Primary students' perceptions of the learning environment as related to teacher training and student achievement

MISSING IMAGE

Material Information

Title:
Primary students' perceptions of the learning environment as related to teacher training and student achievement
Physical Description:
ix, 87 leaves : ; 28 cm.
Language:
English
Creator:
Simmons, Renee, 1956-
Publication Date:

Subjects

Subjects / Keywords:
Academic achievement   ( lcsh )
Teachers -- In-service training   ( lcsh )
School environment   ( lcsh )
School children -- Psychology   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 82-86).
Statement of Responsibility:
by Renee Simmons.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 21206099
ocm21206099
System ID:
AA00011204:00001

Full Text











PRIMARY STUDENTS' PERCEPTIONS OF THE LEARNING ENVIRONMENT
AS RELATED TO TEACHER TRAINING AND STUDENT ACHIEVEMENT










By
RENEESIMMONS


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


1989




























Copyright 1989
by
Renee Simmons













ACKNOWLEDGMENTS


After all of the grueling hours at the word processor, it is with
absolute pleasure that I fumble through this part as well. Although
technically I will receive the credit for this dissertation, it could not have
been completed without the assistance of many of the wonderful people in
my life.
First I would like to thank Dr. Dorene Ross, my committee chair, for her
continued support throughout the dissertation process no matter how long
and demanding it may have been. Her professionalism, astute insights, and
knowledge of the field continually inspired me. Her encouragement,
understanding, and friendship continually motivated me. To her, I promise
never to use the word "while" again. I am grateful to Dr. Regina Weade, my
committee cochair, for her positive reinforcement and reassurance,
especially during the early phases of the process when she made me feel
optimistic at an otherwise frustrating point in the process. I am especially
grateful to Dr. Linda Crocker, who devoted herself above and beyond the call
of duty. She was always there to help and encourage me, especially after
hours of battling with the computer and my statistics. She allowed me to
feel successful at what I was doing even during the most frustrating and
seemingly hopeless moments. I am sure that I would not have made it
through this dissertation without her. I would also like to thank Dr. Linda








Lamme for her friendship and continued professional support and confidence
throughout the dissertation process and all my years at the university. Dr.
Athol Packer and Dr. James Longstreth also served on my committee and
added valuable input into the study.
Next I would like to thank my family and friends, especially my
husband, Scott Simmons, who supported me throughout the dissertation
process, as always, with love, faith, understanding and pride. I now promise
that the dirty dishes will be washed and that we will eat something besides
pizza for dinner. My daughter, Chelsea, was patient and understanding when
I felt like "Oscar" during trying times along the way. Her smile and loving
nature offered me continued motivation to wake up happy and approach
everything with confidence and enthusiasm as she does. My son, Cody, gave
of himself when I didn't have time to play or give him as many hugs as I
wanted to give him. I would also like to thank him for waiting until I
finished my data analysis before coming into the world. He has been
consistently helpful and loving since then.
My parents contributed financial and emotional support that encouraged
their "smart daughter" to consider taking on the Ph.D. process. My sisters,
brother, friends, and the wonderful staff at Holy Trinity Child Caring Center
never gave up on me and confidently trusted that one day I would finally call
and tell them when to come to my graduation. My husband's family was
always there to love and support me and share in the pride he felt for my
accomplishments. My daughter's school, and Cody's babysitters, Angel,
Tammie, and Lyvonne provided lots of "TLC" for them while I was busy
working.








Last, I would like to thank all of the people in the county school
system, especially the teachers that participated in the study, for the
cooperation, friendship, and support that helped make this study possible.













TABLE OF CONTENTS



page
ACKNOWLEDGMENTS.................................................................................... iii
ABSTRACT................................................................................................... viii
CHAPTERS
I INTRODUCTION.......................................................................... 1
The Statement of Purpose................................. ........... 2
Rationale for the Study............................... ........... ....... 3
Definition of Terms........................................................ 5
Limitations of the Study................................................... 6
II REVIEW OF THE RESEARCH............................... ........... 9
Dimensions of the Learning Environment as
Described by the My Class Inventory........................ 9
Students' Perceptions of the Learning
Environment as Related to Student
Achievement.................................................................... 11
Changes in Students' Perceptions of the Learning
Environment as a Result of Changing the
Learning Environment..................................... ........ 14
Review of Staff Development Literature..................... 19
Summary.................................................... ....................... 23
III METHODS.................................................................................. 25
Context and Settings.......................................... ........... 25
S ubjects...................................................................................... 26
Instrumentation..................................................... .......... 28
The My Class Inventory.............................. .......... 28
Reliability for Use of the My Class Inventory
with Kindergarteners......................................... 32
The California Achievement Test............................. 33
Research Design............................................................... 37
Treatment............................................................................ 38
Data Collection Procedures.............................. ....... .. 46
Data Analysis................................. .................................... 47










IV RESULTS AND CONCLUSIONS................................................ 49
Relationship between MCI Scores and Student
Achievement...................................... ........................ 50
Effectiveness of Teacher Training................................. 52
V SUMMARY, DISCUSSION, AND RECOMMENDATIONS......... 59
Discussion and Implications........................... ............ 60
Recommendations....................... .................................... 69
Sum m ary................................................... ........................... 71
APPENDICES
A SUMS OF SQUARES TABLES FOR ANALYSIS OF
COVARIANCE FOR KINDERGARTENERS...................... 74
B SUMS OF SQUARES TABLES FOR ANALYSIS OF
COVARIANCE FOR PRIMARY-GRADE STUDENTS....... 76
C SUMS OF SQUARES TABLES FOR ANALYSIS OF
COVARIANCE FOR KINDERGARTENERS'
ACHIEVEM ENT .......... ....................................................... 78
D SUMS OF SQUARES TABLES FOR ANALYSIS OF
COVARIANCE FOR PRIMARY-GRADE STUDENTS'
ACHIEVEMENT ......................... ................... ................ 80

REFERENCES..................................................................................................... 82
BIOGRAPHICAL SKETCH............................................... .................................. 87













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PRIMARY STUDENTS' PERCEPTIONS OF THE LEARNING ENVIRONMENT
AS RELATED TO TEACHER TRAINING AND STUDENT ACHIEVEMENT

By
Renee Simmons
August 1989

Chair: Dr. Dorene Ross
Cochair: Dr. Regina Weade
Major Department: Instruction and Curriculum

Previous research indicated that (a) there is a relationship between
students' achievement and their perceptions of the learning environment in
the upper elementary grades, and (b) with individualized training, teachers
can alter classroom learning environments. The purpose of this study was
to investigate the relationship between younger students' perceptions of the
learning environment and their reading achievement. The researcher also
tested the effectiveness of a training program to help teachers change
students' perceptions of their learning environment. Outcomes were
measured for 177 kindergarten children and 123 primary-grade children.
Fifteen volunteer elementary teachers from a single school district
were classified into two groups--either kindergarten or primary. Each
was then randomly assigned to the treatment group, which received









training or to the control group, which received no training. The treatment
consisted of six meetings over a five-month period and included activities
and discussions to facilitate the teachers' implementation of the
researcher-written handbook, Creating a Positive Learning Environment: A
Teacher's Handbook. The teachers' students were administered the My Class
Inventory (MCI) in the fall prior to the training and five months later,
subsequent to the training sessions. Subscale scores on the MCI described
students' perceptions of the learning environment in five areas.
Significant Pearson correlation coefficients (p< .01) were found
between the kindergarten and combined kindergarten and primary groups'
perceptions of difficulty of class work and student achievement.
Kindergarten students' perceptions of friction in the classroom and
primary students' perceptions of satisfaction were significantly correlated
to student achievement (p< .05).
The data were also used to test for the effectiveness of the teacher
training. Treatment effects were tested using an analysis of covariance
model with preMCI and prior achievement as covariates and treatment and
teacher-within-treatment as independent variables in the model. The
observed mean differences between treatment and control groups on the MCI
subscales of satisfaction, friction, cohesiveness, and difficulty scores
changed in the expected direction; however, none of these differences were
statistically significant, suggesting the training program was not effective.













CHAPTER I
INTRODUCTION


There is growing interest in the study of classroom learning
environments because the nature of the classroom learning environment has
been shown to be correlated with student achievement (Walberg & Haertel,
1980). For example, positive relationships between students' perceptions of
the learning environment and cognitive or affective learning outcomes have
been reported by Fraser (1979), Fraser and Fisher (1982a), Fraser and
O'Brien (1985) and Walberg (1969a). Results from a meta-analysis
performed by Haertel, Walberg, and Haertel (1981) indicated that students'
perceptions of difficulty, cohesiveness, formality, democracy, and
satisfaction in the class were positively associated with gains in learning
in a variety of subject areas on a variety of measures of achievement.
Competition and friction, however, have a negative relationship to
achievement (Fraser & Fisher, 1982c; Walberg, 1969a; Talmage & Walberg,
1978).
Studies relating to affective development imply students'
self-concepts are more positive in classes that have greater involvement,
affiliation, teacher support, and satisfaction (Galluzi, Kirby, & Zucker,
1980). Cohesiveness is consistently linked with more positive attitudes at
different grade levels and for different subjects (Haladyna, Shaughnessy, &
Shaughnessy, 1983). Furthermore, friction and satisfaction have been





2



reported to have an effect on science interest and activities
(Walberg,1969b).
Based upon the results of these studies, Anderson and Walberg (1974)
and Fraser and O'Brien (1985) recommended making efforts to change
students' perceptions of the learning environment. Although the ultimate
goal was to have a positive impact on learning outcomes, Anderson and
Walberg (1974) stated that assessment of changes in perceptions of the
learning environment was an important first step because

valid and useful differences among educational treatments are
often reflected first and most strongly in changes on student
perceptions of their learning environment. Later, and in moderated
form, these changes also show up in terms of student learning and
other such indicators of outcome. (p. 95)

Thus, Fraser (1986) insisted that "the classroom environment is such a
potent determinant of student outcomes that it should not be ignored by
those wishing to improve the effectiveness of schools" (p.1).


The Statement of Purpose

The purpose of this study was twofold: (a) to investigate the
relationship between students' perceptions of the learning environment and
student achievement for primary-grade students, and (b) to investigate the
effectiveness of a group training program designed to change students'
perceptions of the learning environment.
Specifically, the purpose of this study was to examine the following
research questions:










1. Is student perception of the classroom learning environment
related to student achievement?

a. Is student perception of cohesiveness related to student
achievement?
b. Is student perception of competitiveness related to
student achievement?
c. Is student perception of difficulty related to student
achievement?
d. Is student perception of friction related to student
achievement?
e. Is student perception of satisfaction related to student
achievement?


2. Do students' perceptions of the classroom learning environment change
significantly after teachers receive training on altering the learning
environment?

Rationale for the Study

The majority of the researchers who have reported positive
correlations between students' perceptions of the learning environment and
student achievement studied students in grades four and above. This study
extended that body of research on the relationship between students'
perceptions of the learning environment and student achievement to the
primary grades.
In several intervention studies, researchers have focused on training
teachers to change students' perceptions of the learning environment. In
these small-scale studies (Fraser & Fisher, 1986; Fraser & O'Brien, 1985;
Fraser, Seddon, & Eagleson, 1982) the researchers trained the teachers










individually. The individual teacher received feedback about the students'
perceptions of the learning environment based on a classroom environment
inventory administered to the class. The researchers presented the
teachers with the results of the inventory and informally discussed the
results. After discussing ideas for intervention to facilitate change where
it was warranted, the teachers were encouraged to implement the ideas.
Data on students' perceptions collected two months later documented that
change occurred in most of the areas targeted by the teachers.
These outcomes suggested that it was possible to train teachers to
change students' perceptions of the learning environment; however, the
methods used were unrealistic for training a larger audience of teachers
efficiently. In the present study, the researcher attempted to train
teachers to change students' perceptions of the learning environment by
testing the effectiveness of a staff development program that was more
time- and cost-efficient. Research-based staff development principles
were combined to develop a training method that included assessment,
feedback, reflection and discussion, intervention, and reassessment. This
researcher designed a teacher handbook to provide the rationale and ideas
for facilitating change in students' perceptions of the learning environment,
and to compensate for some of the individualized trainer time in the
previous studies. The reduced need for individual consultation also
decreased the time needed for actual training sessions. Therefore, the
present study was designed to test the effectiveness of this feasible,
large-scale staff development approach in changing students' perceptions of
the learning environment.





5



Definition of Terms

Learning environment refers to the classroom environment
characterized by the following five dimensions of the My Class Inventory
(MCI) described by Fraser (1986):

Cohesiveness-- the extent to which students know,
help, and are friendly towards each
other.
Friction-- the amount of tension and
quarreling among the students of the class.
Difficulty-- the extent to which students find
difficulty with the work of the class.
Satisfaction-- the extent of enjoyment of class work.
Competitiveness-- the emphasis on students competing with
each other (p. 20).


My Class Inventory (Fraser, Anderson, & Walberg, 1982) refers to an
inventory designed to obtain perceptual measures of the classroom learning
environment of elementary school students. A short form, containing
twenty-five dichotomously scored items, was used in this study. The
Actual Form of the MCI refers to the inventory on which the respondents
answer according to how they perceive the existing learning environment.
The Preferred Form of the MCI refers to the inventory on which the
respondents answer as they would prefer the learning environment to be.

