punk“.

t
. U

.wé

.Wrxr‘
(c.2- lih
x3.

”3..

. ..
a.

s. an}?
sfifi.
‘

HR“?

 

3...: .
.19..
: 5110:1121 2.1
It“ 1932.9..1...)
L: .1! 9.. filxi
:xi....rd. :.. 9 l
12:1“! lilh.§:.0‘|.itnml“
. -.,Lla.!.m.1t;l‘hm«n.. .. x.
. 25‘! . Di. 1"

 

l5

5 i193.

.. 5&3...“
. 3am

 

- I.§.xt\t:n Vb...

‘ «.16 . nix :39: Bait;
c3553.; Iii. II\X
3?“!1'
in}!!! ‘3
9:.X\~.I1)I :. 3.9. 5
$171.! Pittalaul finl’.’

.. . 1).}. Iii... . _ ‘A

a. l‘..|l 1.Hvi 5.

.3

a 3'
«ltll.
991‘
.1?

. 13......
A...

.‘S

 

.. ‘ YI V
.2. a... .f:

 

 

M I’ LIBRARY
Michigan State
Universuy A

 

 

This is to certify that the
dissertation entitled

ATTITUDES AND AFFECT:
HOW EARLY CAREER SPECIAL EDUCATION AND
GENERAL EDUCATION TEACHERS RESPOND TO
TEACHING AND THE CONTEXT OF SCHOOLS I

‘.A+_-¢-o-A-.|..O-l-A-J-l-L.--

 

presented by
Nathan D. Jones

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Education

Put; (Aw-7y

Major Prof€§sor’s Signature

7/aI/oq

Date

 

 

 

MSU is an Affinnative Action/Equal Opportunity Employer

 

 

ATTITUDES AND AFFECT:
HOW EARLY CAREER SPECIAL EDUCATION AND GENERAL EDUCATION
TEACHERS RESPOND TO TEACHING AND THE CONTEXT OF SCHOOLS
By

Nathan D. Jones

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Education

2009

ABSTRACT
ATTITUDES AND AFFECT: HOW EARLY CAREER SPECIAL EDUCATION AND
GENERAL EDUCATION TEACHERS RESPOND TO TEACHING AND THE
CONTEXT OF SCHOOLS
By
Nathan D. Jones

This dissertation is a collection of three separate but interrelated studies exploring
the experiences of early career Special education and general teachers as they encounter
their school environments. The first sub-study is an exploration of how teachers spend
their school time and their instructional time more specifically. Additionally, I investigate
the degree to which teachers’ affective responses vary depending on the activity in which
they are engaged. I find that teachers reported higher levels of concentration, challenge,
activation, and self-esteem in school as compared to outside of school. Of a variety of
work-related activities, instruction is associated with the highest levels of positive
emotions. The data also indicate that teaching assignment (i.e., one’s role and
responsibilities) is strongly associated with teachers’ time use and how they responded to
teaching.

The second study analyzes whether momentary emotional responses at school
become aggregated into attitudes about work, namely commitment and burnout. I draw
on momentary affective data and fall and Spring surveys to test whether mean levels of
positive affect, negative affect, skill, and fatigue are associated with teacher attitudes. I
find that negative affect and fatigue are both predictive of teacher burnout, even when

controlling for prior levels of burnout. Also, mean levels of positive affect and skill are

predictive of one’s commitment to their grade/subject area.

The third sub-study compares the experiences of new general education and
special education teachers in their interactions with mentors and social network members.
Drawing on survey data from novice teachers, their mentors, and their key colleagues, I
investigate differences in the two groups of teachers’ abilities to access social capital in
their social networks, and whether these differences contribute to variations in teachers’
commitment to their schools, districts, and profession, as well as their levels of burnout.
My results suggest that support from colleagues plays an important role in influencing

teachers’ career plans, particularly for special education teachers.

ACKNOWLEDGEMENTS

This dissertation was a product, in large part, Of the generous support 1 received
from several individuals throughout my doctoral career. I am incredibly fortunate to have
such exceptional dissertation committee members, Dr. Peter Youngs, Dr. Ken Frank, Dr.
Troy Mariage, and Dr. Barbara Schneider. The risk in completing a joint doctorate in
special education and education policy was always the potential to fall through the cracks
by not belonging to any one program. Instead, I had the rare opportunity to work closely
with tremendous faculty across disciplines, all of whom have been instrumental to my
development as a scholar.

The data collection for this study was funded through the Michigan Indiana Early
Career Teacher Study, a multi-year project directed by Dr. Peter Youngs and Dr. Ken
Frank. Both Dr. Youngs and Dr. Frank have been incredibly generous in extending their
research study to include my own set of research ideas and Objectives. The inclusion of
the Experience Sampling Method, which I had initially intended to be a supplement
dataset to the larger study, moved to the forefront of my dissertation in large part because
of Dr. Youngs’ intellectual guidance and support for the project. Throughout my time
working on the MlECT project, Dr. Youngs has gone out of his way to engage me
directly in the conception and implementation of his research. 1 would also like to thank
Dr. Schneider, whose devotion to my development as a researcher has been invaluable;
her continual press to pursue evidence in educational research has had a profound
influence on my research and academic goals. Finally, Dr. Mariage was a large part of
why I chose to enroll at Michigan State in the first place, and his thoughtful and

insightful approach to special education research has been an incredible resource for me.

iv

I would also like to thank the research assistants who worked with me on the
MlECT study, Mark Low, Ben Pogodzinski, and Rich Holdgreve-Resendez, as well as
Dr. Schneider’s team—in particular Venessa and Jeff Keesler and Christina Mazuka. I
can only hope they survived their time in the Office with me relatively unscathed. Lastly,
I would like to thank my family, particularly my wife Larissa. I cannot say nearly enough

about what they’ve done for me.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................... viii
LIST OF FIGURES ............................................................................................. x
CHAPTER 1
Attitudes and Affect: How Early Career Special Education and General
Education Teachers Respond to Teaching and the Context of Schools ............ 1
Study Design and Methodology ...................................................................... 2
Findings from the Three Studies ..................................................................... 5
Discussion of Findings Across the Sub-studies .............................................. 9
References ....................................................................................................... 13
CHAPTER 2
School Time Use and Affective Responses of General and Special Education
Teachers During Their First Years of Teaching .................................................. 15
Introduction .................................................................................................... l 5
Teachers’ Time Use and Affective Responses to Teaching and Other
Activities .................................................................................................... 17
The Experience Sampling Method ............................................................. 22
Theoretical Framework .............................................................................. 25
Method ............................................................................................................ 28
Study Participants ...................................................................................... 28
ESM Questionnaire .................................................................................... 30
Measures .................................................................................................... 31
Results ............................................................................................................. 32
Time Use .................................................................................................... 32
Variations in Emotions .............................................................................. 38
Discussion ....................................................................................................... 41
Appendix A ..................................................................................................... 53
Appendix B ..................................................................................................... 59
References ....................................................................................................... 61
CHAPTER 3
The Relationship Between Early Career Teachers’ Affective Responses to
Work and Their Levels of Commitment and Burnout ......................................... 68
Introduction ..................................................................................................... 68
Literature Review ....................................................................................... 71
Theoretical Framework .............................................................................. 74
Purpose ....................................................................................................... 76
Method ........................................................................................................... 78
Sample ........................................................................................................ 78
Data and Measures ..................................................................................... 79
Analytic Approach ..................................................................................... 85

vi

Results ........................................................................................................
Discussion .......................................................................................................
Limitations .................................................................................................
Implications ................................................................................................
Appendix A .....................................................................................................
Appendix B: Supplemental Tables and Figures ..............................................
References .......................................................................................................

CHAPTER 5
A Comparison Of the Influence of Social Context on the Commitment and
Burnout of Early Career Special Education and General Education Teachers....
Introduction .....................................................................................................
Literature Review .......................................................................................
Theoretical Framework ..............................................................................
Purpose .......................................................................................................
Method
Sample ........................................................................................................
Data ............................................................................................................
Analytic Approach .....................................................................................
Results .............................................................................................................
Descriptive Results ....................................................................................
Regression Results .....................................................................................
Social Network Analysis Results ...............................................................
Discussion .......................................................................................................
Limitations .................................................................................................
Implications ................................................................................................
Appendix A .....................................................................................................
Appendix B .....................................................................................................
References .......................................................................................................

vii

86
89
93
95
97
100
109

114
114
117
123
127
129
129
131
135
138
138
141
144
145
148
151
153
165
167

LIST OF TABLES

Table 1: Demographics Of the ESM Sub-Sample vs. Full Sample (in 3 Districts) ..... 53
Table 2: Teachers' Waking Hours During the Work Week ......................................... 54

Table 3: Teachers' Waking Hours During the Work Week (by Teacher Category)... 54

Table 4: Teachers’ Use of School Time ....................................................................... 55
Table 5: Teachers’ Use of School Time (by Category) ............................................... 55
Table 6: Teachers’ Use of Class Time ......................................................................... 56
Table 7: Teachers' Use of Class Time (by Category) .................................................. 57

Table 8: Teachers’ Emotional States In and Out of School (Using Context-Specific
Z-Scores) ...................................................................................................................... 58

Table 9: Teachers’ Emotional States At School (Using Context-Specific Z-Scores).. 58

Table 10.1: Predicting Teachers’ Emotional States At School Based on School
Context (Providing Instruction) ................................................................................... 59

Table 10.2: Predicting Teachers’ Emotional States At School Based on School

Context (Talking with Colleagues) .............................................................................. 59
Table 10.3: Predicting Teachers’ Emotional States At School Based on School

Context (Working Alone) ............................................................................................ 60
Table II: Demographic Characteristics of the Districts in the Sample ....................... 97

Table 12: Demographics of the ESM Sub-Sample vs. Full Sample (in 3 Districts)... 97

Table 13: Models for Predicting Teacher Bumout—Spring ........................................ 98
Table 14: Models for Predicting Teacher Commitment—Spring ................................ 99
Table 15: Description of Variables ............................................................................. 100
Table 16: Correlations for Variables Used in the Analysis ......................................... 102

Table 17.1: Models for Predicting Teacher Attitudes Based on Teachers’ Levels of
Positive Affect .............................................................................................................. 103

viii

Table 17.2: Models for Predicting Teacher Attitudes Based on Teachers’ Levels of

Negative Affect ............................................................................................................ 104
Table 17.3: Models for Predicting Teacher Attitudes Based on Teachers’ Levels of
Skill .............................................................................................................................. 105
Table 17.4: Models for Predicting Teacher Attitudes Based on Teachers’ Levels of
Fatigue .......................................................................................................................... 106
Table 18.]: Models for Predicting Teacher Burnout (with Interaction Terms) ........... 107

Table 18.2: Models for Predicting Teacher Commitment (with Interaction Terms)... 108

Table 19: Demographic Characteristics of Districts in the Sample ............................. 153
Table 20: Demographic Characteristics of Teachers in the Sample. .......................... 153
Table 21: Characteristics Of Mentoring Relationships ................................................. 154
Table 22: Models for Predicting Teacher Burnout—Spring ........................................ 157
Table 23: Models for Predicting Teacher Commitment—Spring ................................ 158

Table 24: Models for Predicting Teacher Commitment—Spring (With Interactions) 159

Table 25: Influence Models Predicting ECT Commitment ......................................... 163
Table 26: Influence Models Predicting ECT Perceptions of Fit .................................. 163
Table 27: Influence Models Predicting ECT Perceptions of Trust .............................. 164
Table 28: Influence Models Predicting ECT Perceptions of Collective

Responsibility ............................................................................................................... 1 64
Table 29: Description of Variables ............................................................................. 165

LIST OF FIGURES

Figure 1: Frequency of Interactions Between ECTs and their Mentors ....................... 155
Figure 2: Content of Interactions Between ECTs and their Mentors ........................... 155
Figure 3: Content of Interactions Between ECTs and Colleagues ............................... 156

Figure 4.1: Scatter Plot of the Association Between Perceptions of Fit and
Commitment (by Special Education Status) ................................................................ 160

Figure 4.2: Scatter Plot of the Association Between Perceptions of Relational Trust
and Commitment (by Special Education Status) ......................................................... 161

Figure 4.3: Scatter Plot of the Association Between Perceptions of Collective
Responsibility and Commitment (by Special Education Status) ................................. 162

CHAPTER 1: ATTITUDES AND AFFECT: HOW EARLY CAREER SPECIAL
EDUCATION AND GENERAL EDUCATION TEACHERS RESPOND TO
TEACHING AND THE CONTEXT OF SCHOOLS

This dissertation is a collection of three sub-studies exploring the experiences of
early career special education and general teachers as they encounter their school
environments. The first two of my sub-studies explore teachers’ emotional responses to
different aspects of their work, and I attempt to connect these emotional responses to
broader attitudes about teaching. Both studies are psychological in that they explore how
teachers interact with their work emotionally on a daily basis, but they are not direct tests
of how policy impacts the lives of teachers. The third essay includes a comparison of
formal and informal channels of support in schools, but the focus is on teachers’ social
environments, and how special educators and general educators experience the context of
their schools in very different ways.

At first glance, it is difficult to locate the role Of policy in this dissertation.
However, across the three studies, an argument is advanced that policy plays a critical
role in shaping the working environment that new teachers enter into. The transition into
one’s first teaching position has often been described as a process of “sink-or-swim” or
“trial-by-fire,” (Lortie, 1975; Gold, 1996; Ingersoll & Kralik, 2004). From the onset, the
novice teacher often faces the same duties and expectations as a veteran teacher, but must
simultaneously navigate the unfamiliar terrain of teaching without experience or practical
knowledge to fall back on. A new teacher’s success depends on their ability to achieve
multiple goals: they must acquire instructional and classroom management skills, become

familiar with district curricula, and adapt to the professional norms and procedures of

their individual schools.1 And, the learning curve is steep--they are to acquire their
knowledge through experience, learning what does and does not work through practice.

Historically, this process of learning to teach has been considered a private ordeal,
which Lortie (1975) attributes to the “cellular organization” of schools. However,
research has increasingly suggested that new teachers strongly desire guidance and
assistance from their peers (Kauffman et al., 2002; Kardos & Johnson, 2007), and that
schools as social organizations can have profound influences on teachers’ success in the
classroom (Desimone etal., 2004; Hill, Rowan, & Ball, 2005). Therefore, guiding my
dissertation is the assumption that new teachers’ beliefs-—about effective instructional
practices, about their effectiveness at delivering instruction, and about their commitment
to teaching—are situated within the social context of schooling (Bidwell, 2000; Zhao &
Frank, 2003). Teachers do not struggle through the challenges of learning to teach in
isolation, nor do structural or institutional factors alone dictate how teachers come to
think about themselves as teachers. Instead, the two processes, psychological and social,
interact in a way that Shapes an individual’s beliefs and attitudes about teaching. These
internal and external factors, both of which influence teachers’ attitudes about their work,
are the focus of this dissertation.

