) stressed that the percei'led world is the world that determines behav i or

PAGE 41

33 Stogdill (1974, p. 87) pointed out that characteristics of the followers helped to determine the kinds of leadership characteristics that will be acceptable to the group Reviewing sociometric studies of friendship and leadership, Stogdill concluded that groups tend to choose leaders with attitudes and personalities similar to their own. Marquit (1968) stressed that supervisory stimuli must be received by the teacher if the principal is to be credited with affecting change. A teacher must perceive the stimulus if he or she is to respond to it. Bruner and Tagiuri (1954, p. 98) cautioned that the accuracy in judging others is often a consensual kind, basically in agreement with others regarding a person's characteristics From studies in percep tion they concluded: (1) accuracy is aided by similarity between judge and judged; (2) accuracy is dependent on having cues to work on; therefore, traits with little behavioral manifestation are poorly judged; (3) certain systematic errors in judgment--halo effect, logical error, and the like--account for much of the error involved in judg ment; (4) there are systematic relationships between various person ality variables and judging ability; and (5) a global or intuitive approach seems to improve judgment. Knowledge of the forces at work in the forming of impressions can be an aid in interpersonal behavior. Managers need to know which behaviors need additional exposure and which need to be cur tailed. The adaptive managerial behavior may alter the perception fcr:r:ed by cbservers. March and Simon (1958, p. 79) pointed out that because of perceptual biases, observers cannot be truly objective about any thing. March and Simon explained that perception depends on both

PAGE 42

34 internal states (including values, goals, beliefs, and perceived rela tions between actions and their outcomes) and external states (includ ing upbringing, reading habits, and leisure-time activities). Study after study of managerial behavior has shown that the accuracy of observers' judgments of others is questionable. The descriptions by others of a manager's behavior seem potentially more accurate than self-descriptions as predictors of overall effectiveness (Campbell, Dunnette, Lawler, & Weick, 1970, p. 78). Likert (1961) concluded that "the higher the productivity, the greater the accuracy of perceptions" (p. 49). In a review of admin istrative research, Erickson (1967) found that although teacher per ceptions of administrative behavior cannot be considered completely accurate, they are a vital subject of investigation in that they repre sent one of many criteria in determining overall administrative effectiveness. Employee Morale The nature of morale has been a controversial subject among scholars. There are basically two opposing points of view that have extended over a number of years: one has held that morale is an individual phenomenon (Silverman, 1957, p. 271) and another has treated morale as a group phenomenon (Maier, 1955, pp. 109-111). in the field of education there has been a decided tendency to use the term to apply to both individuals and groups. Maier (1955) expressed the relationship between individual and group morale: In describing an individual, one can speak of his attitude, his motiv.:tlon, and his adjustment. When one attempts to describe a group, one uses the term morale and it communi cates all of these things, but with a group reference This means that the relationship between individuals is also part of the meaning of morale. (p. 111)

PAGE 43

35 There is also a tendency in the field of education to use the term morale interchangeably with job satisfaction and job attitude (Herzberg, Mausner, Snyderman, 1969). In the review that follows, with full recognition of the controversy in the methodological problems, the rubric of teacher morale is treated both with what some authorities would call job satisfaction and others teacher morale as both an indi vidual and as a group phenomenon. The quality of education received by pupils depends in large part upon the teacher in the classroom. Teacher morale is one determinant of the behavior of the teacher. expressed the importance of morale: Ellenburg (1973) It is one of the factors which may determine whether a school functions at its best, demanding and receiving the utmost from its students, or whether it is happy just to see the passing of another day. Call it what you will. It is easy to overlook, yet it can make a school stand ahead of the rest. (p. 7) Herzberg, Mausner, and Snyderman (1969, p. 37) theorized that worker job attitude is determined by two distinct classes of variables: job satisfiers, referred to as motivators or intrinsic factors, and job dissatisfiers, referred to as hygiene or extrinsic factors. According to Herzberg and his colleagues, the dimensions of work, responsi bility, and advancement are major motivators, while company policy and administration, supervision, working conditions, and pay are hygiene factors The motivators seem to be effective in motivating people to effective performance; the hygiene factors prevent dis satisfaction. In the discussion of the Herzberg theory presented by Sergiovanni and Carver (1975), the motivation-hygiene or satisfier dissatisfier theory is deemed applicable to educational organizations for two reasons:

PAGE 44

1. The theory is consistent with the humanistic baiief pattern which forms one dimension of our applied science of educational administration. 2. When the Herzberg model is tested using teachers and other educators as respondents, results similar to those found for other groups are obtained. (p. 72) 36 Maslow (1970, p. 12) and McGregor (1960, p. 27) were inter ested in behavior and attitudes as psychological factors for human existence in work. Maslow noted that a person's needs may be dif ferent at different times and developed a hierarchy of needs to include low order needs such as physiological and safety needs, and high order needs including belonging and affection, ego satisfaction, and self-actualization. McGregor (1960, p. 30) developed what he called "Theory X" and "Theory Y" in which opposite attitudes toward work were described. In "Theory X" McGregor assumed that man was essentially lazy and had a dislike for work and must be coerced into work per formance. In "Theory Y" he assumed that man found work natural and would seek responsibility given pre-per conditions The work accomplished by Maslow, McGregor, and Herzberg described the factors r,ecessary to meet the psychological needs of employees, and it was implied that it was management s responsib i lity to consider these factors when evaluating employee behavior One of the basic needs of an employee was a continuity, a predictability, and a routine of relationships in his life. Fox and Scott (1943, p. 67) reported management s responsibility was to help provide for this need or to risk the consequences, one of which was absenteeism. Argyris (1957, explained that "need" referred to something that existed "in" the personality which required a particular behavior / condition to occur A need was said to be related to all other needs

PAGE 45

37 and to initiate and guide behavior until the goal of the individual was reached. According to Argyris, "the needs to be integrated, to feel successful, to feel wanted and so on are felt by people in our cultura, and come from within their personalities" (p. 31). Using the Purdue Teacher Opinionnaire (PTO) developed by Rempel and Bentley (1967), Henderson (1976) found a positive rela tionship between participation in decision-making and job satisfaction. Teachers' feelings about the professional competency of their princi pal, the principal's interest in teachers and their work, and his or her skill in human relations were also found to have a positive effect upon morale. The PTO measures individual teacher morale, but scores can be used as an aggregate of the group studied. Shealy (1975) studied readily accessible data about classroom teachers to determine the usefulness of such data in predicting teacher morale. The morale score and personal information were obtained from each teacher participating in the study. Shealy concluded that the data which are readily accessible about teachers were indeed useful in predicting teacher morale. A score for the National Teachers Examination, type degree held by the teacher, and the number of education courses taken by the teacher made contributions to the predictive efficiency of teacher morale. Shealy also found that the older, more experienced teachers with more education courses and advanced degrees tended to have higher morale scores. Males tended to score lower than did the females and the elementary teachers had higher morale scores than did the middle and secondary teachers. The locale of the school is another aspect to be considered when studying teacher morale. Cook (1971) used the PTO to measure the morale of teachers in inner-city conventional school settings. He

PAGE 46

38 reported that the mean scores of teachers in i nner-city schools were significantly lower on three dimensions of the PTO. These d i mensions were support among teachers, satisfact i on with teaching, and com munity support of education. Teachers' perceptions of the i r princi pals' consideration affected their satisfaction w i th teaching In another study using the PTO, Herrmann (1971) found positive correlations between teacher morale and teacher involvement in school policy determinat i on He also discovered a posit i ve corre l at i on between faculty involvement and principa l -teacher rapport. When teacher morale and authoritar i anism and the authoritar i an levgl of administrators were examined by Gubser (1969), older teach ers scored higher in morale factors and authoritar i anism than did the younger teachers This led Gubser to conclude pr i ncipals may have an i nd i rect effect on the level of autho ritarianism of their faculties i f they discourage younger, more anti-autho ri ta:-ian teachers from remaining on the i r faculties. (p. 38) Relationships between teacher morale and verbal teacher behavior were investigated by Greenwood and Soar (1973). They used the Rec i procal Category System, a mod i fication of Flanders Interaction Analysis to determine teacher verbal behavior and the PTO to measure teacher morale The product moment correlations developed by these researchers i ndicated : 1. Teacher morale was positively related to satisfaction with teaching, teach i ng load, curriculum i ssues, and support of education by the commun i ty. 2. Teacher morale was negatively related to teacher talk 3. The amount of pup i l talk followed by more pupil talk was positively related to sat i sfaction with teacher salary and teacher load 4 The percentage of teacher acceptance behav i ors was positively related to satisfact i on with teaching and rapport among teachers ( p 105)

PAGE 47

39 Burgett (1977) studied the morale of 400 Montana teachers and found it was possible to predict the morale of those teachers by the use of the teachers' perceptions of the level of agreement between the school board and the superintendent on certain administrative prac tices He also pointed out there were three areas of morale that were most affected by the administrative practices: teacher rapport with the principal, curriculum issues, and community support of education. (p 54) Several other stud i es which used the PTO dealt with t he behavior of principals or supervisors In one study examining selected princi pal behavior as seen by teachers and principals, Pennebaker (1970) suggested these generalizations: 1. Where teachers perceive their principal as setting a good example by working hard, they will exhibit higher morale 2. Where teachers perceive their principal as being highly considerate, they exhibit higher morale. 3. Principals feel that they are more cons i derate than their teachers perceive them to be. 4. Teacher morale is not adversely affected when teachers perceive their principal as being aloof and acting the role of the "straw boss. 11 5. Teachers with high morale are more apt to keep abreast of current professional problems. 6. The more aloof, close supervising princ i pals had teach ers who spent more time preparing for teach i ng. (p. 36) Kokovich (1969) i nvestigated the leader behavior of some second ary school principals to determine whether any relationships existed between perception of leader behavior and teacher morale. He mea sured leader behavior with the LBDQ-Xl I and teacher morale with the PTO. He concluded that the leader behavior most highly associated w i th satisfaction is consideration. Kokovich said, "the behavior of

PAGE 48

40 the principal does have a strong relationship to morale status, but it is not the sole determinant of morale status" (p 24). In another study using the PTO and the LBDQ-XI I, Lambert (1969) found that teachers who perceived their principals as being above the mean on leader behavior perceived their own morale as being significantly higher than was the morale of those teachers who perceived leader behavior of their school principal as being below the mean Perry (1977) used the PTO and the Executive Professional Leadership Scale ( EPLS) to determine whether relationships existed between teacher morale and the principal's attempts to improve teacher performance. Perry found that each of the 10 morale factors of the PTO was significantly correlated with the principal's professional leadership as determined by the EPLS. Blumberg and Weber (1968) studied the possible relationship that might exist between the manner in which a teacher perceives his or her supervisor's behavior and teacher morale. These authors found that the behavioral style of the supervisor as seen by the teacher was related to the morale of that teacher. Numerous other studies dealing with teacher morale which did not employ the PTO were also found. In a study of 776 teachers, Schultz (1952) reported that the administrator is most often identified as the focal point of the teachers' satisfaction or dissatisfaction. Gragg (1955) found that the most frequently mentioned con tributor to teachers' high morale in his study was "confidence in the leadership of my principal and other administrators" (p. 494). An 81-item questionnaire was administered by Silverman (1957) to a group of teachers with instructions to respond in relationship to the way they thought the item would affect morale The teachers indicated

PAGE 49

41 that 85% of the items dealing with a principal's character or behavior would have some effect on teacher morale. Silverman concluded that the principals represented the strongest morale-influencing factor and the principal is in a position to bring about change. The principal's role in promoting teacher morale was also exam ined by Gezi (1962). Subjects were asked to respond to the open end question, "What do you think the principal should do in order to obtain and maintain his teachers' morale?" In discussion sessions, the teachers defined morale as the group's attitude and feelings which are conducive to their working harmoniously and effectively with each other and with the administration so that by achieving the organi zation goals many of their individual goals may also be achieved. (p. 88) Three hundred and eighty suggestions were made for fostering teacher morale in response to the question posed. The findings indicated that in order for the respondents to be emotionally secure and in order for them to gain recognition as members of the school team, they must find in their principal the qualities which they perceived to be necessary for the achievement of their goals and those of the school. Redefer (1962) provided insight concerning the selection of principals. As a result of his studies, he found that principals in high morale schools received their satisfaction from helping the faculty to grow professionally The principals of low morale schools received satisfaction from the power of the position. In a survey of 400 teachers, Prewett (1956) found that many of the barriers contributing to teacher inefficiency and low teacher morale are trivial and could be removed by an administrator who is people conscious and thinks in terms of personalities. Campbell

PAGE 50

42 (1959) found that the highly satisfied teachers he studied consistently made references to certain qualities of their principals such as, "understanding, competent, courteous, makes teachers feel worthy, and guides teachers in such a way as to make it easy for them to teach" (p 2) The dissat i sfied teachers mentioned such factors pertaining to the i r work as, "lack of staff participation, too many cler i cal duties, shortage of supplies, and displeasure with teaching assignments" (p 2) Another study of factors affecting teacher morale was conducted by Suehr (1962), who identified some characteristics of high and low morale teachers. The high morale teachers had taught l onger felt they had fulfilled the i r parents' expectations for them and most often came from upper or upper-middle class homes Those teachers who felt they were not realizing the i r fullest potential were in the low morale group. This group indica t ed that an opposite-sexed parent had influenced them more considered their self-confidence to be greater, and rated their level of ambition high. Principals of open climate schools were researched by T i rpak (1971) and were found t o be warmhearted good-natured, sociable, and attentive to people. The open climate schools were also found to be high in esprit or morale and were s t affed by faculties who charac terized the pr i ncipals by their high degree of emotional stability frustrat i on tolerance, and calm and real i stic approach to life Perceptions of teachers of organizational climate and the rein forc i ng behavior of the principal were studied by Nelson (1972) using the Organizational Climate Descript i on Questionnaire He noted that teachers tended to perceive an open climate in schools led by princi pals whom the teachers perceived to reflect a high l evel of reinforce ment behav i o r

PAGE 51

43 Absenteeism as an Indicator of Organizational Health A certain number of legitimate absences from a job are due to illness, obligations, or emergencies. A high rate of absenteeism, however, may be an indicator of a problem within an organization. Herzberg (1966, p. 113) and Vroom (1964, p. 79) concluded that absenteeism correlates negatively with indices of worker morale Mann and Baumgartel (1953, p. 47) concluded that workers who had low absence rates were relatively more satisifed with their jobs, supervisors, and work associates. Argyris (1964) found that absenteeism is an adaptive activity of the employe~. Just as some employees learn to accept the work environment, some are "also capable of modifying their working world so that they can express some of their frustrations, decrease them, or partially avoid them" (p. 59). Kornhauser (1962, p. 26) found that people who have a high absenteeism rate tend to be frustrated on the job by poor supervision and non-challenging routine work Likert (1967, p. 85) observed that excessive absence tends to be highest when people are free to move under an exploitive-authoritative style of management. Abse,1teeism is likely to be moderately high under a benevolent-authoritative style of management, moderate under consultation-authoritative management, and low under a participative approach to management. Likert further noted than an unfavorable trend in worker attitude will, in turn, be manifested in increased absence. He observed that the factor "time" plays an integral part in the manisfestation of excessive absence. When hostile attitudes, uncooperative motivation, and distrust of management are widespread and deep-seated, it often requires years for even an extremely competent manager to bring about any substantial improvement. (p. 88)

PAGE 52

44 Ronan and Prien (1971, p. 67) four.d that excessive abs:mteeism indicates that an employee will eventually leave the company. Merritt (1974) found that management personnel believed that employee absen teeism was due to changing attitudes and the changing environment. Workers were no longer satisfied with just an adequate income and work in a safe, acceptable working environment. They expect to be able to obtain some job satisfaction for the time and energy they contri bute to the job. (p. 177) Workers were no longer ingrained with the attitude that the job must be accomplished above all else. Metzner and Mann (1953) found that frequency of absence proved to be more successful than the length of absence in yielding relation ships between satisfaction and absenteeism for white-collar workers They concluded that frequency of absence yielded more significant relationships with satisfaction because it minimized the effective 3bsen teeism due to prolonged illness. Metzner and Mann recommended the use of the frequency of absence measure for further research into the relationship between absenteeism and other concepts. Hersey and Blanchard (1972, p. 74), in reviewing research, pointed out that absenteeism is one of the many end-result variables which may be indicators of deterioration of work attitudes. Merritt (1974) observed that government statistics demonstrated that the number of employed workers contributing to absenteeism has increased from 4.3% of the total employed workers in 1967 to 5% in 1972. In other words, 2,800,000 workers were absent at any one time in 1967; 3,500,000 were absent at any one time in 1972, an increase of 700,000 workers.