Student achievement, when discussed as a variable in this study, refers
to the students' scaled reading scores on Form E of the California
Achievement Test (California Achievement Tests, 1986). The kindergarten
reading score is a composite of a vocabulary section and an oral
comprehension section for the Level 10 test. The primary-grade reading










score is a composite score of a vocabulary section and a written
comprehension section for the Level 11 test.

Training program refers to the program designed to introduce teachers
to the elements of their classroom learning environment and to stress the
significance of these elements. Structured learning activities helped
teachers learn how to facilitate change in students' perceptions of the
learning environment. This term was used interchangeably with "staff
development" throughout the text.

Creating a Positive Learning Environment: A Teacher's Handbook is a
67-page, researcher-written handbook designed to help teachers create a
more positive learning environment in their classrooms. The text and
activities were arranged according to the five dimensions of the My Class
Inventory. For the sake of economy, this will be referred to as "the
handbook" throughout this text.

Trainer refers to the person who conducted the meetings and presented
the structured learning activities for the staff development program. In
this study the researcher was the trainer.

Limitations of the Study

Although it was advantageous to have teachers from both the
experimental and training group at several of the same schools to control
for the overall school-climate effect, it was also problematic. The
possibility existed that the two groups interacted and shared information
with one another, thus minimizing the differences between the experimental










and control groups. Both groups, however, were asked not to discuss their
training experiences with anyone outside of the training sessions.
An additional potential threat to the validity of the study was the
possibility of an experimenter effect (Bracht & Glass, 1968). If teachers
had administered the My Class Inventory to their students, it seemed likely
that students might alter their answers to please their teachers. Therefore,
the researcher administered the inventory to all of the classes. Although
the students may have been somewhat intimidated by a stranger
administering the test, they had no preconception of the researcher's
expectations. This procedure also provided consistency in the
administration of the instrument. The procedures used during each of the
administrations of the inventory remained constant. Therefore, any
possible pretest sensitization remained constant over all of the teachers'
classes.
Another possible limitation of this study was the use of a self-report
instrument. Use of such instruments for elementary students has received
some criticism. Some critics contend that student responses are
"irresponsible and unrealistic manifestations of feelings with a strong
emotive bias" (Kuert, 1979, p.180). However, there are proponents of
student inventories and questionnaires who regard them as realistic and
reliable measures of the environment (Kuert, 1979; Fraser & Walberg,1981).
For example, the My Class Inventory has been established as a reliable tool
for older students. However, although the MCI has been proven reliable for
primary-grade elementary students in past studies (Fisher & Fraser, 1981;
Prawat & Solomon, 1981), its use with students in primary grades has





8



been limited. Therefore, there was less evidence of the validity and
reliability of the MCI for primary-grade students.












CHAPTER II
REVIEW OF THE RESEARCH


This review of literature, related to the learning environment and its
relationship to student achievement, is organized according to the following
topics: (a) dimensions of the learning environment as described by the My
Class Inventory. (b) students' perceptions of the learning environment as
related to student achievement, and (c) changing the learning environment to
change students' perceptions of the learning environment. Further, a review
of literature on staff development supporting potentially effective means
for helping teachers learn to change the learning environment of their
classrooms is presented.
Several of the studies reviewed in this chapter were conducted and
reported in countries other than the United States and contained limited
references to the race, age, sex, or socio-economic status (SES) of the
subjects. Where available, however, such demographic information has been
reported.

Dimensions of the Learning Environment as Described by the
My Class Inventory

Influences on the learning environment of a classroom vary, and many
dimensions have been used to describe it (Moos & Trickett,1974; Fraser et
al., 1982; and Fraser, 1986). Elementary school teachers are likely to be
interested in the My Class Inventory for use with their students









(Fisher & Fraser, 1981; Fraser et al., 1982; Prawat & Solomon, 1981)
because the Mv Class Inventory (Fraser et al., 1982) elicits descriptions of
the learning environment through the following five dimensions:
cohesiveness, competition, difficulty, friction, and satisfaction.
All five dimensions have been shown to have a relationship to student
learning outcomes ( Fraser & Fisher, 1982a; Walberg, 1969; Talmage &
Walberg,1978). The degree of difficulty of children's work, for example, can
affect students' perceived level of competence (Jones & Jones, 1981). The
friction between classmates and the amount of cohesiveness in the
classroom affect the students' sense of belonging to the group, which in
turn affects the students' sense of security (Dreikurs,1964). When children
feel secure they are more likely to take risks in attempting to master new
skills (Jones & Jones, 1981).
Competitiveness in a class can have a negative effect on achievement
too. Talmage and Walberg (1978) found that students who perceived their
learning environment as highly competitive showed lower reading
achievement gains than students who perceived their learning environment
as low in competition. Furthermore, when there is more competition in a
class, students tend to have less positive feelings about their teacher than
do students experiencing cooperative instruction (Johnson & Johnson,
1979).
Combs (1982) argued that learning occurred best when teachers were
successful in creating atmospheres that were challenging without being
threatening. Careful manipulation of the amount of difficulty,
competitiveness, satisfaction, friction, and cohesiveness in the classroom









enhances the learning environment and aids in development of the students'
self-concepts.

Students' Perceptions of the Learning Environment as Related to
Student Achievement

The relationship of students' perceptions of the learning environment
to student achievement has received increased attention in the past two
decades. Research in several countries with different age groups, subject
areas, and sample sizes, has revealed consistent results in learning
outcomes. Haertel et al. (1981) conducted a meta-analysis involving 734
correlations obtained from 12 studies of 10 data sets in eight subject areas
encompassing 117,805 students in 823 classes in four nations. They found
that students' varying perceptions of the classroom environment contributed
to variance in learning outcomes. Moreover, this variance was beyond that
accounted for by ability and pretest measures, regardless of the
characteristics of the study. Learning outcomes were found to be positively
associated with certain classroom environment dimensions (e.g.,
cohesiveness, satisfaction, and material environment) and negatively
associated with others (e.g., friction, apathy, and disorganization). The size
and magnitude of the correlations were consistent across studies.
Researchers studying elementary school students generally have
replicated the findings obtained from research conducted with older
students. Ellett, Masters, and Pool (1978) administered the MCI orally to a
sample of 6,151 fourth-grade students from 89 schools in Georgia. They
found that the five MCI dimensions accounted for significant proportions of
the variance in achievement (72%) and attendance (14%). Boulanger (1980)
administered the MCI to a sample of 27 classes in grades four through eight









in schools throughout metropolitan Chicago. Associations between a
reasoning outcome and environment perceptions were found to be
statistically significant for four of the MCl's five scales.
Fraser and O'Brien (1985) conducted a study by administering the MCI, a
word knowledge, and a reading comprehension test to a sample of 758
third-grade students in 32 classes in eight schools from the suburbs of
Sydney, Australia. Using the class mean as the unit of analysis, simple
correlations were statistically significant (p< 0.01) for word knowledge and
comprehension for the cohesiveness, difficulty, friction, and satisfaction
scales. Only the competitiveness scale proved nonsignificant. The multiple
correlation was 0.88 between the total set of five scales and word
knowledge, and 0.85 between comprehension and the set of five scales.
Therefore, students' perceptions of their classroom environment accounted
for 77% of the variance in word knowledge and 72% of the variance in
comprehension. Performance on both the word knowledge and comprehension
measures was greater in classes perceived by students to have more
satisfaction, less friction, less difficulty, and less cohesiveness.
The reported negative relationship between cohesiveness and student
achievement required further inquiry. Friction and cohesiveness appeared
contradictory based on the meta-analysis. The impact of positive group
interactions (Schmuck & Schmuck, 1983) indicated cohesiveness would
facilitate student achievement. One possible explanation for this
discrepancy was that the study reviewed was conducted in Australia, and
the Australian educational system may differ from that of the United States
with respect to cohesiveness in the classroom.









A significant relationship was found in another study of elementary
school students. Talmage and Walberg (1978) studied the
outcome-environment relationship using the My Class Inventory (MCI) among
students participating in a district reading program in Illinois. The sample
consisted of approximately 1,600 students from grades one, two, three, and
six in 60 classes. Reading achievement was measured approximately one
year later using alternate forms of the Science Research Associates Reading
Test. A multiple regression analysis with the class as the unit of analysis
revealed that the set of MCI dimensions accounted for a significant
increment of 11% in the variance in posttest reading achievement beyond
that attributable to pretest reading scores. Talmage and Walberg (1978)
concluded that perceptions of greater classroom competitiveness were
associated with lower reading achievement.
Although these studies involving student perceptual measures in the
elementary school have provided interesting results, such studies continue
to be scarce. Perhaps one reason is that typically multiple classrooms must
be involved for large-scale evaluations of the effect of classroom climate
on achievement. In addition, the more practical and reliable 38-item
version of the MCI (Fraser et al., 1982) was relatively new and infrequently
used, and it may have been too long for use with elementary students. A
shorter form of the MCI, developed and validated in 1982 (Fraser, 1982;
Fraser & Fisher, 1983), may encourage future use of student perceptual
measures of the learning environment in the elementary school grades.
In summary, data from a range of countries, grade levels, and subjects
have suggested that student achievement is positively associated with
cohesiveness and satisfaction and negatively associated with friction. At









the elementary level student achievement has been positively associated
with satisfaction and negatively associated with friction, difficulty,
cohesiveness, and competitiveness. Although the relationship between
friction and student achievement has been consistently observed at the
elementary level, a discrepancy exists in this relationship when reported
for a wider range of grade levels.
Although the findings of the research on correlations of student
achievement with perceptual measures of the learning environment have
been fairly consistent, there are limitations to this body of research: (a) the
instruments used in these studies were all relatively new, (b) the different
instruments assessed different aspects of the learning environment, (c) the
number of studies was small, and (d) generalizability was limited because
individual studies have not been replicated.

Changes in Students' Perceptions of the Learning Environment as a
Result of Changing the Learning Environment

The question of whether changing the students' learning environment
would lead to subsequent changes in students' perceptions was investigated
by Fraser (1986), Anderson and Walberg (1974), and Fraser and O'Brien
(1985) with positive results. These researchers employed a five-step
procedure for changing the learning environment: (a) assessment,
(b) feedback, (c) reflection and discussion, (d) intervention, and (e)
reassessment. Assessment involved collection of data using student
perceptual measures. Teachers received a report of the scores from their
students' inventories. Reflection and discussion were held informally
between the teachers and their peers and between teachers and the
researchers. The intervention involved training teachers in a variety of









strategies aimed at changing selected dimensions of the learning
environment. Reassessment was a readministration of the student
perceptual measure used as the pretest.
Fraser and Fisher (1985) studied an elementary school teacher with a
class of 26 sixth-grade students of lower ability attending a coeducational
government school in a suburb of Sydney. The students were assessed using
the Actual and Preferred forms of the MCI. The teacher was provided with
feedback derived from the students' responses. After reflection and
discussion with the teacher's peers and with the researcher, intervention
was aimed at two specific dimensions on which the teacher chose to
concentrate.
Intervention consisted of a series of meetings between the teacher and
the researchers, which led to the development of a variety of strategies for
facilitating change in the learning environment. Some of these strategies
were suggested by the wording of the items contained in the inventory. For
example, one response, "Some people in my class are not my friends,"
alerted the teacher to give some special assistance to particular students
who seemed to be less well-liked by their classmates.
Following intervention, the teacher concentrated on implementation of
the intervention for two months. At the end of the implementation, the MCI
was readministered to students. The results were reported graphically with
three separate lines: the Actual scores on the pretest, the Preferred scores
on the pretest, and the Actual scores from the posttest. Statistically
significant reductions occurred in actual-preferred discrepancy on the
competitiveness scale (t= 2.4, p< 0.05) and cohesiveness scale (t- 2.6,
p< 0.05), but nonsignificant changes occurred on the other three MCI scales