Study Design and Methodology
I investigated internal and external predictors’ of early career teachers’ attitudes at

school through multiple sources of data. In the first two essays, I explored the dynamic

 

1In addition to these experiences which face all early career teachers, novice special education teachers are
likely to encounter a unique set of challenges and responsibilities associated with teaching special
education. Among others, these include modifying curriculum for specific populations, understanding
federal special education laws, developing effective relationships with paraprofessionals, determining how
to use assistive technology, and documenting student progress on individualized education plans (Boyer &
Gillespie, 2000).

nature of teachers’ experiences at work by making use of a data collection strategy
known as the Experience Sampling Method (ESM). The ESM is an intensive, weeklong
data collection process in which individuals are asked to respond to brief questionnaires
at random intervals throughout their waking hours (7:00 - 11:00 PM). Participants were
asked where they were, what they were doing, what they were thinking about, and how
they felt about the activity in which they were engaged. The ESM differs from other time
diaries because it obtains detailed information about individuals’ time use and their
subjective interpretations of various activities during daily life. A sample of 46 teachers
participated in the ESM portion of the study.

I also collected survey data from early career teachers (ECTs) and their mentors
and key colleagues. The surveys asked questions related to the teachers’ backgrounds,
professional preparation, interactions with others, experiences in their schools, and
commitment to their grade/subject, their school, and their district. Early career teachers
were surveyed in fall of 2007 and were asked to list the names of their mentors as well as
up to eight colleagues with whom they engaged in professional interactions. To allow for
social network analysis, mentors and colleagues nominated by ECTs were asked to
complete surveys in the early winter of 2008 on similar items as those in the fall survey.
Finally, early career teachers were invited to complete follow-up surveys in the spring to
facilitate investigations of change over time. These data were used in the second and third
sub-studies of this analysis.

Despite the frequent use of ESM studies in psychology (e.g., Haworth & Hill,
1992; Csikzentrnihalyi, Rathunde, & Whalen, 1993; Schneider & Waite, 2005), there are

fewer examples in education (DiBianca, 2000; Schweinle & Turner, 2006; Shemoff et al.,

2003; Yair, 2000). Further, the method has not been used to explore how teachers spend
their time at school, or how they respond emotionally to their work. In my first sub-study,
1 therefore explored how the teachers in the ESM sample spent their time at school, as
well as the activities they reported when they were engaged in some form of instruction. I
also collected data on their average emotional states in each category of time use in order
to capture the variation of emotions across context. Finally, I made comparisons between
teachers based on whether they taught special education or general education, whether
they taught in an elementary school or middle school, and whetherthey were in their
first, second, or third year of teaching.

The goal of the second sub-study was to predict whether momentary affective
responses (i.e., local measures of emotions and mood) become aggregated into attitudinal
evaluations of individuals’ work (i.e., global measures Of job satisfaction). To do so, I
developed ESM-based measures of teachers’ levels of positive affect, negative affect,
skill, and fatigue at school, and investigated whether these factors were associated with
survey measures of burnout and commitment in the spring—while controlling for prior
levels of attitude and burnout. I also tested two different conceptual definitions of how
emotions might influence attitudes about work. First, I used person-level means, which
are likely to reflect participants’ overall affective state throughout the week—detached
from any specific events. A second aggregation strategy was to investigate the
relationship between participants’ “peak” emotions and their attitudes about work.

Finally, the third sub-study compared the experiences of beginning special
education and general education teachers in their interactions with mentors and social

network members, as well as their perception Of the collective assets embedded in the

school environment (i.e., collective responsibility, relational trust, and perception of fit).
Regression models were developed for the two groups of teachers, with each model
estimating the effects of support from colleagues on ECTs’ levels of stress and burnout.
Specifically, these models included predictor variables for the importance of colleague
support, as well as perceptions of the collective assets of the school organization (i.e.,
ECTS’ perceptions of trust and collective responsibility, as well as their perceptions of
organizational fit). I also included interaction terms to see whether the sources of
colleague support had a differential impact on general educators and Special educators.
Finally, I used social network analyses to more directly measure the influence of
colleagues’ commitment and perceptions of the school organization on early career
teachers own attitudes.
Findings from the Three Studies

In the first sub—study, my findings showed that teachers spent an average of just
over 50 hours per week at school. When teachers were at school, instruction was the most
frequent activity cited by our participants (20.34 hrs/wk), which is not surprising given
that instruction is a teacher’s primary focus. The relative frequency of activities such as
interacting with colleagues and engaging in non-instructional interactions with students
suggest that teachers spent little of their daily time alone; social interactions played a
large role in their daily work. Regarding instructional time use, instructional practices
varied considerably across the ESM sample, although lecture was the most frequent
classroom activity reported by teachers. Other common forms of instructional activities
included whole-group activities such as reading as a class or engaging in discussion, or

student-centered activities such as one-on-one or small-group instruction.

In comparisons of how different groups of teachers used their school time (i.e.,
general education teachers vs. special education teachers, elementary school teachers vs.
middle school teachers, and first, second, and third year teachers), the most important
distinction appeared to be between general education teachers and special education
teachers. This is likely attributable to the differences in the job responsibilities of general
education and special education teachers. Special educators, for example, spent
significantly fewer hours per week providing instruction to their students than general
education teachers. However, they also spent more hours per week doing paperwork—a
task that played a minimal role in the daily lives of general education teachers. Fewer
important differences were seen between elementary school teachers and middle school
teachers, with middle school teachers using technology more frequently during the school
day. Finally, differences based on year of teaching were not significant for any school or
class activity.

When looking at teachers’ average emotions in specific school contexts, I was
surprised at how positive the experience Of providing instruction was. Teachers reported
higher than average levels of skill, challenge, engagement, self-esteem, and activation
when engaged in instruction. Meanwhile, when doing activities such as planning,
paperwork, or grading, the teachers in this sample reported lower than average levels of
positive affect, enjoyment, and challenge. Interactions with colleagues, meanwhile, were
associated with lower than average levels of being in control and feeling skilled. At the
same time, during such interactions they reported higher levels of concentration and did

not appear to have lower positive affect or enjoyment scores.

While the first sub-study focused on how teachers responded affectively to
various activities throughout the school day, the second study focused on the aggregate
effect of teachers’ emotions at work. Several studies of early career teachers have drawn
connections between challenging work environments and decisions to leave the
profession, but omitted from these studies is the measurement of teachers’ actual
emotional responses to these conditions—as well as the direct consequences of these
decisions for teachers’ plans to stay in their teaching position. The results from my
analysis of ESM and survey data provide evidence of a moderate and negative
association between teachers’ average levels of “negative affect” and “fatigue” and their
levels of burnout with respect to theirjobs. This finding held for both general education
and special education teachers, and the effects of negative affect and fatigue were
important for spring burnout regardless of teachers’ levels of burnout in the fall.

Regarding the relationship between affective reactions to work and early career
teachers’ career plans, my results suggest that “positive affect” and “skill” are both
positively associated with commitment to one’s grade/subject, controlling for prior
commitment as well as teacher demographic characteristics. This suggests that overall
positive experiences at work are likely to increase teachers’ desire to stay in their
teaching positions; in contrast, negative affect was negatively associated with the
commitment variable, although the coefficient for this relationship was only significant at
p _<_ .10. Finally, “mean” measures of affective responses at work were better predictors Of
teacher attitudes than “max” measures, suggesting that specific positive and negative
experiences are less important to teachers’ judgments about their jobs than overall moods

at work.

The findings Of my third sub-study highlight potentially important differences in
the induction experiences of early career special education and general education
teachers; and, these differences appear important for teachers’ plans to stay in teaching.
Descriptive analyses revealed that special education teachers interacted with their
mentors far less frequently than general education teachers, which is likely attributable to
the fact that fewer special education teachers had mentors who were colleagues in their
schools. Both groups of teachers, however, valued the support they received from
colleagues more than the support they received from mentors. This finding was supported
by teachers’ reports of the frequency of interactions with colleagues around instructional
issues such as curriculum, teaching strategies, and student behavior. The frequency of
interactions between special education teachers and their colleagues did not differ
significantly from the colleague interactions of general education ECTs. Overall, the
special education teachers in this sample do not appear more isolated from their
colleagues than general education teachers.

Regression analyses predicting teacher burnout and commitment confirm that for
both groups of teachers, support from colleagues was important for their ability to handle
the stress associated with their jobs (i.e., their levels of burnout). More important was the
positive role of colleagues in predicting teacher commitment, even when controlling for
prior commitment and several teacher demographic variables. The findings also suggest
that perceptions of school—level assets, including professional fit and collective
responsibility, were each positively associated with overall levels Of commitment.

Models including interaction terms for special education teachers and each of these

sources of colleague support showed that, for special educators, higher levels of support
were more closely associated with high levels of commitment.

Finally, the above measures of colleague support were all indirect measures of
support—I was relying on ECTS’ perceptions of the importance of their informal
relationships, rather than attempting to capture the actual influence of colleagues on
ECTs. However, by running social network models predicting teacher attitudes, I found
that ECTS’ attitudes were positively associated with their colleagues’ attitudes about their
work, even when controlling for prior levels of these attitude variables. Collectively,
these results present strong evidence that how early career teachers come to think about
their schools as organizations is influenced by their social networks.

Discussion of Findings across the Sub-studies

My focus shifts from the first and second sub-studies to the third, where I move
from an analysis of internal factors (such as teachers’ emotions and attitudes) to an
analysis of teachers’ broader school context. However, a common goal across each of the
studies is to explicate what happens to teachers as they navigate the uncertainties
surrounding their first teaching positions. The findings from the first sub-study suggest
that it is not instruction itself that is likely to produce negative emotions for early career
teachers; instead, it is activities such as planning, doing paperwork, and grading where
teachers express higher levels of negative emotions. Further, the second analysis answers
important questions about how teachers come to make judgments about their work. The
analysis also suggests that researchers should not Simply draw connections between work
conditions and teacher outcomes, but should consider the mediating role of teachers’ day-

tO-day emotional responses to their work. Based on my findings, I would make the

argument that work context does not directly impact teacher attitudes; instead, it is
through other intermediary factors that these two are linked.

The results from the third analysis also support this conclusion. While my
regression analyses suggest a positive association between one’s social context and their
overall commitment, I believe my findings from the social network analyses are more
persuasive, because they allow me to capture the process by which colleagues’ beliefs
actually influence early career teachers’ beliefs. In this sense, the real contribution of this
study is in the evidence I present regarding processes influencing new teachers’ attitudes,
many of which are Often only assumed in many studies of new teacher induction
experiences. I show that individuals make judgments about their jobs based both on
influence from their colleagues, as well as based on their own interpretations Of events at
school. There appears to be an interplay between teachers’ social contexts and their
psychological responses to work. Or, as Bidwell writes in his description of social
psychology: “The person’s cognitive and emotional states (e.g., thought and feeling) in
some way link social context and individual behavior” (2000, p. 19).

In the case of teachers within schools, they do not develop teaching beliefs (e.g.,
about effective instructional practices, their ability to promote student learning, or their
commitment to teaching) in isolation, nor do structural or institutional factors alone
dictate how teachers develop convictions about teaching. Instead, psychological and
social processes interact in ways that shape an individual’s beliefs and behaviors. It is
important, then, in trying to understand teachers’ behaviors, to take account of a teacher’s

social context as well as their subjective responses to this context.

10

There are also important methodological contributions of this dissertation study.
For one, I present evidence across multiple sources of data. I connect Experience
Sampling Method data to survey data, as well as early career teacher survey data to
colleague survey data. Additionally, in each of my models using survey data, I take
advantage of the longitudinal nature of my data by controlling for prior measures. The
fact that average emotions are associated with survey data suggests that the ESM could
supplement more traditional forms of data collection in future studies. Similarly, the step
Of looking jointly at the experiences of general education and Special education teachers
is an important one. It suggests to policymakers that, within the same districts, general
education and Special education teachers experience induction policies differently,
specifically with respect to formal mentors.

Finally, several of my findings inform prior research on early career special
education teachers (see Billingsley, Carlson, & Klein, 2004; Miller, Brownell, & Smith,
1999); much of this prior research suggests that early career special educators are likely
to feel overwhelmed by their jobs and are likely to feel isolated and confused about their
role in the broader school context. The results of these three analyses present a more
nuanced picture of the experiences of early career special education teachers. While it is
true that special education teachers do not have access to the same formal sources of
support as general education teachers (i.e., in comparison to general educators, special
education teachers have, on average, less frequent interaction with their mentors), they
appear to interact with their colleagues just as frequently as general education teachers
do. Further, the relationship between both positive and negative emotions at work and

attitudinal outcomes in the form of burnout and commitment is likely to function

11

similarly between the two groups of teachers. The more critical distinction appears to be
in how important colleague support is for special education teachers’ levels of
commitment. My data suggest that collective assets of a school (such as relational trust,
collective efficacy, and perception of fit) have a greater effect on commitment for special
educators than they do for general educators. In this sense, the importance of school

context is especially salient for new special educators.

12

References

Bidwell, CE. (2000). School as context and construction: A social psychological
approach to the study of schooling. In M. Hallinan (Ed.), Handbook of the Sociology
of Education, pp.15-36. New York: Kluwer Academic/Plenum Publishes.

Billingsley, B., Carlson, E., & Klein, S. (2004). The working conditions and induction
support of early career special educators. Exceptional Children, 70(3), 333-347.

Boyer, L. & Gillespie, P. (2000). Keeping the committed. Teaching Exceptional
Children, 33(1), 10-15.

Csikszentrnihalyi, M., Rathunde, K., & Whalen, S. (1993). Talented teenagers: The roots
of success and failure. New York: Cambridge University Press.

Desimone, L. M., Porter, A. C., Garet, M. 8., Yoon, K. S., & Birrnan, B. F. (2002).
Effects of professional development on teacher’s instruction: Results from a three
year longitudinal study. Educational Evaluation and Policy Analysis, 24(2), 81-112

DiBianca, R. (2000). Teaching adolescents: Relationships between features of instruction
and student engagement in high school mathematics and science classrooms.
Chicago: University of Chicago.

Gold, Y., Roth, A., Wright, C., & Michael, W. (1991). The relationship of scores on the
Educators Survey, a modified version of the Maslach Burnout Inventory, to three
teaching-related variables for a sample of 132 beginning teachers. Educational and
Psychological Measurement, 5], 429-438.

Haworth, J. T. & Hill, S. (1992). Work, leisure, and psychological well-being in a sample
Of young adults. Journal of Community and Applied Psychology, 2, 147-160.

Hill, H. C., Rowan, B., & Ball, D. L. (2005). Effects of teachers’ mathematical
knowledge for teaching on student achievement. American Educational Research
Journal, 42, 371—406.