PAGE 53

Leadership Behavior, Morale, Job-Related Variables, Personal Characteristic Variables, and Employee Absenteeism 45 In studying the management of organizations, Likert (1967) identified three sets of variables for use in research and operating purposes: causal variables, intervening variables, and end-result variables. The condition of the intervening variables was affected by the causal variables and, in turn, the intervening variables influ enced the end-result variables. Included in the causal variables were those independent variables which influenced developments within the organization and which could be altered or changed by the organization and its management. Among the causal variables, Likert listed managerial skills and mana gerial behaviors. Among the intervening variables, which reflected the internal state of the organization, were worker perceptions and attitudes. Among the end-result variables, the dependent variables, was absenteeism, a performance variable. As managers learn more about the influence of the variables and adapt their behavior accordingly, there should be an improvement in the positive direction of the end-resuit variables (Likert, 1967, p. 121). Li kert added that efforts to improve the end-result variables by altering the intervening variables will usually be less effective than the changing of causal variables. When the effectiveness of an organization is evaluated, it is usually the end result or output alone which is considered (Hersey & Blanchard, 1972, p. 97). Reddin (1970, p 41) stated that the effectiveness of a manager should be measured objectivelv by the

PAGE 54

46 maximum output. It is necessary to measure all three sets of vari ables periodically in an organization to detect trends which may influence output, according to the research done by Likert (1961, p. 101). Li kert (1967, p. 38) cautioned that the causal, intervening, and end-result variables comprise a complex network with many interde pendent relationships, sometimes making it difficult to classify organi zational variables into categories. He added that the employment of managerial principles will prove to be causal in most situations. While researchers concluded that no one causal factor alone could account for differences in employee morale and performance (Fiedler, 1967; Hersey & Blanchard, 1972; Likert, 1967), the identification of those conditions under which certain types of managerial behavior would be more likely to adversely affect worker behavior could pro vide additional insight. Leadership Behavior In describing a case study of management concern about the absenteeism of manufacturing plant workers, Covner (1950, p. 336) reported that when an attitude survey was conducted, the unfavor able attitude of workers toward management was the strongest pre dictor of absenteeism. Mann and Baumgartel (1953, p. 121) indicated that, among plant workers, low absenteeism was noted where workers reported that the foreman (a) created an atmosphere which contri buted to free and easy discussion of work problems, (b) had time to talk to his men about personal problems, (c) held group discussions with his men, (d) could be counted to "go to bat" or "stand up" for his men

PAGE 55

47 Metzner and Mann (1953, p. 481) concluded that satisfaction with leader behavior was negatively related to the absence of white and blue-collar males studied, but had no effect on the absence of white collar female~ examined. Gerstenfeld (1969, p. 55) reported in a study of female production workers that there was a strong relation ship between the worker's attitude toward her immediate supervisor and her absences. Those workers who felt that the boss was fre quently unfair were generally the same workers with poor records of attendanr:e. Johns (1978, p. 443) investigated three factors relating to the influence of the supervisor on the absenteeism of 208 operative employees in a manufacturing plant. Satisfaction with leader behavior was found to be negatively related to absence frequency, but un related to time lost. Consideration, one of the leadership style vari ables analyzed, was negatively related to both absence frequency and time lost. Initiating structure, the other leadership style factor studied, was negatively related to time lost, but unrelated to absence frequency. Much of the research conducted among school employees contra dicts the belief that satisfaction with leader behavior exerts an appre ciable impact on absenteeism. Schroeder (1977, p. 83) reported no significant relationship between 12 patterns of the principal's man agerial behavior and the absence rate of 96 teachers in eight schools randomly selected from a total population of 3,800 teachers in a school system located in metropolitan New Orleans. In addition, satisfaction with supervision, as measured by the Job Description Index, was found to have no relationship to teacher absence frequency. In a study of 1,500 Pennsylvania teachers, Slick (1974, p. 109) found that teacher absence frequently was not significantly related to the level

PAGE 56

of rapport with the principal or the perceived level of aloofness, production emphasis, thrust, or consideration on the part of the principal. Morale There is a large body of existing literature dealing with the re!ationship of morale and employee absenteeism llgen and Hollenbach (1977) noted in their study of these two variables although most of the reseach is correlational, the general model assumes that employees appro : ,ch (attend) jobs per ceived to lead to satisfaction and avoid (are absent from) jobs perceived to lead to dissatisfaction. (p 149) The earliest known study in this area, according to Mochinsky (1977, p. 317) was conducted by Kornhauser and Sharp in 1932 No statistical analyses were included in their results, which focused on a sample of female factory workers, but they concluded that a slight relationship existed between job attitudes and lost time. Likewise, Noland (1945, p. 508) found a negative relationship existed between overall morale and the absence rate among a sample of industrial workers, although again no statistical procedures were presented Kerr, Koppelmeier, and Sullivan (1951, p. 54) noted that the morale of employees in manufactur i ng departments was negatively related to unexcused absenteeism, but positively related with total absenteeism When frequency of absence was used in their study, Metzner and Mann (1953, p. 483) reported a negative correlation between overall morale and the absence of blue and white-collar workers. When absence was measured by a count of actual days lost, no relationship was found between these two factors. They contended that these findings would most likely appear when absence indices are used which increase the weighting factors of persons who have irregular attendance patterns and decrease the weighting factors of absences

PAGE 57

49 caused by illness. Morale was negatively related to absenteeism in studies by van Zelst and Kerr in 1953 (for manufacturing workers), Fleishman, Harris, and Burtt in 1955 (for production workers), Harding and Bottenberg in 1961 (for airmen), Lundquist in 1958 (for Swedish factory workers), and Talacchi in 1960 (for office workers) Bass (1965) wrote that certain work groups typically are absent more frequently than others. In particular, he noted that poor morale traditionally has been related to the absenteeism of "lower-skilled" employees, but not of more highly-skilled white-collar workers or women. A possible explanation for this finding, he explained, is that high status white-collar employees have more freedom to use other forms of withdrawal when dissatisfied, like taking extra long coffee breaks or three-hour lunch periods (p. 107) llgen and Hollenback (1977, p. 159) found that, for the most part, morale was unrelated to the absenteeism of a sample of 164 clerical workers at Purdue University. The results of a study by Garrison and Muchinsky (1977) involving 174 accounting department workers produced mixed results. Overall morale, as measured by the Job Description Index, was found to be a significant negative pre dictor of absenteeism without pay, but it had no relationship to absenteeism with pay. In contrast to the wealth of research conducted on morale and absenteeism in business and industry, only three researchers have investigated this relationship as it affects educational personnel Teacher morale was found to be significantly related to teacher absen teeism by Coller ( 1975, p. 131) He learned that the low absence teachers he studied tended to have higher teacher morale scores than did the high absence teachers. In a sample of 1,500 Pennsylvania teachers, Slick (1974, p. 90) found an inverse relationship between

PAGE 58

so teacher absence frequency and the perceived level of teacher morale. Douglas (1976, p. 83) reported that morale was found to be a signifi cant predictor of absenteeism in a study of 154 teachers in Central Ohio. Job-Related Variables The job-related variables of work unit size and type of school were found to relate consistently with worker absenteeism. Covner (1950, p. 336), who studied 38 groups of plant and office workers, reported a positive relationship existed between work unit size and employee absenteeism. Argyle, Gardner, and Cioffi (1958, p. 36), who studied workers in 86 production departments, and lndik and Seashore (1961, p. 22), who studied 91 groups of factory workers, found similar positive correlations. One of Gibson's (1968) conclusions in his study of staff absence in Boston, Massachusetts, area schools was that: In small systems as compared with large systems, absence is more a function of the total social system while in large systems, absence is more associated with the characteristics of the subsystem work group (p. 7) According to Bridgt!s and Hollin an ( 1978, p 41), subunit size and work system interdependence had direct, independent effects on the absenteeism of teachers in 57 California and Wisconsin elementary schools. Subunit size was positively associated with absenteeism and work system interdependence was negatively related to absenteeism. In the Chicago, Illinois, public schools, staff absenteeism in elementary schools was lowest in the smallest schools in 1959-1960, but no other consistent relationship seemed to occur (Chicago Board of Education 1960, p. 12). Data for February, 1960, indicated that the lowest average amount of sick leave taken was for elementary

PAGE 59

51 schools with a staff of less than 10 (0.56 days) Elementary school teachers in eight larger staff size categories used between 0.89 and 0. 72 sick leave days. A relationship may be found between the type of school in which a teacher works and his or her absence rate The State of New York Office of Education Performance Review (1974) reported that the absence rate for New York City teachers in Title I elementary schools was 29 percent higher than in non-Title I elementary schools during the 1972-1973 school year As the report concluded This leads to the inference that where the educational need is greatest, teacher absenteeism is highest, especially absence over which the teacher has control. (p. 14) Douglas (1976) attempted to relate various social-psychological factors, generally considered to be sources of teacher stress, to work atten dance and the excessive use of sick leave Twenty-seven predictor variables were analyzed by stepwise multiple regression using the Wherry program MULREG. He discovered that inner-city vs. sub urban school was one cf nine variables found to be predictors of absenteeism when added in stepwise regression Personal Characteristic Variables The impact of personal characteristic variables on employee absenteeism has been investigated by behavioral scientists. Literature related to an employee's sex, marital status, level of education, and employment experience as potential variables impacting absenteeism was reviewed and is reported in the paragraphs which follow. The available research indicates that female employees have higher rates of absence than males, although men seem to be absent for longer periods of time than women Hedges ( 1973, 1975) reported

PAGE 60

52 that data contained in the Bureau of Labor Statistics Current Popula tion Surveys indicated that the absence rate for females was approxi mately twice that for males, in part-week absences in 1972 and 1974. Females were also absent significantly more often than males for full-week absences in 1967, 1972, and 1974. Hedges (1977, p 22) and Taylor (1979, p. 51) found that the incidence of absence and the proportion of available time lost was much higher for females than males in May 1976 and May 1978. But where time lost by absent workers was measured, female workers lost a smaller proportion of their usual weekly hours than males The results from three studies were mixed. Kerr, Koppelmeir, and Sullivan (1951, p. 52) indicated that female production workers were absent more than males for total absences and certified (pre viously approved) absences, but that males were absent more than females for uncertified absences. Garrison and Muchinsky (1977, p 228) found that female white-collar workers had significantly more paid absences than males, yet there was no correlation between sex and upaid absence. Johns (1978, p. 442) reported that women manu facturing workers studied had a significantly higher frequency of absence than men, but that there was no relationsihp between sex and total time lost. However, he concluded from regression analysis that sex was the single best predictor of absence among job satisfac tion, six personal characteristics variables, two leadership style ~ariables, and six job content variables examined. In a study of workers at a light engineering factory, Martin (1971, p. 86) reported that the demographic factors of sex and age seemed more important than education or marital status when correlated with absenteeism. In all categories of c:1bsence, females were absent more often than males

PAGE 61

53 The Philadelphia Suburban School Study Council (1970, p. 15) con ducted a survey on teacher absenteeism covering school years 19681969 and 1969-1970. Fifty-six Pennsylvania school systems in five study councils participated in the study. The researchers found that women were absent more than men in the "average" district. Female teachers averaged 6. 95 days of leave, with pay and without pay in 1968-69 and male teachers, 7.54 days. Females also had a higher mean rate of absence--on the average, females were absent 3. 73% of the total work days, compared to a rate of 1. 92% for males. Female teachers were absent significantly more than male teachers according to Marlin's (1976, p. 78) study of 425 teachers in a semi rural school system. The Pennsylvania School Boards Association (1978, p 23) found that female teachers were absent .97% more often than male teachers in school year 1977-1978. A Dade County, Fiorida, Public Schools Office of Management and Budget study (1978, p. 18) contained data indicating in school year 1977-1978 female teachers took approximately 70% of their sick and personal leave, where male teachers took approximately 30%. Redmond (1978, p. 111) found that of 10 demographic variables studied, only gender was significantly related to the absence of professional personnel in the Fort Madison ( Iowa) Community School District, i.e., women were absent more frequently than men. In conjunction with his study of the effects of stress on absence, Sy I wester ( 1979, p. 20) reported absence rates by sex for a sample of 335 Oregon elementary and secondary school teachers and adminis trators. Male educators were absent an average of 3. 6 days in school year 1977-1978, compared to 5 4 days for female educators. The average number of days absent for the total group was 4. 7 days.

PAGE 62

54 Although research findings appear to link sex and absenteeism care should be taken in interpreting these findings. The United States Department of Labor Statistics Report warned that other factors may influence the sex-absenteeism relationship, such as age, marital status, and occupation Occupation is especially critical to this relationship. Traditionally, more females have been new hires in the lower skilled, lower pay positions, two factors regularly associated with relatively high rates of absence. Similar groupings of employees should be considered when analyzing sex and absence Fewer women are employed in high salary positions usually associated with lower absence. "Sex d i fferences in absence rates narrow when comparisons are made within a particular occupation group," observed Hedges (1973), a Bureau of Labor Statistics economist, "even though within the group men tend to occupy the better pay i ng jobs" (p. 28). lsambert-Jamati (1962) found that, even if they have several children to raise, highly-trained women in responsible positions are absent infrequently Researchers have found a relationsh i p between absence and an employee's marital status Jackson (1944 p 293) reported that married men with several dependents had a steadier attendance record than either single or married men with no children. Shepherd and Walker (1957, p. 271) found that single males were absent the most among iron and steel workers studied Hedges (1973, p. 28) found that married males had a lower part-week absence rate than single males, but married females had a higher rate than single females. Taylor (1979, p. 51) reported from Bureau of Labor Statistics data that men who were never married had a h i gher absence rate than married men in May, 1978 ; the reverse was true for women dur i ng

PAGE 63

55 this period. Married women alsc :-.ad a higher percent of time lost for both illness and injury and miscellaneous reasons than did single women. "Family duties may be one factor causing their higher absence rates," Taylor noted, "although absences due to child care activities and other family-related responsibilities could be expected to appear as miscellaneous reasons" (p. 51). Sharples (1973, p. 98), conducting research on classified civil service personnel in Ohio, found that both the high absence male and high absence female categories studied contained a higher percent of married employees than the low absence groups for males and females Martin (1971, p. 87) reported that, of the light engineering workers he studied, single workers were absent more than married workers. Marital status was found to be a significant negative predictor of absence without pay, but not with pay, in a study of accounting workers by Garrison and Muchinsky (1977, p. 228). Lee (1960, p. 23), in a study conducted in the Wichita, Kansas, public schools for school year 1959-1960, reported that 34.6% of the male teachers, 25 3% of the single female teachers, and 16.5% of the married female teachers had perfect attendance records While more married females than single females were absent less than 5 days, more single females were absent 15 days or more. Coller (1975, p. 93) found that married teachers tended to have lower absence records than single teachers in the Livonia, Michigan, public schools. However, Marlin (1976, p. 88) reported that the mean absence rate for married teachers was higher than for unmarried teachers in a semi-rural Mississippi school system. No relationship was found between marital status and teacher absence in Bridges and Hollinan's (1978) study in California and Wisconsin, Bundren's (1974) study in

PAGE 64

56 Clark County, Nevada, Marchant's (1976) study in Richmond, Virginia, or Redmond's (1978) study in Iowa Education level, as a personal characteristic variable which may i mpact employee absenteeism, has been reported in numerous studies focusing on business and industrial employees with no consistent relationsiiip emerging Noland (1945, p 508) reported that absen teeism among a sample of industrial workers was inversely related to the workers' education level. However, neither the absence measure used nor the range of education was described. Raouf (1973, p. 41) found that i n the companies he surveyed i n Windsor, Ontario, employ ees without high school diplomas generally had the highes t rate of absence Sharples (1973, p. 99) found that high absence females in his study of classified civil service personne l in Ohio had more formal education than low absence females Martin (1971, p 92) reported that light engineer i ng workers w i th more than eight years of educa tion appeared to have more absences of all types than workers with less educat i on Citing data from the Current Population Survey of the Bureau of Labor Statist i cs Taylor (1979, p 52) reported that a negative rela tionship existed between education level and absenteeism in the United States work force in May 1978. The incidence rate (percentage of workers absent) was highest for workers who completed elementary school (8.3%), followed by h i gh school graduate; (6 6%), and college graduates ( 4 1%). The inactivity rate (percentage of aggregate time lost) for workers with an elementary school education was 4. 9%; for high school graduates, 3 5%; and for college graduates, 2 1% Few studies focusing on the relat i onship of education level and teacher absenteeism have been reported Lee (1960, p 8), i n a

PAGE 65

57 study conducted in the Chicago, Illinois, public schools during February 1960, indicated that absenteeism decreased as the amount of education increased. Teachers with a bachelor's degree on the average took 0. 78 days of sick leave; teachers with a master's degree, 0. 73 days; teachers with a master's degree plus 30 hours, 0.67 days; and teachers with a doctor's degree, 0.12 days. The caution was offered that the type of school, age, and sex may have complicated these findings. "Academic degree" was one of nine variables that Douglas (1976, p 87) found to be predictors of absenteeism when added in a stepwise multiple regression. However, there was no correlation between education level and absenteeism of professional personnel in the Fort Madison ( Iowa) Community School District (Redmond, 1978, p. 113). Conflicting results have been reported from studies focusing on the relationship of employment experience to absenteeism. A negative correlation existed between length of employment and absence of white-collar males in a study by Metzner and Mann (1953, p. 483), but there was no relationship between these variables for blue-collar males. Hill and Trist (1955, p. 148) reported that no relationship existed between employment experience and the absence rate of factory workers in a longitudinal study. Kahn (1957, p. 92) found that the longer the period of service in an organization, the lower the absence frequency, regardless of age. Martin (1971, p. 89) noted that, for the British light engi neering workers studied, a positive relationship between employment experience and absence was found for males; however, no significant correlation was noted for females. Raouf (1973, p. 41) noted that the highest absence rates began after the first six months of employment

PAGE 66

58 and ended after the third year on the job. Little absence was found in the early days of employment in his sample of factory workers in Windsor, Ontario. Sharples (1973, p. 108) indicated that, in his study of classified civil service personnel employed at Ohio University, high absence females spent fewer years on the job than low absence females. Nicholson and Goodge (1976, p. 252) indicated that younger, short-service female food processing employees were most likely to have high levels of casual and unsanctioned absence Two measures of sickness absence were found to increase with longer service among the oldest group and decrease with longer service among the youngest group. After the effects of pay and age were partialled out, a negative relationship between employment experience and absence was found in Bernardin's (1976, p 90) study of 109 male white-collar sales workers. Garrison and Muchinsky (1977, p. 227) found a significant positive relationship existed between length of employment and the paid absences of 195 accounting workers; however, c: significant negative relat i onship was found between length of employment and unpaid absences. Studies have also been conducted to measure the impact of the years of employment experience on the absence rates of educational personnel. Gibson and Lafornara (1972, p. 17) reported that in a 30-year longitud i nal study of teacher absenteeism in a single school system from 1938-1939 to 1968-1969, "newcomers" (teachers with up to 10 years of service) were absent more often than "continuing" teach ers (those with 10 or more years of service) Newcomers were seen as the bearers of "creeping legitimacy, 11 whereby absence norms shifted from illness to other reasons as newcomers entered the school system and influenced the absence patterns of continuing personnel

PAGE 67

59 Stallings (1959, p. 62) reported little difference in the use of sick leave by permanent and probationary teachers in 16 Southern California school systems. For the 1955-1956 school year, permanent teachers used about a half a day or more sick leave than probationary teachers, on the average. An examination of sick leave days taken by teachers in Chicago, Illinois, in February, 1960, showed that the amount of sick leave taken generally increased with experience. Teachers with 1-2 years of experience averaged about 50 days of sick leave during this month; teachers with 4 years and those with 10-20 years of experience averaged 75 days; teachers with 21-35 years of experience averaged over .80 days; and teachers with 36 or more years of experience averaged 1 day of sick leave (Chicago Board of Education, 1960, p. 12). In his sample of teachers in Livonia, Michigan, Coller (1975, p. 94) found that teacher absenteeism was significantly related to years of teaching experience in a curvilinear fashion. Teachers with 2-4 and 23-25 years of teaching experience tended to have low absence records Marlin (1976, p 87) reported that the mean rate of absenteeism for teachers with four or more years of experience was slightly high~r than for teachers with less than four years of experience in a semi rural Mississippi school system. Douglas (1976, p. 89) reported that "years of teaching experience" was one of nine variables that was found to be a predictor of teacher absenteeism when added in a stepwise regression, and one of five variables included in a "predictb "' profile" of Ii kely high-absence teachers In 1977-1978, teacher absenteeism in Dade County, Florida, was influenced by employees' years of experience. Teachers with 1-3 and 7-9 years of experience used an average of 15% of their sick and

PAGE 68

60 personal leave. Teachers with 47 years of experience used slightly more than 20% of this leave, and teachers with 11 or more years of experience, about 45% (Dade County, Florida, Public Schools Office of Management and Budget, 1978, p. 12). In contrast, teacher absenteeism was not found to be significantly related to length of continuous employment in Clark County, Nevada, according to Bundren (1974, p. 121), nor to previous educational experience, according to Marchant (1976, p 98). Redmond (1978, p. 114) concluded that neither the amount of teaching experience in the district nor total teaching experience was significantly related to the absenteeism of professional personnel in the Fort Madison ( Iowa) Community School District.