(friction, satisfaction and difficulty). The two scales for which significant
changes were reported were those areas that the intervention was designed
to affect.
Fraser and Fisher (1984) performed another study using the same
five-step procedure in a science class of ninth-grade students of mixed
ability in a government school in Tasmania. The assessment instrument in
this study was the short form of the Classroom Environment Scale (CES)
(Moos and Trickett, 1974). The Preferred form was answered first, and the
Actual form was administered in the same time slot one week later. The
feedback was in the form of a profile of mean classroom environment
scores. Reflection and discussion occurred among the teacher, the
researcher, and the teachers' peers.
The main criteria used for selection of dimensions for change were (a)
existence of a sizeable actual-preference discrepancy, (b) teacher concern
about the reported differences, and (c) teacher desire to reduce the
discrepancy. For more concentrated efforts, it was recommended that the
teacher choose only two areas to change. Intervention was a variety of
strategies which originated from discussions with the teacher's peers and
from ideas contained in the individual CES items. Reassessment involved a
second administration of the student Actual form of the CES in order to
determine whether the students perceived their classroom environment
differently than before the intervention.
The five-step procedure in this study yielded positive results. Tests of
statistical significance indicated that pretest-posttest differences were
significant in improving the classroom environment in (a) Teacher Support









and (b) Order and Organization, the two dimensions that the teacher had
attempted to change.
A third small-scale study was conducted by Fraser et al. (1982) in a
private secondary school in suburban Sydney. The sample consisted of a
class of 31 seventh-grade boys of mixed ability who were studying English,
mathematics, and history with the teacher involved in the study. The study
employed the five-step procedure used in the previous studies to address
improvement in the learning environment.
The student Actual and student Preferred forms of the Individualized
Classroom Environment Questionnaire (ICEQ) were used to assess initial
perceptions. The Actual form of the ICEQ was administered to the students
by the teacher during the first week, and the Preferred form in the same
time slot one week later. The teacher received feedback about student
responses to the ICEQ. The data were analyzed by university staff and
presented to the teacher in the form of computer-generated profiles
representing class means of students' Actual and Preferred environment
scores. The university staff helped interpret the results for the teacher.
Reflection and discussion about the profiles took place through private
reflection, as well as through discussions between the teacher and the
university staff. The criteria for selecting the area of focus for change
were the same as in the studies discussed above. The intervention, which
lasted approximately one month, consisted of a variety of strategies, some
of which originated during meetings between the teacher and university
personnel and others which were suggested by university personnel
examining the content of the inventory items. The reassessment was a









readministration of the student form of the ICEQ at the end of the
intervention.
The t tests for dependent samples for the significance of
pretest-posttest changes in discrepancy between PreMCI and PostMCI scores
on each scale were calculated. Because only a single assessment of
Preferred environment was made, these t tests for pretest-posttest
changes in discrepancy scores were equivalent to t tests for pretest-
posttest changes in Actual scores. A large and statistically significant
reduction occurred in actual-preferred discrepancy on the scales during the
time of the intervention. On the other hand, nonsignificant changes were
observed for the other three ICEQ scales. Thus, the procedure proved
successful in changing the learning environment in targeted areas.
Upon inspection of the studies reviewed in this chapter, several
similarities were noted. First, in each of the studies, researchers reported
success in changing the learning environment. Second, each of the studies
involved one classroom teacher and the teacher's students. Third, the
teachers received individualized attention from the researcher. Fourth, the
effects of the intervention were all assessed immediately after the
intervention.
These common features are important to consider when planning future
investigations. The results were statistically significant in all of the
studies, so the procedure was proven to be effective. However, because
there has been no long-term reassessment, it is difficult to determine the
long-term effects of intervention. Each study involved intense,
individualized researcher participation, so the practicality of these
procedures is open to question. The cost of providing such expertise to









large numbers of teachers would be prohibitive, and the implementation of
the procedure on a large scale following the exact format appears unlikely.
Although the immediate success of the studies reviewed was
encouraging, the following factors affect the applicability of this research
to regular classroom situations: (a) lack of investigation of long-term
results, (b) limited generalizability because of the small sample size and
concentrated area of study, and (c) the impracticality of replicating the
study on a large scale.
A comprehensive investigation of class environment requires the
availability of the long-term results. In addition, whether the training
process could be altered to accommodate the training of many teachers
using the same basic principles remained undetermined. The researcher in
the present study, concerned with the development of such a process,
designed a training program combining this successful five-step procedure
with effective staff development principles. The balance of the literature
review supports the design of the training program.

Review of Staff Development Literature

A review of the general literature relating to effective staff
development practices supported the method of teacher training used in the
present study. This staff development program was designed to reach a
larger audience than that reached by the previously described studies.
Voluntary participation is a key principle of effective staff
development. Although the use of volunteers for experiments often raises
questions as to the results for teachers in general (Gage, 1985), it has been
found to be related to positive outcomes of specific programs (Griffin,
1982). A willingness to learn and a higher motivation to improve were









often displayed by volunteers (Gage, 1985; Roberts, 1984). Although the use
of volunteers may affect the generalizability of results, their participation
may also serve to motivate others to participate. Thus, the use of
volunteers was generally recommended.
Once enrolled in a training session, several factors affect the degree to
which teachers adopt new behaviors. Teachers were more likely to adopt
new ideas if they were specifically stated (Anderson, Evertson, & Brophy,
1979), cost-effective in terms of time and energy, and consistent with the
teacher's own role definition (Gage & Giaconia ,1981; Doyle & Ponder,
1977). Teachers consider the amount of role change involved in
implementing the suggested recommendations as well as the philosophical
and practical value of making the change (Mohlman, Coladarci & Gage, 1982).
Those recommendations seen as practical were more likely to be
incorporated into the teachers' plans (Doyle & Ponder, 1977). Therefore, the
way recommendations were stated, the practicality of their
implementation, and their congruence with existing values all affected a
teacher's decision to implement change.
Because congruence with existing values is important for acceptance,
it is important to provide alternative choices. Rubin (1978) stated teacher
training should emphasize instructional alternatives rather than mandatory
methods. Speiker (1978) said teachers will receive greater benefits from
programs in which they are allowed to choose goals and activities for
themselves than from programs with predetermined activities and goals.
Though it is important to respect and encourage individual choices
when designing a staff development program, it is also important to
encourage shared efforts. The recommendation that teachers work in small









groups was repeated throughout the literature (Wood & Thompson, 1980;
Farley, 1982; Speiker, 1978). Such interaction encouraged discussion and
allowed teachers to clarify their ideas and mutually support one another.
Finally, in addition to carefully planning the actual delivery style of
the material, some type of feedback was encouraged (Joyce & Showers,
1980). Programs that emphasized demonstrations, supervised trials and
included feedback were more likely to accomplish their goals (Speiker,
1978). Oliver (1980) also found it important to help teachers determine the
rationale for their behavior and the effects of their actions on classroom
processes. A continuous, guided reflection where teachers were encouraged
to ask questions and examine their experiences from a variety of views was
recommended by Sprinthall and Thies-Sprinthall (1983) on the basis of
initial cognitive-developmental studies. Moreover, documentation of
activities, including planning, implementation, and outcome was
recommended (Howey & Vaughan, 1983). Thus, feedback in any of several
forms should be included in an effective staff development program.
Several important recommendations have been supported throughout the
effective staff development literature. These recommendations include
(a) sharing information specifically, (b) having the information be
practically applicable, (c) having ideas that are congruent with teachers'
existing values, (d) encouraging small-group interaction during the training,
(e) providing individual choices of goals and activities, and (f) including
some form of feedback where the participants can judge their own progress.
A study discussed by Mohlman et al. (1982) supported the design of the
present training program. The trainers in that study created an interactive,
supportive workshop environment where teachers could communicate with









the trainer and their peers about their class experiences. Furthermore,
activities were designed to encourage discussion of difficulties,
disadvantages, or philosophical objections to the information posed. It was
the goal of the present study to achieve that same environment through the
design of the workshops.
Because the trainer was the leader at the four meetings, members of
the treatment group interacted more with the trainer than did the control
group. As recommended by Rubin (1978) teachers were allowed continuous
access to an available trained technical resource.
Gage and Coladarci (1984) concluded that the teachers needed more
personal contact or a stronger rationale to adopt the principles proposed in
the training. To encourage adoption of recommended procedures, the trainer
provided lectures on the rationale behind the principles and emphasized
their practicality. Moreover, the trainer led several small-group discussion
activities because peer interaction was repeatedly recommended throughout
the teacher training literature (Speiker, 1978; Farley, 1982; Wood, 1980).
The major resource for the intervention was a teacher's handbook. The
handbook included information related to all five scales and was provided to
the teachers at the beginning of the training. A handbook, or similar manual,
has been used successfully in prior studies (Anderson et al., 1979; Good and
Grouws, 1979).
A minimal intervention study by Anderson et al. (1979) involved first-
grade teachers who received a manual presenting an instructional model.
The researchers met with teachers in schools in October of the school year
and described the purpose of the study. The teachers who agreed to
participate were included as the treatment group and read the manual









describing the instructional model. They met again with the researchers to
discuss the manual. There was no further training conducted during the year
to boost the treatment. Although the training was minimal, there was a
significant treatment effect favoring the experimental group on a reading
composite score.
Good and Grouws (1979) investigated the effectiveness of an
experimental teaching program in fourth-grade classrooms. The program
was presented in a 45-page manual similar in nature to the handbook
designed for the present study. There were two 90-minute training
sessions in which the purpose was to entertain questions about the meaning
of certain teaching behaviors and to respond to any difficulties the teachers
may have encountered. Observational measures revealed that teachers
generally implemented the treatment, and analyses of the product showed
that students of treatment teachers generally outperformed those of the
control group on both standardized and content tests.


Summary

Learning in the classroom was positively correlated to satisfaction
with the learning environment and negatively correlated to perceptions of
difficulty, friction, and competitiveness in the learning environment.
Contradictory data with regard to the relationship of cohesiveness to
student achievement indicates the need for further research. Further
research on the relationship between student achievement and various
dimensions of the learning environment will broaden the bases for
investigating correlations.









Consistent findings of a relationship between student achievement and
several dimensions of the learning environment indicated the need for
facilitating development of a constructive learning environment. Therefore,
principles of effective staff development were reviewed to establish the
method for adapting this procedure to a larger audience.
The following principles were derived from the literature for
developing a training program on classroom environments: (a) focus on
behaviors familiar to teachers and consistent with their own role
definition, (b) give specific assistance in implementing behavior change
that is complex or novel, (c) make recommendations easily understood,
explicit, and in operational language, (d) provide a rationale based on other
classroom procedures or student outcomes, and (e) develop desired
behaviors that are cost effective in terms of time and energy.
The research on effective staff development provided a strong basis
from which to design effective training programs. The staff development
literature reinforced these findings and added several other
recommendations: (a) use small groups to encourage peer teaching,
(b) provide opportunities for participants to practice what they are learning
in real work situations as part of their training, (c) allow teachers to
choose their own goals and activities, (d) divide material into segments, and
(e) encourage guided reflection. The treatment in the present study
combined what we know about facilitiating change in the learning
environment for individual teachers with elements of effective staff
development in an attempt to facilitate and maintain positive changes in the
learning environment in primary-grade classrooms.











CHAPTER III
METHODS


The methodology for this study was designed to investigate the
relationship between students' perceptions of the learning environment and
their achievement, and to determine if change had occurred in students'
perceptions of the learning environment as a result of their teachers'
participation in a training program. In the sections of this chapter the
context and setting, subjects, instrumentation, research design, treatment,
data collection, and analysis procedures are described.

Context and Settings

The county where the study was conducted contained 21 elementary
schools, grades kindergarten through five. The kindergarten students are
required by state law to be five years of age by September 1 of the year in
which they enter school. Likewise, the first graders are required to be six
years old by September 1 of the year in which they enter school. Of the
thirteen classes involved in this study, eight were kindergarten classes.
This North Central Florida school district includes predominantly
suburban schools. Fifteen are located within the city limits of a small
metropolitan area. Six of the schools, however, serve a predominantly rural
population from small towns outside the city. The volunteers for the study
came from ten different schools--eight of the schools are in the small










metropolitan area and two are from small towns outside the city. The
extent of participation in staff development activities differed for each
teacher in the county. Though no district-level requirements exist, the
state requires 120 hours of staff development or six college credits for
teachers to renew their licenses for a five- or ten-year term (depending
upon when they were first licensed). The requirements for recertification
changed within the time frame of the present study and this may have
influenced enrollment in the training program. Five teachers participating
in the study reported that they took the training so they could meet the new
state requirements. However, a major state university is located in the
study county, and the college credit option is used often. Moreover, many of
the county's teachers are working on advanced degrees and have met the
state's requirements while pursuing courses in the College of Education.


Subjects

Although the target population of the present study included all
elementary school teachers in the school system, it was impossible to
include the whole of this population. The experimentally accessible
population, therefore, included teachers who volunteered for participation
during a county inservice training program conducted by the researcher.
Of the 15 teachers who volunteered for the training program, eight
taught kindergarten, one taught a kindergarten/first-grade combination
class, three taught first-grade, one taught a first/second-grade
combination class, and two taught second-grade. Ten of the classes were










heterogeneously grouped, and five of the classes were homogeneous
remedial classes. The teaching experience, level of college degree, and
experience in the present district or building varied among volunteers.
Because the MCI had not previously been used with kindergarten
students, the researcher decided to treat kindergarten and primary-grade
teachers as separate strata. The kindergarten stratum included eight
kindergarten teachers with a total of 177 students, and the primary-grade
stratum included one teacher of a kindergarten/first-grade combination
class, three first-grade teachers, one teacher of a first/second-grade
combination class, and two second-grade teachers with a total of 123
students for the primary-grade stratum.
Once divided into groups by grade, teachers from each group were
randomly assigned to one of two conditions: the treatment group or the
control group. As a result, eight teachers were assigned to the treatment
group and seven teachers to the control group. Of the eight classes in the
treatment group, five were kindergarten classes, two were first-grade
classes and one was a second-grade class. One of the treatment group
classes was a homogeneous remedial class. Of the seven control group
classes, three were kindergarten classes, one was a kindergarten/first-
grade combination class, one was a first-grade class, one was a
first/second-grade combination class, and one was a second-grade class.
Four of the control group classes were homogeneous remedial classes.
A form was completed and sent to the University Human Subjects
Review Board to assure that the subjects were protected during their










participation. Although it is usually necessary to obtain informed parental
consent to test the students participating in research studies, this
requirement was waived, as the training program conducted was part of the
county's master staff development program and the information obtained
from the students was a requirement of the county staff development
program.