Ingersoll, R., & Kralik, J. (2004). The impact of mentoring on teacher retention: What the
research says. Denver, CO: Education Commission of the States. Available at
http://www.ecs.org/clearinghouse/50/36/5036.htm

Kardos, S. & Johnson, S. (2007). On their own and presumed expert: New teachers’
experience with their colleagues. Teachers College Record, 109 (12).

Kauffman, D., Johnson, S. M., Kardos, S. M., Liu, E., & Peske, H. G. (2002). "Lost at

sea": New teachers' experiences with curriculum and assessment. Teachers College
Record, 104(2), 273-300.

13

Lortie, D. C. (1975). Schoolteacher: A sociological study. Chicago: University Of
Chicago Press.

Miller, M.D., Brownell, M.T, & Smith, SW. (1999). Factors that predict teachers staying
in, leaving, or transferring from the special education classroom. Exceptional
Children, 65(2), 201 -2 I 8.

Schneider, B., & Waite, LJ. (Eds). (2005). Being together, working apart: Dual-career
families and the work-life balance. Cambridge, UK: Cambridge University Press.

Schweinle, A., & Turner, J. (2006). Striking the right balance: Students’ motivational
experiences and affect in upper elementary mathematics classes. Journal of
Educational Research, 99(5), 271-293.

Shemoff, D., Csikszentmihalyi, M. Schneider, 3., & Steele-Shemoff, E. (2003). Student
engagement in high school classrooms from the perspective of flow theory. School

Psychology Quarterly, 18, 158-176.

Yair, G. (2000). Educational battlefields in America: The tug-of-war over students’
engagement with instruction. Sociology of Education, 73(4), 247-269.

Zhao, Y., & Frank, K. A. (2003). Factors affecting technology users in schools: An
ecological perspective. American Educational Research Journal, 40, 807-840.

14

CHAPTER 2: SCHOOL TIME USE AND AFFECTIVE RESPONSES OF GENERAL
AND SPECIAL EDUCATION TEACHERS DURING THEIR FIRST YEARS OF
TEACHING
Introduction

The period at the beginning of a teacher’s career has been described in a variety
of ways. One well-known portrayal is Fuller and Bown’s (1975) conceptual model of
teacher development, which distinguishes among three phases of teaching, based on the
concerns that characterize a teacher’s main focus and efforts. The first phase of teaching
— when teachers transition from teacher preparation programs/pathways into their first
teaching positions — is typically marked by concerns about their own adequacy and
survival. From a developmental perspective, the changes in teachers’ concerns can be
described as a shift from a self orientation to a pupil orientation. This viewpoint is
supported by Veenman (1984), who in a review of the literature on beginning teachers,
cited several challenges that novices frequently face (e.g., classroom discipline,
motivating students, organizing instruction, insufficient materials and supplies, and
dealing with problems of individual students), many of which can be classified as
problems of survival.

The beginning years Of teaching have also been described as a process of “sink-
or-swim” or “trial-by-fire” (Gold, 1996; Ingersoll & Kralik, 2004; Lortie, 1975); from the
onset, the novice teacher faces the same duties and expectations as a veteran teacher, but
must simultaneously navigate the unfamiliar terrain of teaching with less experience or
practical knowledge. Success depends on achieving multiple goals: novices must acquire

instructional and classroom management skills, become familiar with district curricula,

and adapt to the norms of their individual schools. Facing theses challenges, new teachers

15

are likely to face high rates of anxiety, frustration, and inadequacy (F eiman-Nemser,
Schwille, Carver, & Yusko, I999).

The challenges associated with the beginning years of teaching have
consequences for the retention of early career teachers. Data from the 1999-2000
Schools and Staffing Survey (SASS) indicate that close to 30% of first-year general
education and special education teachers either left teaching at the end Of the year or
migrated to other schools or districts (Smith & Ingersoll, 2004). Further, research has
found teacher attrition in the first five years among general and special education teachers
to be 30% or greater (Henke, Chen, & Geis, 2000).

In recent years, a number of research studies have investigated associations
between the context of learning to teach and a variety of new teacher outcomes. Some
researchers have employed interviews and observations to examine possible associations
between school or district context and beginning teachers’ experiences (Achinstein,
Ogawa, & Speiglman, 2004; Author, 2007; Grossman & Thompson, 2004; Kardos et al.,
2001). Others have used large-scale survey data and student achievement data to
consider the impact of mentoring and other induction activities on teacher retention and
student performance (Fletcher, Strong, & Villar, 2008; Glazerman et al., 2008; Kapadia,
Coca, & Easton, 2007; Smith & Ingersoll, 2004).

But few studies have investigated what actually happens to teachers in the first
years of teaching. How do early career teachers spend their time, both in terms of in-
school experiences (such as planning classroom activities, engaging in specific forms of
instruction, monitoring student behavior, and interacting with colleagues) and out-of-

school activities (including work-related, family, and leisure activities)? How much of

16

teachers’ time outside Of school is devoted to working or thinking about school? Further,
given that early career teachers’ experiences are often characterized by struggles and
uncertainty, what are teachers’ emotional responses to the first years of teaching, and
what kinds of instructional and non-instructional experiences are likely to elicit negative
emotions among teachers?

This paper addresses these questions using a unique time sampling method known
as the Experiencing Sampling Method (ESM); using the ESM, we collected data on the
time use and subjective well-being of 42 early career teachers in Michigan and Indiana.
In the first section of the paper, we briefly review the research literature on teachers’ time
use and their affective responses to teaching. The second section introduces the ESM and
reviews research on its use in studying adolescents and adults. Third, we describe the
theoretical framework that shaped our research design and analysis. In the fourth section,
we discuss our method, including sample, ESM questionnaire, and measures used in the
analysis. The fifth and sixth sections feature the results of this analysis and some
interpretation of the study findings. Finally, we conclude by discussing the implications
of this study for research, policy, and practice.

Teachers’ Time Use and A flective Responses to Teaching and Other Activities

Teachers’ Time Use. Until recently, there had been limited research on how
much time US. teachers devoted to work-related and other activities, either in or outside
of school. With the advent of the American Time Use Survey (ATUS) in the early-
20005, though, national data are now available on the work patterns of teachers and other
professionals. The ATUS employs a random sample drawn from households that

participated in the US. Current Population Survey (CPS) and is administered in the form

17

of computer-assisted telephone interviews (Hamermesh, F razis, & Stewart, 2005). Each
ATUS participant is asked to complete a time diary by describing all Of their activities for
a 24-hour period (i.e., from 4 AM the previous day until 4 AM the day of the interview);
only their primary activities, however, are recorded and coded. After the time diary is
complete, participants are asked a series of follow-up questions about child care, paid
work, and volunteering (Hamermesh, Frazis, & Stewart, 2005).

Using ATUS data from 2003 through 2006, Krantz-Kent (2008) reported a
number of noteworthy findings about teachers’ time use and how it compared to that Of
other professionals. For example, teachers were more likely than other professionals to
do some work at home, to work on Sundays, and to hold more than one job at the same
time. At the same time, teachers spent fewer minutes than other professionals working
on weekdays or Saturdays (Krantz-Kent, 2006). In addition, teachers generally worked
less in the summer months than during other times of the year and Older teachers (ages 40
and Older) generally worked more hours each week than teachers aged 20 to 39. Finally,
as compared to other professionals, teachers spent less time working and more time doing
household activities (Krantz-Kent, 2006).

The ATUS data represent a significant contribution to what is known about
teachers’ time use (as well as that of other professionals) in the US. At the same time,
these data have some limitations. First, they provide little detailed information about the
types of primary activities teachers engage in during the work day (e.g., large-group
instruction, small-group instruction, planning, meeting with colleagues, extra-curricular
activities, etc.) or their secondary activities (e.g., child care, watching television, eating)

(Horrigan & Herz, 2005). Second, they do not differentiate among early career, mid

18

career, and late career teachers. Third, the ATUS data are based on participants’ ability
to recall activities that occurred up to 36 hours earlier. Finally, these data are limited by
the fact that the time diaries are only completed for a single 24-hour period.

Recent pioneering research by Rowan and colleagues at the University of
Michigan has employed teacher instructional logs to investigate teachers’ curricular and
instructional practices in literacy and mathematics (Rowan, Cambum, & Correnti, 2004;
Rowan, Harrison, & Hayes, 2004). Unlike most surveys of instruction, the teacher logs
focus on instruction provided to individual students and are completed by teachers
multiple times for each sampled student. In an analysis of elementary math instruction,
for example, Rowan, Harrison, and Hayes (2004) analyzed data from 19,999 logs
completed by 509 teachers in 53 schools. In this study, teachers who participated in
every session were expected to complete approximately 60 to 70 logs, or about eight or
nine for each sample student. Due to student and teacher absences and unusable data, the
researchers obtained an average of 39-40 days of log data on instruction for each teacher.
Similarly, in their analysis of elementary literacy practices, Rowan, Cambum, and
Correnti (2004) had an average of 30 to 35 usable logs for each study participant.

The use of teacher logs has several advantages over traditional surveys of
instruction. First, they provide much richer and more frequent data about individual
teachers’ classroom practices. Second, they can reveal variations in instruction within
schools, within classrooms across students, and within classrooms over the course of a
school year. Third, they address the issue of recall that can pose methodological
challenges for annual or biannual surveys. At the same time, for the purposes of the

analysis presented in this paper, instructional logs have two main shortcomings. First,

19

they are not designed to collect data on teachers’ time use throughout the school day or
outside of school; instead, they obtain detailed data during single periods of instruction.
Second, the logs are not intended to reveal information about teachers’ emotional
responses to instructional and other work-related activities.

Teachers’ Emotional Responses to Work-Related Activities. Over the past 15
years, several researchers in Canada, the US, England, and Australia have begun to
investigate the emotions of teaching, teachers’ emotional responses to key aspects of their
work, and how these responses can vary based on career stage, teaching context, and
other factors (Day & Leitch, 2001; Dinham & Scott, 1998, 2000; Hargreaves et al., 2006;
Jeffrey & Woods, 1996; Little, 1996). A leading contributor in this area has been Andy
Hargreaves, who has theorized and empirically examined emotional geographies of
teaching (2001). In particular, Hargreaves directed a major study of teachers’ emotional
responses to their work in Ontario, Canada in the 19905. Drawing on interview data with
elementary and secondary teachers, he identified and illustrated “sociocultural, moral,

professional, political, and physical distance as five key emotional geographies of
teaching” (2001, p.106l). For example, when students and families came from cultural,
lingUistic, and/or sociO-economic backgrounds that differed from those of their teachers,
according to Hargreaves, this sometimes led to stereotyping, lack of emotional
understanding, and readiness to assign blame on the part of teachers.

As part of their research in Ontario, Hargreaves and colleagues interviewed
teachers to learn about their emotional responses to interactions with students
(Hargreaves, 2000) and parents (Hargreaves, 2001; Lasky, 2000), and to serving as

dcPal‘tment chairs (heads) in secondary schools (Schmidt, 2000). In addition, researchers

20

from the US. have employed interviews (Little, 1996) and interviews and surveys
(Lasky, 2005) to probe the emotional responses of secondary school teachers to efforts to
engage in comprehensive high school reform. These studies represent important
advances in research on teachers’ emotions because they distinguish affective responses
among teachers based on a) schooling level, b) work context (e.g., interacting with
students versus interacting with parents), c) role (e.g., teacher, department chair, etc.),
and d) reform context.

At the same time, researchers have acknowledged some limitations in the use of
interviews to analyze teachers’ emotional responses to their work (Hargreaves, 2001). In
response, scholars have employed narrative, autobiographical accounts (Day & Leitch,
2001) and long-term observation (Jeffrey & Woods, 1996) to examine teachers’
emotional experiences during professional development activities andschool inspections.
Further, researchers have also used surveys to investigate teacher job satisfaction, morale,
and commitment (see, e.g., Dinham & Scott, 1998, 2000; Ebmeier, 2003; Evans, 2000;
Ingersoll, 2001; Kapadia, Coca, & Easton, 2007). In our view, interviews and one-time
surveys have limited capacity to provide detailed information about teachers’ affective
responses to the many different activities in which they engage in and outside of school.
And while narrative autobiography and long-term observation Offer fine-grained data
concerning teachers’ emotions, they have limitations as well. In particular, narrative
accounts are also based on teachers’ ability to recall past events while long-term
observation requires many researchers and/or observations for each study participant.

In the next section, we describe an alternative way to collect detailed data on

teachers’ time use and their emotional responses to the activities in which they engage —

21

the Experience Sampling Method (ESM). The ESM has been used widely in research in
the areas of psychology and human development (Csikszentmihalyi,, 1997;
Csikszentmihalyi & Schneider, 2000; Schneider & Waite, 2005) but much less frequently
in educational settings. Compared to interviews and other forms of data collection, it
enables researchers to obtain in-the-moment data on teachers’ primary and secondary
activities throughout several days over a week or more.
The Experience Sampling Method

Developed by Mihaly Csikszentmihalyi and his colleagues (Prescott,
Csikszentmihalyi, & Graef, 1981; Csikszentmihalyi & Larson, 1984; Csikszentmihalyi,
1997), the ESM is a week-long data collection activity in which participants wear a
signaling device such as a beeper, watch, or personal digital assistant (PDA). The
signaling devices are pre-programmed to beep eight times a day for seven days, randomly
within two-hour time intervals. That is, they emit signals each day eight times across a
16-hour period (from 7:00 am. to 1 1:00 pm), with no two beeps occurring less than 30
minutes apart. The ESM differs from other time diaries because it also asks individuals
what they are thinking about at a particular moment and obtains detailed information
about subjective interpretations of various activities during daily life. Also important is
that ESM data provide a random sample of an individual’s daily activities, giving the
researcher “a relatively complete and undistorted picture of daily life, and (these data)
Often provide unexpected glimpses into what one would never have expected” (Hektner,
Schmidt, & Csikszentmihalyi, 2007).

Previous research has shown that ESM measures and data are reliable and valid

compared with data collected through other survey methods (Csikszentmihalyi & Larson,

22

1984; Robinson, 1985). Additionally, by having multiple responses for each individual
across a variety of variables (there can be as many as 56 data points for every
participant), researchers can develop aggregate scores at the person level that take into
account the ways in which emotions vary across situations. For example, the metric used
to determine a person-level characteristic such as happiness is a mean or sum score based
on multiple responses across multiple contexts and experiences; this aggregate happiness
measure can then be used in a regression analysis in conjunction with variables from
other surveys. An additional benefit of the ESM is that it does not suffer from the same
bias of recall that characterizes other methods, which require a participant to reconstruct
at a later point what they were doing at a previous time and how they felt about it.
Instead, when responding to the ESM, individuals record their activities and emotional
responses to these activities as they occur.