PAGE 69

CHAPTER Ill PRESENTATION AND ANALYSIS OF DATA The purpose of the study was to determine the extent which the unexpected absenteeism of teachers in selected elementary schools within the Orange County, Florida, Public School System and the unexpected absenteeism of individual teachers couid be explained by teachers' perceptions of the leadership style of principals, teacher morale, and selected teacher-related and school-related characteristics. There were five basic questions which gave direction to the study. The first three questions were preliminary and used single variant analysis to determine if there were differences among certain subsets of what were thought to be major presumed independent variables i n regard to the dependent variable of teacher absenteeism. Specifi cally, question one was designed to determine if there were differ ences in teachers' perceptions of leadership style of the principals on each of the 12 dimensions of the LBDQ-XI I among 15 selected elemen tary schools with high, average, and low unexpected teacher absen teeism. Question two was designed to determine if there were differ ences on each of 10 dimensions of teacher morale as measured by the PTO among the 3 groups of elementary schools The intent in ques tion three was to determine if there were differences in the teachers' perceptions of the leadership style of principals in the 5 schools with the highest morale scores and the 5 schools with the lowest morale scores from the 15 schools selected for the study Questions four and five related to the extent to which the presumed i ndependent variables 61

PAGE 70

62 relating to the teacher's perception of the system's orientation of the principal, the personal orientation of the principal, the teacher's total morale score, and teacher-related and school-related demographic variables could be used to explain individual teacher's unexpected absenteeism Questions four and five were different in that morale was considered an independent variable in question four and was held as a covariant in question five As previou:;ly noted, two basic statistical techniques were utilized--one-way analysis of variance and stepwise multiple regress i on. From the 67 elementary schools in the district, those five with the highest percentage of unexpected teacher absentee i sm, those five with the lowest percentage of unexpected teacher absenteeism, and those five nearest the mean percentage of unexpected teacher absen teeism rate for all elementary schools for 1978-1979 school year were selected to participate in the study The researcher solicited coop eration from the 496 teachers employed in the 15 schools and 423 teachers provided usable data The required materials for the research included two instru ments-the Leadership Behavior Description Questionnaire-Form XII ( LBDQXI I) and the Purdue Teacher Opinionnaire (PTO). The LBDQXII was used to measure the independent variables related to teachers' perceptions of the leadership style of a principal, and the PTO was used to measure the independent variable of teacher moraie. The school-related and teacher-related demographic variables were obtained from teachers by means of a researcher-developed personal data sheet.

PAGE 71

63 The focus of the present chapter is a presentation and analysis of the data in regard to the aforementioned questions. However, before turning directly to the results of the statistical analysis, in order that the reader may have a perspective for understanding the analysis, descriptive data relative to the independent and dependent variables are presented This is followed by comparisons in relation to the leadership style of the principals and teacher morale among elementary schools having high, average, and low unexpected teacher absenteeism. The third section is focused upon comparisons between high and low morale elementary schools relative to teachers' percep tions of the leadership style of the principals, and the fourth section deals with the prediction of individual unexpected teacher absenteeism from the presumed independent variables. In Table 1 each school is described by number of certificated staff members in the school, the number of study participants, the mean unexpected absenteeism rate for the school, and whether it was a Title I or non-Title I school As denoted in the table, the range by school was 20-47 certificated staff members with 8 of the schools classified as small and 7 as large schools. The number of voluntary participants at the 15 schools varied from 19 to 40, with the percen tage of participants ranging from 70% in school 10 to 100% in school 12. Overall, 423 of the 496 teachers (85%) employed in the 15 schools participated. (To strengthen validity, a t test was utilized to deter mine if there was a significant difference between the mean absenteeism rate of respondents and non-respondents. No significant difference was found at the 05 level of confidence.) The mean unexpected teacher absenteeism ranged from a low of 3 61 days in school 3 to

PAGE 72

Table 1 Basic Information Relative to the 15 Elementary Schools Selected for the Study School Size SNcuhmoobel ra Certificated Study Staff Members Participants Mean Unexpected Absenteeism Rate (in days) Title I School 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Total 47 (l)b 34 (L) 41 (L) 32 (S) 37 (L) 32 (S) 33 (S) 34 (L) 20 (S) 33 (S) 25 (S) 28 (S) 37 (L) 27 (S) 36 (L) 496 40 28 38 28 26 25 26 29 19 23 22 28 35 24 32 423 7.37 4.67 3.61 4.83 4.05 7.03 4.83 6.84 11.10 7 08 11 52 12.00 6.97 10.70 13.70 X X X X X X X X X X X 11 64 Non-Title School X X X X 4 aThe school number i s merely a designation used to protect the anonymity of the school. bThe L designation indicates the school was classified as Large for pur poses of the study; the S indicates the school was classified as Small.

PAGE 73

65 13. 70 days in school 15. Eleven of the 15 schools were classified as Title I. In order to evaluate properly the findings, it was necessary to collect descriptive data on the participants as well as compute stan dard descriptive statistics for the variables. The next section is focused on a descriptive analysis of the participants, the independent and dependent variables, and the schools. Descriptive Data by Schools Relative to Teacher-Related Variables To obtain descriptive statistics, the data were first aggregated by the subgroups--high absenteeism schools, average absenteeism schools, and low absen"i:eeism schools. The means and standard deviations for each dimension of the LBDQ-XI I and PTO were com puted. Also determined were distribut i ons by teacher sex, education level, marital status, and experience in the school district. In the paragraphs that follow, these data are presented for high absenteeism schools, average absenteeism schools, low absenteeism schools, and for the total group of schools. Presented in Table 2 are data regarding the independent vari able, leadership style of the principal as perceived by the teachers. Data for each of the 12 dimensions of the LBDQ-XI I and for the system and personal orientation categories are presented. The reader is reminded that the systems orientation consisted of the following five dimension scores--demand reconciliation, initiation of structure, production emphasis, integration, and superior orientation

PAGE 74

Table 2 LBDQ-XII Dimension Means, Orientation Means, and Standard Deviations for the 15 Elementary Schools Included in the Study LBDQ-XI I Dimension Means and Standard Deviations Demand Tolerance Representation Reconciliation of Uncertainty Persuasiveness School Standard Standard Standard Standard Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation High 9 13 68 3 98 12.89 1.85 25 37 3 35 33.21 4.14 Absenteeism 11 14 81 1.56 13.23 1 69 27 32 2 42 35.23 2.52 Schools 12 17. 71 4 31 14 25 5 34 31 50 7 99 29.61 8 71 14 13 62 1 66 12.91 1 91 28 29 2 53 30 42 6.39 15 14.25 2.42 12.97 2 02 26.81 2.67 32.34 3.33 Average 1 15.45 3 67 13 43 3.81 27.98 6.58 33 00 6.68 Absenteeism 6 15 48 4.90 15 84 4 88 33 16 6 87 31.08 8.09 Schools 8 20 38 3. 12 21 07 3 63 37 86 6 83 39 41 6.81 10 19.35 3 23 15 43 5.34 28 91 6. 74 28 30 8.25 13 15.42 1. 97 12 77 1 59 28 31 2 08 34 71 2.72 Low 2 17 75 2.88 16.29 4 79 30 71 7 29 35.89 6 34 Absenteeism 3 15.68 2.34 12.32 1.69 28 50 2.83 35 13 3.47 Schools 4 14 75 6 25 15 32 5 38 26 93 9.94 27 96 11.3 1 5 19.96 3 17 19 73 3 22 36 31 5 18 37.92 5 11 7 15 00 5 02 16.04 5.81 32 23 8 89 32.23 9.21 Total 16.22 4 08 14 87 4 52 29 98 6 82 32.22 7.24 a, a,

PAGE 75

Table 2 (continued) LBDQ-XI I Dimension Means and Standard Deviations Initiation of Tolerance of Role Structure Freedom Assumption Consideration School Standard Standard Standard Standard Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation High 9 30.79 4.33 32 37 5.33 26.16 2.41 26.42 3.44 Absenteeism 11 32.95 2.84 33 77 3.53 26 64 1 87 26.73 2.99 Schools 12 32.18 9.69 36 68 8.73 28.18 10 34 34 96 8.43 14 26 75 5. 10 31 29 5 54 25.83 2.35 25.08 3.03 15 30.06 3.44 30 84 5.54 27.78 3.32 27.38 3.03 Average 1 33 48 7. 19 33 13 8 30 27.07 6.51 28 55 7 .16 Absenteeism 6 33.48 8 38 38 72 4.24 31. 16 10.04 31.60 8 86 Schools 8 40 86 4.72 42 83 4.68 41.69 6 45 41.24 5.60 10 32.83 8.73 33 43 7.36 30.87 8.16 31.57 7.29 13 33.54 2.75 32.89 5.47 25.63 2.49 28.60 2.49 Low 2 37.54 6.56 36.64 5 12 35.61 7 94 33.93 7 02 Absenteeism 3 33.97 3.36 34.68 5.00 26.79 2 55 26.45 2.65 Schools 4 32.18 10. 16 28.29 12 57 28 14 12.78 28.25 11.82 5 41.00 4.51 38.46 7 08 40.92 6 63 38.96 5.57 7 33.81 8.44 30 38 8.88 32 12 6.52 26 58 10.72 Total 33.79 7.24 34.28 7 72 30.17 8 40 30.38 8.03

PAGE 76

Table 2 (continued) LBDQ-XII Dimension Means Production Predictive Emphasis Accuracy School Standard Standard Category Number Mean Deviation Mean Deviation High 9 30 32 3.28 17 53 2 91 Absenteeism 11 29 23 3 56 17.64 2 01 Schools 12 28 57 7.75 15.11 4 31 14 25 42 4.46 16.88 2 56 15 29.81 2 83 19.06 2.03 Average 1 29 .4 8 6.83 17 .4 0 3 43 Absenteeism 6 30 36 7 62 16.32 4.91 Schools 8 32 45 4 59 19 86 2 81 10 30 35 7 73 13 96 3 80 13 31 31 3.49 19.03 2 02 Low 2 31 64 6 76 16 61 3 82 Absenteeism 3 30 68 3 17 18.76 2.19 Schools 4 27 79 9 07 13 64 5 49 5 33 88 5 67 18 15 2 84 7 27 77 5.26 16.31 4 25 Total 29 99 6 01 17.22 3.80 and Standard Deviations Superior Integration Orientation Standard Standard Mean Dev ia tion Mean Deviation 14 05 2.22 32 00 4 08 14 27 1.55 34.36 3.09 15 07 4 89 26.89 11 35 13 25 2 36 30.25 4 73 14.28 1.94 32.03 3 06 14 93 4 23 34.75 6 83 15.56 5.32 29.12 10 50 21.10 4 43 34 59 9 79 13 83 4.46 30 35 11 80 15.17 1 64 33.43 3.16 16 75 3 87 30 43 7.01 15 82 1 96 32.66 3.84 16.21 3 93 25 29 9.79 18.85 3 54 34 73 8.03 14 08 5 53 23 62 11.25 15 61 4.12 31. 15 8.35 m ex,

PAGE 77

Table 2 (continued) LBDQ-XI I Orientation Means and Standard Systems Orientation Personal Orientation School Standard Standard Category Number Mean Deviation Mean Deviation High 9 120.05 9.73 174.74 14 23 Absenteeism 11 124 05 8 43 182. 14 7.45 Schools 12 116.96 35 36 193.75 47.76 14 108 58 14.39 171 42 15.82 15 119 16 8.45 178.47 12.09 Average 1 126.05 26.42 182.58 38.46 Absenteeism 6 124 36 31 02 197 52 38 89 Schools 8 150.07 20.32 243.28 29.25 10 122 78 32.50 186.39 38.76 13 126.23 8.29 184 60 10.33 Low 2 132.64 17 90 207 14 27 57 Absenteeism 3 125.45 9.77 186 00 11 91 Schools 4 116.79 33 60 167.96 66.78 5 148. 19 16 45 230.69 25 42 7 115 32 30.94 184 84 47 48 Total 125 42 24.48 191.48 38.73 Deviations Total Standard Mean Deviation 262.79 19.81 271 81 12.50 283 82 73.37 249.75 24.64 265.59 15 77 273.88 58.22 292.76 62 53 358.76 40 45 278.83 61.58 277.40 14 66 309.36 43.90 278.79 17.01 259.46 92 98 344 15 34.53 276 54 68.59 285. 75 56.24 0,
PAGE 78

70 The personal orientation consisted of the remaining seven dimension scores--representation, tolerance of uncertainty, persuasiveness, role assumption, consideration, predictive accuracy, and tolerance of freedom. It can be seen in Table 2 that mean scores for schools in the high absenteeism group for the representation dimension of leadership were, with the exception of one school, below the mean score for the total population surveyed. Scores for three of the five schools in the average and low aosenteeism groups were also below the mean scores for this dimension. Mean scores for all schools in the high absen teeism group for the demand reconciliation dimension were below the mean for the total population, while scores for three of the average absenteeism schools and four of the low absenteeism schools exceeded the total population mean score. Four schools in the high absenteeism group were below the total population mean for the dimension toler ance of uncertainty, while two schools in the average absenteeism group and three schools in the low absenteeism group were above the total mean Three schools in each group were above the total popu lation mean on the persuasiveness dimension. All schools in the high absenteeism group and four schools in the average absenteeism group were below the total population mean on the initiation of structure dimension, while four schools in the low absenteeism group exceeded the total mean score. Four schools in the high absenteeism group and three schools in the average absenteeism group were below the total mean on the tolerance of freedom dimension, and three schools in the low absenteeism group were above the total mean All schools in the

PAGE 79

71 high absenteeism group were below the total mean on the dimension role assumption, while three schools in the average absenteeism group and three schools in the low absenteeism group were above the mean for the total population. Four schools in the high absenteeism cate tory were below the total mean for the consideration dimension and three schools in the average absenteeism group and two schools in the low absenteeism group were above the total mean Four schools in the high absenteeism group also were below the total mean on the production emphasis dimension while four schools in the average absenteeism g r:iup and three schocls in the low absenteeism group exceeded the total mean score. However, for the predictive accuracy dimension, three of the high absenteeism schools and average absen teeism schools were above the total mean, and three of the low absen teeism schools were below the total mean. All schools in the high absenteeism group and three schools in the average absenteeism group were below the total mean on the integration dimension while four of the low absenteeism schools exceeded the total population mean. Three of the schools in the high absenteeism and the average absen teeism categories were above the total mean on the superior orienta tion dimension while three of the low absenteeism schools were below the total mean. For the systems orientation of the principal dimension all of the schools in the high absenteeism group were below the total population mean. Three of the schools in the average absenteeism group and three in the low absenteeism group were above the total mean for this orientation. For the personal orientation of the principal, four schools

PAGE 80

72 in the high absenteeism group and three schools in the average absenteeism group and three schools in the low abse:nteeism group were below the total population mean. Examining the LBOQ-Xl I total mean scores for each group it can be seen that all schools in the high absenteeism group were below the total population mean score, while one school in the average absenteeism group and two schools in the low absenteeism group were above the total population mean score. The mean leadership score for each dimension was low compared to the scores obtained by the groups upon whom the instrument was field tested. Presented in Table 3 are data regarding each of the 10 dimen sions of the PTO and a total morale score for each school is shown. It can be noted that one school in each category--school 12 in the high absentee i sm category, school 8 in the average absenteeism cate gory, and school 3 in the low absenteeism category--consistently scored in a dissimilar manner from other schools in their categories. For the teacher rapport with principal dimension, four schools in the high absenteeism category were below the mean score for the total population, three schools in the average absenteeism category were above the total mean, and four schools in the low absenteeism cate gory were above the total mean. Four of the high absenteeism cate gory schools were below the total mean, and three of the average absenteeism category schools and four of the low absenteeism category schools were above the total mean for the satisfaction with teaching dimension Four of the high absenteeism category schools were below the total mean on the rapport among teachers dimension, while three

PAGE 81

Table 3 PTO Dimension Means, Total Means, and Standard Deviations for the 15 Elementary Schools Included in the Study PTO Dimension Means and Standard Deviations Teacher Rapport Satisfaction Rapport Teacher with Principal with Teaching Among Teachers Salary School Standard Standard Standard Standard Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation High 9 21 42 2.39 20.68 3 67 17.47 2.01 6.89 1.70 Absenteeism 11 20.05 3 23 23.59 2 84 17 32 2.17 6.73 2.00 Schools 12 58.90 11 36 66.18 9 17 39. 71 6.05 17.46 3.97 14 21 88 3.84 22.29 2 39 15.08 2.87 7.83 1.27 15 22.81 2 28 21.34 3 24 17 .16 3.34 7.50 1.14 Average 1 27 70 15.97 29.38 15.37 21.85 9 93 9 43 4.32 Absenteeism 6 42.24 21 08 48.40 22.32 33.56 15 39 13.60 5.79 Schools 8 69.28 10 90 70 41 6 91 50.41 5.34 17 66 2 83 10 50.22 12.78 63 17 9.17 37.30 7 18 16 70 3.78 13 22 60 3 36 23 51 3.04 17.86 2 61 7.46 1 20 Low 2 53. 71 21.13 54 11 20.32 38.46 14 94 15.07 5 70 Absenteeism 3 26 24 2.77 24.84 2.40 19.39 1.72 8.63 1.28 Schools 4 55.29 11.72 62 43 12.30 45.04 5.33 19.29 3 85 5 65.50 11. 18 64 92 11.16 44.81 7. 15 17.38 4.43 7 45.69 19.63 66 58 14. 41 39.85 11.08 16.73 4 80 Total 39.74 20.91 43.29 22.59 29.95 14.33 12 40 5.78 ..... w