Instrumentation

The My Class Inventory
The My Class Inventory (MCI) was used to assess teachers' and students'
perceptions on scales related to five dimensions of the learning
environment: cohesiveness, competitiveness, difficulty, friction, and
satisfaction. A copy of the short form of the instrument was obtained from
the book, Classroom Environment by Fraser (1986) and Fraser and Fisher
(1986).
The short form of the MCI consists of 25 statements. The 25 items are
arranged so that every fifth statement assesses the perception of one of the
five dimensions. For example, perceptions of satisfaction are determined
from responses to statements 1, 6, 11, 16, and 21. Each of the five scales
consists of five statements. Scale descriptions and representative items
selected from the five scales are presented in Table 3-1 to illustrate both
the format and content of the inventory. Note that the word "pupil" was
deleted from items in favor of the more desirable term "children" (Prawat &
Solomon, 1981).











Table 3-1
Sca e Descriptions and Sample Items of t y


(lkan Invc~ntnnJ (MC~I~


Scale

Cohesiveness


Competitive-
ness


Difficulty


Friction


Satisfaction


Scale Description

extent to which students know, help,
and are friendly toward each other

emphasis on students competing with
to each other


extent to which students find difficulty
with the work of the class

amount of tension and quarreling
among students

extent of enjoyment of class work


Sample Item

In my class everybody is my
friend.

Most children want their work
be better than their friend's
work.

In our class the work is hard
to do.

Children in our class like to
fight.

Children seem to like the
class.


The Actual and Preferred Forms of the MCI

The two forms of the MCI, Actual and Preferred, differ only in the

directions given to the respondent. That is, when administering the Actual

Form of the MCI, the directions are, "Respond to the items as they reflect

how you perceive the learning environment in your classroom to be." When

administering the Preferred Form of the inventory, however, the directions

are, "Respond to the items as they reflect how you would prefer the learning

environment in your classroom to be." The format and the scoring are

identical for the two forms.

Scoring. The inventory is designed for hand scoring because it is

generally used by individual teachers. Twenty of the items are scored with

three points for yes, one point for no, and two points for an invalid or


. .. ..III v I v W ,i li


I










omitted response. Five items, however, are worded so that they are scored
in reverse and are marked with a line under the number to remind the scorer
to score the items in reverse--one point for yes and three points for no.
Therefore, questions 6, 9, 10, 16, and 24 need special attention during
scoring.
Although the inventory was easily scored by hand, it was used in many
classrooms, and was administered to approximately 1,000 children.
Consequently, adaptations in scoring were necessary. The questions were
reprinted in a larger format for use by younger elementary school children
and were printed on a general purpose computer scanning sheet so the
numbers of yes, no and invalid responses could be calculated by computer.
Validity. Concurrent validity is a measure of how the test scores of a
group of subjects relate to a criterion measure administered at the same
time or within a short interval of time (Borg & Gall, 1979). The correlation
of the long and short forms provided an estimate of concurrent validity for
the short form of the MCI. Researchers validating the short form of the MCI
have reported correlations of .91 .97 for the different scales (Fraser,
1982; Fraser & Fisher, 1983). To establish construct validity, information
was gathered from a variety of sources to insure that the construct was not
substantially correlated with instruments known to measure different
constructs (Ary, Jacobs, & Razavieh, 1979). In measuring the discriminant
validity of the short form of the MCI, researchers have used the mean
magnitude of the correlation of a scale with other scales in the same
instrument as a convenient index (Fraser, 1982; Fraser & O'Brien, 1985).










The mean correlations for the five scales of the short form of the MCI
ranged from .11 .31. These values suggested that scores on the short form
have some degree of discriminant validity (Fraser, 1986). That is, the
scales in each instrument measure distinct, although somewhat overlapping,
aspects of classroom environment.
Reliability. The reliability of a measuring instrument is the degree
of consistency with which the instrument would yield the same results
applying it repeatedly to the same individual. (Ary et al., 1979; Babbie,
1986). The long form of the MCI was field-tested on a large sample of
2,305 seventh-grade students in Australia (Fraser, 1982). The shorter
25-item version of the MCI was developed recently based on item analyses
of data collected using the MCI's long form. In addition to these statistical
criteria, items were selected carefully for each short scale to provide
adequate coverage of the universe of ideas included in the corresponding
long form. The data for the reliability study of the short form of the MCI
was based on a sample of 758 third-grade students in 32 classes in eight
schools in an outer suburb of Sydney, Australia (Fraser & O'Brien, 1985).
This age group more closely resembled that of the students in the present
study. The internal consistency reliability coefficients for the five scales
ranged from .58 -.81 for the short form.
Available research using the My Class Inventory and correspondence
with Professor Fraser, indicated the instrument had not been used with
kindergarten age children. Therefore, the researcher established reliability










for use of this instrument with kindergarten age children before including
these children in the study.

Reliability for Use of the My Class Inventory with Kindergarteners

Because the test-retest technique had been recommended for shorter
instruments where consistency of the subjects' scores over time was of
interest (Ary, Jacobs & Razavieh, 1979), the researcher chose test-retest
techniques for establishing reliability in the present study. The MCI was
administered to the same two groups of individuals on two occasions and
the paired scores were correlated. The subjects were 45 children in two
kindergarten classes from two different schools. The groups were
heterogenous in socioeconomic makeup. The test was administered in
groups of four students. The test was read and the students' responses were
recorded by the researcher. The retest was given two weeks later at the
same time of day of the initial test in the same manner with the same
four students in each group. This same testing procedure was later used in
the study for kindergarten children.
The teachers were told that the inventory tested students' perceptions
of the classroom learning environment, but were not shown the instrument
before either administration. The withholding of exact content of the
inventory was to eliminate the possibility of any contamination of the
second test results. The researcher shared the teachers' individual class
results with the teachers after the second administration. The researcher
presented the teachers with a graph of the students' class mean scores and










the teachers' Actual and Preferred scores, and interpreted the results for
them.
The test-retest scores from both classes were paired and a reliability
coefficient was calculated using the Pearson correlation for each of the
five scales of the MCI. Although the competitiveness scale appeared to
exhibit limited reliability with a .17 correlation, the remaining scales had
reliability coefficients that ranged from .30 .60. Although these values
are too low for individual score interpretation, this relatively low
reliability is less serious for scores to be used in creating class means,
especially for scales containing only five items. Table 3-2 shows the
test-retest correlations for each of the five scales.


Table 3-2
Test-retest Correlations for the Five Scales of My Class Inventory (MCI)

Satisfaction Friction Comweti- Difficulty Cohesiveness
tiveness
Test-retest 0.6015 0.4225 0.1718 0.3031 0.4334


The California Achievement Test

The California Achievement Test (CAT), Form E, was used to assess
students' reading achievement. The complete battery was administered to
all elementary school students in the county in May of the school year.
However, only the total reading scores were relevant to this study.
The kindergarteners were administered Level 10, Form E of the CAT.
The total reading score for Level 10 of the CAT is based on the vocabulary










and oral comprehension tests. The first-grade students were administered
Level 11, Form E of the CAT and the second-grade students were
administered Level 12, Form E of the CAT. The total reading score for
Levels 11 and 12 of the CAT are based on the vocabulary and comprehension
(of written material) tests.
The vocabulary test in the CAT contains 30 items that measure the
ability to identify a word that is associated with an orally presented
category or definition. The test also includes items that measure a
student's ability to identify words with similar meanings and to identify
words in context. The comprehension test incorporates three types of
comprehension: literal, inferential, and critical. The comprehension test for
Level 10 contains 30 items for which the students identify the picture that
shows the meaning of sentences or stories. The Level 11 and 12
comprehension tests each contain 30 items. Some of the items ask the
student to identify a picture that shows the meaning of a written sentence
and others ask the student to identify details and ideas in written passages.
Scoring. The students' total reading score was reported, for purposes
of this study, in scale scores. Scale scores are units of a single,
equal-interval scale that are applied across all levels of the CAT Form E,
regardless of grade or time of year of testing. These scores are expressed
in numbers that range from 0 through 999. They are used primarily to
provide a basis for deriving normative scores to describe test performance.
The scale scores are obtained by converting the "number-correct" scores to
scale scores using a conversion table published in the Norms Books for the










CAT. Scale scores for one content area cannot be compared with the scale
scores for another area.
The current scores for students were obtained by the researcher from
the county school board offices. The county staff also provided the
researcher with the students' identification information, teacher
identification, and the total reading scale score from the previous school
year for students in grades 1 and 2.
Validity. Content validity is used to decide whether or not a test
actually measures what it is supposed to measure. Content validity
depends to a large extent on whether or not the items in a particular test
accurately represent the subject matter that the test was designed to
cover. Therefore, the content validity can be checked by comparing the
content descriptions and the test items to particular curricular objectives.
Test developers consulted curriculum guides and major textbook series in
designing the test. Although the test reflects standardization procedures
that attend to the general population rather than to the local situation per
se, the local district has selected the CAT, presumably on the basis of
adequate content validity.
Reliability. Reliability refers to the consistency of test results. A
frequently used measure of internal consistency, the Kuder-Richardson
formula 20 (KR 20), was applied to the CAT Form E. Based on a single
administration of a test, this formula provides a reliability estimate that
equals the average of all split-half coefficients that would be obtained on
all possible divisions of the test into halves. The technical reports for the










CAT indicated KR 20 scores of .82 for Level 10, .93 for Level 11, and .95 for
Level 12.
Administration. The two subtests of the CAT relevant to the present
study were administered during the first two days of a week-long test
period. The vocabulary and comprehension tests, from which the total
reading score was derived, were administered as two different subtests.
Before the formal testing procedures began, a practice test was
administered so that children could become familiar with the testing
procedure. For kindergarteners, the goal of the practice test was to
familiarize them with listening to directions and marking their papers
correctly. For the first- and second-grade students the goal was to
familiarize them with the mechanics of marking their answers
appropriately.
The tests were administered to the entire class at once. The
classroom teacher read aloud the general directions for the vocabulary test,
and then administered two sample questions. The students were encouraged
to ask questions during this time. The teacher and a proctor were available
in each classroom for supervision and assistance. When the students
signalled that they were ready to proceed, the teacher read the question
items and the students responded on their answer forms.
The comprehension test was administered in much the same way.
However, the kindergarten students were not required to read. Each item
was based on a sentence or passage read aloud by the teacher. The students
listened, then chose an answer from three pictured answer choices. On the










first-grade test, students were asked to read a short sentence and choose
from three pictured answer choices. The second-grade students read the
questions and chose their answers independently.


Research Design

The purpose of this study was to investigate (a) the relationship
between students' perceptions of the learning environment and student
achievement and (b) the effectiveness of a training program designed to
change students' perceptions of the learning environment. The first
question was addressed using data obtained from the participants and their
students. To answer the second question, participating teachers were
divided into two groups. The treatment group received training to help them
change students' perceptions of the learning environment. Scores of
students in treatment teachers' classrooms were compared to those of
students in control group teachers' classrooms on a measure of students'
perceptions of the learning environment. Thus, the research design selected
for this study was a pretest-posttest control group design (Campbell &
Stanley, 1963).
The pretest-posttest control group design allowed the researcher to
test the effect of an experimental stimulus on a dependent variable through
the use of protesting and posttesting students of teachers who were
randomly assigned to treatment or control conditions. A pretest-posttest










control group design with the control group later receiving a delayed
treatment was used, as shown below:


Treatment group pretest treatment posttest
Control group pretest posttest (treatment)

With this design, most potential threats to internal validity are
controlled through the use of random assignment to the two groups.
However, demoralization may have occurred if the control group teachers
felt abandoned because of the late date when their training began.
As true with any study involving a pretest, a potential threat to
external validity is the possibility of an interaction between the pretest
and the experimental treatment (Huck, Cormier, & Bounds, 1974). However,
in the present study, the pretest was considered to be a part of the
experimental treatment and was expected to have an effect on the results.
Consequently, the issue of the generalizability of the results of the
experimental group to groups not having had the pretest was moot. A more
serious threat to external validity may have been the use of all volunteers
in both treatment and control groups.


Treatment

The Treatment Group
The teachers in the treatment group attended an introductory meeting
designed to familiarize them with the concept of learning environment and
its relationship to student achievement. Teachers also learned how










important their role as the classroom teacher could be in facilitating
greater achievement. In addition, the teachers were informed about their
role in the training program, what the program entailed, and how it was to
be conducted.
After careful consideration of the recommendations for effective staff
development programs, the researcher designed a training program for the
present study that followed a five-step procedure including assessment,
feedback, reflection and discussion, intervention, and reassessment.
Both the Actual and Preferred Forms of the MCI were used in assessing
perceptions of the learning environment. Both students and teachers
completed the Actual Form of the MCI, which tests for perceptions of five
dimensions of the classroom learning environment. The teachers also
completed the Preferred Form of the MCI. The five scale scores reported
correspond to the five dimensions of the learning environment incorporated
in the inventory. The researcher presented the teachers with their own
scores on the My Class Inventory, as well as their classes' scores. The
results were presented in graph form at a subsequent meeting so that the
teachers could readily see the discrepancies between these scores.
Comparison of the two scores provided information which the teachers used
to understand their students' perceptions as well as to clarify their own.
This information provided teachers with the feedback recommended for
effective staff development (Joyce & Showers, 1980 ).