Research studies in education that have used the ESM have most frequently
investigated student engagement in schooling, as well as how specific teacher practices
elicit different behaviors and emotions in students (see, e.g., DiBianca, 2000; Schmidt,
Shumow, & Kackar, 2007; Schweinle & Turner, 2006; Shemoff & Schmidt, 2008;
Shemoff & Vandell, 2007; Shemoff et al., 2003; Yair, 2000). In one study, for example,
Shemoff and Schmidt (2007) used the ESM to investigate similarities and differences in
achievement, engagement, and quality of experience among 586 white, black, Latino, and
Asian high school students. In a second study employing ESM data, Shemoff and
Vandell (2007) examined students’ engagement levels in after-school programs during

different types of activities and with various partners. Finally, Shemoff et al. (2003)

23

drew on ESM data to consider how adolescents Spent their time in high school and the
conditions under which they reported being engaged.

A number of researchers have also used the ESM to examine adults’ emotional
responses to work and non-work activities. In the 500 Family Study, Schneider and
Waite (2005) collected detailed information, including ESM data and parent and
adolescent surveys, from over 500 middle-class, dual-earner families in eight
communities across the US. The purpose of the study was to obtain and analyze
extensive data on parents’ home and work experiences, adolescents’ home and school
experiences, and both groups’ affective responses to their experiences. In a smaller
study, Williams and Alliger (1994) used the ESM with 41 parents to explore their
subjective experiences with work and parenting and whether their emotional responses in
one sphere of responsibility carried over to the other sphere. Further, Fisher (2000)
employed the ESM with 121 working adults in order to consider the relationship between
real-time emotions during the work day and overall job satisfaction.

In sum, the ESM has been shown to be a valid, reliable instrument for measuring
adults’ activities and their in-the-moment affective responses to their activities. A
growing number of studies have used this research method to examine student
engagement, adolescents’ experiences at school and home, and adults’ experiences at
work, at home, and when they are/are not parenting. At the same time, though, no
existing ESM studies have focused on how teachers spend their time, as well as how they
respond emotionally to experiences in their daily lives, such as trying to help students
master a difficult math topic, asking one’s mentor for advice, or grading tests in front of

the television at home. In this manuscript, we report findings from a study that employed

24

the ESM with 42 early career teachers, thereby contributing to the research literature on
teachers’ time use and their affective responses to their experiences.
Theoretical Framework

In this study, we employ a theoretical framework that is based on a social
psychological approach to the study of schooling (Bidwell, 2000; Bryk & Schneider,
2002; Frank, 1998). From this perspective, an individual’s psychological state is in part
shaped by their social context, which in turn influences their behavior. Or as Bidwell
writes: “The person’s cognitive and emotional states (e.g., thought and feeling) in some
way link social context and individual behavior” (2000, p. 19). In the case Of teachers
within schools, they do not develop teaching beliefs (e.g., about effective instructional
practices, their ability to promote student learning, or their commitment to teaching) in
isolation, nor do structural or institutional factors alone dictate how teachers develop
convictions about teaching. Instead, psychological and social processes interact in ways
that shape an individual’s beliefs and behaviors. It is important, then, in trying to
understand teachers’ behaviors, to take account of a teacher’s subjective responses to
their social context.

Our conceptualization of a teacher’s focuses on their social position within the
school organization (e.g., whether they teach at the primary or secondary level, their
content area and grade level, the population of students they teach). The basic question of
what it looks like to be a teacher is profoundly influenced by each dimension of their
particular social context. Positional variation, such as whether a teacher works with
elementary school students or middle school students, is likely to shape their definition of

what their role should be in the lives of their students, while the distinction between

25

teaching special education versus general education may have profound consequences for
how novice teachers think about attending to individual needs of students compare to
those a class as a whole.

Variations in individuals’ teaching contexts are likely to dictate the kinds of
activities in which teachers engage on a daily basis. And, given the dynamic and complex
nature of a teacher’s work life, the range of activities related to teaching are likely to
produce a variety of emotional responses from moment to moment, including happiness,
frustration, anxiety, and excitement; and two different individuals may respond to the
same circumstances in two different ways. Consequently, the study of how emotions vary
by context and by person can help move us toward a more accurate representation of
individuals’ experiences in their everyday lives, as well as how these momentary
experiences are likely to influence future attitudes and behaviors. As Csikszentmihalyi
and Larson (1987) write, the objective of this research is to “identify and analyze how
patterns in people’s subjective experience relate to the wider conditions of their lives” (p.
527). Or, how are momentary emotional responses associated with individuals’ global
psychological states?

In the field of organizational science, researchers have paid increasing attention to
the influence of current reports of mood and emotion at work on critical outcomes such
as job performance and satisfaction (Brief & Weiss, 2002; Fisher, 2000; Fisher &
Ashkanasy, 2000). Much of this work draws on a theoretical model developed by Weiss
and Cropanzano (1996), known as the Affective events theory (AET). The AET model
provides a framework for understanding how momentary emotions may produce “affect-

driven behaviors,” which, in the aggregate, can influence attitudes and behaviors about

26

work. Relevant to teachers is the question of whether their momentary emotional
responses predict more global job attitudes such as self-efficacy, motivation, and
commitment to one’s school and to the teaching profession, each of which is likely to
influence teachers’ quality of instruction and decisions related to retention.

In sum, the theoretical framework guiding this study suggests that momentary
emotional responses to teaching provide a theoretical link between a teacher’s social
context — including the design of their teaching environment and their interactions with
colleagues, students, and other school personnel — and their attitudes and behaviors
related to their teaching position.

In this study, we employed the Experience Sampling Method (ESM) with more
than 40 early career teachers to investigate the social-psychological processes of learning
to teach and whether/how these processes varied among different groups of teachers.
That is, we studied whether these processes differed among a) general education and
Special education teachers, b) elementary and middle school teachers, and c) first-year
versus second- or third-year teachers. In particular, the study was designed to address the
following research questions:

1. What activities do early career teachers engage in throughout the course of

the work week, both during their time at school and when they are

specifically engaged in instruction Does this time use vary by category of

teacher, including whether an individual teaches special education or

general education, whether they are in an elementary school or a middle

school, and whether they are in their first, second, or third year of

teaching?

27

2. When teachers say they engage in instruction, how is this instructional
time used? Does it vary by category of teacher?
3. What are teachers’ emotional responses to work? How do emotional
responses vary by school activities?
Method

Study Participants

For this study, we recruited 42 teachers to participate from three school districts in
Michigan and Indiana. The Experience Sampling Method (ESM) study reported on here
is part of a larger, three-year study of early career teachers in several urban Michigan and
Indiana districts. Funded by Carnegie Corporation of New York, the larger study
examines how mentoring, social networks, and district policies affect early career
teachers' commitment, retention, and instructional practices, and the achievement gains of
their students. The study has been taking place since 2006-07 and continues through the
2008-09 school year; it features approximately 300 general education and special
education teachers in school districts in Michigan and Indiana.2

In selecting first-, second-, and third-year teachers to participate in the ESM study
described here, we focused on elementary and middle school general and special
education teachers. In terms of the general education teachers, we only invited those in
core content areas in grades 1-8 to participate in the ESM study. This included
elementary school teachers (grades 1-5) and middle school teachers in the areas of

language arts, history/social studies, mathematics, and science. With regard to the special

 

2 In 2006-07, the sample for the larger study included first- and second-year teachers; in 2007-08, the
sample for the larger study included first-, second-, and third-year year teachers; and, in 2008-09, the
sample for the larger study includes first-, second-, third-, and fourth-year teachers.

28

educators, we invited teachers who were responsible for providing academic instruction
to students in grades 1-8, but excluded those who did not provide instruction (e.g., school
psychologists, speech pathologists, social workers). All first-, second-, and third-year
teachers who met these criteria were invited to participate, but they had to be teaching
full-time, have earned a standard teaching certificate, and have completed university-
based teacher preparation.

In addition, the student demographics in the study participants’ classrooms and
schools had to be consistent with those throughout their district. That is, we wanted to
ensure that a given novice’s experiences were not significantly shaped by having much
higher or lower percentages of low-income students than other teachers in the study. In
sum, the criteria for selecting new teacher participants included a) being responsible for
academic instruction/teaching in a core content area, b) teaching full-time, c) having
earned a standard teaching certificate and completed university-based teacher
preparation, and d) having demographics in their classrooms and schools that were
consistent with those throughout the three districts.3

In the three districts chosen for the ESM study, we selected 42 teachers to
participate. Of these 42 teachers, 27 were teaching in elementary schools and 15 were
teaching in middle schools. Twenty-six of the teachers taught general education, while 16
taught special education. The years of experience of teachers in the sample varied, with
11 first-year teachers, 22 second-year teachers, and 9 third-year teachers participating in
the ESM study. The majority of the teachers in the sample were white females, with only

seven males out of the total number of participants in the study. As seen in Table 1, the

 

3 Table I Illustrates how the ESM sample aligns with the full sample for the three districts.

29

sample of ESM teachers is largely representative of the full sample of teachers from the

three districts participating in the MlECT study. However, the ESM sample has a higher

proportion of special education teachers than the full sample, as well as a higher

proportion of second-year teachers than is present in the full three-district sample.
[Insert Table 1 Here]

ESM Questionnaire

We adapted our questionnaire and response procedures from Schneider and
Waite’s 500 Family Study (2005). When a participant was beeped, they documented the
time they were initially beeped as well as when they were able to respond. Participants
described how they were spending their time by reporting where they were, what activity
they were engaged in, what else they were doing, as well as what was on their minds.
Two trained coders, using coding schemes adopted from previous ESM studies, coded
each of the open-ended questions about participants’ activities and locations.4 Inter-rater
reliability for the coding of ESM activities ranged from .75 to .90.

Participants were asked who they were with based on a list of categories of
school- and home-based individuals. They also reported characteristics of their
psychological states at the time they were beeped, responding to a set of Likert-type
questions such as “Did you have the abilities to deal with the situation?” and “Was the

activity important to you?” These were followed by a battery of emotions such as

. . 5
“cheerful,” “worrred,” and “rsolated.”

 

4 These categories (as well as the categories for instructional time use) were developed based on pilots of
the ESM in 2007-08 as well as in-depth interviews with pilot participants. See Author (2008) for more
details on the pilot.

5 The validity of these items is described at length in Csikszentmihalyi and Larson (1987), and in Hektner,
Schmidt, and Csikszentmihalyi (2007).

30

Measures

In the analysis presented here, three categories of measures are used: activity
when beeped, hours of time engaged in activity, and emotional state during activity. For
the first category of measures, we were interested in three levels of activities. First, and
most generally, we looked at teachers’ total waking hours during the school week, and
focused on the amount of time teachers spent at school, at home, and in public. Next, we
looked at what activities teachers were engaged in during school; these activities included
instruction, planning, non-instructional time with students, talking with colleagues,
talking with administrators, talking with parents, doing paperwork, grading, disciplining
students, professional development, extra-curricular activities, and using the computer.
Of these, we only present findings on the categories that consumed greater than one
percent of teachers’ time in school (on average, across the sample).6 Finally, to determine
how teachers spent their class time on average, we developed categories for the following
classroom activities: lecture, working with a single student, reading as a class, engaging
in class discussion, working in small groups, administering quizzes/exams, monitoring
student behavior, using technology, engaging in games/activities, engaging in classroom
routines, monitoring student seatwork, and working with students on functional skills
(i.e., those skills not directly related to the curriculum, but that will help students in their
everyday lives).

To determine the amount of time devoted to each of these activities, we calculated
a proportion score by dividing the number of beeps for each person while in a particular

context by their total number of beeps. “Total beeps” in this analysis refer to the person’s

 

The followrng actrvrtres were dropped from the analysrs: talking wrth administrators, talking wrth
parents, disciplining students, professional development, and extra-curricular activities.

31

total number of responses during the work week (Monday through Friday). We have
excluded data from the weekend in this analysis because of our focus on activities related
to teaching; therefore, the total number of beeps possible was 40, or 8 beeps a day over 5
days. From these proportion scores, estimates of hours of time use were derived by
multiplying the proportion score by the total number of waking hours in the work week.
For example, during their waking hours in the work week, the teachers in our sample
spent on average 63% of their time at school. Multiplying this by their total waking hours
(16 hours per day x 5 days) gives us .63 x 80, or 50.38 hours per week at school.

We addressed the question of how teachers’ psychological states varied
throughout their teaching day by including emotional and cognitive items that were likely
to be impacted by teachers’ social environments. For each teacher in the sample, we
calculated a person-level average based on their beep-level responses for a given
emotion. Eight single measures were included: a) Did you feel skilled in the activity? b)
Did you enjoy what you were doing? c) How well were you concentrating? (I) Did you
feel challenged by the activity? e) Did you feel stressed? I) Did you feel in control of the
situation? g) Did you feel energetic? and h) Did you feel active. Additionally, we
included two composite variables that have been used frequently in ESM research; see
Hektner, Schmidt, and Csikszentmihalyi (2007) for examples. These include self-esteem
(the scale variables included living up to one’s own expectations, living up to others’
expectations, control, feeling good about self, and succeeding); positive affect (the scale
variables included happy, cheerful, friendly, relaxed); activation (the scale variables

included strong, active, and excited); and engagement (the scale variables included

32

concentrating, enjoy, interesting). Each of these composite variables had an internal
consistency (Alpha) of higher than .75.

Results
Time Use

School Week Time Use. The first goal Of the time-use analysis was to determine
how many hours per five-day week teachers spent at school in comparison to time spent
at home or in public (outside of school). Table 2 presents the results of this analysis,
revealing that on average, the teachers in our study spent 50.38 hours per week at school.
The remainder Of their waking hours was divided between being at home (18.57 hrs/wk)
and being in public (10.94 hrs/wk); the high standard deviations for each of these
variables suggests that teachers varied considerably with regard to how they distributed
their time outside of school.

[Insert Table 2 here]

We were also interested in whether certain categories of teachers spent more or
less time at school relative to other locations; for example, did first-year teachers spend
more time at work than second- and third-year teachers? To determine whether there
were significant differences in time use based on assignment (general education vs.
special education, elementary vs. middle school), we conducted independent samples t-
tests on mean differences in weekly hours spent at school, in public, and at home. One-
way ANOVAs were conducted to determine whether there were significant differences
based on years of experience (first vs. second vs. third). Few of these comparisons were
significant (see Table 3), although amount of time spent at school varied between special

education and general education teachers (GE = 51.99 hrs/wk, SE = 47.78 hrs/wk;

33

p<.05), as well as between elementary school teachers and middle school teachers. (ES =
51.61 hrs/wk, MS = 47.92 hrs/wk; p<.05). There were no significant results when
comparing teachers by years of experience.