PAGE 82

Table 3 (continued) PTO Dimension Means and Standard Deviations Curriculum Community Support Teacher Load Issues Teacher Status of Education School Standard Standard Standard Standard Category Number Mean Deviation Mean Deviation Mean Deviation Mean Deviation High 9 23 89 3 20 9 89 2.11 9.42 1.80 8.42 2.65 Ab~ enteeism 11 23.77 2 89 9.91 1 11 9 86 1.52 9.14 2 92 Schools 12 27 68 5.49 15 50 3 44 24 14 5.25 14.93 2.65 14 23 04 2 56 10 38 2.10 8.83 1.86 7.63 3.50 15 22.84 3 19 9 81 2.02 9 09 2.02 7 47 2 59 Average 1 24 08 5 63 10 65 3.13 11.58 5.01 9.60 2.37 Absenteeism 6 25.12 6.83 11.88 4.77 16 68 8.01 8 72 3 0::i Schools 8 31.97 4 92 18 10 1.84 22.90 5.21 15.07 2.43 10 26 87 7.46 14 61 3 31 20 43 6.60 13 26 2.90 13 23.51 3 11 9.49 1.63 10.31 1.55 9 06 2 15 Low 2 27 25 4 80 14.54 3.62 18 07 7.48 9 64 3 29 Absenteeism 3 23.53 2.39 9 74 1.46 10 32 1 47 9.21 1.74 Schools 4 29 21 7.04 16.07 3.23 23 32 7.07 15 64 2.59 5 28. 15 5 86 15.35 3.19 20 81 6 63 13 77 3.36 7 30. 77 7.43 14.54 4 65 23 38 7 99 13 58 4.01 Total 26.01 5 79 12.58 4.06 15 72 7 83 10.95 3 95 '-I A

PAGE 83

Table 3 (continued) PTO Dimension Means and Standard Deviath:ms School Facilities Community Total and Services Pressures Mean Morale School Standard Standard Standard Category Number Mean Deviation Mean Deviation Mean Deviation High 9 5 47 0 84 8.21 2.10 111.79 12.63 Absenteeism 11 5.50 0.96 8.00 1.89 E3 86 10.01 Schools 12 14.61 2 73 14.43 1.48 293.54 36.88 14 5.46 0 66 8 08 2.12 130.50 12.28 15 5.56 1.32 8 00 1.81 131.59 12 63 Average 1 6.55 4.27 9 33 2.44 160.13 63.00 Absenteeism 6 10 16 4.57 12.97 3.98 223.32 86.01 Schools 8 17 14 2.46 15.76 2.23 328.69 27.64 10 14 65 2 53 14 83 2.52 272.04 46 73 13 5 66 1.06 7.94 1.75 137.40 11.70 Low 2 13 04 4.89 13.64 4.05 257.54 80 37 Absenteeism 3 5.58 0 72 7.50 2.00 144.97 7.39 Schools 4 14.21 5 60 13.68 3.61 294 18 49.97 5 14 12 3.59 15.00 2.83 299.81 44.36 7 13.46 4 43 15.54 3.48 280.12 60.61 Total 9.89 5 35 11.35 4 16 211 90 86 71 -..J c.n

PAGE 84

76 schools in the average absenteeism category and four schools in the low absenteeism category were above the total population mean. For the teacher salary dimension, four of the high absenteeism category schools were below the total maan, while three of the average absen teeism category schools and four of the low absenteeism category schools were above the total mean Four of the high absenteeism category schools and three of the average absenteeism category schools were below the total mean on the teacher load dimension, while four of the low absenteeism category schools were above the total popula tion mean. For the curriculum issues dimension four of the high absenteeism category schools and three of the average absenteeism category schools were below the total mean, and four of the low absenteeism category schools were above the total mean. Four of the high absenteeism category schools were below the total mean, while three of the average absenteeism category schools and four of the low absenteeism category schools were above the total population mean. For the community support of education dimension, four of the high absenteeism category schools and three of the average absenteeism category schools were below the total mean, while three of the low absenteeism category schools were above the total mean. Four of the high absenteeism schools were below the total mean for the school facilities and services dimension and for the community pressures dimension, and three of the average absenteeism schools and four of the low absenteeism schools were above the total mean. The low absenteeism category schools exhibited, as a group, the highest total morale scores as measured by the PTO. Average absenteeism cate gory schools had the second highest group morale scores, with high

PAGE 85

77 absenteeism category schools attaining the lowest morale scores If the aforement i oned schools--school 12, school 8, and school 3--were treated as outliers, the within and between group comparisons would become even more distinct. Shown in Table 4 are data regarding each of the four teacher related independent variables--sex, education level, marital status, and experience in the school district. As one reviews the table, it can be seen that 44 of the teacher participants in the study were male and 379 female. Two hundred and seventy had bachelor's de grees, 151 had master's degrees, 1 had a specialist's degree, and 1 had a doctorate. One hundred and twenty were single and 303 were married. One hundred and thirty-two teachers had 1-5 years experi ence in the school district, 123 had 6-10 years experience, 74 had 11-15 years experience, and 94 had more than 15 years experience in the district. In terms of the biographical characteristics of the teachers, the three categories of schools were similar, and within group differences were more pronounced than differences among groups. In terms of teacher characteristics the high absenteeism category schools were most different from average absenteeism and low absenteeism category schools in the percentage of teachers with 15+ years in the district Twelve percent of the teachers in the high absenteeism schools had 15+ years of experience, compared to 24% of the teachers in the average absenteeism schools, and 29% in the low absenteeism schools. Comparisons Among Elementary Schoois with High, Average, and Low Unexpected Teacher Absenteeism As has been noted, one-way analysis of variance ( F test) was used to analyze the extent of the differences in principals' leadership

PAGE 86

Table 4 Number of Teacher Participants, Male-Female Ratios, Distribution by Education Level, Marital Status, and Experience in the School District for the 15 Elementary Schools Included in the Study Teachers' Number of Teachers Education Level of Teachers Marital Status School Category Number Male Female Total Bachelors Masters Spi:.,cialist Doctorate Single Married High 9 2 17 19 15 4 0 0 3 16 Absenteeism 11 1 21 22 14 8 0 0 5 17 Schools 12 0 28 28 18 10 0 0 6 22 14 2 22 24 18 6 0 0 11 13 15 4 28 32 24 7 0 1 5 27 Average 1 5 35 40 17 23 0 0 10 30 Absenteeism 6 4 21 25 15 10 0 0 7 18 Schools 8 0 29 29 15 14 0 0 6 23 10 5 18 23 18 5 0 0 8 15 13 6 29 35 25 10 0 0 9 26 Low 2 4 24 28 21 7 0 0 15 13 Absenteeism 3 4 34 38 27 11 0 0 10 28 Schools 4 2 26 28 17 11 0 0 7 21 5 2 24 26 10 15 1 0 9 17 7 3 23 26 16 10 0 0 9 17 T0tal 44 379 423 270 151 120 303

PAGE 87

Table 4 (continued) Teachers' Exeerience in School District School Category Number 1-5 years 6-10 years 11-15 years 15+ years High 9 7 8 3 1 Absenteeism 11 4 5 10 3 Schools 12 5 9 7 7 14 14 5 1 4 15 16 11 5 0 Average 1 10 11 10 9 Absenteeism 6 8 5 4 8 Schools 8 3 8 4 14 10 12 9 0 2 13 9 14 9 3 Low 2 9 12 4 3 Absenteeism 3 16 11 5 6 Schools 4 4 3 3 18 5 7 6 7 6 7 8 6 2 10 Total 132 123 74 94

PAGE 88

80 style and teacher morale, among the three groups of schools. The focus herein is on those comparisons relative to mean scores on the LBDQ-XI I and the PTO for schools grouped by high, average, and low unexpected teacher absenteeism. Table 5 displays indicators of the teachers' perceptions of the principals' leadership style among the three groups of schools. Statis tical comparisons among the groups, with F values and level of sig nificance, are shown for each dimension of the LBDQ-XI I instrument. It can be seen that a significant statistical difference, at the 05 level of confidence, was found for 10 of the 12 dimensions when groups of schools were compared. As noted previously, when a significant difference was found among the groups, the Tukey test was used to locate the differences. Two dimensions--persuasiveness and pre dictive accuracy--were not statistically different among groups at the .05 level. For the representation dimension, the high absenteeism schools were significantly different at the .05 level from schools with average or low absenteeism There was no significant difference between average and low absenteeism schools on this dimension The same pattern existed among the three groups for the dimensions of demand reconciliation, tolerance of uncertainty, initiation of structure, role assumption, and integration. The difference between high absenteeism schools and low absenteeism schools was not significant at the 05 level for the tolerance of freedom dimension. Schools with average teacher absenteeism had a higher mean score on this dimension which was significantly different from both high absenteeism and low absenteeism schools For the consideration and production emphasis

PAGE 89

LBDQ-XII Dimensions Representation Demand Reconciliation Table 5 LBDQ-XII Dimension Means, F Values, and Observed Alpha Level for Schools Classified as High, Average, Low in Teacher Absenteeism Mean Schools with High Absenteeism 14.92 13 28 Mean Schools with Average Absenteeism 16 98 15 43 Mean Schools with Low Absenteeism 16 54 15.63 F Value 9.83* 11 56* Observed Alpha Level .0001 0000 Tolerance of Uncertainty 28.01 30 93 30 67 7 67* 0005 Pe:-suasiveness 32.16 33.30 33 83 2 60 0749 Initiation of Structure 30 52 34.80 35.53 20. 16* .0000 Tolerance of Freedom 32.98 35 88 33.73 5.53* .0042 Role Assumption 27 04 30.77 32.20 14.15* 0000 Consideration 28.37 31 94 30 47 6.96* 0011 Production Emphasis 28 66 30.74 30 36 4 59* .0106 Predictive Accuracy 17.24 17 31 16 69 1 37 .2539 Integration 14.22 16 09 16 30 10.65* .0000 Superior Orientation 30 94 32 82 29 57 5.81* .0032 *e (05

PAGE 90

82 dimensions there was no significant difference between high absenteeism schools and low absenteeism schools, nor was there a s i gnificant difference between average absenteeism schools and low absenteeism schools. However, for both of these dimensions there was a sign i ficant difference at the OS level between average absenteeism schools and high absentee i sm schools There was no significant difference between h i gh and low absenteeism schools on the superior orientat i on dimension, but there was significant difference at the 05 level between average and l ow absenteeism schools Shown in Table 6 are the results of a one-way analysis of vari ance which prov i ded comparative statistics among the three groups of schools on the bas i s of unexpected teacher absenteeism category and teachers' scores on the PTO The reader is reminded that the PTO instrument was used to measure teacher morale. Mean scores, F values, and sign i ficance levels are shown for each d i mension of the instrument. The differences among groups for each of the 10 d i men sions were found to be significant at the 05 level of confidence Again, where signif i cant differences were found the Tukey test was used to locate the differences For the dimensions, teacher r apport with principal, satisfaction with teaching, rapport among teachers, teacher salary, curr i culum issues, teacher status, community support of educat i on and school facilities and services significant differences at the 05 level were found between the following pairs : high absenteeism schools--low absenteeism schools, high absenteeism schools--average absenteeism schools, and low absenteeism schools--average absenteeism schools. For the dimensions teacher load and community pressures, significant

PAGE 91

Table 6 PTO Dimension Means, F Values, and Observed Alpha Level for Schools Classified as High, Average, Low in Teacher Absenteeism Mean Schools Mean Schools Mean Schools with High with Average with Low Observed PTO Teacher Teacher Teacher F Alpha Dimensions Absenteeism Absenteeism Absenteeism Value Level Teacher Rapport with 30.01 40.25 47 53 26.59* .0000 Principal Satisfaction with Teaching 31 86 44.09 52 23 31 51* .0000 Rapport among Teachers 21.88 30 64 36.13 39 71* .0000 Teacher Salary 9.56 12.32 14. 91 33 16* .0000 Teacher Load 24 28 26.04 27.44 10 46* .0000 Curriculum Issues 11 22 12.60 13.72 13 58* .0000 Teacher Status 12 60 15.62 18.49 20 89* .0000 Community Support of 9 60 10.92 12 11 14.41* .0000 Education School Facilities and 7.54 10.18 11 58 21 49* .0000 Services Community Pressures 9.48 11.66 12 63 21.79* .0000 *e <.o5 co (.)

PAGE 92

84 differences were found between high absenteeism schools and low absenteeism schools and high absenteeism schools and average absen teeism schools. No significant difference was found between low absenteeism schools and average absenteeism schools for these two dimensions. Mean scores in low absenteeism schools were higher on each of the 10 dimensions than were the means in the average and high absenteeism schools. The average absenteeism schools also had higher means on every dimension than did the high absenteeism schools The greatest difference in mean scores occurred between high absen teeism schools and low absenteeism schools for the dimensions of teacher rapport with principal (30.01--47 53), satisfaction with teach ing (31.86--52.23), rapport among teachers (21 88--36 13), and teacher salary (9.56--14.91). Comparisons Among High and Low Morale Elementary Schools Relative to Teachers' Perception of the Leadership Style of the Principal As has been noted, one-way analysis of variance ( F test) was used to analyze the differences in the teachers' perceptions of the leadership style of principals in high and low morale elementary schools. The schools were grouped according to the five highest morale schools and the five lowest morale schools. ( Four of the five low morale schools were also four of the five schools with the highest unexpected teacher absenteeism.) Where a significant difference at the predetermined level (the .05 level of confidence) was found, the Tukey test was employed to locate the difference within the set

PAGE 93

85 Displayed in Table 7 are F values and dimension means, orienta tion means, and total mean scores for the two groups of schools--high morale and low morale--on the LBDQ-Xl I. It may be observed that four of the high morale schools were above the total population mean (based on data from all 15 schools) while all of the low morale schools were below the total mean on the representation and demand reconcil i tion dimensions. Four of the high morale schools and three of the low morale schools were above the total population mean on the toler ance of uncertainty dimension. However, on the persuasiveness dimension three of the high morale schools were below the total mean, while four of the low morale schools were above the total mean Three of the high morale schools and all of the low morale schools were below the total mean on the initiation of structure dimension For the tolerance of freedom and role assumpt i on dimensions, three of the high morale schools were above the total population mean, wh i le all of the low morale schools were below the total mean. Four of the high morale schools were above the total mean on the consideration dimension and all of the low morale schools were below. Three of the high morale schools were above and three of the low morale schools were below the total population mean on the production emphas i s dimension. The high morale schools were below the low morale schools on predictive accuracy, with three high morale schools being below the total mean score and four of the low morale schools being above Three high morale schools were above the total mean while all of the low morale schools were below the total mean on the integration dimen sion On the final dimension, superior orientation, two of the high morale schools and four of the low morale schools were above the total

PAGE 94

School Category Number High 4 Morale 5 Schools 8 10 12 Low 9 Morale 11 Schools 13 14 15 Total Populationa F Value b Observed Alpha Level *e < 05 Table 7 F Values and LBDQ-XII Dimension, Orientation, and Total Means for Schools Classified as High and Low in Teacher Morale Dimension Means Demand Tolerance of Representation Reconciliation Uncertainty Persuasiveness 14.75 15.32 26 93 27.96 19.96 19 73 36.31 37 92 20 .3 8 21 07 37.86 39.41 19.35 15.43 28 91 28 30 17. 71 14.25 31 50 29 61 13.68 12.89 25 37 33 21 14. 81 13.23 27 32 35.23 15 42 12.77 28 31 34 71 13 62 12 91 28 29 30 42 14.25 12.97 26.81 32.34 16.22 14 87 26 98 32 22 9 83* 11 56* 7.67* 2.61 .0001 0000 0005 .0749 Cl) Ol

PAGE 95

Table 7 (continued) Dimension Means School Initiation of Tolerance of Role Category Number Structure Freedom Assumption Consideration High 4 32 18 28.29 28.14 28 25 Morale 5 41.00 38.46 40 92 38.96 Schools 8 40.86 42 83 41 69 41 24 10 32.83 33.43 30 87 31.57 12 32 18 36 68 28 18 34 96 Low 9 30.79 32 37 26 16 26.42 Morille 11 32.95 33 77 26.64 26.73 Schools 13 33.54 32 89 25.63 28 60 14 26 75 31 29 25.83 25 08 15 30.06 30.84 27.78 27 38 Total Population 33 79 34.28 30.17 30.38 F Vdlue 20 16* 5.53* 14. 16* 6.97* Observed Alpha Level .0000 .0042 .0000 0011 *e (.05

PAGE 96

School Production Category Number Emphasis High 4 27 79 Morale 5 33.88 Schools 8 32 45 10 30.35 12 28.57 Low 9 30.32 Morale 11 29 23 Schools 13 31 31 14 25 42 15 29 81 Total Population 29.99 F Value 4 60* Observed Alpha Level .0106 *e < o5 Table 7 (continued) Dimension Means Predictive Accuracy 13 64 18.15 19.86 13 96 15. 11 17.53 17.64 19.03 16.88 19.06 17 22 1.38 .2539 Integration 16. 21 18 85 21 10 13 83 15.07 14 05 14.27 15.17 13.25 14.28 15.61 10.66* .0000 Superior Orientation 25.29 34.73 34.59 30.35 26.89 32.00 34.36 33.43 30 25 32.03 31.15 5.82* .0032 CD CD

PAGE 97

Table 7 (continued) Orientation Means School Category Number Systems Personal High 4 116 79 167.96 Morale 5 148 19 230 69 Schools 8 150 07 243 28 10 122 78 186 39 12 116 96 193.75 Low 9 12 0. 05 174 74 Morale 11 124.05 182 14 Schools 13 126 23 184 60 14 108 58 171.42 15 119 16 178 47 Total Populat i on 125 42 191.48 F Value 14 02 30 94 Observed Alpha Level 0002 .0000 aRefers to the mean computed for all 1 5 schools in the study Total Means 259 46 344 15 358 76 278.83 283 82 262 79 271. 81 277.40 249 75 265.59 285 75 33 97 .0000 bRefe r s to the computed F value when the high and low morale schools were compared e (05