Reflection and Discussion
The purpose of the second meeting of the training program was to set
the foundation for the training. A discussion was held about how to
interpret the scores and how to decide what they meant to individual
teachers and their students. The feedback was individualized to help the
teachers examine their classroom goals and begin to set personal goals.
Individual goals were the focus of intervention as recommended by Speiker
(1978). Based on assessment and feedback from the MCI, the teachers chose
two areas of concentration for the intervention. Their decisions were based
on reflection about their individual goals for their students as well as the
discrepancy between their scores and their students' scores. As a result,
three teachers chose the competitiveness and friction and cohesiveness
scales, and five teachers chose satisfaction and friction and cohesiveness.
Intervention
After providing feedback at the second meeting, the researcher
presented a lecture on the importance of promoting cohesiveness and
minimizing friction and competitiveness in the learning environment. The
researcher stressed the relationship of these variables to student
achievement. The teachers also received a handbook to use as a guide in
focusing their efforts to change students' perceptions of the learning
environment.
The five dimensions tested by the MCI were incorporated into the
handbook. However, the handbook contains only four sections because
activities for friction and cohesiveness were so similar it seemed prudent
to combine the two scales into one section. The organization of the










handbook by dimensions was motivated by Palmer (1978), who recommended
that material be broken up into segments that could be assimilated more
easily. After reviewing five experiments, Mohlman et al. (1982) found that
teachers were more likely to adopt new teaching practices if they were
convinced that the time and effort spent in incorporating the changes into
their daily classroom procedures would improve student achievement. As a
result, each section contained a brief narrative about how that dimension
related to student achievement as well as a rationale for the need for
continual improvement of that dimension. Doyle and Ponder (1977) stressed
the importance of practical activities if implementation was to be
expected. Therefore, in each of the sections, following the text, a number
of practical activities were listed to facilitate change in that dimension of
the learning environment. The activities were designed to provide choices
so the teachers could match activities to their individual styles of teaching.
Congruence between ideas presented and teaching style was encouraged by
Doyle and Ponder (1977) as a way of facilitating adoption of ideas.
The documentation of planning, implementation, and outcomes of
activities was stressed by Howey and Vaughan (1983). Furthermore, Oliver
(1980) believed that documentation helped teachers make better use of
research findings on teacher effectiveness. Therefore, teachers were given
planning sheets which included questions to stimulate guided reflection of
the implementation of the plans. The use of the planning sheets was
encouraged throughout the intervention so that teachers could reflect on
their actions to determine their effectiveness.










The handbook contained the majority of the information to be shared
with the teachers. Training sessions reinforced the information through
modeling implementation of the activities, afforded practice in analyzing
the potential benefits and problems with implementing the activities, and
encouraged incorporation of general strategies into specific classroom
teaching behaviors.
These goals were met through a variety of teaching techniques. The
trainer modeled the implementation of many of the activities in the
handbook throughout the training sessions. After participation in several of
the lessons, the teachers were involved in large-group discussions about the
lessons' potential effects on students. Because the teachers had
experienced the lesson from a student's point of view, they had a basis upon
which to evaluate the students' reactions.
The nature of several of the activities indirectly encouraged peer
interaction. Small-group exercises were employed to encourage peer
interaction and several other small-group activities were designed to
stimulate teachers to incorporate the general strategies and learning
activities into individual classroom teaching behaviors.
Specifically, the third meeting of the training program was held two
weeks after the second meeting, and lasted approximately one and one-half
hours. The purpose of this meeting was to facilitate implementation of the
activities by actively involving the teachers in small group discussions to
determine what factors enhanced their success or caused their failures
during implemention. Large-group discussion encouraged teachers to










examine the common causes of success and failure so they might decide
what future steps to take to successfully implement the activities. The
trainer led several activities involving teacher participation in activities
described in their handbook. In this manner the teachers had a model, as
well as a chance, to experience the benefits from the exercises.
The fourth meeting for the treatment group was held eight weeks later,
the week after school resumed following winter break. The longer time
between meetings allowed teachers to enjoy their usual holiday activities
without feeling pressured to incorporate new skills and activities into their
programs at a time of year generally filled with extracurricular activities.
The purpose of the fourth meeting was to enable teachers, through
small group interaction, to share their experiences with one another,
provide support for one another, and explore two new dimensions of the
learning environment--difficulty and satisfaction. Information about the
two new dimensions of the learning environment was shared through
lecture, and the trainer modeled the implementation of several of the
activities and strategies presented in the handbook. The trainer also led an
activity encouraging the teachers to consider how they might apply the
general strategies listed in the handbook to their specific classroom. The
teachers were instructed to concentrate their efforts on implementing the
activities associated with one of the two new dimensions.
The fifth meeting, held two weeks following the fourth meeting,
focused on the last two dimensions. Small-group activities designed to
enable teachers, among their peers, to practice implementing some of the










ideas in the handbook served to help meet this goal. Active participation in
the exercises facilitated their implementation by providing opportunity for
the teachers to implement some of the ideas immediately with materials
that related to their individual current curricula.

Reassessment
The Actual Form of the MCI was readministered at the end of the
intervention. Before the conclusion of the fifth meeting the teachers signed
up for a posttest date. The posttesting began three weeks later and lasted
for two weeks. The results of the posttest were sent to the teachers in the
form of a graph depicting changes from their students' initial class mean
scores on the MCI to their present scores. With a graphic representation of
the scores, changes could be easily identified by the teachers. This
information allowed teachers to evaluate the success of their efforts during
the training. The need for evaluation was repeated throughout the literature
on staff development (Rubin, 1978).

The Control Group
The teachers in the control group attended the same introductory
meeting as the treatment group teachers. After completing necessary
paperwork and reviewing the procedures for the training, their meeting was
devoted to an overview of the learning environment and its relationship to
student achievement. The information provided to the teachers was general
in nature to avoid contamination of their students' pretest scores.
The control group teachers were notified that after their students were
pretested, their next contact with the researcher would be sixteen weeks










later to schedule their classes' posttest. Furthermore, the teachers in this
group did not see the test that was given to their students, so as not to
influence their actions before their students took the posttest. By contrast,
the treatment group took the inventory when their class was pretested.
They also interacted with the researcher during training sessions before the
posttesting.
Posttesting the control group served two purposes. Although the
teachers in the treatment group were finished with their training at this
point, the control teachers were just beginning theirs. The scores served as
a pretest for these teachers to use in their training sessions, and, for
research purposes, the scores served as control group posttest measures.
The control group met the following week to discuss the dimensions of
the learning environment, and then began the staff development program
following the same training program as the treatment group. Because no
holidays occurred during implementation, the time span of the training
program was shorter for the control group than for the treatment group.
Therefore, the last meeting of the control group took place twelve weeks
following the administration of the first posttest, and concluded two
weeks after a second posttest was administered. Both the control and
treatment groups were present together at this last meeting.










Data Collection Procedures
Teachers
Data were collected by issuing a packet to the eight teachers in the
experimental groups upon completion of the protesting of their students.
The packets included a direction sheet, two forms of the MCI (Actual and
Preferred), and a return envelope. All completed inventories were returned
to the researcher by a date specified on the direction sheet. The teachers in
the control group did not receive a packet. They were informed that they
would receive the inventory when their students were tested again eighteen
weeks later.

Students

Kindergarten classrooms. The researcher visited the first and second
grade classrooms between 9:00 a.m. and 1:00 p.m., depending upon when the
teacher scheduled the testing. Groups of four children were called to meet
with the researcher in an area of the classroom that was set aside from the
rest of the class so as not to interfere with the teacher and other classroom
activities. After an introduction, the researcher began to read the items
one at a time. After each statement was read, the researcher asked the
children individually for their responses and recorded them on answer
sheets. The children were encouraged to think independently and were
verbally reinforced for doing so. The composition of each group was
recorded so that the groups could be replicated as closely as possible during
the posttesting. A total of 177 kindergarten children were tested in this
manner.










Primary classrooms. The researcher visited the first- and second-
grade classrooms between 9:00 a.m. and 1:00 p.m., depending upon when the
teacher chose to schedule the testing. Groups of four children were called
to meet with the researcher in an area of the classroom that was set aside
from the rest of the class so as not to interfere with the teacher and other
classroom activities. After an introduction, the researcher distributed the
inventory and a pencil appropriate for computer scanning. The researcher
then read the directions to the children. When the children understood the
directions, each item on the inventory was read aloud and the children
recorded their responses. The researcher was in close proximity to insure
correct procedure for responding and to encourage independent responses.
The students were reassured that their answers would not be shared with
anyone. A total of 123 children were tested. This procedure was repeated
in February for both groups, and again in May.
The data collected during the administration of the short form of the
MCI to students served two purposes in the present study. First, students'
perceptions as measured by the MCI served as the dependent variable in the
study. Second, the data were provided as feedback in the teachers' training
workshops. Only those students who took both the September pretest and
the February posttest were included in the data analysis.

Data Analysis

To investigate the relationship between students' perceptions of the
learning environment and student achievement, the researcher applied
Pearson correlation procedures (Agresti & Agresti, 1979) between the










students' posttest scores and their achievement scores. In addition,
multiple regression (Pedhazur, 1973) was used to test for significance of
these relationships when controlling for previous levels of student
achievement.
To determine whether the teacher training was effective, descriptive
data were examined. Further investigation included an analysis of
covariance (Pedhazur, 1973) in which the postMCI (MCI administered at the
end of the treatment) score was the dependent variable. The two
independent variables were treatment group and teacher (teachers were
nested within treatments). For kindergarteners the covariate was the
preMCI, and for the primary-grades, the covariates were the preMCI and
preAchievement. Assumptions of a fixed independent variable, linear
relationships, homoscedasticity, and normal distributions of error were
made. Because the multiple regression and covariance analyses were
repeated for each of the five subscales on each sample, the alpha levels for
the tests of significance were set at .01.















CHAPTER IV
RESULTS AND CONCLUSIONS


This study was designed to investigate the effectiveness of a staff
development program that trained teachers to facilitate change in students'
perceptions of the learning environment. The first purpose of the
investigation was to determine whether students' perceptions of the
classroom learning environment were related to student achievement. The
students' posttest MCI scores and achievement scores were studied to
investigate this relationship. The researcher used the data from
kindergarten and primary classes to apply Pearson correlation procedures
and a regression model in two separate analyses to address this issue.
The second purpose was to determine whether there was a significant
difference between treatment and control groups on posttest MCI scores. An
analysis of covariance was used, controlling for initial level of achievement
and pretest MCI scores where they were applicable. After protesting all
participating teachers' classes on the MCI, the researcher provided training
in altering learning environments to teachers in the treatment group.
Teachers in the control group did not receive the training. The posttest
scores of students of both groups were used to investigate the effects of
the treatment. Using the students' posttest scores and their current










achievement scores, the relationship between students' perceptions of the

learning environment and achievement was investigated.


Relationship between MCI Scores and Student Achievement

To test for the relationship between students' perceptions of the
learning environment (using posttest MCI scores) and student achievement,
Pearson product correlations were computed for three groupings: 164
kindergarten students, 82 primary students, and the combined group of 246
kindergarten and primary students. As indicated in Table 4-1, the

kindergarten students' reading achievement scores were negatively
correlated with friction, competitiveness, and difficulty, while satisfaction

and cohesiveness were positively correlated. Only two of these
correlations were statistically significant. The negative correlation
between achievement and perceived difficulty of work was significantly
greater than zero (p< .01). In addition, the students' perceptions of the
amount of friction between classmates was significantly associated with
their achievement scores (p< .05) in a negative direction.


Table 4-1
Pearson Correlation Coefficients Between Post MCI Scores and
Achievement for Kindergarteners (N=164)

sat fric comp diff coh

Ach 0.092 -0.198 -0.052 -0.289 0.037
0.257 0.014* 0.518 0.000** 0.644
< .05
** 1< .01












The analysis for the primary group revealed only one statitistically

significant relationship between year-end achievement and MCI scale

scores. As indicated in Table 4-2, a significant negative correlation was

found between satisfaction and student achievement test scores at p< .05.


Table 4-2
Pearson Correlation Coefficients Between Post MCI Scores and Achievement
for the Primary Group (N=82)

sat fric comp diff coh

Ach -0.233 0.031 -0.067 -0.202 -0.206
0.036* 0.786 0.550 0.070 0.065

*.< .05


The correlations for the total group revealed no significant

relationships between achievement and four MCI scales (satisfaction,

friction, competitiveness, and cohesiveness). There was, however, a

statistically significant (p< .01) negative relationship between perceived

difficulty of class work and student achievement, as indicated in Table 4-3.


Table 4-3
Pearson Correlation Coefficients Between Post MCI Scores and
Achievement for the Combined Group (N=246)

sat fric comp diff coh

Ach -0.030 -0.109 -0.064 -0.264 -0.048
0.647 0.095 0.327 0.000* 0.459

1< .01










To further investigate the relationship between environment and
achievement, a regression model was used to test whether the relationship
between MCI scales and achievement could be different for different
teachers. This question was examined by testing for a significant
interaction between teacher and posttest MCI. The model was
PostAch = PostMCI + Tchr + PostMCI Tchr.
The data were analyzed for three separate groupings of students:
kindergarten students, primary students, and the combined group. One
significant interaction (PR>F=.0044 with alpha at .01) occurred for the
friction scale of the MCI for the kindergarten group. No other interactions
were significant. Thus, the failure to find relationships between
achievement and perceptions of classroom environment could not be
attributed to the fact that these relationships varied over different
teachers.