[Insert Table 3 here]

Time Use at School. Person-level means were calculated for beeps in which
participants indicated that they were at school. The person-level aggregates were then
averaged across the entire sample of teachers to determine average time use while in
school. As expected, the majority of teachers’ time in school was devoted to instruction
(20.36 hrs/wk), which far exceeded all other categories of time use (see Table 4). The two
other major sources of time use during school included planning (6.59 hrs/wk) and non-
instructional time spent with students (6.39 hrs/wk); the category of non-instructional
time with students consisted of classroom interactions unrelated to instruction, as well as
non-classroom interactions, such as monitoring hallways and walking students to lunch.
Slightly less time was devoted to talking to colleagues (4.34 hrs/wk), but such
conversations were likely to occur more frequently as secondary activities, while teachers
were engaged in other tasks. Overall, there was substantial variation in how the teachers
in the sample spent their time.7

[Insert Table 4 here]

As in our analysis of group differences in time spent at school versus at home or

in public, we conducted independent samples t-tests on mean differences in weekly hours

spent on school activities, investigating whether there were significant differences in time

 

7 A relevant check on the validity of these time estimations is the amount of time spent each week eating
lunch. Teachers in the sample reported spending, on average, approximately 2.24 hours per week eating
lunch; or, roughly a half hour per day. This estimate of time use fluctuated very little across each of the
categories of teachers.

34

use based on assignment. One-way ANOVAS were also conducted to determine whether
there were significant differences in the amount of time spent on various school activities
based on years of experience (first vs. second vs. third).

[Insert Table 5 here]

Of these distinctions, the most significant differences with regard to time use in
school appeared to be between general education teachers and special education teachers.
Throughout their days at school, special education teachers spent significantly less time
than their general education peers engaged in instruction (General Education (GE) =
23.74 hrs/wk, Special Education (SE) = 14.88 hrs/wk; p<.01) and grading (GE = 2.89
hrs/wk, SE = 1.15 hrs/wk; p<.05), although they spent much more time interacting with
colleagues (GE = 3.14 hrs/wk, SE = 6.78 hrs/wk; p<.05). Significant differences in
school time use were also present in amount of time devoted to paperwork (GE = .60
hrs/wk, SE = 4.87 hrs/wk), which was likely due to federal and state policy regulations
placed on special education teachers for completing paperwork for individual students.

There were fewer significant differences between elementary school teachers and
middle school teachers. These teachers spent similar amounts of time on instruction,
talking with colleagues, and talking with students; however, differences emerged in how
other non-instructional time was used — middle school teachers spent more of their time
grading (Elementary School (ES) = 1.46 hrs/wk, Middle School (MS) = 3.60 hrs/wk;
p<.05), for example. The greatest differences between the two groups were in the number
of hours spent using computers and other forms of technology for work-related tasks,
with middle school teachers spending significantly more time on computers than

elementary school teachers (ES = 1.42 hrs/wk, MS = 4.15 hrs/wk; p<.01).

35

Results of the ANOVAs show no significant effects for the comparisons of school
activity based on years of experience (all F’s < 2.48, n.s.). However, several potentially
important trends are worth noting. Teachers with more years of experience spent less
time on instruction (first-year teachers = 20.21 hrs/wk; second-years = 20.93 hrs/wk; and
third-years = 18.79hrs/wk) and on planning (first-year teachers = 7.17 hrs/wk; second-
years = 6.74 hrs/wk; and third-years = 5.50 hrs/wk). However, the amount of time spent
talking to colleagues increased with years of experience (first-year teachers = 2.96
hrs/wk; second-years = 3.93 hrs/wk; and third-years = 7.02) as did the amount of time
talking to students about non-instructional topics (first-year teachers = 5.32 hrs/wk;
second-years = 5.87 hrs/wk; and third-years = 8.96 hrs/wk).

Time Use During Instruction. We were interested both in determining how
teachers spent their time devoted to instruction, as well as whether significant differences
emerged among the relevant categories of teachers. AS with the analysis of school time
use, we aggregated the beep-level data into person-level means for moments when
teachers indicated that they were engaged in instruction. Means were then calculated
across the sample as a whole, revealing large variations in how instructional time was
used. The most frequent classroom activity was lecture (4.18 hrs/wk), followed by
instruction devoted to a single student (3.25 hrs/wk), working in small groups (2.08
hrs/wk) and reading as a class (2.04 hrs/wk). On average, other activities, such as
classroom routines and technology-based instruction, occurred less frequently.

[Insert Table 6 here]
Differences in time use by teaching assignment were tested using independent

samples t-tests, while one-way ANOVAS were conducted to determine whether there

36

were significant differences by years of experience. Results of these analyses are
presented in Table 7.
[Insert Table 7 here]

The greatest differences in time use during instruction were between general
education teachers and special education teachers. General education teachers were more
likely than their special education peers to spend their instructional time lecturing.
Special education teachers devoted on average 1.41 hrs/wk to lecture, versus 5.89 hrs/wk
for general education teachers (p<.01), reflecting differences in the classroom dynamics
between the two settings. In comparison to general education teachers, special education
teachers devoted virtually no time to classroom routines (GE = 1.22 hrs/wk, SE = 0.13
hrs/wk, p<.05). They were also much more likely to focus on functional skills instruction
than their general education peers (GE = 0.74 hrs/wk, SE = 3.92 hrs/wk; p<.05).

There were fewer meaningful differences with regard to how elementary school
teachers and middle school teachers spent their instructional time. For one example,
middle school teachers spent more of their time using technology in the classroom (ES =
0.47 hrs/wk, MS = 1.77 hrs/wk; p<.05). There were also significant differences in time
devoted to classroom routines (ES = 1.13 hrs/wk, MS = 0.22 hrs/wk; p<.10) and in time
in small groups (ES = 1.76 hrs/wk, MS = 0.86 hrs/wk; p<.10); however, given the
number of t-tests conducted, reports of significance at the p<.10 level should be
interpreted with caution.

The ANOVA tests for differences in years of experience yielded no significant
results. Generally, it appeared that third-year teachers devoted a greater proportion of

their classroom time to direct forms of instruction (such as lecture, discussion, and

37

quizzes and exams) and less of their time to classroom routines and monitoring student
behavior. This is not surprising; as teachers gain experience, they are likely to focus less
on classroom management and more on the quality of their instruction.

Finally, although we were interested in how the groups of teachers differed in
their time use while engaged in instruction, there were no significant differences between
categories. This is likely a result of small cell sizes; for many instructional activities (e.g.,
administering an exam, reading as a class, student seatwork), the frequency of their
occurrence ranged widely within groups, and for some participants, certain activities
either did not occur or occurred no more than a few times. This, coupled with the fact that
the analysis divided the sample into smaller groups of teachers (e.g., for general
education, N = 26; for special education, N = 16), likely explains our lack of significant
findings.

Variations in Emotions
Emotional Responses In and Outside of School. In our analysis Of variations in

emotional states during the work week, we compared average emotional states at school

to emotional states when not at school, using paired samples t-tests.8 Results are
presented for our nine emotion measures of interest, represented by context-specific z-
scores that indicate emotions at school relative to person-level averages. The use of
context-specific z-scores allowed us to account for differences in how individuals used
the Likert-type scales that documented their emotion at the moment they were beeped.

[Insert Table 8 here]

 

The “Not at school” beeps included all beeps when a teacher was at home or in public during the work
week.

38

Our results reveal important differences in emotions in and outside of school. In
general, it appears that school provided teachers with higher levels of intellectual
stimulation, relative to time spent at home or in public. While in school, teachers reported
being more challenged, more active, and concentrating more than on average; they also
had higher levels of self-esteem. Conversely, when teachers were not at school, they were
less stressed and enjoyed what they were doing more. Given these findings, it is
surprising that neither engagement nor positive affect vary much between settings.

Emotional Responses During School Activities. We next investigated teachers’
emotional states while at school, and how teachers’ affective responses varied by school
activity. We focused specifically on six school activities that occurred most frequently
across teachers: planning, instruction, talking with colleagues, non-instructional time with
students, grading, and paperwork. For each of these categories, we developed person-
level z-scores for our nine emotion measures of interest. Thus, for each activity,
participants had mean z-scores on the nine emotions, which could then be compared to
their average emotional states across all activities.

[Insert Table 9 here]

Of the six school activities of focus, instruction was associated with the highest
levels of positive emotions; in comparison to their average emotional states, teachers
reported significantly higher than average levels of skill, activation, concentration, self-
esteem, and engagement (all p—values <.01) when teaching. Instruction was the one
activity where teachers also reported higher than average levels of enjoyment (p<.05).
Other activities involving students and colleagues were associated with more mixed

emotions. When interacting with their colleagues, teachers reported higher than average

39

levels of concentration and challenge (p<.01), though they felt significantly less skilled
and in control of the situation (p<.01). Non-instructional activities with students (such as
lining students up to go home or monitoring them at lunch) were associated with higher
levels of activation (p<.01) and positive affect (p<.05) but lower levels of feeling
challenged (p<.05).

As might be expected, lower levels Of positive emotions were experienced by
teachers when they were engaged in activities by themselves; when teachers were
planning, grading, or doing paperwork, they experienced significantly lower levels of
enjoyment (p<.01). Grading was associated with the lowest levels of positive affect
(p<.01). Paperwork was also associated with significantly higher average levels of
challenge (p<.01) and stress (p<.05), but lower levels of skill (p<.01). Meanwhile, when
they were grading, the teachers reported lower than average levels of feeling activation
(p<.05), but higher than average levels Of skill (p<.01). Of the Six activities, teachers felt
least engaged when planning, reporting significantly lower than average levels of
engagement (p<.05).9

For almost all comparisons of teachers in their affective responses to school-based
activities, we saw no significant effects. However, there were two notable exceptions.
First, in comparison to second- and third-year teachers, first-year teachers had
significantly higher activation z-scores during instruction (p<.05), meaning that when
teaching they varied more from their average emotional states than did either second- or

third- year teachers. Second, there were also significant differences in affective responses

 

9 To account for within-person variation in emotions while in various school contexts, multilevel models
were run using beeps at Level I and persons at Level 2. Separate models controlled for three different
contexts: instructing, talking with colleagues, and working alone (planning, doing paperwork, or grading).
Results from these analyses mirrored those presented in Table 9. The final HLM models can be found in
the appendix in Tables 10.1-10.3.

40

between special education and general education teachers during non-instructional
interactions with students. Relative to general education teachers, special education
teachers reported higher levels of challenge (p<.05), enjoyment (p<.05), and engagement
(.01).

Given the lack of significant findings for differences between categories of
teachers in instructional time use, we did not conduct additional investigations into
variations in affective response by instructional activity.

Discussion

This paper set out to address three questions related to how early career teachers
experience the beginning years of teaching: a) How do teachers spend their time
generally throughout the work week?, b) How much time do they devote to different
school and class activities?, and c) How do they react emotionally to key experiences
during their jobs? Our first goal was largely exploratory; we wanted to examine how
early career teachers divided their work week between time at school, time at home, and
time in public. On average, the teachers in our sample were at school for 50.38 hours per
week, with the rest of their waking hours divided between being at home and in public
(i.e., not in school). General education teachers spent an average Of four additional hours
per week at school relative to special education teachers, while elementary school
teachers spent significantly more time at school than middle school teachers.

These differences in time use at school do not necessarily suggest that teachers
who spend less time at school work fewer hours than their peers. The current analyses did
not capture how much time teachers Spent on school-related tasks when they were at

home (either during the work week or on the weekend). It could be, for example, that

41

middle school teachers were able to do more of their work at home than were elementary
school teachers. Additionally, our measure Of time spent at school probably captured
some activities unrelated to instruction, suggesting that the number Of hours a week spent
working likely varied from the time use estimates presented here.

How did the teachers in our sample spend their time in school? Our findings
suggest that teachers’ daily work is complex and requires that they manage a variety of
responsibilities and relationships. Instruction was the most frequent activity cited by our
participants (20.34 hrs/wk or approximately 4 hrs/day), which makes sense given that
instruction is a teacher’s primary focus. Yet at the same time, instruction accounted for
less than half of teachers’ time at school, with the rest of the day divided among
necessary (though perhaps less desirable) tasks like grading, planning, and completing
paperwork. Data on the other major categories of time use (i.e., non-instructional time
with students and talking with colleagues) suggest that teachers spent little of their daily
time alone; social interactions played a large role in their daily lives.

Turning our attention to teachers’ instructional time, we found that lecture was
the most frequent classroom activity; other forms of instruction, including whole-group
activities such as reading as a class or engaging in discussion, as well as student-centered
activities such as one-on-one or small-group, occurred for roughly two to three hours
each. Thus, while the teachers in our sample spent the greatest proportion of their
instructional time using lecture to present information, our evidence suggests that lecture
may not play the dominant role that Goodlad (1984) and Sizer (1984) found it did 25
years ago. The instructional practices used by teachers in our study were highly varied,

and very often consisted of the kinds of activities that would facilitate give-and-take

42

between teachers and students. Further, it appeared that first-year teachers were just as
likely to integrate these kinds of practices into their instruction.

On a related note, we had expected that for a sample of early career teachers,
activities such as monitoring student behavior or disciplining students would have been
observed more frequently. However, each of these activities occupied less than an hour of
a given teacher’s instructional time per week. Perhaps teachers did not see discipline as
their main activity, and were more likely to think of it as a temporary interruption in their
instruction. One advantage of the ESM is that it allows us to investigate teachers’
secondary activities in addition to their primary activities. In future analyses, it will be
possible to identify how frequently teachers documented either Of these tasks as
secondary activities, as well as what their primary activities tended to be when they were
monitoring student behavior or disciplining students.

We hypothesized that time use was likely to vary depending on years of
experience as well as job assignment (including whether they taught special education or
general education, and whether they were an elementary school or middle school
teacher). Further, we hypothesized that affective responses to teaching were likely to vary
depending on the activities in which teachers were engaged. Our results largely
confirmed these predictions, revealing significant differences in time use by teacher
category and affective response by activity.

When comparing how different groups of teachers varied in their school and
instructional time use, we found that the special education/general education distinction
appeared to be the most important. Much of this variation is likely attributable to the

unique role that special education teachers play in schools. Their job responsibilities

43

require that they focus their instruction on meeting the needs of individual students,
which often means that less time is devoted directly to instruction, and more time is spent
on paperwork and other tasks. In fact, the amount of time required to complete such tasks
has been cited in several studies as a major reason why special education teachers leave
the classroom (Billingsley, 2004). Interestingly, special education teachers reported
spending more time than general education teachers talking with colleagues. This may
have been because special education teachers at times may have been providing
instruction to students with disabilities who were mainstreamed, requiring them to work
in general education teachers’ classrooms.