PAGE 98

90 population mean. Two of the high morale schools and one of the low morale schools were above the total population mean on the systems orientation section of the LBDQ-XI I Three high morale schcc!:; w ere above the total mean on the personal orientation section while all of the low morale schools were below this mean. Two of the high morale schools were above the total population mean and all of the low morale schools were below the mean on the total instrument score. The high morale schools were significantly different at the 05 level from the low morale schools on 10 of 12 dimensions of the LBDQ-Xll--representation (F value--9 83), demand reconc i liation (F value--11 56), tolerance of uncertainty (F value--7.67), initiation of structure (F value--20.16), tolerance of freedom (F value--5 53), role assumption (F value--14.16), consideration (F value--6.97), production emphasis (F value--4 60), integration (F value--10.66), and superior orientation (F value--5 82). In each instance the means in the high morale schools were higher No significant difference at the 05 level between the groups was found for the pers.iasiveness ( F value--2. 61) and predictive accuracy ( F value--1 38) dimensions The scores in the high morale schools on the systems ( F value--14. 02) and personal ( F value--30. 94 ) orientations were significantly higher at the 05 level than the scores in the low morale schools The total LBDQ-XII mean scores were also significantly higher in the high morale schools than in the low morale schools ( F value--33 97) Prediction of Unexpected Teacher Absentee i sm From Selected Variables Presented in Table 8 are the results of the regression analysis using all 10 independent variables to predict the extent to wh i ch

PAGE 99

Table 8 Results of Multiple Regression Analysis of 10 Independent Variables as Predictors of Individual Teacher Unexpected Absenteeism Standard Error Regression of Regression Beta Variable Coefficient Coefficient Weight Systems Orientation of Principal -0 108 0.020 -0.095 Personal Orientation of Principal -0 088 0.014 0.045 Teacher PTO Total Score -0.039 0.004 0. 110 Teacher Education Level Bachelor -0.034 Master 0.037 0.507 0 024 Specialist -0.032 Doctorate 0 010 4.816 -0.021 Teacher Sex Female -0.029 0.766 -0 042 Male 0 029 Teacher Marital Status Single -0.174 0.537 -0 090 Married 0.174 Teacher Experience in the District 1-5 years -0.173 0.622 -0. 153 6-10 years 0.060 11-15 years 0.037 0 713 -0 015 15+ years 0.094 0.691 0.063 Title I School -0. 108 0.629 0.149 Non-Title I School 0 108 F Value 0.96 0.18 2.63 0 25 0.21 0.87 3 66 7.42 0.08 1.28 7 .12 U)

PAGE 100

Variable School Size School Absentee ism Category Notes : Multiple R = 42 R 2 = 18 Small Large Low Average H i gh F with 16, 406 df 1.69 e =( 195, Table 8 (continued) Regression Coefficit,nt 0 1 ~ 7 -0 157 -0 222 -0.033 0 266 Standard Error of Regression Coefficient 0.10 0.581 0 829 Beta Weight 0 197 0. 131 0.396 F Value 6 31 5.83 28 90

PAGE 101

93 individual teacher unexpected absenteeism could be predicted. As reported in Chapter I the value of the independent variables making a contribution to the presumed dependent variable will be in part deter mined by the degree of intercorrelatior: among the independent vari ables. The intercorrelations are shown in Appendix F As can be seen in the appendix, two of the independent variablP.s, low unexpected absenteeism schools and high unexpected absenteeism schools, corre lated significantly (e < Q.S, n = 423). The 10 independent variables in the regression analysis were the teacher's perception of the system's orientation of the principal, the teacher's perception of the personal orientation of the principal, the teacher's total morale score on the PTO, the teacher's level of educa tion, the teacher's sex, the teacher's marital status, the teacher's experience in the school district, whether the teacher was employed in a Title I or a non-Title I school, whether the teacher was employed in a large or a small elementary school, and whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absenteeism. As one reviews Table 8, it can be seen that the overall F value of 1.69 was not significant at the set alpha level of .05. In fact, the alpha level for an F value of 1 69 is .195. Examining the independent variables separately, PTO total score was significant at the 05 level with an F value of 2. 63, single marital status with an F value of 3. 66, 1-5 years experience in the district with an F value of 7. 42, T i tle I school assignment with an F value of 7 12, small school size with an F value of 6.31, average absenteeism rate with an F value of 5 83, and high absenteeism rate with an F value of 28.90. The next

PAGE 102

~4 highest F value of 1.28 was insignificant for 15+ years experience in the district at the 05 Alpha level. The multiple correlation coefficient was 42 indicating that the independent variables accounted for 17. 72% of the total variance (Appendix G). Presented in Table 9 are the results of the regression analysis of nine independent variables as predicters of unexpected teacher absenteeism with a PTO total score treated as a covariate. The predetermined Alpha level was again .05. It can be seen that the overall F value of 1. 69 was not significant at the 05 level, and that the PTO total score treated as a covariate produced no significant change in the F values or statistical significance of the independent variables. Those independent variables which produced significant interaction when the PTO total score was used as an independent variable remained significant when the PTO total score was treated as a covariate in the regression equation (Appendix H).

PAGE 103

Table 9 Results of Multiple Regression Analysis of Nine Independent Variables as Predictors of Individual Teacher Unexpected Absenteeism when the Teacher's Total Morale Score (PTO) is Treated as a Covariate Standard Error Regression of Regression Beta Variable Coefficient Coefficient Weight Systems Orientation of Principal -0 108 0 020 -0 094 Personal Orientation of Principal -0 088 0.014 0.041 Teacher Education Level Bachelor -0.034 Master 0.037 0 506 0 026 Specialist -0 032 Doctorate 0 010 4 815 -0.021 Teacher Sex Female -0 029 0.766 -0.043 Male 0 02fl Teacher Marital Statu s Single -0 174 0 537 -0.090 Married 0 174 Teacher E x perience i n the District 1-5 years -0 173 0 621 -0 152 6-10 years 0 060 11-15 years 0 037 0. 713 -0.015 15+ years 0.094 0.688 0.059 Title I School -0 108 0.629 0 149 Non Title I School 0 108 F Value 0.94 0.15 0 31 0.21 0 90 3 66 7 43 0.08 1 13 7 22 ID (/1

PAGE 104

Variable School Size School Absentee ism Category Notes: Multiple R = .42 R 2 = 18 Small Large Low Average High F with 16, 406 df 1.69* *e =(195 Table 9 (continued) Regression Coefficient 0.157 -0 157 -0.222 -0.033 0.266 Standard Error of Regression Coefficient 0.808 0.579 0.828 Beta Weight 0 202 0.128 0.395 F Value 6 65 5.57 28 79

PAGE 105

CHAPTER IV SUMMARY, CONCLUSIONS, AND DISCUSSION Summary The problem in the study was to determine the extent to which the unexpected absenteeism of teachers in selected elementary schools within the Orange County, Florida, Public School System and the unexpected absenteeism of individual teachers could be explained by teacher perceptions of the leadership style of principals, teacher morale, and selected teacher-related and school-related characteristics. Specifically, answers to the following questions were sought: 1. Were there differences in teacher perceptions of the leadership style of the principals on each of the 12 dimensions of the Leader Behavior Descript : on Questionnaire-Form XII (LBDQ-XII), among elementary schools with high, average, and low unexpected absenteeism? 2. Were there differences in the 10 dimensions of the teacher morale as measured by the Purdue Teacher Opinionnaire (PTO), among elementary schools with high, average, and low unexpected absenteeism? 3. Were there differences in the teachers' perceptions of the leader ship style of principals of high and low morale elementary schools? 4. To what extent could individual teacher unexpected absenteeism be predicted from a. the teacher's perception of the systems orientation of the principal, 97

PAGE 106

98 b the teacher's perception of the personal orientation of the principal, c. the teacher's total morale score on the PTO, d. the teacher's level of education, e. the teacher's sex, f. the teacher's marital status, g. the teacher's experience in the school district, h. whether the teacher was employed in a Title I or non-Title I school, i. whether the teacher was employed in a large or a small elementary school, and j. whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absenteeism? 5. When the teacher's total morale score was treated as a covariate, to what extent could individual teacher unexpected absenteeism be predicted from a. the teacher's perception of the systems orientation of the principal b the teacher's perception of the personal orientation of the princ i pal, c. the teacher's level of education, d the teacher's sex, e. the teacher's marital status, f the teacher's experience in the school district, g. whether the teacher was employed in a Title I or non-Title I school,

PAGE 107

99 h. whether the teacher was employed in a large or a small elementary school, and i. whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teachef" absenteeism? The study was confined to teachers employed in 15 selected elementary schools in the Orange County, Florida, Public School System. The 15 schools were selected from the 67 elementary schools in the district and included those 5 with the highest percentage of unexpected teacher absenteeism, those 5 with the lowest percentage of unexpected teacher absenteeism, and those 5 nearest the mean percen tage of unexpected teacher absenteeism for all elementary schools during the 1978-1979 school year All teachers in each of the 15 schools (a total of 496 teachers) were asked to participate in the study and 423 voluntarily participated. ( External validity was strengthened by administering a t test to determine if a significant difference in the rate of unexpected teacher absenteeism occurred between respondents and non-respondents. No significant diiference was found at the 05 level of confidence ) The independent variable in question one, teacher perceptions of the leadership style of the principal, was measured by the Leader Behavior Description Questionnaire-Form XII (LBDQ-XII). Each participant indicated the extent the behavior described by each item characterized the behavior of the principal of the school where the participant was employed. The LBDQ-XI I measures 12 dimensions of leader behavior which may be grouped into 2 general categoriessystems orientation and personal orientation. The dimensions grouped

PAGE 108

100 as systems orientation are demand reconciliation, initiation of struc ture, production emphasis, integration, and superior orientation. Those grouped as personal orientation are representation, tolerance of uncertainty, persuasiveness, role assumption, consideration, predictive accuracy, and tolerance of freedom. Data on unexpected teacher absenteeism, the depel"'dent variable, were obtained from the school district personnel and payroll departments. Teacher morale, the independent variable in question two, was measured by the Purdue Teacher Opinionnaire (PTO) Each study participant completed the instrument which contains 100 items and provides a total morale score derived from the subscores on 10 dimen sions of morale The dimensions measured were teacher rapport with the principal, satisfaction with teaching, rapport among teachers, teacher salary, teacher load, curriculum issues, teacher status, community support of education, school facilities and services, and community pressures. The teacher-related and school-related charac teristics which, along with leadership orientation and teacher morale, served as independent variables in questions four and five were obtained from school district records and from a researcher-developed personal data sheet which was completed by each participant. Two basic statistical techniques were utilized to answer the questions posed in the statement of the problem. To answer ques tions one, two, and three, the basic technique utilized was the one way analysis of variance ( F test) followed where significant difference at the .OS level was found, by the Tukey test. To answer questions four and five, stepwise multiple regression was used to determine the contributions of the independent variables to the dependent variable.

PAGE 109

101 Relative to demographic data regarding teacher-related and school-related characteristics, it was found that: (a) 31% of the participants had 1-5 years experience in the district, 29% had 6-10 years experience, 17% had 11-15 years experience, and the remaining 22% had more than 15 years experience in the district; (b) 379 of the 423 participants were female; (c) 270 of the participants had bachelor degrees as the highest earned, 151 had masters degrees, 1 had a specialist degree, and 1 held a doctorate; (d) 120 of the participants were single and 303 were married; (e) average unexpected teacher absenteeism ranged from 3. 61 days per teacher in the lowest absentee ism school to 13. 70 days per teacher in the highest absenteeism school, with a mean rate of 7.75 days per teacher for all 15 schools; (f) 11 of the schools were Title schools; and (g) 7 of the schools were defined as large schools and 8 as small schools for the study. In regard to teacher perceptions of the leadership style of the principals among elementary schools with high, average, and low unexpected teacher absenteeism (question one), the major findings were as fol lows: 1. The majority of the schools in the high absenteeism category were below the total population mean on 10 of the 12 dimensions of the LBDQ-XI I (exceptions were predictive accuracy and superior orientation). 2. The majority of the schools in the average absenteeism category were above the total population mean on 7 of the 12 dimensions of the LBDQ-XII (exceptions were representation, tolerance of uncertainty, initiation of structure, tolerance of freedom, and integration).

PAGE 110

102 3. The majority of schools in the low absenteeism category were above the total population mean on 8 of the 12 dimensions of the LBDQ-XI I (exceptions were representation, consideration, pre dictive accuracy, and superior orientation). 4. All of the schools in the high absenteeism category were below the total population mean for the total LBDQ-XI I score Two of the schools in the average absenteeism category and two of the schools in the low absenteeism category were above the total population mean for the LBDQ-XI I total mean score. 5 Significant statistical difference, at the 05 level of confidence, was found for 10 of the 12 dimensions when categories of schools were compared. For the representation dimension of the LBDQ-X 11, the high absenteeism schools were significantly different from schools with average or low absenteeism. (There was no signifi cant difference between average and low absenteeism schools on this dimension.) The same pattern existed among the three categories of schools for the dimensions of demand reconciliation, tolerance of uncertainty, initiation of structure, role assumption, and integration. For the tolerance of freedom dimension, schools with average teacher unexpected absenteeism achieved a higher group mean and they were significantly different from both h i gh absenteeism and low absenteeism schools. A significant difference was found between high absentee i sm schools and average absen teeism schools for the consideration and production emphasis dimensions. For the superior orientation dimension, a significant difference was found between average and low absenteeism cate gory schools For two dimensions--persuasiveness and predictive

PAGE 111

103 accuracy--there were no significant differences (at the 05 level) among the three categories of schools. Relative to the teacher morale among elementary schools with high, c:verage, and low unexpected teacher absenteeism (question two) the major findings were as follows: 1. Four of the five schools in the high absenteeism category were below the total population mean on each of the 10 dimensions of the PTO and on the toti!I PTO score. (The same school was the exception on each dimansion and that school scored above the total population mean on each dimension and on the total score.) 2. Three of the 5 schools in the average absenteeism category were above the total population mean on 9 of the 10 dimensions of the PTO (the teacher load dimension was the exception). 3. Four of the 5 schools in the low absenteeism category were above the total population mean on 9 of the 10 dimensions of the PTO (the community support of education dimension was the exception) 4. Four of the five schools in the high absenteeism category were below the total population mean for the PTO total score, while three of the five schools in the average absenteeism category and four of the five schools in the low absenteeism category were above the total population mean for the PTO total score. 5. The greatest range in mean scores occurred between high absen teeism schools and low absenteeism schools for the dimensions of teacher rapport with principal (30 01--47. 53), satisfaction with teaching (31.86--52.23), rapport among teachers (21.88--36 13), and teacher salary (9.56--14.91).

PAGE 112

104 6. Significant statistical difference, at the .OS level of confidence, was found among the three categories of schools for each of the 10 dimensions of the PTO. For the dimensions of teacher rapport with principal, satisfaction with teaching, rapport among teach ers, teacher salary, curriculum issues, teacher status, com munity support of education, and school facilities and services, significant differences were found between the following pairs : high absenteeism schools--low absenteeism schools, high absen teeism schools--average absenteeism schools, and low absenteeism schools--average absenteeism schools. For the dimensions of teacher load and community pressures, significant differences were found between high absenteeism schools and low absenteeism schools and high absenteeism schools and average absenteeism schools In regard to teachers' perceptions of the leadership style of principals in high and low morale elementary schools (question three), the major findings were as follows: 1 The high morale schools were significantly different, at the 05 level of confidence, from the low morale schools on 10 of 12 dimensions of the LBDQ-XI 1--representation, demand reconcili ation, tolerance of uncertainty, initiation of structure, tolerance of freedom, role assumption, consideration, production emphasis, integration, and superior orientation 2. No significant statistical differences between the groups were found for the persuasiveness and pred i ctive accuracy dimensions. 3. The high morale schools' scores on the systems and personal orientations of the school principals were significantly higher

PAGE 113

105 than the scores for the low morale schools. The total LBDQ-XI I scores were also significantly higher, at the 05 level, for the high morale schools than for the low morale schools. 4. Four of the five schools with the lowest total PTO scores were also among the five schools with the highest unexpected teacher absenteeism. Relative to the regression analysis used to answer question 4, regarding the extent to which individual teacher unexpected absen teeism could be predicted from the 10 independent variables, the following major findings emerged: 1. Using all independent variables--teacher's perception of the systems orientation of the principal, teacher's perception of the personal or i entation of the principal, the teacher's total morale score, the teacher's level of education, the teacher's sex, the teacher's marital status, the teacher's experience in the school district, whether the teacher was employed in a Title I or a non-Title I school, whether the teacher was employed in a large or a small elementary school, and whether the teacher was employed in a school with high unexpected teacher absenteeism or a school without high unexpected teacher absenteeism--the overall F value of 1.69 was not significant at the OS level. Examining the independent variables separately, PTO total score was significant with an F value of 2. 63, single marital status with an F value of 3 66, 1-5 years experience in the district with an F value of 7. 42, Title I school assignment with an F value of 7. 12, small school size with an F value of 6. 31, average absenteeism with an F value of 5. 83, and high absenteeism rate with an F value of 28.90.

PAGE 114

106 2. The independent variables accounted for 18% of the variance ir. individual teacher unexpected absenteeism. In regard to the regression analysis used to answer question five, where the teacher's total morale score was treated as a covariate and the researcher sought to find the extent to which individual teacher unexpected absenteeism could be predicted from a series of nine independent v2!"iables, it was found that the overall F value of 1 69 was not significant at the 05 level and that the PTO total score treated as a covariate produced no significant change in the F values or statistical significance of the independent variables when they were examined separately. Conclusions In regard to the major question which gave d i rection to the investigation (the extent to which the unexpected absenteeism of teachers in selected elementary schools and the unexpected absentee ism of i ndividual teachers cou l d be explained by teacher perceptions of the leadership style of principa l s, teacher morale, and selected teacher-related and school-related characteristics), the findings seem to warrant the following major conclusions: 1. Teacher percept i ons of the leadership style of principals were different i n schools with high unexpected teacher absenteeism when compared to schools with average and low unexpected teacher absenteeism However, there were no such differences between schools w i th average unexpected teacher absenteeism and schools with low unexpected teacher absenteeism.