Effectiveness of Teacher Training

In the investigation of the effectiveness of the training, data were
analyzed separately for kindergarten students and primary students. From
the means and standard deviations shown in Table 4-4, the following
observations can be made: (a) satisfaction increased for both the treatment
and control groups, (b) friction decreased for the treatment group while
increasing for the control group, (c) competitiveness increased for the
treatment group and decreased for the control group, (d) difficulty
decreased more for the treatment group than it did for the control group,







53




and (e) cohesiveness increased for the treatment group and decreased for

the control group.


Table 4-4
Means and Standard Deviations on the Pretests and Posttests on the
MCI for the Treatment and Control Groups (for the Kindergarten Group)


presat
postsat

prefric
postfric

precomp
postcomp

prediff
postdiff

precoh
postcoh


treatmenta

mean

12.31
12.96

9.77
9.11

11.01
11.72

8.78
6.85

11.98
12.26


std dev

2.19
2.35

3.04
2.84

3.00
2.68

2.24
2.13

2.83
2.87


mean

12.09
13.15

10.17
11.79

12.24
11.79

9.34
7.90

12.50
11.98


std dev

2.09
2.31

2.81
2.87

2.94
2.67


2.44
2.30

2.25
3.05


aN=108 students
bN53 students


Although three of the five observed mean differences were in the predicted

direction to support inference of treatment effectiveness, the magnitude of

these observed differences was small.

To determine whether these observed differences were statistically

significant, further analysis was conducted using analysis of covariance

(ANCOVA) with a two-factor design--the two independent variables were

treatment group and teacher nested within treatment. For kindergarten










students the covariate was preMCI only. A separate analysis was conducted
for each of the five scales of the MCI using an alpha level of .01 for the test
of significance. In each analysis an initial test was made for

covariate-by-treatment interaction. Because none were found, the
assumption of homogeneous slopes for regression lines of posttest on

pretest appeared to be met.
The nested factor was considered random and was thus the appropriate
error term with which to test the significance of treatment effect. The
F-ratios and probabilities from these analyses for the kindergarten students

are presented below. In Table 4-5, PreMCI denotes the pretest scores for
the MCI, PostMCI denotes the posttest scores for the MCI, Trt denotes
treatment, and Tchr:Trt denotes the teacher-nested -within-treatment
effect. (Complete sums-of-squares tables for these analyses are presented
in Appendix A.) To summarize, there were no significant effects of the

treatment on the kindergarten classes' MCI posttest scores.


Table 4-5
Results of Analysis of Covariance For Kindergarten Students
Model Dependent variable F for treatment PR> F
PostMCl= PreMCI + Trt + Tchr:Trt sat 0.00 0.9713

fric 2.47 0.1469

comp 0.02 0.9010
diff 0.00 0.9817
coh 0.78 0.3975












Because there was no treatment effect on students' PostMCI scores, an
additional analysis was performed to determine if there was a treatment
effect on students' achievement. This analysis was conducted using
analysis of covariance with a two-factor design--the independent variables

were treatment group and teacher nested within treatment. The covariate
was PreMCI scores. The nested factor was considered random and was used
as the error term with which to test for significance. Again, no significant
effects were found for the treatment on student achievement posttest
scores. Results of this analysis are presented in Table 4-6.


Table 4-6
Results of Analysis of Covariance For Kindergarten Students'
Achievement
Model Dependent variable F for treatment PR > F
PostAch= PreMCI + Trt + Tchr:Trt sat 0.12 0.7413

fric 4.15 0.0812

comp 0.05 0.8235

diff 2.36 0.1683
coh 0.15 0.7092


The means and standard deviations for the primary students (grades
1-2) are shown in Table 4-7. From the data in Table 4-7, the following
observations can be made: (a) satisfaction decreased for both groups,
(b) friction increased for both groups, (c) competitiveness increased for the











treatment group and decreased for the control group, (d) the perceived

difficulty decreased for both groups, and (e) cohesiveness decreased

slightly for the treatment group yet decreased substantially for the control

group. Although the difficulty scores changed in the preferred direction, the

overall observed changes did not support treatment effectiveness.


Table 4-7
Means and Standard Deviations on the Pretests and Posttests on the MCI for
the Treatment and Control Groups (for the Primary Group)


treatmenta controlb

mean std dev mean std dev

presat 12.35 2.45 11.76 2.61
postsat 12.00 2.31 11.24 2.69

prefric 8.60 3.24 10.09 3.34
postfric 10.05 3.30 10.95 2.59

precomp 11.75 2.51 12.14 2.58
postcomp 11.90 2.32 11.71 2.49

prediff 9.05 2.42 8.81 2.06
postdiff 8.55 2.77 8.48 2.79

precoh 12.20 2.67 11.90 3.10
postcoh 12.15 2.90 9.90 3.19

aN=40 students
bN=42 students

Further investigation was conducted using the same analysis as with

the kindergarten group's data. The two independent variables again were

treatment group and teacher nested within treatment. For the primary-

grade students, however, PreMCI scores and previous achievement were the










covariates. A separate analysis was conducted for each of the five scales

of the MCI using an alpha level of .01 for the test of significance.

Table 4-8 contains the F-ratios and probabilities for each scale for

the primary classes. In Table 4-8, PreMCI denotes the pretest scores for

the MCI, PreAch denotes the students' scaled reading scores for the previous

year, PostMCI denotes the posttest scores for the MCI, Trt denotes

treatment and Tchr:Trt denotes the teacher nested within treatment effect.

(Complete sums-of-squares tables for these analyses are presented in

Appendix B.) There were no significant differences between the treatment

and control groups on any of the primary students' five scales of the MCI.

Therefore, no evidence suggested that the teacher training had a significant

effect on changing the primary students' perceptions of the learning

environment as measured by the MCI.


Table 4-8
Results of Analysis of Covariance For Primary Students

Model Dependent variable F for Trt PR > F

PostMCl= PreAch + PreMCI + sat 0.04 0.8503
Trt + Tchr: trt
fric 0.00 0.9733

comp 0.59 0.4844

diff 1.32 0.3152

coh 2.04 0.2268


An additional analysis was performed to determine if there was a
treatment effect on students' achievement. This analysis again was an











analysis of covariance with a two-factor design. The independent variables

were treatment group and teacher nested within treatment. The covariates

were previous level of achievement and PreMCI scores. As indicated in

Table 4-9, no significant treatment effect was found between the treatment

and students' achievement.


Table 4-9
Analysis of Covariance for Post Achievement for the Primary Grades

Model Dependent variable F for Trt PR> F

PostAch= PreAch + PreMCI + sat 0.53 0.5079
Trt + Tchr: trt
fric 0.24 0.6467

comp 0.32 0.6022

diff 0.37 0.5752

coh 0.45 0.5384












CHAPTER V
SUMMARY, DISCUSSION, AND RECOMMENDATIONS


The purpose of this study was to determine (a) whether training
teachers in classroom environment strategies could change students'
perceptions of the learning environment, and (b) whether students'
perceptions of the learning environment are related to student achievement.
The posttest results of the students' perceptions were analyzed using
analysis of covariance to establish if there was a significant difference
between students' Mv Class Inventory scores for the treatment group (those
whose teachers received training) and the control group (those whose
teachers did not receive training). The students' perceptions of the learning
environment were collected before and after teacher training. Year-end
mean reading achievement scores for the students in the classes of
treatment and control teachers were also compared. Finally, posttest MCI
scores were also analyzed to determine their relationship to the students'
achievement scores.
In the present study a five-step training procedure was adapted from
previous studies by expanding its implementation from a one-to-one
training program to training that was delivered to larger groups. Staff
development research literature provided a basis for development of the
materials and structure of the training sessions.








No significant differences were evident between the treatment and
control group students' scores on the posttest of any of the five subscales
of the MCI, after taking into account students' pretest scores and initial
achievement. This was true of kindergarten students and primary students.
As measured by the MCI, evidence did not suggest that teacher training had a
significant effect on changing the students' perceptions of their learning
environment. Similarly, no evidence was found to indicate that the teacher
training had any impact on student reading achievement test scores.
The relationship between students' perceptions of the learning
environment and student achievement was also examined. Tests of
significance for Pearson correlation coefficients between students'
perceptions of their learning environment and student achievement revealed
a statistically significant negative correlation (p< .01) between
kindergarten students' perceptions of the difficulty of their work and
student achievement. The amount of friction they perceived to be in their
class was also significantly (p< .05), and negatively related to student
achievement. An additional significant negative correlation (p< .05) was
found between perceived satisfaction and student achievement for the
primary-grade students. However, most of the MCI subscales were
unrelated to student achievement in reading, even when prior level of
reading ability was controlled.


Discussion and Implications

Previous studies report success with using the five-step procedure for
training teachers to change students' perceptions of the learning








environment (Fraser, 1986; Anderson & Walberg, 1974; Fraser & O'Brien,
1985). This study does not. Explanations of these discrepant findings
require an examination of how the present study differs from the previous
ones. To begin, one of the differences is in the grade levels of the students.
The students in the present study were kindergarten and primary-grade
students rather than the upper elementary or older students in previous
studies ( Fraser & Fisher, 1985; Fraser et al., 1982). Because the reliability
of the MCI was not as well established for use with primary age students, it
is possible that there was a change in students' perceptions of the learning
environment, but the instrument did not measure it. Second, there was a
difference in the student populations of the classes involved in the studies.
Because the make-up of the classes may affect the teachers' ability to
change students' perceptions of the learning environment, there is a
possibility that the training did lead to changes in teacher behavior, but the
nature of the student group meant that there was little change in students'
perceptions. Third, deviations in the training procedures of the current
study from the design of the previous studies may have decreased the
training effectiveness in the present study. Perhaps these changes meant
that the teachers were not trained well enough to facilitate change in
students' perceptions of the learning environment. Last, the training was
extended to a group rather than to individual teachers. Since there is less
accountability in a group situation, we must examine the possibility that
the training had the potential to lead to change in teacher behavior but the
teachers did not follow through on the recommendations presented in the
training.








In the following discussion these differences between similar previous
studies and the present study will be addressed. One concern is there may
have been a change in students' perceptions of the learning environment, but
the instrument did not measure it. It is possible that younger students'
perceptions are less stable than those of upper elementary students. It may
be that primary-grade students' responses are not as reliable as those of
older students because they are not as socially, emotionally, and cognitively
mature. This was informally observed with regard to the age differences
even within the students participating in the present study. For instance, a
larger number of statements on the inventory needed to be restated for the
kindergarten students than for first- and second-grade students. In
addition, five questions on the inventory are worded such that an
affirmative reaction to the statement called for a negative response to the
question. It was obvious that many more of the younger children were
confused by these statements, although the researcher tried to be
consistent with the rewording and explanation of the questions for all of
the students tested. Furthermore, because younger children are more
concrete than their older counterparts, the younger students' responses may
have reflected their perceptions for the day of the testing only. Because of
their less mature social, emotional, and cognitive development, it may be
inappropriate to expect the primary students to generalize their responses
beyond an immediate time frame.
Another possible influence on the difference in reliability of the
instrument between primary grade students and upper elementary students
was in the nature of the administration of the inventory itself. Since the








kindergarten students' responses needed to be recorded by the
administrator, they were tested in small groups to save time in
administration and to insure an accurate record of their perceptions.
Despite the fact that independent responses were encouraged throughout the
testing process, it was possible that students responded to an item strictly
because the peer answering prior to them gave that same response.
Although the first and second grade students had individual answer sheets,
they were still tested in small groups where they could view their peers'
responses. The upper elementary grade students were administered the test
while seated in their usual seating arrangement with little influence from
their peers. However, the treatment had no effect on achievement. If the
treatment had been effective, previous studies indicate there would have
been some impact on achievement. Although there are some questions about
use of the instrument with young children, the data seems to suggest other
explanations for the lack of significant findings.
When the present study was designed to train a group of teachers
rather than an individual teacher as in previous studies, it became
necessary to solicit volunteers. Consequently, there was no control over the
number of participants in the study, the age of the students, or the class
population. Further, when the size of the volunteer group is small,
statistical significance is more difficult to establish regardless of what is
studied. The low number of volunteers leads one to question why so few
teachers were interested in the learning environment of their classrooms.
Because there was no control over the age of the students or the class
populations, all of the subjects' classes were kindergarten and primary








grades, leaving no representation from the upper elementary grades. In
addition, four of the seven classes in the control group and one of the
treatment groups were remedial groups. Teachers of remedial classes
taught approximately half as many students as teachers of regular
heterogeneous classes, and remedial groups were more homogeneous in
ability than student groups in previous studies. Given the lower number of
students per class and the lower abilities of the majority of the control
group classes, it is possible that inclusion of these classes in the study may
have influenced the results of the study. It seems possible that the age of
the students, the socio-economic make-up of the class, and the academic
grouping of the students within the classes may affect teachers' ability to
change students' perceptions of the learning environment. One wonders
whether there may be some contexts in which it is more difficult to change
students' perceptions of the learning environment than others.
Several elements of the design offer possible explanations for the
minimal effectiveness of the training in the present study. Some of the
elements of the training were altered from previous studies, while others
remained the same. These deviations may have limited the teachers'
abilities to make changes in their learning environments.
The changes in the design included extending the time span for the
training and limiting the trainer interaction with the participants. The
longer span of the present study was designed to allow time for the
teachers to digest what was learned at the sessions, implement the few
activities they saw modeled, and try some other activities as they
corresponded with their class' schedule and ability levels. However, it may








be that the time span left too much time between sessions, allowing time
for teachers to procrastinate and simply forget about the concentration on
the content of the training sessions. Alternately, it is possible that the
study would have yielded different results if the posttests were performed
after a longer interval. It may be that the effects of the teachers' behavior
and implementation of new ideas and activities would take longer to be felt.
If a longer time span was needed for changes to take place, the changes
would not have been realized in the present analysis.
One of the elements of this study's training replicated from the
previous studies (Fraser, 1983; Fraser et al., 1982) was the emphasis on
one or two of the individual scales of the MCI. The teachers were directed
to concentrate their efforts on implementing activities toward changing one
or two specific scales. None of the teachers chose to concentrate their
efforts on the difficulty scale. Since perceived difficulty of a students'
work was significantly correlated with kindergarten students' acheivement,
change may have been more likely if teachers had concentrated their efforts
in this area. However, since there was no concentration in this area which
affects the curriculum itself, it may have reduced the probability of
achieving a significant relationship between treatment and student
achievement. The researcher included the choice of areas of concentration
because it was part of the design of previous studies reporting success
(Fraser, 1986; Anderson & Walberg, 1974; Fraser & O'Brien, 1985), and
because staff development research encourages trainers to allow teachers
to choose their own goals (Speiker, 1978). It is possible, however, that it







66

may have been detrimental to require the teachers to focus on one or two

areas on the MCI.