Student needs likely dictated why Special education teachers differed from general
education teachers in their instructional time use as well. Lecture was the most common
form of instruction in general education classrooms, while special educators spent
significantly more of their time reading as a class. Working with students on functional
skills was also a more frequent occurrence in special education classrooms, while it rarely
occurred in general education teachers’ classrooms. The opposite was true with respect to
time spent on classroom routines, which occurred less frequently in special education
classes.

In our comparisons of time use based on instructional level, few meaningful
differences were observed between elementary and middle school teachers. We had
assumed that, much like the special education/ general education distinction, differences
in job specification would result in significant variation between these two categories of
teachers. Elementary school teachers, for example, are often only responsible for one

group of students and must prepare instructional material across multiple subject areas,

44

while middle school teachers generally have fewer “prep periods” but more students. Our
results do reflect these differences, to a limited degree; middle school teachers spent more
than twice as much time grading as elementary school teachers, and elementary school
teachers spent more time planning, although this finding was only significant at the p<.10
level.

The most significant distinction between elementary and middle school teachers,
both in terms of school time use and instructional time use, was the number of hours
spent using computers and other technology. On average, middle school teachers spent
4.15 hours per week using technology, while elementary school teachers only devoted
1.42 hours per week; the greater use of technology by middle school teachers carried over
to instructional time as well. These findings likely reflect the increased focus on
academic content at the middle school level, where technology could have served as an
important tool in math and science instruction. To determine that these differences were
not simply a result Of middle school teachers having more time to themselves (and thus
more time to check email and browse the intemet), it would be important, in future
analyses, to use more refined measures of technology use. However, given that these
findings held across school- and instructional-time use, we suspect that technology
played an integral role in middle school teachers’ instruction.

Time-use comparison between first-, second-, and third-year teachers yielded no
significant findings. This may be attributable to small sample sizes; dividing the sample
into three groups left one category (third-years) with only 9 teachers. Alternatively, we
could view these findings as representative of an extended induction phase in which,

rather than make drastic changes from year to year, teachers gradually adapted their

45

practice as they gained more experience. Several trends in our data support this
conclusion (although none of these differences were significant): teachers with more
experience spent less time planning, more time talking with colleagues, and less time
directly engaged in instruction. Related to instructional time use, experienced teachers
spent less of their time using lecture to deliver classroom material, but spent more time
providing one-on-one instruction; they also devoted fewer hours to classroom routines.
Although these results should be interpreted with caution, they lend support to findings in
extant literature.

Our final goal for this paper was to determine the degree to which early career
teachers’ affective states were likely to fluctuate depending on their context at school, as
well as which situations during the work week were likely to elicit the strongest
emotions. When comparing the emotional states of the teachers in our sample when in
and outside of school, we found that being at work was associated with higher levels of
intellectual stimulation, as evidenced by their higher levels of concentration, challenge,
activation, and self-esteem. These heightened emotions, however, were also accompanied
by less enjoyment and increased stress, suggesting that the work experience was not
uniformly positive. That being said, these findings raise the question of whether these
early career teachers’ emotions are what we would expect given previous
conceptualizations of the beginning years of teaching as “sink or swim” or “trial by fire.”

In fact, when we looked more closely at teachers’ experiences at school, we found
that across multiple categories, instruction was associated with higher than average levels
of positive emotions. When engaged in instruction, teachers felt more challenged and

more skilled, they reported higher levels of self-esteem and enjoyment, and they felt both

46

more active and more engaged. Rather than being associated with anxiety and
uncertainty, it appears that instruction was actually a time of heightened positive
emotions. The tasks and responsibilities that are mandatory parts Of a teacher’s work day,
however, generally reflected a “comedown” from the positive emotions associated with
instruction. In these instances, teachers were less likely to enjoy their activity and often
felt less active and less engaged than their average emotional states. Meanwhile, when
teachers were engaged in social interactions other than instruction, their affective
responses were not as clear-cut. When talking with colleagues, for example, the teachers
in our study reported lower than average levels of being in control and feeling skilled. At
the same time, during such interactions they were also concentrating more and did not
appear to have lower positive affect or enjoyment scores.
Conclusion

In conclusion, this paper builds on recent research in Canada, the US, England,
and Australia that has examined the emotions of teaching and teachers’ affective
responses to their work (e.g., Day & Leitch, 2001; Dinham & Scott, 1998; Hargreaves et
al., 2006). In our analysis, we used a social-psychological framework (Bidwell, 2000;
Frank, 1998) and the Experience Sampling Method (ESM) to analyze extensive data on
42 early career teachers’ activities and their emotional responses to their activities. One
of our main findings was that teachers reported higher levels of concentration, challenge,
activation, and self-esteem in school as compared to when they were outside of school.
In addition, when involved in instruction at school (as compared to other work-related
activities), teachers reported feeling more active, more engaged, more challenged, and

more skilled; and they had higher levels Of self-esteem and enjoyment. At the same time,

47

the teachers in this sample also reported feeling less active, less engaged, and less
enjoyment when they were involved with non-instructional work-related activities at
school, such as planning, grading, or paperwork.

These findings represent an important contribution to the growing body of
literature on teachers’ emotions. Previous research has employed interviews to learn
about teachers’ affective responses to a range of work-related roles and activities,
including interactions with students and parents; serving as department chairs in
secondary schools; and responding to comprehensive reform efforts (Hargreaves, 1998;
Lasky, 2000; Little, 1996; Schmidt, 2000). In addition, scholars have used narrative
accounts and long-term observation to consider teachers’ emotional experiences in
professional development and during school inspections (Day & Leitch, 2001; Jeffrey &
Woods, 1996). But few studies have obtained data on teachers’ multiple activities during
the school day and how they respond emotionally to these various activities and the
people with whom they interact. By collecting ESM data from more than 40 teachers
over several days of teaching, this study was able to analyze teachers’ emotional states
across a range Of work-related activities. Further, by creating person-level z-scores for
our nine emotion measures, we were able to compare participants’ mean z-scores for the
nine emotions with their average emotional states across all activities.

A second key finding from this study involved the nature of the general education
teachers’ instructional practices and the relatively low amount of attention they devoted
to managing student behavior (at least as a primary activity). While we found that the
general education teachers engaged most frequently in lecture (5.89 hours/five-day

week), this represented less than half of their instructional time. In addition to lecture,

48

the teachers reported allocating substantial amounts of time to working with a single
student (3.61 hours/week), reading as a class (2.85 hours/week), and working with
students in discussion (1.74 hours/week) and small groups (1.56 hours/week). Further,
they reported spending less time monitoring student behavior (i.e., 1.43 hours/week) than
any of these instructional activities. In contrast to scholarship and research findings from
the 19803 (e.g., Goodlad, 1984; Sizer, 1984), these results suggest that the general
education teachers in this study devoted a significant amount of time to a range of one-
on-one and group activities in which students were actively engaged in reading and
discussion (Cohen, McLaughlin, & Talbert, 1993; Fenstermacher & Richardson, 2005).

Previous research has documented the general processes by which many
educators move from the survival stage of teaching (characterized by a preoccupation
with oneself, the delivery of one’s lessons, and managing student behavior) to the
mastery stage, in which teachers grow more concerned with curriculum, pedagogy, and
their own effects on student learning (Author, 2007; Burden, 1990; Fuller & Bown, 1975;
Huberrnan, 1993). Given that our sample consisted of first-, second-, and third-year
teachers, the analysis presented here suggests that as a group, the general education
teachers were moving from the survival stage to the mastery stage. In addition, this study
provides evidence that the ESM can be used to document early career teachers’
instructional and other work-related activities, thereby providing detailed data about the
specific processes and emotions that novices experience as they move towards mastery
teaching.

A third finding of importance had to do with variations between general education

teachers and special education teachers in this study. In particular, the ESM data indicate

49

that teaching assignment (i.e., one’s role and responsibilities) was strongly associated
with how the teachers spent their time and how they responded to their teaching. We
found that special educators spent significantly less time than general educators engaged
in instruction, but much more time interacting with colleagues and completing
paperwork. With regard to instructional activities, the special education teachers in the
study spent much less time lecturing (1.41 hours/week compared to 5.89 hours/week),
devoted less time to classroom routines (0.13 hours/week compared to 1.22 hours/week),
and allocated more time to functional skills instruction than the general education
teachers (3.92 hours/week compared to 0.74 hours/week). Finally, while there were few
differences between the groups in terms of emotional responses to school-based
activities, during non-instructional interactions with students, special education teachers
reported higher levels of enjoyment, engagement, and challenge.

These results build on other research that has examined the nature of beginning
special education teachers’ induction experiences and how the expectations placed on
novice special educators can differ from those faced by new general educators (Author,
under review; Billingsley, Carlson, & Klein, 2004; Whitaker, 2000). More specifically,
the findings from this ESM study suggest that novice teachers in both groups reported
more positive emotions (i.e., with regard to self-esteem, enjoyment, and feeling active,
engaged, skilled, and challenged), but that new special educators spent less time than
general educators on instruction (and correspondingly more time on paperwork and
interactions with colleagues). Given that instructional time seems to be associated with
higher levels of positive emotions (and perhaps other key teacher outcomes), it could be

useful for administrators and policy makers to examine ways to provide beginning special

50

educators with more opportunities to engage in instruction while reducing the amount of
time they must spent on paperwork and/or meetings with colleagues.

With regard to limitations, the size of our sample (n=42) prevented us from
analyzing and comparing various groups of teachers’ emotional responses (i.e., general
education vs. special education; elementary vs. middle school; first-year vs. second- and
third-year). Thus, one implication for future research on teachers’ affective responses
would be to include sufficiently large samples to examine differences between and
among groups of teachers. Second, our findings revealed that on average, general
education elementary teachers spent more time at school each day than general education
middle school teachers or special education teachers. But the data did not indicate
whether general education elementary teachers were actually devoting more time to
work-related activities at school than the other groups. Therefore, a second direction for
future research would be to investigate whether certain groups allocate more time to
work-related activities at school or at home, and what factors account for these
differences.

Third, our data indicated that the teachers spent relatively small amounts of time
managing student behavior, but the analyses reported here did not include data on the
secondary activities in which teachers were engaged. That is, teachers may have actually
devoted a much greater percentage of time to disciplining students and/or reinforcing
positive behavior, but they may have engaged in these activities at the same time that
they were lecturing, working with students one-on-one and in small groups, and/or
leading class discussion. Thus, future research could productively capitalize on data on

teachers’ secondary activities to probe the extent to which they manage student behavior

51

as a secondary activity. Finally, key research questions involving both the social and
psychological aspects of teaching remain unanswered. In particular, this study did not
address whether a) teachers’ time use, b) who they were with, and c) their emotional
responses to their activities were associated with key outcomes, such as self-efficacy,
commitment, and retention. Therefore, it will be important for future research that
employs the ESM with beginning teachers to test associations between such outcomes

and teachers’ time use, who they are with, and their affective responses.

52

Table 1: Demographics of the ESM Sub-Sample vs. Full Sample (in 3 Districts)

Appendix A

 

Demographic Characteristics of ESM Sample

Demographics of Full Sample in 3

 

Districts
Total n=42 Total n=92
Gender Gender
Female 35 83% Female 75 8 1%
Male 7 1 7% Male 1 7 19%
Grade Level Grade Level
Elementary School 27 64% Elementary School 58 63%
Middle School 15 36% Middle School 34 37%
Special Ed. Status Special Ed. Status)
General Education 26 62% General Education 67 73%
Special Education 16 38% Special Education 25 27%
Year of Teaching Year of Teaching
First 1 1 26% First 30 34%
Second 22 52% Second 38 41%
Third 9 21% Third 24 25%
Race/Ethnicity Race/Ethnicity
African-American 2 5% African-American 6 7%
Asian 1 2% Asian 1 1%
White 39 93% White 85 92%
District District
Daus (Michigan) 7 17% Daus (Michigan) 17 18%
Kaline (Michigan) 8 19% Kaline (Michigan) 20 22%
Engram (Indiana) 27 64% Lngram (Indiana) 55 60%

 

53

225.8 8. v a: we. v a... Haaz
August? 285%?

L8 3.2 9 mozmzflmi BEE: e5 m<>OZ< 83.0.8 33.6.8... 03 .8288 mo motowfiao oz: 55 0.58 mew—8&8 555 .n

.2805:me 20>» 82—23.

5252. Seances—G 550:3 2.158% 9 3.26:8 203 $33 0383. Eugene—o5 £2.88 mo 810380 95 wztanEoo 555 .m

 

8. :2 8.: 8.2 2.. 8.: 3.2 8.- 8.2 8.2 23.5 5
2. 8.2 8.8 8.8 3.. 8.8 8.: 2:- 8.8 3.: use: 2
S. 2 .8 Ne. _ m 82:. .. 8.8- 3:. S. _ m .38 8.2. 8. _ m .83 .<
.2 an: 8n: :1. A 2H2 8n: 2 21,2 81.
2 MKEKUBNN ufiN~§UBN~ hkmfibfimk MLE‘UGNN «(NQUBN'N B HLWQUGNB hkfimbfimyfi

.2 2...: .2 3.8% i 5.2 Sn .32 52 3 3%. .3 Se

 

a 3.8930 Losoaoh. .26 33>) 2.83 2: ”5:5 ”:8: wee—:3 .anoaoh. an 03.?

5:250 Ewe—Sm n Om .e

.289 «.59. e3 x33

:83 05 wESu 82o; main? :39 .8282
05 .3 Dome—8 some 5 2:: mo SEQ—9a 05
mix—3:3: .3 36.3.8 2m 8:2. wet—e3 .a

 

 

 

23 3.2 23.... 5
82 3.: use: 2
86 8.8 32.8 E
a Qm. =33

.. :83 is: 2: mats:

9:5: Mac—a3 3.2.235. "N 035—.

54

225.3 8. v a: He. v a... 9oz

 

 

8. 8.8 8.8 8.8 as. 8.8 8.8 E. 8.8 8.8
8. m: 2.8 3.8 :88 m 2. 8.. .8. 8.8 $8
8. 2.. $8 a. _A 8. _- 8.. EN :84“ S... 8.
8. a: 8.8. 88 .58 88 2... .88 m: 8.8

8.8 82 8.». 8.8 88 8.4 8.... .48- 2e 3....

8; 8.” 2% 8H 8.- Ea m3 8.. 3m :2
8. own is :2 2. _- as. 82 8. 23 as
2. 2.2 8.8 8.8 .2- 8.2 8.8 :22 8.: 8.8

k 222.638 .23er 222938 2 22.388 22338 N 28388 2833.5
.2 2...: .2 3.8% .2 E: 2.8. s: :52 em 8% E 5e

Rog—3:21—
zwo_o::oo._.\.§:anU

{cabana

8620
8:83:00 9 95:8.

3535?
dogma—:82
math—BE

c2838:—

 

Abowugau be 255. 325 he on: 3.85.35. "m 03:...