PAGE 115

107 2. There was a difference in teacher morale among schools with high, average, and low unexpected teacher absenteeism. While the most significant differences occurred between schools with high unexpected teacher absenteeism and those with low unex pected teacher absenteeism (all 10 dimensions on the PTO were significant between these groups), significant differences also occurred between schools with average unexpected teacher absen teeism and those with low unexpected teacher absenteeism on 8 of 10 dimensions of the PTO. 3 Differences existed in the teachers' perceptions of the leadership style of principals in high and low morale elementary schools. This generalization is supported by the differences, significant at the 05 level, found for 10 of the 12 dimensions of the LBDQ-XI I. Also, the high morale schools' scores on the systems orientation, personal orientation, and total LBDQ-XII instrument were significantly higher than scores for the low morale schools. 4 The independent variables of employment in a school with high or average unexpected teacher absenteeism, employment in a small school, 1-5 years experience in the district, employment in a Title I school, single marital status, and PTO total score were significant predictors of individual teacher unexpected absentee ism. (The combined effect of these variables accounted for 17 72% of the variance in teacher unexpected absenteeism.) While these independent variables were statistically significant in the study, only two of them--the school norm of high unexpected teacher absenteeism and teacher morale ( PTO score)--had prac tical application for decreasing absenteeism. The remaining four

PAGE 116

108 variables could not be altered without impacting the school system beyond the local school 5. Treating the teacher's total morale score as a covariate made no significant difference in the predictive value of the independent variables on individual teacher unexpected absentee i sm from that found when the total morale score was treated as an independent variable Discussion In the study there was an effort to answer some of the questions which existed about the relationships among leadership style of the principal, teacher morale, teacher-related and school-related charac teristics, and unexpected teacher absenteeism. The results of the study are discussed in two ways. First, is a discussion in relation to the justification for the study and the relation to the findings of other studies included in the literature review. Second, there is a discussion of practical application of the findings. Results of the Study in Relation to Its Justification and Previous Research One of the reasons for conduct i ng the study was to provide additional knowledge concerning the influence of leadership behavior on unexpected teacher absenteeism. From the measurement of leader ship behavior it was found that, as a group, the schools in the high absenteeism category scored below the total population mean on 10 of 12 dimensions of the LBDQ-Xll--representation, demand reconcil i ation, tolerance of uncertainty, persuasiveness, initiation of structure tolerance of freedom, role assumption, consideration, production

PAGE 117

109 emphasis, and integration. The group scored above the total popula tion mean on predictive accuracy and superior orientation. This supported the previous research by the Pennsylvania School Board Association (1978) which found increased absenteeism by teachers was associated with low administrative leadership behavior scores. It has been noted that schools with high unexpected teacher absenteeism scored significantly lower on the systems orientation and the personal orientation of the LBDQ-XI I than did schools with average or low unexpected teacher absenteeism. This finding supported general leadership effectiveness research conducted by Blake and Mouton (1964), Halpin (1966), Huse and Bowditch (1963), and Stogdill (1974), who concluded that organizations are most effective under the inte grating approach which emphasizes both task orientation and people orientation. However, previous research by Schroeder (1977) which reported no significant relat i onship between 12 patterns of the prin cipal's managerial behavior and the absence rate of 96 teachers in 8 schools randomly selected from a population of 3,800 teachers was not supported by the present study Another reason for the study was to provide additional knowl edge concerning the influence of teacher morale on unexpected teacher absenteeism. From the measurem~nt of teacher morale it was found that schools in the high unexpected teacher absenteeism category, as a group, were below the total population mean on each of the 10 dimensions of the PTO while the schools in the average and low absenteeism categories, as a group, were above the total mean score on each dimension. This supported the research by Bamber ( 1979) who found teacher absenteeism a result of low morale The greatest

PAGE 118

110 range in mean scores occurred between high absenteeism schools and low absenteeism schools for the PTO dimensions of teacher rapport with principal, satisfaction with teaching, rapport among teachers, and teacher salary. This finding supported the assertion by Herzberg, Mausner, and Snyderman (1969) that motivators--dimensions of work, responsibility, and advancement--are effective in motivating people to performance (i.e. report to work in the study reported herein) while hygiene factors--company policy and administration, supervision, working conditions, and pay--prevent dissatisfaction (exhibited herein by high absenteeism). Fox and Scott's (1943) contention that manage ment's responsibility is to help provide a continuity, a predictability, and a routine of relationships in the life of an employee or face unpleasant consequences, one of which was absenteeism, was also supported by the findings. Coller (1970) learned that the low absence teachers he studied tended to have higher teacher morale scores than did the high absence teachers, a finding supported by the present study. The locale of the school, although not a major issue in the study reported herein, appeared to affect teacher morale. The four inner city schools in the study had mean PTO total scores significantly lower than the total population mean. This finding supported previ ous research by Cook (1971) who used the PTO to measure the morale of teachers in inner-city school settings. He reported that the mean scores for teachers in inner-city schools were significantly lower on the PTO than for teachers in schools located in suburban settings. A third reason for the study was to determine to what extent selected teacher-related and school-related variables affected absenteeism.

PAGE 119

111 Seven of the 10 variables entered into the regression analysis were significant as predictors at the 05 level: employment in a school with high unexpected teacher absenteeism, employment in a school with average unexpected teacher absenteeism, employment in a small school, one-five years experience in the district, employment in a Title I school, single marital status, and total morale score. The combined effect of these variables on the variance in unexpected teacher absenteeism was 17 72%. Examining the contributions of the independent variables, it may be concluded that the findings relative to employment in a small school did not support previous research by Argyle, Gardner, and Cioffi (1958), Covner (1950), and lndik and Seashore (1961) who found positive relationships existed between work unit size and employee absenteeism A possible explanation for the difference in the findings between the previous research cited and the present study may be the definition of work unit size. While small groups in the previous studies were defined as 10 or fewer participants, small groups in the study reported herein were defined as 33 or fewer participants (the reader is reminded that the smallest school studied produced 19 participants from the 20 teachers employed at the school). The findings relative to employment in a Title I school (i.e. employees in Title I schools had a higher rate of absenteeism) supported a State of New York Office of Education Performance Review (1974) study, which reported a 29% higher absence rate for teachers in Title I schools when compared to teachers in non-Title I schools. Previous studies conducted to measure the impact of years of employment on the absence rate of teachers have produced conflicting results.

PAGE 120

112 Gibson and Lafornara (1972) reported that teachers with fewer than 10 years of service were absent more often than teachers with more than 10 years of service. In the study reported herein, that finding was supported for teachers with 1-5 years of experience, but teachers with 6-10 years of experience were not found to be absent significantly more than other groups of teachers. Single marital status as a sig nificant predictor of teacher absenteeism supported previous research by Martin (1971) and Garrison and Muchinsky (1977). Low teacher morale scores as a significant predictor of unexpected teacher absen teeism supported previous research by Coller (1970), who found that the low absence teachers he studied tended to have higher morale scores than did the high absence teachers, and Douglas (1976) who found morale to be a significant predictor of absenteeism in a study of 154 teachers. Teacher sex was used as an independent variable but did not emerge as a significant predictor of absenteeism. This finding was in contradiction to previous research by Sylwester (1979) and Redmond (1978) who found the absence rates for females significantly lower than the absence rates for males. Considering the deviation from previous findings on the relationship of teacher sex to absenteeism, the possibility must be entertained that the major reason for this deviation was the low number of males to females ( 44 males, 379 females) in the study. It may also be explained by Hedges (1973) findings that sex differences in absence rates narrow when com parisons are made within a particular occupation group, as they were in the present study

PAGE 121

113 Few studies focusing on the relationship of education : ,wel and teacher absenteeism have been reported, and those have produced mixed results. Education level, when entered into the multiple re gression in the present study, was not significant at the 05 level of confidence. This supported previous research by Redmond (1978) who found no significant correlation between educational level and absenteeism of professional personnel in the Fort Madison ( Iowa) Community School District. The finding in the present study did not support previous research by Douglas (1976) who found education level was one of nine variables significant as a predictor of teacher absenteeism. The researcher anticipated that treating the PTO total score as a covariate in the multiple regression analysis (question five) would produce signficantly different predictor values than did the regression analysis (question four) where the PTO score was entered as an independent variable. However, no significant difference occurred Those 7 predictors of teacher absenteeism which were signficant at the .05 level when PTO score was entered as 1 of 10 independent variables were the same 7 predictors which were statistically signifi cant when PTO score was treated as a covariate. A possible expla nation for th i s unanticipated occurrence may be found in Likert's (1967) study of three sets of variables for use in research and oper ating purposes: (a) causal variables--those which influenced devel opments within the organization and which could be altered or changed by the organization and its management; (b) intervening variablesthose which reflected the internal state of the organization; and (c) end-result variables--those which reflected performance. Likert

PAGE 122

114 found the condition of the intervening variables was affected by th~ causal variables and, in turn, the intervening variables influenced the end-result variables. PTO score (morale), as an intervening variable in the present study, interacted in the regression analysis with causal and end-result variables and, while PTO score was a signficant predictor of unexpected teacher absenteeism when entered as an independent variable in the regression analysis, it had the lowest F value at the 05 level of the seven significant predictors. Practical Apolication of the Study Results A conclusion of the present study was that teacher perceptions of the leadership style of principals were different between schools with high unexpected teacher absenteeism and schools with average and low unexpected teacher absenteeism Effective leadership behavior in the areas of representation, demand reconciliation, tolerance of uncertainty, initiation of structure, role assumption, and integration was identified as a significant contributor to low unexpected teacher absenteeism. Pre-service and in-service tra i ning for administrators, with particular emphasis on those leadership behavior characteristics which discourage unexpected absenteeism, could improve teacher attendance and lower costs to the school district for substitute teachers. A second conclusion of the present study was that teacher morale was different among schools with high, average, and low unexpected teacher absenteeism. Again, an awareness by administra tors of those factors which foster positive teacher morale could improve attendance and lower substitute teacher costs. Since four of the five schools with the lowest morale scores were also four of the five schools

PAGE 123

115 with the highest unexpected teacher absenteeism it may be inferred by this ex post facto study that improving morale would decrease absenteeism A third conclusion in the present study was that differences did exist in the teachers' perceptions of the leadership style of principals in high and low morale elementary schools. The high morale schools' scores on the systems orientation, personal orientation, and total LBDQ-XI I instrument were significantly higher than score~ "'.'Ir the low morale schools. One may infer that leadership behavior ... oes signifi cantly affect teacher morale. Two of the dimensions on the PTO which had the greatest range in mean scores between high absenteeism and low absenteeism schools were teacher rapport with principal and rapport among teachers Emphasis on improving communication among individuals could have a positive effect on rapport and, subsequently, could decrease teacher absenteeism. The reader of literature in the field of education may observe that rapport among those individuals involved in the formal education process, under the rubric of humanism, has advanced at a more accelerated rate between teacher and student than between teacher and profess i onal colleague A fourth conclusion in the present study was that the inde pendent variables of employment in a school with high or average unexpected teacher absenteeism, employment in a small school, one five years experience in the district, employment in a Title I school single marital status, and PTO total score are sign i ficant predictors of individual teacher unexpected absenteeism. From the perspective of seeking practical ways of reducing absentee i sm, however, only two of the predictors--the unexpected teacher absenteeism rate for the

PAGE 124

116 school and teacher morale (PTO total score)--may be changed within the school and not require alteration to other subsystems of the district school system. Schools categorized as high absenteeism schools were placed in that category as a result of high absenteeism among the individual teachers within the school Since one-five years experience in the district was also a significant predictor of high individual teacher unexpected absenteeism, it could also be inferred that less experienced teachers who enter the system and find the school norm one of high absentee i sm may tend to accept that norm and be i nfluenced by it. When the reader notes that h i gh absentee ism rate had the highest F value of any predictor in the regression equation, it may be inferred that a school system which established as a goal the reduction of unexpected teacher absenteeism would be well-advised to concentrate on changing the norm of high absenteeism within schools.

PAGE 125

APPENDIX A LEADER BEHAVIOR DESCRIPTION QUESTIONNAIRE-FORM XII ( Items grouped by dimensions) I Representation 1 He acts as the spokesman of the group A B C D E 11. He publicizes the activities of the group A B C D E 21. He speaks as the representative of the group A B C D E 31 He speaks for the group when visitors are present A B C D E 41. He represents the group at outside meetings A B C D E 11. Reconciliation 51. He handles complex problems efficiently A B C D E 61. He gets swamped by details A B C D E 71. He gets th i ngs all tangled up A B C D E 81. He can reduce a madhouse to system and order A B C D E 91. He gets confused when too many demands are made of him A B C D E 11 1 Tolerance of Uncertainty 2 He waits patiently for the results of a decision A B C D E 12 He becomes anxious when he cannot find out what is coming next A B C D E 22 He accepts defeat in stride A B C D E 32. He accepts delays without becoming upset A B C D E 42. He becomes anxious when waiting for new developments A 8 C D E 52. He is able to tolerate postponement and uncertainty A B C D E 62. He can wait just so long, then blows up A B C D E 72. He remains calm when uncertain about coming events A B C D E 82 He is able to delay action until the proper time occurs A B C D E 117

PAGE 126

118 APPEND,x A (continued) 92. He worries about the outcome of any new procedure A B C D E IV. Persuasiveness 3. He makes pep talks to stimulate the group A B C D E 13. His arguments are convincing A B C D E 23. He argues persuasively for his point of view A B C D E 33. He is a very persuasive talker A B C D E 43. He is very skillful in an argument A B C D E 53. He is not a very convincing talker A B C D E 63. He speaks from a strong inner conviction A B C D E 73. He is an inspiring talker A B C D E 83. He persuades others that his ideas are to their advantage A B C D E 93. He can inspire enthusiasm for a project A B C D E V. Structure 4. He lets group members know what is expected of them A B C D E 14. He encourages the use of uniform procedures A B C D E 24 He tries out his ideas in the group A B C D E 34. He makes his attitudes clear to the group A B C D E 44. He decides what shall be done and how it shall be done A B C D E 54. He assigns group members to particular tasks A B C D E 64 He makes sure that his part in the group is understood by the group members A B C D E 74. He schedules the work to be done A B C D E 84. He maintains definite standards of performance A B C D E 94. He asks that group members follow standard rules and regulations A B C D E

PAGE 127

119 APPENDIX A (continued) VI. Tolerance of Freedom 5. He allows the members complete freedom in their work A B C D E 15. He permits the members to use their own judgment in solving problems A B C D E 25. He encourages initiative in the group members A B C D E 35. He lets the members do their work the way they think best A B C D E 45. He assigns a task, then lets the members handle it A B C D E 55. He turns the members loose on a job, and lets them go to it A B C D E 65. He is reluctant to allow the members any freedom of action 75. He allows the group a high degree of initiative 85. He trusts the members to exercise good judgment 95. He permits the group to set its own pace VI I. Role Assumption A B C D E A B C D E A B C D E A B C D E 6. He is hesitant about taking initiative in the group A B C D E 16 He fails to take necessary action A B C D E 26. He lets other persons take away his leadership in the group A B C D E 36. He lets some members take advantage of him A B C D E 46. He is the leader of the group in name only A B C D F. 56. He backs down when he ought to stand firm A B C D E 66. He lets some members have authority that he should keep A B C D E 76. He takes full charge when emergencies arise A B C D E 86. He overcomes attempts made to challenge his leadership A B C D E 96. He is easily recognized as the leader of the group A B C D E

PAGE 128

120 APPENDIX A (continued) VI 11. Consideration 7. He is friendly and approachable A B C D E 17. He does little things to make it pleasant to be a member of the group A B C D E 27. He puts suggestions made by the group into operation A B C D E 37 He treats all group members as his equals A B C D E 47. He gives advance notice of changes A B C D E 57. He keeps to himself A B C D E 67. He looks out for the personal welfare of group members A B C D E 77. He is willing to make changes A B C D E 87. He refuses to explain his actions A B C D E 97. He acts without consulting the group A B C D E IX. Production Emehasis 8. He encourages overtime work A B C D E 18. He stresses being ahead of competing groups A B C D E 28 He needles members for greater effort A B C D E 38. He keeps the work moving at a rapid pace A B C D E 48 He pushes for increased production A B C D E 58. He asks the members to work harder A B C D E 68. He permits the members to take it easy in their work A B C D E 78. He drives hard when there is a job to be done A B C D E 88. He urges the group to beat its previous record A B C D E 98. He keeps the group working up to capacity A B C D E X. Predictive Accurac:i::: 9 He makes accurate decisions A B C D E

PAGE 129

121 APPENDIX A (continued) 29. He seems able to predict what is coming next A B C D E 49 Things usually turn out as he predicts A B C D E 59. He is accurate in predicting the trend of events A B C D E 89. He anticipates problems and plans for them A B C D E XI. Integration 19. He keeps the group working together as a team 39. He settles conflicts when they occur in the group 69. He sees to it that the work of the group is coordinated 79 He helps group members settle their differences 99 He maintains a closely knit group XI I. Superior Orientation 10. He gets along well with the people above him 20. He keeps the group in good standing with higher authority 30 He is working r.::-d for a promotion A B C D E A B C D E A B C D E A B C D E A B C D E A B C D E A B C D E A B C D E 40 His superiors act favorably on most of his suggest i ons A B C D E 50 He enjoys the priv i leges of his position A B C D E 60. He gets his superiors to act for the welfare of the group members A B C D E 70. His word carries weight with his superiors A B C D E 80 He gets what he asks for from his superiors A B C D E 90 He is working his way to the top A B C D E 100 He maintains cordial relations with super i ors A B C D E A = Always B = Often C = Occasionally D = Seldom E = Never Copyright, 1968, by the Ohio State University Permiss i on granted for inclusion in dissertation

PAGE 130

APPENDIX B MEANS AND STANDARD DEVIATIONS OF SUBSCALE LBDQ-XII SCORES FOR SELECTED LEADERS IN LEADERSHIP POSITIONS Army Highway Aircraft Subscale Division Patrol Executives Mean SD Mean SD Mean SD 1 Representation 20.0 3.0 19.9 2 8 19.8 2 8 2 Demand Reconciliation 19.2 2.8 3. Tclerance Uncertainty 36.2 4.7 35 6 4 6 33.2 6 2 4. Persuas i veness 38 3 6 2 37 9 5 9 36.5 5.5 5 Initiating Structure 38.6 5.7 39.7 4.5 36.6 5 .4 6 Tolerance Freedom 35.9 6.5 36 3 5.3 38.0 5 9 7. Role Assumption 42 7 6 1 42 7 5.3 40.9 5 6 8. Cons i deration 37. 1 5.6 36.9 6.5 37 1 5.8 9 Production Emphasis 36 3 5 1 35.8 5 7 36. 1 5.6 10 Predictive Accuracy 19. 1 2. 1 1 7 8 2 1 19 2 2.6 11 Integration 19.5 2.6 19. 1 2.7 12 Superior Or i entat i on 39 9 4 9 39 1 5 1 38 6 4.2 Number of Cases 235 185 165 122