It also appears that is might be beneficial to eliminate choice from the

design of the workshop when considering the relationships among the MCI

scales for the students' posttests. Investigation of the interrelatedness of

the scales, using Pearson correlation procedures, revealed significant

relationships between many of the scales. Table 5-1 shows those

relationships.


Table 5-1
Relationship between MCI Scales for Posttest Scores

PostSat PostFric PostComp PostDiff PostCoh

PostSat 1.000 -0.398 -0.120 -0.226 0.551
0.000 0.000* 0.059 0.000* 0.000*

PostFric 1.000 0.266 0.212 -0.372
0.000 0.000* 0.001* 0.000*

PostComp 1.000 0.096 -0.116
0.000 0.131 0.067
PostDiff 1.000 -0.124
0.000 0.051
PostCoh 1.000
0.000
*significant at p<.01

Because of the interrelatedness among the scales on the MCI, perhaps

there is no reason to single out any one or two areas for concentration.

Perhaps study of the learning environment is best approached in a more

wholistic way, where no one area warrants more or less attention.








Another factor that may have been affected by design changes was
the motivation for teachers to implement the training. In the previous
studies, the participants worked individually with the researcher. It
seems reasonable to assume that this type of intimate interaction would
encourage the teachers to be more prepared for discussion and to
implement more activities than they did in the present study. It is
possible that because the small group setting provided less intimate
interaction with the trainer, the teachers in the present study were less
motivated than those in previous studies to use the activities presented
in the training sessions.
Teachers in the present study reported only moderate effort in
implementing the activities from the training. Therefore, it seems
possible that the treatment was never fully implemented. The use of
volunteers generally causes some concern when interpreting the findings
from a study because the motives of the volunteers are unknown. As is
true in many staff development workshop settings, some participants may
be motivated by the opportunity to acquire inservice points for
recertification; some may not be motivated because of a lack of specific
interest in the designated topic.
Although some teachers may have been especially interested in the
topic and enthusiastic about their implementation, additional contextual
factors may have influenced implementation. First, the curriculum in the
county where the study was conducted is relatively stringent in its
expectations of teachers and leaves little room for flexibility. Therefore,
the teachers hesitated to spend time on anything extraneous, and reported








feeling frustrated when trying to find time to plan and implement new
activities. Having seen several activities modeled in the training
sessions, the teachers could have implemented those activities easily.
However, the training sessions were meant to be facilitating and
motivating rather than an end in and of themselves.
Another factor influencing implementation is the time of the year in
which the training took place. Much of the training took place during the
fall when there are many holidays. Kindergarten and primary-grade
teachers generally have several extra activities to celebrate the holidays.
As a result, there may have been even less time for the implementation of
the activities from the training sessions.
Perhaps a stronger system of accountability was needed to encourage
the teachers to implement more of the activities. It is possible that if
there were incentives to assure implementation and continuation of
activities, then more positive results might have ensued. In the studies
after which the present study was modeled, the participating teachers
met with the researcher individually. It seems reasonable to assume that
personal interaction may have encouraged teachers to be more
accountable for the implementation of the training. Because one of the
goals of the change from individual to group training sessions was to
make the training more economically feasible, trainer participation was
kept to a minimum. Therefore, observation in classrooms was discarded
as an element of the training. The potential loss of these benefits was
projected and staff development research was consulted so as to
incorporate into the training sessions, as many incentives for








implementation as possible. However, it appears that the incentive and
responsiveness in a group setting may have been insufficient to meet the
teachers' needs. Perhaps it is the motivation of the teacher, not the
content or process of staff development that is the key factor in
teachers' ability to change students' perceptions.
In addition to motivation affecting teachers' ability to change
students' perceptions, teachers' entering perspectives may also have an
effect. In their studies attempting to change student teachers'
perspectives, defined as "a coordinated set of ideas and actions a person
uses in dealing with some problematic situation (Zeichner, Tabachnick, &
Densmore, 1987)," Ross, Ashton and Mentonelli and Zeichner et al. found
that teachers elaborate their initial perspectives and that the
perspectives may be clarified and elaborated but they do not change.
Moreover, teachers select ideas and practices that match their initial
perspectives. This suggests that the teachers in this workshop may have
implemented only those activities that were consistent with their
existing perspective. Since the learning environment is so closely related
to a teachers' perspective, change in students' perceptions of the learning
environment may not be possible unless there is change in the teachers'
perspective. Previous studies in which teachers were successful in
changing students' perceptions of the learning environment emphasized
individual development of teachers and this individualized program may
have been more likely to influence their perspectives. Future studies
should investigate ways to help teachers move beyond their entering
perspectives to allow them to change their behavior.










Because of the aforementioned factors, it seems possible that
implementation of the ideas and activities presented in the training
sessions in the present study was limited. Future researchers
investigating staff development activities, and other studies where the
outcome relies on change in teachers' behavior, should incorporate some
means for assessing the implementation of the training. The assessment
could most readily be done through observation. Although it may be
costly, the benefits of a more informed, consistent, interpretation of test
results would be worth the expense. Peer observation is another possible
means for assessing implementation. Although this method would be less
costly, it would also be less precise and consistent. Another possibility
would be to hold informal discussions with the trainer, as was done in
previous studies.
In the previous studies, motivation appeared to be less of a
consideration. Because the teachers in the previous studies worked
individually with the researcher, it may not have been the content of the
training, itself, that encouraged positive results, but simply the result
of the intense attention paid to the teacher during training. Such
attention and reassurance during individual training sessions may have
reinforced the teachers' self-confidence so that the teachers provided a
better environment simply because they they were more confident
teachers. The question is raised, then, as to how we can provide a greater
sense of motivation and reassurance in a small-group setting.






71
Recommendations
After consideration of the factors influencing the results of the
present study, the researcher has several recommendations for further
research.
1. Replicate this study with teachers of older children (where the
reliability of the MCI would be higher), and analyze effectiveness of
teacher training and the relationship between students' perceptions of the
learning environment and student achievement.
2. Follow the five-step procedure used in the teacher training program
with one kindergarten or primary-grade teacher, to determine if the
significant variable is the training program or the age of the students
responding to the MCI.
3. Replicate the study with the same-aged students, but delay
administration of the posttest several months to allow for long-term
effects.
4. Administer the MCI to a large group of kindergarten students to test
for relationships between students' perceptions of the learning environment
and student achievement with a larger, more heterogenous sample.
5. Add items and reword items of the MCI to establish reliability for
use of the MCI with kindergarten and primary-grade students.
6. Use a different evaluative measure to study the classroom learning
environment of kindergarten and primary classes before and after
implementation of the teacher training program used in the present study.
7. Replicate the study with the same-aged students but include some
sort of accountability to assess implementation of the ideas and activities
in the training.








8. Replicate the study with the same-aged students using the five-step
procedure. However, vary the requirements for areas of concentration on
the MCI scales (ranging from having the subjects concentrate on all of the
scales to having them concentrate on either difficulty or friction).
9. It seems important to note that any treatment effects designed to
affect class mean performance require fairly large samples of teachers for
observed differences to be statistically significant. This fact suggests that
the study should be replicated with larger samples.


Summary
In summary, the purpose of this study was to determine (a) whether
training teachers in classroom environment strategies could change
students' perceptions of the learning environment, and (b) whether students'
perceptions of the learning environment have a relationship to student
achievement. Although observed means indicated there was a change in
students' perceptions of the learning environment as a result of the
treatment, there were no significant differences between the treatment and
control teachers' students' scores on the posttest for any of the five
subscales of the MCI. There are several possible explanations for the lack
of treatment effect in the present study. Because the reliability of the MCI
was not as well established for use with primary-age students, it is
possible that there was a change in students' perceptions of the learning
environment but the instrument did not measure it. In addition, the
differences in student population between the present and previous studies
may have been such that the training did lead to changes in teacher behavior,








but the nature of the student group meant that there was little change in
students' perceptions. Moreover, deviations in the procedures of the current
study from the previous studies may have meant that the teachers were not
trained as well to facilitate change in students' perceptions of the learning
environment. Finally, since the training was extended from individualized
to group training where there is less accountability, it is possible that the
training had the potential to lead to change in teacher behavior but the
teachers did not follow through on the recommendations presented in the
training. There are several recommendations for future research that are
related to these potential explanations.
Investigation of the relationship between students' perceptions of the
learning environment and student reading achievement yielded three
significant correlations. However, more sensitive analysis failed to
establish a statistically significant relationship between students'
perceptions of the learning environment and student reading achievement.
Because significant correlations did exist between some of the scales on
the MCI and students' achievement, further investigation of this relationship
is encouraged.


