 

 

Ned wNN
«mm Sam
36. m~.N
mad mNN
5.? end
cod end
emfi and
5m.» emdm

QM. to»:

732225.—
awe—occoohtoSquU
Eczema;

wits—O

8:88:00 o. wage...
macaw-am?»
42527.52
w£=§E

:28:me

 

2:8 .923 3 a: .2225 e. o... a...

55

Table 6: Teachers' Use of Class Time

 

Lecture

Single Student
Reading as a Class
Discussion

Small Group
Quiz/Exam
Monitoring Student
Behavior
Technology
Game/Activity
Classroom Routines
Student Seatwork
Functional Skills

 

Mean SD

4.18 4.08
3.25 3.79
2.04 2.88
1.24 2.40
2.08 3.39
1.72 2.93
1.27 2.22
.94 1.75
.56 1.82
.80 1.67
.69 1.80
.84 2.57

 

56

..B..s-~. .... v ...... .8. v .... 582

 

 

2.. MN. 3.. 2.. 3.8.. m... m. .2:- mo... E.
8. on. m... .E. 8.. E. 8. 3.. w... ...-u.
no. 8. 8. o... E..- S. m... .m..~ 2. 8..
E. 8. m... o... ...... 2.. o... 8.. 8. a.
E. on. 8. 8.. bk E. E... 6.. a. R.
.... Em. 8.. an E.- E. .2 8m. 5.. a...
a: 2: as o... 8.- a... E. $.- 8.~ 8..
z... 2.. .3... «2 E... e... Ea 3... SN 8..
EN .....N Em. E. ... a. .N. E. m... E..
SN 2% 2.. 0.2 2.. E. 2N .2..~ E. a:
.3. a... ....m 2.. mm- 93 2;. E. S... .3
9... 3m 3a a... .m... 8a E... :8... .... 92
R .2238.“ Ema-gawk .239ka h Ema-onus 2233.5 & flux-dag ...-3.?ka
.8. BE .... 3.8% .: as... .3 3:. sum. 3 88-4 3 .28

2.5 .222:

foéwom
Ecuam
8525M
Eco-.330

b52353
awe—0.58%

832.5 2.255
wctozco—z

Eaxmhmso
anew =aEm

:28285

3.20
m 8 wczuaom

.5th 29.5

2303

 

3.89:5 m5 cam-n. mud—U he 3.: .Eogoaoh. "b 0...:-

57

Table 8: Teachers’ Emotional States In and Out of School
(Using Context-Specific Z-Scores) a, b
At School Other

 

 

Mean Mean T
Skilled in Activity .02 -.03 .618
Enjoying Activity -.09 .17 -279"
Concentrating .1 1 -. 16 2.55.
Challen ..
Activityng by .21 -.36 8.14
Stressed .10 -.16 2.59.
In Control -.04 .06 -1.1 1
Positive Affect .01 -.03 .384
Self-Esteem .08 -.14 2.24.
Activation .13 -.22 3.53"
Engagement .00 .01 -.12

 

a. For every individual, context-specific z-scores were calculated
by taking their average z-score in that setting.

b. When comparing differences in emotional states by context,
paired sample t-tests were conducted.

Note: *p < .05; "p < .01 (2-tai1ed).

Table 9: Teachers’ Emotional States At School (Using Context-Specific Z-Scores) '

 

Talking w/ Non-lnstuct

 

£41212:ng 13:23am" Colleagues w/Students gigging fiiirwork
Mean Mean
Skilled in Activity .00 .25” -.31” .00 .30” -.16“
Enjoying Activity -.27" .18‘ .06 -.10 -.56“ -.69”
Concentrating .07 .30" .48" -.1 1 .30 .60'
$252?“ by .07 .50” .43‘ -.22‘ .00 .95"
Stressed .10 .07 .21 -. 12 .09 .60‘
In Control .14 .13 -.48‘ .10 .01 .03
Positive Affect -.21 .16 .19 .31‘ -.48“ -.30
Self-Esteem .19 .31” -.22 .09 .24 .19
Activation -.07 .29” .18 .33" -.50‘ .17
Engagement -.19’ .33“ .28 -.18 -.20 -.24

 

a. For each school context, paired sample t-tests were conducted between a person’s situation-specific
emotional state and their average emotional state (by definition, the person-level average results in a z-

score of 0).
Note: *p < .05; "p < .01 (2-tai1ed).

58

 

COZEOEOU

 

 

 

 

 

 

mm. om. om. m m. .N. em. mm. 5N. .- .. w _. 3208.:—
G... .N... G... .N. .. E... ...... .2... ...... E... .2... 3.8.2:.
:8.— nnowd ......QN :3... SEN-N ......K. tam.— ..imo. 1.3.— ......om.~
.E... a... 6.... .... .. ...... ...... .... .. ...... a... ...... mo$3.3“.
.2. oo. N..- 2. ...... .m: m ..- Na. 2.. no_. mm: 5.3 mix—a...
EoE Eoofim Sofia. .0323. .3 mafia b_>_8< 3.3.3..
-owawcm coza>co< ....om 03.69. .8280 5 “.0325 vows-2.20 2:00:00 wEAoPm E 3:65
30:30:90 5.3 ”Ex—ah. .5280 .02—om .5 cow-am .02—um 3. 823m .acozefim .anoaoh 959:6...— fié— «SF—- 9
5
...-0.....-N. .... v ...: .m... V .... no...z
. . . . . . . . . . 5:29:00
mm am mm mm m. cm mm mm c. C 822......
G... .N... 3.... .N... E... ...... a... ...... E... a... .3225
1&0.— ..iwwd Sac-N 3.26 fine—Au tummy. inc—.— ..:wmw— 1&5.— .3..ch
.2... ...... .2... E... 3.... G... E... ...... 6.... ...... 8.8.52.
......wm. *2. «*NN. c—. ...m — . ha. 3.. ......MN. *ncm. nix-N. WEEZEA—
EoE Eooam woo-w: .0353. .3 war... $2.3. 3353‘
-owuwcm 5.33.3. ...—om 02:85. 35.80 E vommobm vow—5:20 -..—00:00 waioqcm E 005%

 

933:5»:— ufiEZth. .5280 32.5 .3 63am .eesom 3. 335 ...:ozefim ...-Banach. ”.5330...— 36— «Zah-

mlfinlcgqfl

 

 

 

. . . . . . . . . . cons—9:00

om om om mm mm mm mm R E 2 $208.:—

....... .N... ...... .. ... ...... ...... a... ...... .0... a... .5222.
EEO.— **oo.m Eye—N ......hmé .3..de ..imw. from.— **No._ 33a.— fro—N

....w .....w .m... 89W .8. .5... .....N .8. .....N ...... 252 2:83
......am 3mm co 2.; 3 S no 3 1.3 cc
E2: 808mm 800$. .338... .3 mix: b_>_uo< .0353.
-omawcm 22:333. ....om 358.. 35:00 E commobm woman—=20 -3380 9.30.8.“ E cosy—m

 

A283. 9.3.83. 3.8.30 .255 no can: 32.9w 3. 3.5m ...:ozofim .2263? ”58:5...— "2: 03a...

60

References

Achinstein, B., Ogawa, R.T., & Speiglman, A. (2004). Are we creating separate and
unequal tracks of teachers? The impact of state policy, local conditions, and teacher
characteristics on new teacher socialization. American Educational Research Journal,
41(3), 557-603.

Author (2007).
Author (2008).
Author (under review).

Bidwell, CE. (2000). School as context and construction: A social psychological
approach to the study of schooling. In M. Hallinan (Ed.), Handbook of the Sociology
of Education, pp.15-36. New York: Kluwer Academic/Plenum Publishes.

Billingsley, B. (2004). Special education teacher retention and attrition: A critical
analysis of the research literature. Journal of Special Education, 33(1), 39-55.

Billingsley, B., Carlson, E., & Klein, S. (2004). The working conditions and induction
support of early career special educators. Exceptional Children, 70(3), 333-347.

Brief, A.P., & Weiss, H.M. (2002). Organizational behavior: Affect in the workplace.
Annual Review of Psychology, 53, 279—307.

Burden, RR. (1990). Teacher development. In W.R. Houston, M. Haberman, & J. Sikula
(Eds.), Handbook of research on teacher education (pp.3l 1-328). New York:
Macmillan.

Bryk, A.S., & Schneider, BL. (2002). Trust in schools: A core resource for improvement.
New York: Russell Sage Foundation.

Cohen, D.K., McLaughlin, M.W., & Talbert, J.E. (Eds). (1993). Teaching for
understanding: Challenges for policy and practice. San Francisco: Jossey—Bass.

Csikszentmihalyi, M. (1997). Finding flow: The psychology of engagement with everyday
life. New York: Basic Books.

Csikszentmihalyi, M., & Larson, R. (1984). Being adolescent: Conflict and growth in the
teenager years. New York: Basic Books.

Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the Experience
Sampling Method. Journal of Nervous and Mental Disease, 1 75(9), 526-536.

6]

Csikszentmihalyi, M., Rathunde, K., & Whalen, S. (1993). Talented teenagers: The roots
of success and failure. New York: Cambridge University Press.

Csikszentmihalyi, M., & Schneider, B. (2000). Becoming adult: How teenagers prepare
for the world of work. New York: Basic Books.

Day, C., & Leitch, R. (2001). Teachers’ and teacher educators’ lives: The role of
emotion. Teaching and Teacher Education, 17(4), 403-415.

DiBianca, R. (2000). Teaching adolescents: Relationships between features of instruction
and student engagement in high school mathematics and science classrooms.
Chicago: University of Chicago.

Dinham, S., & Scott, C. (1998). An international comparative study of teacher
satisfaction, motivation, and health: Australia, England, and New Zealand. Paper
presented at the annual meeting of the American Educational Research Association,
San Diego.

Ebmeier, H. (2003). How supervision influences teacher efficacy and commitment: An
investigation of a path model. Journal of Curriculum and Supervision, 18(2), 110-
141.

Evans, L. (2000). The effects of educational change on morale, job satisfaction and
motivation. The Journal of Educational Change, I (2), 173—192.

Feiman-Nemser, S., Carver, C., Schwille, S., & Yusko, B. (1999). Beyond support:
Taking new teachers seriously as learners. In M. Scherer (Ed.), A better beginning:
Supporting and mentoring new teachers. Alexandria, VA: Association for
Supervision and Curriculum Development.

Fensterrnacher, G.D., & Richardson, V. (2005). On making determinants of quality in
teaching. Teachers College Record, 107(1), 186-213.

Fisher, CD. (2000). Mood and emotions while working: Missing pieces of job
satisfaction? Journal of Organizational Behavior, 21 185-202.

Fisher, C.D., & Ashkanasy, NM. (2000). Emotions in organizations. Journal of
Organizational Behavior, 21, 121—234.

Fletcher, 8., Strong, M. & Villar, A. (2008). An investigation of the effects of variations
in mentor-based induction on the performance of students in California. Teachers

College Record, [10(10), 2271-2289.

Frank, K.A. (1998). The social context of schooling: Quantitative methods. Review of
Research in Education, 23, 171-216.

62

Frank, K.A., Zhao, Y., & Borman, K. (2004). Social capital and the diffusion of
innovations within organizations: The case of computer technology in schools.
Sociology of Education, 77(2), 148-171.

Fuller, F.F., & Bown, O. (1975). Becoming a teacher. In K. Ryan (Ed.), Teacher
education (74’h yearbook of the National Society for the Study of Education, pp. 25-
52). Chicago: University of Chicago Press

Glazerman, S., Dolfin, S., Bleeker, M., Johnson, A., Isenberg, E., Hugo-Gil, J. Gruder,
M., Britton, E., Britton, E., & Ali, M. (2008). Impacts of comprehensive teacher
induction: Results from the first year of a randomized controlled study. Washington,
DC: US. Department of Education, Institute of Education Sciences.

Gold, Y. (1996). Beginning teacher support: Attrition, mentoring, and induction. In J.
Sikula, T. J. Buttery & E. Guyton (Eds.), Handbook of research on teacher education
(2nd ed., pp. 548-594). New York: Macmillan.

Goodlad, J .I. (I 984). A place called school: Prospects for the future. New York:
McGraw-Hill.

Grossman, P.L., & Thompson, C. (2004). District policy and beginning teachers: A lens
on teacher learning. Educational Evaluation and Policy Analysis, 26(4), 281-301.

Hamermesh, D., Frazis, H., & Stewart, J. (2005). Data watch: The American Time Use
Survey. Journal of Economic Perspectives, 19(1), 221-232.

Hargeaves, A. (2000). Mixed emotions: Teachers’ perceptions of their interactions with
their students. Teaching and Teacher Education, 16(8), 81 1-826.

Hargreaves, A. (2001). The emotional geographies of teaching. Teachers College Record,
103(6), 1056-1080.

Hargreaves, A., Beatty, B., Lasky, S., Schmidt, M., & Wilson, SJ. (2006). The emotions
of teaching. San Francisco: Jossey-Bass.

Hektner, J .M., Schmidt, J .A., & Csikszentmihalyi, M. (2007). Experience sampling
method: Measuring the quality of everyday life. Thousand Oaks, CA: Sage .

Henke, R.R., Chen, X., & Geis, S. (2000). Progress through the teacher pipeline: 1992-
93 college graduates and elementary/secondary teaching as of 1997 (N CES No.
2000152). Washington, DC: US. Department of Education, National Center for
Education Statistics.

Horrigan, M., & Herz, D. (2005). A study in the process of planning, designing, and

executing a survey program: The BLS American Time-Use Survey. In D. Hamermesh
& G. Pfann (Eds), The Economics of Time Use. Amsterdam: North-Holland.

63

Huberman, M. (1993). Steps towards a developmental model of the teaching career. In L.
Kremer-Hayon, J .H.C Vonk, & R. Fessler (Eds), Teacher professional development:
A multiple perspective approach (pp.93-I 18). Amsterdam: Swets & Zeitlinger.

Ingersoll, RM. (2001). Teacher turnover and teacher shortages: An organizational
analysis. American Educational Research Journal, 38(3), 499-534.

Ingersoll, R., & Kralik, J. (2004). The impact of mentoring on teacher retention: What the
research says. Denver, CO: Education Commission of the States. Available at
http://www.ecs.org/cIeariighouse/S0/36/5036.htm

Jeffrey, B., & Woods, P. (1996). Feeling deprofessionalised: The social construction of
emotions during an OF STED Inspection. Cambridge Journal of Education, 26(3),
325-343.

Johnson, S.M., & Birkeland, SE. (2003). Pursuing a "sense of success": New teachers
explain their career decisions. American Educational Research Journal, 40(3), 581-
617.