PAGE 131

123 APPENDIX B (continued) Community Corporation Subscale Ministers Leaders Presidents Mean SD Mean SD Mean SD 1. Representation 20.4 2.4 19.6 2.4 20.5 1.8 2. Demand Reconc i liat i on 19 8 3 1 19.7 3.3 20.6 2.7 3 Tolerance Uncertainty 37 5 6.3 37 7 5.6 35.9 5 4 4. Persuasiveness 42 1 4.7 39 5 5 5 40.1 4.2 5. Initiat i ng Structure 38 7 4 9 37 2 5 7 38.5 5.0 6. Tolerance Freedom 37.5 6.0 36 .4 5 0 38 9 4.9 7. Role Assumpt i on 41.5 5.4 39.8 5.6 42.7 3.5 8. Consideration 42.5 5.8 41 .1 4.7 41.5 4.0 9. Product i on Emphasis 34.9 5 1 35 4 6.8 38 9 4.4 10. Predictive Accuracy 20.5 2.3 19.8 2.5 20 .1 1.8 11. Integration 12. Superior Orientation 43.2 3. 1 Number of Cases 103 57 55

PAGE 132

124 APPENDIX B (continued) Labor College Subscale Pres i dents Presidents Senators Mean SD Mean SD Mean SD 1. Representat i on 22.2 2.2 21.4 1 9 20.7 2 5 2. Demand Reconciliation 21.5 3 2 20.7 3 5 3. Tolerance Uncerta i nty 40.4 5.6 37 2 5 5 35 .3 7 6 4. Persuas i veness 43 1 4.8 41 1 4 2 42.5 4 6 5. Init i ating Structure 38.3 5.6 37 7 4.2 38.8 5 .5 6 Tolerance Freedom 38.0 4 0 39 6 3 9 36 6 6 2 7 Role Assumption 43.3 5.5 43.5 4.5 41 0 5.7 8 Cons i deration 42.3 5.5 41.3 4 1 41. 1 5 9 9. Product i on Emphas i s 36 0 5 0 36 2 5 0 41 2 5.2 10 Predict i ve Accural:y 20 9 2.0 11. Integ r ation 12. Super i or Or i entat i on 42.9 2.9 Number of Cases 44 55 44

PAGE 133

APPENDIX C RELIABILITY COEFFICIENTS OF SUBSCALES ON THE LBDQ-XII (Modified Kuder-Richardson) Army Highway Aircraft Community Subscale Division Patrol Executives Ministers Leaders 1. Representation .82 .85 .74 .55 .59 2 Demand Reconciliation .73 .77 .58 3. Tolerance Uncertainty 58 .66 82 .84 .85 4 Persuasiveness 84 .85 .84 .77 79 5. Initiating Structure .79 .75 .78 .70 72 6. Tolerance Freedom 81 .79 .86 .75 .86 7 Role Assumption 85 .84 .84 .75 .83 8. Consideration .76 .87 .84 .85 .77 9 Production Emphasis .70 .79 .79 .59 .79 10. Predictive Accuracy .76 .82 91 .83 .62 11 Integration 73 .79 12. Superior Orientation .64 .75 81 125

PAGE 134

126 APPENDIX C (continued) Corporation Labor College Subscale Presidents Presidents Presidents Senators 1. Representation .54 .70 66 .80 2. Demand Reconciliation .59 81 81 3 Tolerance Uncertainty .79 .82 .80 83 4. Persuasiveness 69 .80 76 .82 5 Initiating Structure .77 .78 .80 72 6. Tolerance Freedom .84 58 .73 64 7 Role Assumption .57 .86 .75 65 8. Consideration .78 .83 76 85 9. Production Emphasis .71 .65 74 .38 10. Predictive Accuracy .84 .87 11 Integration 12. Superior Orientation .66 .60

PAGE 135

PLEASE NOTE: Copyrighted materials in this document have not been filmed at the request of the author. They are available for consultation, however, in the author's university library. These consist of pages: University Miadllms International 127-133 300 N Z EEB AD A NN AAS O A M l 48 1 06 3131 76 1 4 70 0

PAGE 136

APPENDIX E TEST-RETEST CORRELATIONS FOR PURDUE TEACHEP OPINIONNAIRE FACTOR AND TOTAL SCORES Factor (N:::3023) 1. Teacher rapport with principal 2. Satisfaction with teaching 3. Rapport among teachers 4. Teacher salary s. Teacher load 6. Curriculum i ssues 7. Teacher status 8. Community support of education 9 School facilities and services 10 Community pressures 134 orrelation .88 .84 .80 81 .n 76 81 78 .80 .62 Total Score 87

PAGE 137

APPENDIX F CORRELATION MATRIX FOR INDEPENDENT AND DEPENDENT VARIABLES OAYSAdS !.TAt-F SYSTf J14S PF.RSCNAL PTOTOTAL SMALLS CH LARGESCH LUAOS "1048S HIGHAB$ NONTITLE TITLE OAYSAUS t.Olll.>O -O. 12'4.fl?. -0.10798 -O.Od8I" -0.039 OotS6<>2 -o. 15652 -Oo222l8 -0.03298 00266 0 Ool08 7 -001011 7 sr A.FF -0.12912 I.OOJOO O. IJOl:J O. OS et I -00095911 -Oo7]62S Oo 73626 Oo 1All3 -0061612 Oo 2 8111 -Oo2 8ld SWSTl'MS -Oo 107'18 O. IJOl:J lo 00000 Oo 31549 o ., -Oo264SU o.26ss o. 0592] Ooll7J8 -Oo 20618 -o. 0615"1 0006157 f>f.l>SONAL -0.0:3RIQ Oo 05811 o. 87549 l000000 OoSllUO -0021516 0021516 OoOS4l6 Oo 1207l -Oo 18]60 -Oo 12362 0012362 f>TOTOT AL -0. Ol'l4 l Oe OYS )U o. 41241 O.SJ3'10 I 000000 o.1ss~3 -0.15553 o. 29272 0002149 -Oo 32760 -o. 36854 Ool685 S14ALLSCH O. l~b~2' -Oo7l62S -O. lbe -0.215tu Oe 15553 I 000000 -l000000 -0. ll27l -0021816 Oo 36771 -Oo O 18 Oo40 16 LARGESCII -o 15652 o. 13625 Oo2b458 Co2ISl6 -0016553 -1oooooo> Io 00000 O. l.327.J 0021e16 -Oo l61'7 I Oo O u1 -Oo40 18 LOW ABS -0.2223:-:1 o. 18.JlJ Co0S923 Co0S l6 0029272 -Oo ll273 o. t:]273 Io 00000 -Oo54l72 -Oo41'020 -Oo 041171 0.04111'7 MIOAOS -0.01298 o. 40444 0.13731:S o. 12 073 0002149 -0021811, 0021816 -Oo54l72 Io 00000 -0048505 -0.09980 o.o99tto HIGHAOZ, :) 200 0 -0.61(112 -0. 20<,l 8 -O. I f! l60 -0.32760 Oo ]6771 -Oo3677 I -0.47020 -0.48505 Io 00000 Oo 15577 -0015577 MONTI TL~ O. lOH47 o.2e1 e -O.Ot,157 -o. I 2 l<> 2 -0036854 -0040418 0040418 -0.04877 -0.09980 Oo I 551'7 Io 00000 -1 oOOOOO TITLE -o. 1Od41 -0.24818 Oo 06157 C.t Z3f>2 0.36b54 0.40418 -0040418 Oo 04 B77 0009980 -OolS571' -1.00000 loOOOOO FL:NALC. -o. 02~4 'ii -o. 055<>7 -0.04614 -(.0]0<.9 Oa04440 0001994 -O.tll994 o.ooJo -Oo 06161 00061'93 -Oo 02056 0002056 116ALE O. OZ949 o. 1 .l5Sti7 o. 04(;14 C. OJ C69 -0.04448 -0.01994 Oo0l994 -Oo 00304 Oo 06161 -0.06793 0002056 -0.0205(1 SINGLE" -Oo 17)62 -0.00257 -o. 00942 -0.00378 -OoOo>IOO 0000716 -o 000716 0.09.66 -0.0.1411 -Oo 06277 -Oo I 0271 0010271 14AR"IEO 0.11110 Oo 00211 -OoCOIOO -Oo008J2 -o 00072;> -0.002ll 0000231 -0009063 0002704 Oo06600 0.10580 -Ool05d0 UACHl:L0"1 -O.l>J415 -Ooll~ll -o.oc:..!>2tt -0008622 -0000595 0.06447 -0.06 -0002268 -0007200 Oa099.34 Oo 091911 -0009198 14ASTC:Q o.oJ,,sz Oa l 1607 o. 0672] OaO8l96 Oa08721 -Oo05553 00055,;] Oa0l953 Oo07960 -Ool0 05 -000961'9 000961'9 SPEC -o. OlZ4il \). 0214 7 o. 00714 CaO76tl8 0003491 -0.04502 0 004502 Oo06705 -o. 03646 -Oo0Jl53 -0.03111' Oo03ll7 DOCTOR 0.0101) OaOl.365 -o. 02472 -Oo 03237 -0 a0449 l -Oo04S02 0004502 -Oo035l4 -Oo036 6 0007516 Oo 07602 -0.07602 c.., TRSIT05 -1\.17?.56 -O ..lOt.67 o. 02193 Ca040lb -o. I l 1-12 0.01116 -0.01116 -0.01675 -0.05117 Oo07820 Oo 07435 -Oo01' l5 CJ1 YR6TOIO o. 059tl5 0.00109 -o. Ol:l-t99 -0.0)573 -0.07791 -0006999 0.06999 -0.04877 0.03039 Oo 018115 0.071 -Oo07146 TAIIT0l5 OeOJ652 0.0120 -o.ooees -C.06l.l0 -0.074-06 ... 0.0~135 0 .05135 -o. 05944 0000530 000563<, 1100066l -Oo006til TAU,T055 Oo 09356 o. llo!a64 0.070,19 (005029 0.28153 0.110:?7 -0011027 0012625 0002634 -Oo 15925 -Oo l6696 0016696 FE14ALE '4ALE' SINGLE MARRIED eACHELOR MASTEi> Sf>EC DOCTOR 'l'RSI T05 '1'"6TOIO YRIIT015 'l'Rl6T055 OAYSAdS -0.02949 0.02-ilfl9 -0.17)62 c.11710 -OoOHl5 0.03652 -0.03249 0.01013 -0.1?256 0.05985 0.03652 0.09356 Sf A.FF ... O.OS"ib7 0.05567 -0.00257 0.00211 -Ooll9ll Oollb07 00021 7 o.-11Ja5 -0.00667 OoOOI 119 0001204 -0.00564 ST STEMS -0.04h14 :>.04614 -o. 00 )42 -0000100 -o 006526 0.06723 0 o 00714 -0.02472 0002193 -Oo 08499 -Oo 008115 0.076-.9 f>ERSONAL -Oo 030~9 O.Ol0<,9 -Oo 00378 -o. 008]2 -0 00862 2 Oo0Rl96 0007688 -Oo 0]237 0004016 -o. 0.1573 -Oo 06130 0.050~9 PTOTOT4L -o.oe -OoOOIOO -0.00122 -0.08595 0008121 Oo0l491 -0.04491 -0.111 -Oo 07791 -o. 01'906 o.2a15:, S14ALLSCH 0.0199 -0001994 0.00116 -Oo002JI 0.06447 -0005553 -0004502 -0004502 Oo 0111'6 -0006999 -Oo0Sll5 Oo II 027 LAA ~o o.02ost. -0.10271 o., 0560 Oe09198 -0.09679 -0003117 0007602 0.07435 Oo01'U6 o. 00"61 -Oo 16696 TITLE 0.1>20Sb -0002056 001oz11 -o. a osao -0.09198 Oe09b79 0003117 -Oo07o02 -0007435 -0007146 -o. 00661 0.1669b FEMALE uoooo -1 .ooo >o o. 00828 -c. 01005 0.00137 -O.OOo\74 0.01659 0001659 O.Ol22l 0001355 -Oo 02655 -0.004 lo\ '4ALE -l000000 1.00000 -o. 00ts2~ c.01005 -0000137 0.0047,\ -Oo0l6S9 -0001659 -Oo 01221 -Oo0l35S 0002655 Oo00 14 SINGLE O.OOR2~ -0000828 Io 00000 -0.99421 OoOIS3l -Oo00916 -0.03063 -0.03063 Oe232b3 -Oo 05651 -0.09653 -0.,09:,2 M-'AAlt.:O -0.01 ,lQ'.j 0.0100! -O.~'H21 t.00000 -0001923 0.01304 o.olOff l 000308I -Oo 2285] Oo 05972 0o09&10 Oo09927 IIACHELOR 0.00137 -0.00137 0.01533 -0o0l92l 1.00000 -0.9B978 -Oo06 67 -0.06467 0020967 -0002720 -0.11958 -0009468 MAST CM -0.00414 0.00,1. -0.00916 Oo Ol 304 -0.98978 1.00000 -0.03627 -Oo03627 -Ooll428 0002273 o.12a 0.10023 SPEC o. 0165'1 -0.016!>9 -Oo0:J06:J o.03oe1 -0.0646'1 -Oo0l627 Io 00000 -0.00237 0007228 -Oo031l7 -0002242 -0002602 ODCTOP Oo 0l65Q -lil.Olb5Q Oo 03063 Oo0J08I -0.06o\61 -i>.03627 -Oo002l7 lo 00000 -OoOJ2; Oo 01'602 -0.02242 -0002602 WRSITn5 0.01221 -0001221 0.2:1261 -0022853 0.20967 -Oo21 28 0.0122a -0.0)279 Io 00000 -Oo ll25 -Oo 31013 -0.36000 t~6f0l0 o.o, .} !i ,i -0001355 -0.05651 c. C54-72 -0002720 0.02273 -0003117 Oo 07602 -0.4.ll2S loOOOOO -0.29485 -0034226 YRI ITOl5 -0.026';5 0.02,.:.55 -O. OllbSl Co 09870 o. 11951'1 o.12a -0002242 -0.02242 -Oo3I Oil -o. 29485 Io 00000 -Oo2 61l YUl6T055 -.:,. 00 1 0000414 -o. l09J2 o. 09927 -0.09>9 Ool002l -o o 0260 2 -0.02602 -Oo ]6000 -Oo 34226 -0024613 loOOOOO

PAGE 138

(,) Ol OS1.IABLE ~TO TOT AL ttl
PAGE 139

YAA IAIILE PTOTOTAL HIGHABS YASIT05 SINGLE MI0ABS TITLE S"ALL5CH UTAFF NUMBBl OF YAl6T055 SYSTEMS FEMALE MASTEA DOCTOR PEASO-.AL YA IITOI S I CONSTANT I w '1 APPENDIX H RESULTS OF MULTIPLE REGR E SSION OF INDEPENDENT VARIABLES ON UNEXPECTED TEACHER ABSENTEEISM WITH TEACHER MORALE TREATED AS A COVARIATE SUMMARY TAlll.lE IIULTIPLE A A SOUAAE ASQ CHANGE S1'4PLE A B 0.03943 0.00155 0.00155 -0.03943 o.647646,m-02 0.21111 0.01353 0.07198 0.26640 4.44Sll
PAGE 140

REFERENCES Argyle, M., Gardner, G., & Cioffi, F Supervisory methods related to productivity, absenteeism, and labor turnover. Human Relations, 1958, .!!, 23-40. Argyr i s, C. Personality and organizations. New York: Harper, 1957. Argyris, C. Integrating the individual and the organization. New York: John Wiley, 1964. Bamber, C Student and teacher absenteeism Bloomington, Ind.: Phi Delta Kappa Educational Foundation, 1979. Bass, B. Organizational psychology. Boston : Allyn & Bacon, 1965 Bernard i n, H. J. The influence of reinforcement orientation on the relationship between supervisory style and effectiveness criteria (Doctoral dissertation, Bowling Green State University, 1976) Dissertation Abstracts International, 1976, 37, 3702B. ( University Microfilm No. 37-33,915) Bernardin, H. J., & Alvares, K. M. The managerial grid as a predictor of conflict resolution method and managerial effectiveness. Administrative Science Quarterly, 1976, 21, 84-92 Blake, R R., & Mouton, J. S. The managerial grid. Houston: Gulf Publishing, 1964. Blumberg, A & Weber, W Teacher morale as a funct i on of perceived supervisor behavior style. The Journal of Educational Research 1968, 76, 109. Bridges, E., & Hallinan, M T Sub-unit size, work system inter dependence, and employee absenteeism Educational Administration Quarterly, 1978, 14, 24-42. Brown, A. F Reactions to leadership Educational Administration Quarterly, 1967, l, 62-73. Bruner, J. S Personality dynam i cs in the process of perceiving In J M. Anglin (Ed.), Beyond the information given: Studies in the psychology of knowing New York: Norton, 1973. Bruner, J. S., & Tagiuri, R. The perception of people. In G. Lindzey (Ed ), Handbook of Social Psychology (Vol 2). Reading, Mass : Addison-Wesley, 1954. Bundren, D L The influence of situational and demographic factors on the absentee patterns of teachers ( Doctoral dissertation, University of Southern California, 1974). Dissertation Abstracts International, 1974, 35, 3504A (University Microfilm No 35-20, 289) 138

PAGE 141

139 Burgett, K. Administrative practices as they relate to teacher morale in Class II school districts in Montana ( Doctoral dissertation, University of Montana, 1977). Dissertation Abstracts International, 1977, 37, 5469A. (University Microfilms No. 77-31,216) Campbell, J. P., Dunnette, M. D., Lawler, E. E. & Weick, K. E. Managerial behavior, performance, and effectiveness. New York: Harper & Row, 1970. Campbell, M. V. Teacher-principal agree~ent on the ceacher's role. Administrator's Notebook, 1959, l, 2. Chicago Board of Education. Study of employee absenteeism in Chicago Public Schools. Chicago, Illinois: Author, 1960. Coller, R. An analysis of teacher absenteeism and its relationship to teacher morale and demographic characteristics of teachers (Doctoral dissertation, University of Michigan, 1975). Disser tation Abstracts International, 1975, 36, 6638A. ( University Microfilms No. 70-31,226) Combs, A. W. The human aspect of administration. Educaticnal Leadership, 1970, 28, 197-205 Cook, R. P. The relationship of principal leader behavior and teacher morale to certain other variables in selected urban elementary schools (Doctoral dissertation, St. Louis University, 1971) Dissertation Abstracts International, 1971, 32, 5063A. ( University Microfilms No. 71-94, 149) Covner, B. J. A study of industrial absenteeism. American Psychologist, 1950, ?., 336. Croft, J. C. Dogmatism and perceptions of leader behavior. Educational Administration Quart~, 1965, .l, 60-71. D~de County, Florida, Public Schools Office of Management and Budget. Absenteeism: A study of leave taken by Dade County Public School employees. Miami: Author, 1978. Davis, K Evolving mode!s of organizational behavior Academy of Management, 1968, .l.l, 27-38. Douglas, S A. Social-psychological correlates of teacher absenteeisma multi-variate study (Doctoral dissertation, Ohio State University, 1976). Dissertation Abstracts International, 1976, 37, 3711A. (University Microfilms No. 76-49, 187) Ellenburg, F. C. Factors affecting teacher morale. The Education Digest, 1973, 38, 7 Erickson, D. A. The school administrator. Review of Educational Research, 1967, 37, 417-432.