APPENDIX A

SUMS OF SQUARES TABLES FOR ANALYSIS OF COVARIANCE
FOR KINDERGARTENERS


MODEL PostMCl scale= PreMCl scale

scale source

sat presat

trt

tchr: trt

residual


scale source

fric prefric

trt

tchr: trt

residual


scale

comp


source

precomp

trt

tchr: trt

residual


+ Trt

df

1

1

7

151


df

1

1

7

151


df

1

1

7

151


+ Tchr: trt

type III SS

2.781

1.575

93.442

760.139


type III SS

74.192

94.990

160.369

1060.047


type III SS

101.506

1.530

199.850

859.679


F value


PR>F


0.12


0.7413


4.15


0.0812


0.05


0.8235













dicale
diff


calh
coh


source

prediff

trt

tchr: trt

residual


source

precoh

trt

tchr: trt

residual


2.36


type III SS

9.464

23.608

69.987

671.592


tpe III SS

85.154

1.289

59.777

1178.645


0.15


PR>F



0.1683







PR>F



0.7092



















APPENDIX B

SUMS OF SQUARES TABLES FOR ANALYSIS OF COVARIANCE
FOR PRIMARY-GRADE STUDENTS

MODEL PostMCI scale= PreAch + PreMCI + Trt + Tchr: trt


se source df type III SS F value PR>F

sat preAch 1 73343.640

presat 1 26147.985

trt 1 20091.992 0.53 0.5079

tchr (trt) 4 152347.109

residual 56 371184.229


scale source df type Ill SS F value PR>F

fric preAch 1 71231.503

prefric 1 8064.131

trt 1 14279.276 0.24 0.6467

tchr (trt) 4 233255.398

residual 56 389268.083


cale source df type III SS Fvalue PR>F

comp preAch 1 78310.149

precomp 1 4.010

trt 1 18030.045 0.32 0.6022

tchr (trt) 4 225836.259

residual 56 745136.359












Asale source df e III SS F value PR>F

diff preAch 1 77416.919

prediff 1 6937.593

trt 1 21482.425 0.37 0.5752

tchr (trt) 4 231461.409

residual 56 390394.621


scale source df type III SS F value PR>F

coh preAch 1 77062.772

precoh 1 17211.205

trt 1 18011.147 0.45 0.5384

tchr (trt) 4 159502.422

residual 56 380121.009















APPENDIX C

SUMS OF SQUARES TABLES FOR ANALYSIS OF COVARIANCE
FOR KINDERGARTENERS' ACHIEVEMENT

MODEL PostAch = PreMCI + Trt + Tchr: trt


type III SS

623.723

41001.215

225647.767

603527.586


type III SS

3533.461

37783.101

226536.424

600617.848


tvpe III SS

5339.125

32013.292

226433.836

598812.184


F value


PR>F


0.2966


F value



1.17


PR>F



0.3157







PR>F


source

presat

trt

tchr: trt

residual


source

prefric

trt

tchr: trt

residual


source

precomp

trt

tchr: trt

residual


0.3530


Wsale

sat


cale

fric


0.99


comp











l source df type III SS F value PR>F

diff prediff 1 40494.302

trt 1 25389.406 0.81 0.3968

tchr: trt 7 225647.767

residual 144 563657.007


scale source df ye III SS F value P>F

coh precoh 1 628.267

trt 1 39068.454 1.39 0.2768

tchr: trt 7 196639.062

residual 144 603523.042


















APPENDIX D

SUMS OF SQUARES TABLES FOR ANALYSIS OF COVARIANCE
FOR PRIMARY-GRADE STUDENTS' ACHIEVEMENT


MODEL PostAch= PreAch

scale source

sat preAch

presat

trt

tchr: trt

residual


scale

fric


+ PreMCl scale +

df

1

1

1

6

56


source

preAch

prefric

trt

tchr: trt

residual


source

preAch

precomp

trt

tchr: trt

residual


scale
comp


Trt + Tchr: trt

type III SS

73343.640

26147.985

1723.434

153317.730

371184.229


tvpe III SS

71231.503

26147.985

1723.434

153317.730

389268.083


type III SS

78310.150

4.010

596.139

234261.897

397328.204


0.07


0.8038


0.07


0.8038


F value


PR>F


0.02


0.9057










81


sall source df type I SS F value PR>F

diff preAch 1 77416.919

prediff 1 6937.593

trt 1 1205.986 0.03 0.8676

tchr: trt 6 239008.552

residual 56 390394.621



se source df type III SS F value PR>F

coh preAch 1 77062.772

precoh 1 17211.205

trt 1 83.961 0.00 0.9585

tchr: trt 6 170825.500

residual 56 380121.009















REFERENCES


Agresti, A. & Agresti, B. F. (1979). Statistical methods for the social
sciences. San Francisco, CA: Dellen.
Anderson, L., Evertson, C., & Brophy, J. (1979). An experimental study of
effective teaching in first-grade reading groups. The Elementary School
JournaL Z9(4), 193-223.
Anderson, G. J., & Walberg, H. J. (1974). Learning environments. In H.J.
Walberg (Ed.), Evaluating educational performance: A sourcebook of
methods. instruments, and examples (pp. 81-98). Berkeley, CA:
McCutchan.
Ary, D., Jacobs, L. C., & Razavieh, A. (1979). Introduction to research in
education. New York: Holt, Rinehart and Winston.
Borg, W. R. & Gall, M. D. (1979). Educational research: An introduction. New
York: Longman.
Boulanger, F. D. (1980). Relationship of an inservice program to student
learning: Naturalistic documentation. Science Education, 64, 349-355.
Bracht, G. H., & Glass, G. V. (1968). The external validity of experiments.
American Educational Research Journal, 5, 437-474.
California Achievement Tests Forms E and F. (1986). Monterey, CA:
CTB/McGraw-Hill.
Campbell, D. T., & Stanley, J. C. (1963). Experimental and
quasi-experimental designs for research on teaching. In N. L. Gage (Ed.),
Handbook of research on teaching (171-246). Skokie, IL: Rand McNally.
Combs, A. W. (1982). Affective education: Or none at all. Educational
Leadership. 39(3), 494-497.
Doyle, W., & Ponder, G. A. (1977). The practicality ethic in teacher
decision making. Interchange. 8(3), 1-12.
Dreikurs, R. (1964). Children: The challenge. Dulton, NY: Hawthorn.











Ellett, C. D., Masters, J. A., & Pool, J. E. (1978). The incremental validity and
student perceptions of school characteristics. Paper presented at
Annual Meeting of Georgia Educational Research Association, Atlanta.
Farley, J. (1982). Raising student achievement though the affective
domain. Educational Leadership. 9(7), 502-503.
Fisher, D. L., & Fraser, B. J. (1981). Validity and use of my class
inventory, Science Education. 6, 145-156.
Fraser, B. J. (1979). Assessment of learning environment in
elementary-school classroom. The Elementary School Journal.
Z9(5), 297-300.
Fraser, B. J. (1982). Development of short forms of several classroom
environment scales. Journal of Educational Measurement. 19, 221-227.
Fraser, B. J. (1986). Classroom environment. London: Croom Helm.
Fraser, B. J., Anderson, G. J., & Walberg,H. J. (1982). Assessment of
learning environments: Manual for learning environment inventory (LEI)
and my class inventory (MCI) (3rd version). Perth: Western Australian
Institute of Technology.
Fraser, B. J., & Fisher, D. L. (1982a). Effects of classroom psyhosocial on
student learning, British Journal of Educational Psychology. 52.
374-377.
Fraser, B. J., & Fisher, D. L. (1982b). Predicting students' outcomes from
their perceptions of classroom psychosocial environment. American
Educational Research Journal. 12, 498-518.
Fraser, B. J., & Fisher, D. L. (1982c). Predictive validity of my class
inventory. Studies in Educational Evaluation. 8, 129-140.
Fraser, B. J., & Fisher, D. L. (1983). Development and validation of short
forms of some instruments measuring student perceptions of actual
and preferred classroom environment. Science Education. 6Z, 115-131.
Fraser, B. J., & Fisher, D. L. (1986). Using short forms of classroom climate
instruments to assess and improve psychosocial environment. Journal
of Research in Science Teaching. 23(5), 387-413.
Fraser, B. J., & O'Brien, P. (1985). Student and teacher perceptions of the
environment of elementary school classrooms. The Elementary School
Journal. 5(5), 567-580.










Fraser, B. J., Seddon, T. & Eagleson, J. (1982). Use of student perceptions in
facilitating improvement in classroom environment. Australian Journal
of Teacher Education, 2, 31-42.
Fraser, B. J., & Walberg, H. J. (1981). Psychosocial learning environment in
science classrooms: A review of research. Studies in Science Education.
8. 67-92.
Gage. N. L. (1985). Hard gains in the soft sciences. Bloomington,IN: Phi Delta
Kappa's Center on Evaluation, Development and Research.
Gage, N. L., & Coladarci, T. (1984). Effects of a minimal intervention on
teacher behavior and student achievement. American Educational
Research Journal. 21(3), 539-555.
Gage, N. L. & Giaconia, R. (1981). Teaching practices and student
achievement: Causal connections. New York University Education
Quarterly. 12(3), 2-9.
Galluzzi, E. G., Kirby, E. A. & Zucker, K. B. (1980). Students' and teachers'
perceptions of classroom environment and self- and others-concepts.
Psychological Reports. 46, 747-753.
Good, T. L. & Grouws, D. A. (1979). The Missouri mathematics effectiveness
project: An experimental study in fourth-grade classrooms. Journal of
Educational Psychology. 71(3), 355-362.
Griffin, G. A. (1982). Staff development. Paper prepared for the NIE Teaching
Synthesis Conference, Airlie House, Virginia.
Haertel, G. D., Walberg, H. J. & Haertel, E. H. (1981).
Socio-psychological environments and learning: A quantitative
synthesis. British Educational Research Journal. 7. 27-36.
Haladyna, T., & Shaughnessy, J., & Shaughnessy, J. M. (1983). A causal
analysis of attitude toward mathematics. Journal of Research in
Mathematics Education. 14, 19-29.
Howey, K. R., & Vaughan, J. C. (1983). Current patterns of staff development.
In G. Griffin (Ed.), Staff Development. The 82nd Yearbook of
the National Society for the Study of Education,Pt. 2 (pp. 92-117).
Chicago: University of Chicago Press.
Huck, S. W., Cormier, W. H., & Bounds, Jr., W. G. (1974). Reading statistics and
research. New York: Harper & Row.










Johnson, D. W., & Johnson, R. T. (1979). Cooperation, competition, and
individualization. In H.J. Walberg (Ed.), Educational environments and
effects: Evaluation. policy, and productivity (pp.101-119). Berkeley,
CA: McCutchan.
Jones, V. F., & Jones, L. S. (1981). Responsible Classroom Discipline:
Creating positive learning environments and solving problems. Boston:
Allyn and Bacon.
Joyce, B., & Showers, B. (1980). Improving inservice training: The message
of research. Educational Leadership, 3Z(5), 379-385.
Kuert, W. P. (1979). Curricular structure. In H.J. Walberg (Ed.),
Educational environments and effects: Evaluation. policy and
productivity (pp.180-199). Berkeley,CA: McCutchan.
Mohlman, G., Colardarci, T., & Gage, N. L. (1982). Comprehension and attitude
as predictors of implementation of teacher training. Journal of Teacher
Education, 33, 31-36.
Moos, R. H., & Trickett, E. J. (1974). Classroom environment scale manual.
Palo Alto, CA.: Consulting Psychologists Press.
Oliver, B. (1980). Action research for inservice training. Educational
Leadership 3Z(5), 394-395.
Pedhazur, E. J. (1973). Multiple repression in behavioral research. New York:
Holt, Rinehart & Winston.
Prawat, R. S., & Solomon, D. J. (1981). Validation of a classroom inventory
for use at the early elementary level. Educational and Psychological
Measurement. 41, 567-573.
Roberts, D. L. (1984). Ideal teaching/learning environments and effective
supervisory climate: Contradictory or comparable? Education, 10(2),
173-179.
Ross, D. D., Ashton, P., & Mentonelli, C. (1989). Developing reflective
teachers: The connections between university coursework and
elementary classrooms. Paper presented at the annual meeting of the
American Educational Research Association, San Francisco.
Rubin, L. (1978). Some postulates and principles on the continuing
professional education of teachers. In L. Rubin (Ed.), he In-Service
Education of Teachers: Trends. Processes and Prescriptions
(pp.293-303). Boston: Allyn and Bacon.










Schmuck, R. A., & Schmuck. P. A. (1983). Group processes in the classroom.
Dubuque, Iowa: Wm. C. Brown.
Speiker, C. A. (1978). Do staff development practices make a difference? In
L. Rubin (Ed.), The In-Service Education of Teachers: Trends. Processes
and Prescriptions (pp.254-261). Boston: Allyn and Bacon.
Sprinthall, N. A., & Thies-Sprinthall. L. (1983). The teacher as an adult
learner: A cognitive-developmental view. In G. Griffin (Ed.), Staff
Development. The 82nd Yearbook of the National Society for
the Study of Education, Pt. 2. (pp. 13-35). Chicago: University of Chicago
Press.
Talmage, H., & Walberg, H. J. (1978). Naturalistic decision-oriented
evaluation of a district reading program. Journal of Reading Behavior.
j0, 185-195.
Trickett, E. J., & Moos, R. H. (1973) Personal correlates of junior high and
high school classrooms. American Journal of Community Psychology, 2,
1-12.
Walberg, H. J. (1969a). The social environment as mediator of classroom
learning. Journal of Educational Psychology. 6., 443-448.
Walberg, H. J. (1969b). Predicting class learning: A multivariate approach to
the class as a social system. American Educational Research Journal. 4,
529-542.
Walberg, H. J., & Haertel, G. D. (1980). Validity and use of educational
environment assessments. Studies in Educational Evaluation, 6,
225-238.
Wood, F. H., & Thompson, S. R. (1980). Guidelines for better staff
development. Educational Leadership, Z3, 374-378.
Zeichner, K. M., Tabachnick, B. R., & Densmore, K. (1987) Individual,
institutional, and cultural influences on the development of teachers'
craft knowledge. In J. Caulderhead (Ed.), Exploring Teachers' Thinking.
(pp. 21-59). London: Cassell Education Limited.













BIOGRAPHICAL SKETCH


Renee Natalie Simmons was born on October 20,1956, in Miami Beach,
Florida, and obtained her primary and secondary education there. She
graduated from the University of Florida with a Bachelor of Arts in
education in 1978, and with a Master of Arts in education in 1980.
Renee was a classroom teacher for three years in Marion County before
becoming a university instructor at the P.K. Yonge Laboratory School in
Gainesville, Florida. She also taught parent effectiveness classes
throughout this period. After teaching at the laboratory school for three
years she became a graduate assistant at the University of Florida, where
she supervised student teachers and taught a course in classroom
management. After the birth of her first child, she returned to teaching
kindergarten part-time while continuing work on her graduate studies.
For the past two years Renee has been the educational coordinator for
two day-care centers in Gainesville, where she writes and supervises the
curriculum. She also conducts staff development workshops and teaches a
parent effectiveness class twice yearly.
Renee is married to Scott Simmons, and they have two children,
Chelsea and Cody. Her hobbies include attending Gator football and
basketball games, exercising, playing tennis, cooking, and studying
nutrition.










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 Philosophy.



Dorene D. Ross, Chair
Associate Professor of Instruction and
Curriculum
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 Philosophy.



Regina Weade, Cochair
Assistant Professor of Instruction and
Curriculum
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 Philosophy.



Linda M. Crocker
Professor of Foundations 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 Philosophy.



AthoB. Packer
Associate Professor of Instruction and
Curriculum











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 Philosophy.


Associate F
Leadership


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 Philosophy.



Linda L. Lamme
Professor of Instruction and Curriculum

This dissertation was submitted to the Graduate Faculty of the College
of Education and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.


August 1989


Dean, College of Education


Dean, Graduate School









































UNIVERSITY OF FLORIDA
III l262 08554 6777IIIII
3 1262 08554 6777