PAGE 142

140 Fiedler, F E. A thE,Ory of leadership effectiveness New York: McGraw-H i ll, 1967. Fleishman, E. A., Harr i s, E. F., & Burtt, H. E Leadership and supervision in industry. (Research Monograph No 33). Columbus : Bureau of Educational Research, Ohio State Univerity, 1955 Fox, J., & Scott, J. Absenteeism : Management s problem Boston : Harvard Graduate School of Business Administration, 1943 Garrison, K. R & Muchinsky, P M Attitudinal and biographical predictors of incidental absenteeism. Journal of Vocational Behavior, 1977, 10, 221-230 Gerstenfeld, A EmployeP. absenteeism: New insights Business Horizons, 1969, 12, 51-57. Getzels, J. W. A psychosociological framework for the study of admin i stration Harvard Educat i onal R e view, 1952, 22, 235-246. Gez i, K I The pr i ncipal's role i n developing staff morale High School Journal, 1962, 46, 388 Gibson, R. 0. Absence, legitimacy, and system size Presented at the annual meeting of the Educat i onal Research Association of New York State, New York, November, 1968. (ERIC Document Reproduction Service No. ED 027 614). Gibson R. 0 & Lafornara, P. Collect i ve legitimacy and organiza t i onal attachment : A long i tudinal case study of school personnel absences Presented at the annual meeting of the American Educational Research Association, Chicago, April, 1972 (ERIC Document Reproduction Serv i ce No ED 063 674) Gragg, W L Teacher morale: Ithaca survey finds teachers agree more on causes of high morale t han of low morale. Clear i nghouse 1955, 24 494 Greenwood G E., & Soar, R. S morale and teacher behav i or ~ 1973, 94 105. Some relationships between teacher The Journal of Educational T~chGubser M. M Author i tarianism among teachers and school principals and i ts poss i ble r elationsh i p to faculty morale The Journal of Educational Research, 1969 66, 36-38 Halpin, A W The leadership behavior of school superintendents. (2nd ed.) Chicago : Midwest Administration Center Univers i ty of Ch i cago, 1959 Halpin, A. W Theory and research in administration New York: Macmillan, 1966. Halpin, A. W., & Croft, D B. The organizat i onal climate of schools Chicago : M i ~ w est Administration Center, University of Chicago, 1963

PAGE 143

141 Harding, F. D., & Botten berg, R. A Effect of personal character istics on relationships between attitudes and job performance Journal of Applied Psycho l ogy, 1961, 45, 428-430. Hedges, J. N. Absence from work--A look at some national data. Monthly Labor Review, 1973, 96, 24-30 Hedges, J N. Unscheduled absence from work--An update Monthly Labor Review, 1975, 98, 36-39. Hedges, J N. Absence from work--Measuring the hours lost Monthly Labor Review 1977, 100, 16-23. Henderson, L. F Elementary teacher satisfaction and morale and perceived participat i on in dec i sion-mak i ng ( Doctoral d i ssertat i on, Ohio State University, 1976). Dissertation Abstracts International, 1976, 37, 2535A (University Microfilms No. 76-21,459) Herrmann W. M. Relationship of faculty morale to faculty involvement in school policy determination in selected count i es of Mercer and Mahoney Counties (Doctoral dissertation, University of Ill i nois, 1971) Dissertation Abstracts Internat i onal, 1971 32, 2950A. (University Microfilms No. 71-10,900) Hersey, P Management concepts and behav i or: Programmed i nstrJct i on for managers. Little Rock : Marvern, 1967 Hersey, P & Blanchard, D. Utilizing human resources. Hall, 1972 Management of organ i zational behavior: Englewood Cliffs, N J .: PrenticeHerzberg, F. Work and the nature of man New York : Thomas Crowell, 1966. Herzberg, F Mausner, B., & Snyderman, D. The motivation to work. New York: John Wiley, 1969 Hill, J. M & Tr i st, E L Changes in accidents and other absences with length of serv i ce: A further study on the i r inc i dence and relation to each other in an iron and steel works. Human Relations, 1955, ~ 121 152 Huse, E. F., & Bowditch, J L. Behavior in organizations: A systems approach to manag i ng. Reading, Mass .: Addison Wesley, 1973. llgen, D R ., & Hollenback, J. The role of job satisfaction in absence behavior Organizational Behavior and Human Perfor ~, 1977, 19, 148-161. lnd i k, B. P & Seashore, S. Effects of or anization size on member attitudes and behavior. ( Research Monograph No 12 Ann Arbor : Survey Research Center of the Inst i tute for Social Research, Univers i ty of M i chigan, 1961.

PAGE 144

142 lsambert-Jamati, V Absenteeism among women workers in industry. International Review, 1962 85, 248-261 Jackson, J. H. Factors involved i n absenteeism. Personne l Journal, 1944, 22, 289-295 Jacobson, P B., Reavis, W. C., & Logsdon, J. D The effective school principal. (2nd Ed ). Englewood Cliffs, N. J. : Prentice Hall, 1963 Johns, G. Att i tudinal and nonattitudinal predictors of two forms of absence from work. Organizational Behavior and Human Perfor ~, 1978, 22 431-444. Kahn, H R. Don't take older workers for granted Harvard Bus i ness Review, 1957, 35, 90-94. Kerlinger, F Foundations of behavioral research (2nd ed ). New York: Holt, Rinehart & Winston, 1973 Kerlinger, F. N., & Pedhazur E J Mult i ple regression in behav i oral research New York : Holt, R i nehart & Winston, 1973. Kerr, W A Koppelmeier, & Sullivan, J. L. Absenteeism, turnover, and morale in a metals fabrication factory. Occupational Psycho l ogy, 1951, 25, 50-55 Knezev i ch, S. J Administration of public educat i on New York : Harper and Row, 1975. Kokovich, S A study of the relationship between perceptions of leader behavior and certain dimensions of teacher morale (Doctoral dissertation, North Texas State University, 1969). Dissertation Abstracts, 1969, 28 2482A. (University Microfilms No 69-18, 523) Kornhauser A Mental health of the industrial worker : A Detroit study Detroit: Department of Psychology, Wayne State University, 1962. Lambert D. B. A study of the relationships between teacher morale and the school principal's leader behav i or (Doctoral dissertation, University of Pittsburgh, 1969) Dissertation Abstracts, 1969, 28, 2483A (University Microfilms No. 69-16,259) Leavitt, H. J. Managerial psychology : An i ntroduction to ind i vidua l s, pairs, and groups and organizations. Chicago University of Chicago Press, 1972. Lee, B L Teacher absences and cost of substitute services. (Research Memorandum No. 1960-35). Washington, D C .: National Education Association Research Division, 1960 Likert, R New patterns of management. New York: McGraw-Hill, 1961.

PAGE 145

143 Likert, R. The human organization: Its management and value. New York: McGraw-Hill, 1967. Lundquist, A. Absenteeism and job turnover as a consequence of unfavorable job adjustment. Acta Sociologica, 1958, ~, 119-131. Maier, N. Psychology in industry. (2nd Ed.) New York: Houghton Mifflin, 1955. Manganiello, L. P. A study to determine the relationship of teacher self-acceptance and other selected variables to teacher absence behavior. (Doctoral dissertation, University of Miami, 1972). Dissertation Abstracts International, 1972, 34, 3401 A. ( University Microfilms No 72-11, 101) Mann, F. C. & Baumgartel, H. Absences and employee attitudes in an electric power company. Ann Arbor: Institute for Social Research, University of Michigan, 1953. March, J. G., & Simon, H. A. Organizations New York: John Wiley, 1958. Marchant, R. E. The effects of selected variables on teacher absen teeism (Doctoral dissertation, University of Virginia, 1976). Dissertation Abstracts International, 1976, 37, 3710A (University Microfilms No. 76-37,286) Marlin, H. B. An analysis of absenteeism and utilization of sick leave by selected full-time ten-month professional personnel i n a semi rural school system. (Doctoral dissertation, University of Southern Mississippi, 1976). Dissertation Abstracts International, 1976, 37, 3709A. (University Microfilms No. 76-20,947) Marquit, L. J. Perceptions of the superv i sory behavior of secondary school principals Presented at the annual meeting of the American Educational Research Association, Chicago, February, 1968. (ERIC Document Reproduction Service No. ED 020 579). Martin, J. Some aspects of absence in a light engineering factory Occupational Psychology, 1971, 45, 77-89. Maslow, A Motivation and personality (2nd Ed.) New York: Harper & Row, 1970 McGregor, D The human side of enterprise. New York : McGraw-Hill, 1960. Merritt, W. D. Absenteeism in industry today Man/Society/Technology, 1974, 33, 177-178. Metzner, H. & Mann, F. C. Employee attitudes and absences Personnel Psychology, 1953, 467-485. Miles, M B. Some properties of schools as social systems. In G Watson (Ed.), Change in school systems. Washington, D. C : National Training Laboratories, National Educational Association, 1967.

PAGE 146

144 Mochinsky, P. M. Employee absenteeism : A review of the literature. Journal of Vocational Behavior, 1977, 10, 316-340. Moore, R. Personnel and life management: Partners in absence control. Personnel Journal, 1973, 52, 115. Nelson, R. H Relationship between teacher perceptions of reinforcing behavior of the principal and the organizational climate of elemen tary schools. ( Doctoral dissertation, Arizona State University, 1972). Dissertation Abstracts International, 1972, 33, 1210A. (Universi~y M crofilms No. 72-07,810) Nicholson, N & Goodge, P. M. The influence of social, organizational and biographical factors on female absence. Journal of Management Studies, 1976, 13, 234-254. Noland, E. W. appraisal. Attitudes and industrial absenteeism: A statistical American Sociological Review, 1945, 10, 503-510. Orange County Planning Department. prospectus Orlando, Florida: Orange County growth management Author, 1979. Orange County Public Schools. report. Orlando, Florida: Orange County Public Schools statistical Author, 1979. Pennebaker, W K. The correlation of teachers' and principals' perceptions of principals' behavior and its effect on teacher morale (Doctoral dissertat i on, Un i versity of Michigan, 1970). Dissertation Abstracts International, 1970, 30, 3696A. ( University Microfilms No. 70-16,350) Pennsylvania School Board Association, Inc Professional staff absence study. Philadelphia: Author, 1978. Perry, C M. The relationship between teacher morale and the prin cipal's attempts to improve teaching performance (Doctoral dissertation, Loyola University of Chicago 1977) Dissertat i on Abstracts International, 1977 37, 2948A. (University Microfilms No. 77-09,952) Philadelphia Suburban School Study Council. Teacher absenteeism and related policies for supplemental remunerat i on Philadelphia: Author, 1970 Prewett, C R. Let's remove the barriers to good teaching. The School Executive, 1956, ZJ., 84. Raouf, B Profile of an absentee. Industry Week, 1973, 178, 41. Reddin, W. J. Managerial effectiveness. New York: McGraw-Hill, 1970. Redefer, F L. The principal and teacher morale. The American School Board Journal, 1962, 145, 5.

PAGE 147

145 Redmond, M. E. The relationships of selected factors to illness absences in the Fort Madison Community School District--A case study. (Doctoral dissertation, The University of Iowa, 1978). Dissertation Abstracts International, 1978, 38, 3906A. (University Microfilms No. 78-41,223) Rempel, A., & Bentley, R. Manual for the Purdue Teacher Opinionnaire West LaFayette, Indiana: University Bookstore, 1967. Rogers, C Client-centered therapy. Boston: Houghton-Mifflin, 1951. Rogers, E M. Diffusion of intervention. New Yori" Free Press of Glencoe, 1962. Ronan, W. W & Prien, E. P. Perspectives on the measurement of human performance. New York: Appleton-Century-Crofts, 1971. Schroeder, J. H. The relationships between teacher perception of managerial behavior, teacher satisfaction, and teacher absenteeism. (Doctoral dissertation, University of New Orleans, 1977) Dissertation Abstracts International, 1977, 38, 3808A. ( University Microfilms No. 77-21,911) Schultz, R. E. Keeping up teacher morale. The Nation's Schools, 1952, 50, 53. Sergiovanni, T J., & Carver, F D. The new school executive : A theory of administration. New York: Dodd, Mead, 1975. Sharples, D. K. A study of the predictive relationship between absenteeism of classified civil service personnel and job satisfac tion. (Doctoral dissertation, Ohio University, 1973). Dissertation Abstracts International, 1973, 34, 3405A. (University Microfilms No. 73-18,810) Shealy, L. F. A study of the relationship of selected variables to teacher morale (Doctoral dissertation, Purdue University, 1975). Dissertation Abstracts International, 1975, 35, 6997 A ( University Microfilms No. 75-12,384) Shepherd, R. D & Walker, J. Absence and the physical conditions of work. British Journal of Industrial Medicine, 1957, 14, 266-274. Silverman, M. Principals--What are you doing to teacher morale? Educational Administration and Supervision, 1957, 43, 270. Slick, G H The relationship of organizational climate factors which influence morale and other selected variables to absence frequency. ( Doctoral dissertation, Temple University, 1974) Dissertation Abstracts International, 1974, 35, 3508A. (University Microfilms No. 74-11,712)

PAGE 148

146 Smardon, Raymond A. Some cures for chronic absenteeism. Supervisory Management 1974, 19, 12-15. Smith, D. C., Kendall, L. M., & Hulin, C. L. The measurement of satisfaction in work and retirement: A strategy for the study of attitudes. Chicago : Rand McNally, 1969. Stallings, J. W. Sick leave for teachers. California Journal of Educational Research, 1959, 10, 60-63. State of New York, Office of Education Performance Review Teacher absenteeism in New York City and the cost effectiveness of substitute teachers. Albany, New York: Author, 1974. Stogdill, R M. Manual for the Leader Behavior Description Question naire-Form XI I. Columbus: Bureau of Business Research, Ohio State University, 1963. Stogdill, R M. Handbook of leadership: A survey of theory and research New York: Free Press, 1974 Stogdill, R. M., & Coons, A. E. (Eds.). Leader behavior: Its descrip tion and measurement. (Research Monograph No. 88). Columbus: Bureau of Business Research, Ohio State University, 1957 Stryker, J. R. Problems confronting the principal ship today. National Association of Elementary School ?rincipals. 1977, 38, 22-24 Suehr, J. H. A study of morale and education utilizing incomplete sentences The Journal of Educational Research, 1962, 56, 79. Sylwester, R. Educator absences and stress. Oregon School Study Council Quarterly Report, 1979, 19, 18-21 Tagiuri R. Person perception In G. Lindzey and E. Aronson (Eds.), The Handbook of Social Psychology {Vol. 3) (2nd ed.). Reading, Mass.: Addison-Wesley, 7969 Talacchi, S Organization size, individual attitudes and behavior : An empirical study. Administrative Science Quarterly 1960, ~, 398-420. Taylor, D E. Absent workers and lost work hours, May, 1978. Monthly Labor Review, 1979, 102, 49-53. Tirpak, R. D. Relationship between organizational climate of elemen tary schools and personal characteristics of the schools' principals. ( Doctoral dissertation, Fordham University, 1971) Dissertation Abstracts International, 1971, 32, 1451 A. ( University Microfilm No 71-11,442) United States Government Printing Office. United States Department of Labor Statistics, Div i sion of Economic Studies. Washington, D. C.: Author, 1978.

PAGE 149

147 .,;in Zelst, R. H., & Kerr, W A. Workers' attitudes toward merit rating. Personnel Psychology, 1953, 2, 159-172. Vroom, V. H. Work and motivation. New York: John Wiley, 1964 Washington, R., & Watson, H. F. Positive teacher morale--The principal's responsibility. National Association of Secondary School Principals Bulletin, 1970, 60, 4-7 Zaleznik, A., & Moment, D. The dynamics of interpersonal behavior. New York: John Wiley, 1964. Zimet, M. Decentralization and school effectiveness: A case study ~f the 1969 decentralization law in New York City. New York: Teachers College Press, Columbia University, 1973.

PAGE 150

BIOGRAPHICAL SKETCH James Donald Shaw was born in Orlando, Florida, on November 26, 1940. He attended the public schools of Orange County, Florida. He received a degree of Bachelor of Arts in Education in June, 1962, from the University of Florida, Gainesville, and the Master of Arts degree in June, 1968, from Rollins Coliege, Winter Park, Florida. From 1962-1968 he was employed as an English teacher in the public schools of Orange County, Florida. From 1968-1974 he served as an elementary school principal in that district. He served as Administrative Assistant to the Superintendent in Orange County from 1974-1977. From 1977-1980 he worked as the district's Assistant Superintendent for Elementary Education Since April, 1980, he has served as the Deputy Superintendent for Instruction for the public schools of Orange County, Florida. 148

PAGE 151

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Education. airman I I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Education. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Education. This dissertation was submitted to the Graduate Faculty of the Depart ment of Educational Administration and Supervision in the College of E-::lucation and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Education. December, 1980 Dean, Graduate School