Johnson, S.M., Kardos, S.M., Kauffman, D., Liu, E., & Donaldson, M.L. (2004). The
support gap: New teachers’ early experiences in high-income and low-income
schools. Education Policy Analysis Archives, 12(61). Retrieved from
http://epaa.asu.edu/epaa/v12n6l/.

Kapadia, K., Coca, V., & Easton, J .Q. (2007). Keeping new teachers: A first look at the
influences of induction in the Chicago Public Schools. Chicago: Consortium on
Chicago School Research, University of Chicago.

Kardos, S.M., Johnson, S.M., Peske, H.G., Kauffman, D., & Liu, E. (2001). Counting on
colleagues: New teachers encounter the professional cultures of their schools.
Educational Administration Quarterly, 37(2), 250-290.

Koh, C. (2005). The everyday emotional experiences of husbands and wives. In. B.
Schneider & L. J. Waite (Eds), Being together working apart: Dual-career families
and the work-life balance (pp. 169-189). New York: Cambridge University Press.

Krantz-Kent, R. (2008). Teachers' work patterns: When, where, and how much do U.S.
teachers work? Monthly Labor Review, [31(3), 52-59.

Lasky, S. (2000). The cultural and emotional politics of teacher-parent interactions.
Teaching and Teacher Education, 16(8), 843-860.

Lasky, S. (2005). A socio-cultural approach to understanding teacher identity, agency,
and professional vulnerability in a context of secondary school reform. Teaching and
Teacher Education, 21(8), 899-916.

64

Little, J .W. (1996). The emotional contours and career trajectories of (disappointed)
reform enthusiasts. Cambridge Journal of Education, 26(3), 345-359.

Lortie, DC. (1975). Schoolteacher: A sociological study. Chicago: University of Chicago
Press.

Prescott, S. Csikszentmihalyi, M., & Graef, R. (1981). Environmental effects on
cognitive and affective states: The experiential time sampling approach. Social
Behavior and Personality: An International Journal, 9(1), 23-32.

Robinson, J.P. (1985). The validity and reliability of diaries versus alternative time use
measures. In F. T. Juster & F. P. Stafford (Eds), Time, goods, and well-being (pp. 33-
62). Ann Arbor, MI: University of Michigan, Survey Research Center, Institute for
Social Research

Rowan, B., Cambum, E., & Correnti, R. (2004). Using teacher logs to measure the
enacted curriculum: A study of literacy teaching in third-grade classrooms.
Elementary School Journal, 105(1), 75-101

Rowan, B., Harrison, D.M., & Hayes, A. (2004). Using instructional logs to study
mathematics curriculum and teaching in the early grades. Elementary School Journal,
105(1), 103-127.

Schmidt, M. (2000). Role theory, emotions, and identity in the department headship of
secondary schooling. Teaching and Teacher Education, 16(8), 827-842.

Schneider, B., & Waite, L.J. (Eds.). (2005). Being together, working apart: Dual-career
families and the work-life balance. Cambridge, UK: Cambridge University Press.

Schweinle, A., & Turner, J. (2006). Striking the right balance: Students’ motivational
experiences and affect in upper elementary mathematics classes. Journal of
Educational Research, 99(5), 271-293.

Shemoff, D., Csikszentmihalyi, M. Schneider, B., & Steele-Shemoff, E. (2003). Student
engagement in high school classrooms from the perspective of flow theory. School
Psycholog/ Quarterly, 18, 158-176.

Shemoff, D., & Schmidt, J .A. (2008). Further evidence of an engagement-achievement
paradox among US. high school students. Journal of Youth and Adolescence, 37(5),
564-5 80.

Shemoff, D., & Vandell, D.L. (2007). Engagement in after-school program activities:
Quality of experience from the perspective of participants. Journal of Youth and
Adolescence, 36(7), 891-903.

65

Schmidt, J .A., Shumow, L. & Kackar, H. (2007). Adolescents' participation in service
activities and its impact on academic, behavioral, and civic outcomes. Journal of
Youth and Adolescence, 36, 127-140.

Sizer, T.R. ( I 992). Horace ’s compromise: The dilemma of the American high school.
Boston: Houghton Mifflin.

Smith, T.M., & Ingersoll, RM. (2004). Reducing teacher turnover: What are the
components of effective induction? American Educational Research Journal, 41(3),
681-714.

Veenman, S. (1984). Perceived problems of beginning teachers. Review of Educational
Research, 54(2), 143-178.

Weiss, H.M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion
of the structure, causes and consequences of affective experiences at work. Research
in Organizational Behavior, 8, 1-74.

Whitaker, SD. (2000). Mentoring beginning special education teachers and the
relationship to attrition. Exceptional Children, 66(4), 546-566.

Williams, K.J., & Alliger, G.M. (1994). Role stressors, mood spillover, and perceptions
of work-family conflict in employed parents. Academy of Management Journal,
3 7(4), 837-868.

Yair, G. (2000). Educational battlefields in America: The tug-of-war over students’
engagement with instruction. Sociology of Education, 73(4), 247-269.

66

CHAPTER 3: THE RELATIONSHIP BETWEEN EARLY CAREER TEACHERS’
AFFECTIVE RESPONSESE TO WORK AND THEIR LEVELS OF COMMITMENT
AND BURNOUT
Introduction

In addressing the high rates of attrition in the teaching workforce, researchers
have increasingly looked to the loss of beginning teachers, or what Ingersoll refers to as
the “revolving door” of teaching (2001a, 2001 b). Using nationally representative data
from the 1993-94 Schools and Staffing Survey (SASS), he found that in addition to
personal characteristics (such as age and teaching assignment), aspects of a teacher’s
work environment are also significant predictors of turnover. As Johnson and her
colleagues have found, many teachers enter the profession with a “tentative commitment”
to teaching; decisions regarding teaching as a career are based in large part on their
experiences within their school environment, including whether they feel supported by
colleagues and administrators (Johnson et al., 2004; Peske, Liu, Johnson, Kauffman, &
Kardos, 2001).

In recent years, several empirical studies have attempted to establish connections
between specific characteristics of schools and districts and relevant outcomes such as
teacher commitment, teacher retention, and student performance. One principal area of
interest is the role of mentoring programs--the establishment of a formal relationship
between a beginning teacher and an experienced teacher in order to aid in the novice’s
development. In their analysis of 1999-2000 SASS data, Smith and Ingersoll (2004)
found that first-year teachers who participated in formal mentoring programs were less

likely to leave the profession. In addition, using data from the California Department of

Education, Strong and Fletcher (2004) found a positive association between having full-

67

release mentors and student learning gains. In contrast, Glazerman and colleagues tested
the effects of two comprehensive teacher induction programs on teacher retention,
teacher practices, and student test scores for teachers in 418 schools in 17 large urban
districts (Glazerman et al., 2008); central to both induction programs was a “carefully
selected and full-time mentor.” The authors found no significant differences between
schools implementing comprehensive induction programs and the control schools on any
of the three outcomes of interest, raising doubts about the potential impact of formal
induction programs.

Glazerman’s findings are in line with research suggesting that induction programs
relying on formal mentoring may not capture the important ways that other colleagues
within schools play a role in the socialization of teachers. A recent study conducted by
the Consortium on Chicago School Research found a stronger effect for informal support
from colleagues than formal mentoring programs on novice teachers’ career decisions
(Kapadia, Coka, & Easton, 2007). And, in addition to their findings regarding formal
mentoring, Smith and Ingersoll also suggest that forms of induction that take into account
the social relationships accessible to new teachers (e.g., having a common planning
period with other colleagues, collaborating with other teachers on instruction) increase
the rate of retention of new teachers.

Research at the local level confirms the importance of informal support from
colleagues. Using interviews with first- and second-year teachers in Massachusetts,
Johnson and Birkeland (2003) followed teachers across three years of teaching, finding
that teachers were more likely to stay in the profession if their schools were organized in

a way that supported their teaching. Schools which promoted frequent interactions

68

between colleagues were found to be more successful at retaining teachers. In other
analyses with the same data, Kardos and her colleagues (2001) found that in schools with
integrated professional cultures —-where beginning teachers received frequent support
and interacted with colleagues across experience levels—novice teachers felt most
supported. In both of the above studies, the participants stressed the importance of
support from principals, a finding which is in line with other research on new teachers
(Youngs, 2007).

In sum, existing research suggests that there are numerous ways in which work
conditions can play a role in early career teachers’ retention decisions. However, little
evidence exists which explores how teachers experience these conditions, as well as how
these experiences in turn influence career decisions. It is conceivable, for example, that
two teachers could react to the same set of organizational conditions in two different
ways; whereas colleague support may be the critical factor for one teacher, feelings of
success related to student achievement may be more important for another. The question
then becomes how do organizational features translate into judgments about staying in
teaching (as well as staying within a school more specifically)? What is the process by
which teachers interpret aspects of theirjob environment and then make decisions about
whether to stay or leave?

This paper advances a framework for understanding the process through which
the organizational conditions of schools influence teacher outcomes, drawing on Weiss
and Cropanzano’s (I996) Affective events theory (AET). I argue that existing research
ignores an important variable that may mediate the relationship between school

characteristics and teacher retention decisions. Specifically, in addition to the analysis of

69

features of teachers’ social environments, researchers should also consider how teachers’
experience these conditions in their daily lives, as well as how they respond emotionally
to these experiences. From this perspective, aspects of the school organization (such as a
formal mentoring program or an informal support network) do not influence retention
decisions directly, but instead do so by shaping teachers’ experiences at school. Their
affective responses to daily events—when aggregated—are hypothesized to interact with
teachers’ attitudes about their jobs, and eventually, influence their decision whether or
not to stay in the profession.

To test this framework, I collected data on the affective states and attitudes of 42
early career teachers in Michigan and Indiana. I investigated teachers’ momentary
affective responses to experiences in schools by using a unique time sampling method
known as the Experiencing Sampling Method (ESM). These data were then used to
predict survey responses measuring teacher attitudes, including burnout and commitment
to one’s current teaching position and to the profession.

In the next section of the paper, I provide an overview of existing literature on the
relationship between teachers’ emotional and attitudinal responses to their work and
relevant outcomes such as student achievement and teacher retention decisions.
Following this section, I provide an overview of Weiss and Cropanzano’s Affective
events theory and describe how the theory helps to potentially explain the relationship
between workplace conditions, teachers’ emotions and teachers’ summative attitudes
about their work. In the third section, I discuss my method, including sample, ESM
questionnaire, survey questionnaire, and measures used in the analysis. The fifth and

sixth sections feature the results of this analysis and some interpretation of the study

70

findings. Finally, I conclude by discussing the implications of this study for research,
policy, and practice.
Literature Review

Research on teachers’ affective responses to their work has typically focused
either on the direct emotional consequences of aspects of teaching, such as how teachers
respond to new school policies or interactions with students, or on teachers’ attitudinal
evaluations of their work (e.g., their reflections on job manageability, stress and burnout,
and commitment to the job). In a sense, these strands of research can be thought of as the
short- and long-term expressions of teachers’ affective states, or alternatively, as local
and global dimensions of attitudes about work. A primary goal of both lines of research
has been to understand how teachers respond to specific aspects of their work conditions,
as well as how their evaluations about teaching translate into important teacher and
student outcomes.

Teachers’ Emotional Responses to Work. Hargreaves and his colleagues have
contributed several important studies on how teachers respond emotionally to features of
their work environment—as well as how responses vary by career stage, teaching
context, and other key factors (Day & Leitch, 2001; Dinham & Scott, 1998, 2000;
Hargreaves, 2001; Hargreaves et al., 2006; Jeffrey & Woods, 1996; Little, 1996). Using
interviews with elementary and secondary school teachers in Canada, Hargreaves (2001)
introduced the concept of “emotional geographies of teaching” as a means for
understanding the ways in which teachers respond emotionally to the conditions and
interactions of their work; Hargreaves defines these emotional geographies as: “the

spatial and experiential patterns of closeness and/or distance in human interactions and

71

relationships that help create, configure, and color the feelings and emotions we
experience about ourselves, the world and each other” (2001, p.106 1 ). Thus, emotional
geographies shape the kinds of interactions teachers have with other individuals at school
and have important consequences for the nature of their work. For example, Hargreaves
found that whether teachers and their students’ parents were from similar cultural,
linguistic, and/or socio-economic backgrounds had important consequences for how
teachers responded emotionally to interactions with parents, which in turn shaped the
nature and frequency of future interactions.

Hargreaves and colleagues have extended the study of teachers’ emotions to other
aspects of teachers’ work lives, including their emotional responses to interactions with
students (Hargreaves, 2000) and to serving as department chairs (heads) in secondary
schools (Schmidt, 2000). In addition, researchers have investigated secondary school
teachers’ responses to comprehensive school reforms (Lasky, 2005; Little, 1996); Lasky,
for example, found that these reforms frequently constrained teachers’ sense of agency.
These studies represent important advances in research on teachers’ emotions because
they acknowledge that teachers’ affective responses are likely to vary depending on their:
a) schooling level, b) work context (e.g., interacting with students versus interacting with
parents), c) role (e.g., teacher, department chair, etc.), and d) reform context.

Teachers ’ Attitudes about Work. Separate from research on teachers’ emotions,
several studies have attempted to tease out the causes and consequences of teachers’
summative attitudes about their work. A common response to overwhelming or negative
job circumstances is a high level of stress, which when unaddressed over a long period of

time, can undermine teachers’ satisfaction with their work. The term “burnout” has been

72

used to describe the results of this process. According to Maslach and Jackson (1 981),

burnout can be defined along three dimensions: emotional exhaustion, depersonalization,

and reduced personal accomplishment“). Their Maslach Burnout Inventory has been a
useful tool for analyzing how a negative work experience impacts teachers, and allows
researchers to place teachers on a continuum of burnout. Research confirms that stress
and burnout are problematic for the physical and psychological health of teachers
(Chemiss, 1995; Gold, Roth, Wright, & Michael, 1991; Guglielni & Tatrow, 1998).
Additionally, recent studies confirm that for experienced teachers (Weisberg & Sagie,
1999) and, for early career teachers (Goddard & Goddard, 2006), burnout is associated
with intentions to leave the profession. And for special education teachers, burnout is
predictive of attrition (Miller, Brownell, & Smith, 1999; Singer, 1993).

A second relevant attitudinal outcome in educational research is commitment—
both to one’s organization and to the profession. Ingersoll (2001b) argues that
commitment is a more relevant outcome variable for policymakers than retention itself,
because commitment more closely reflects attitudes about one’s job; retention decisions,
meanwhile, may result from circumstances not having to do with work itself (e.g., the
desire to move out of the area, start a family, or to pursue additional schooling). That
being said, commitment has been shown to be predictive of career decisions (Weisberg &
Sagie, 1999; Weiss, 1999). Commitment also has been a useful construct for
understanding how characteristics of individuals’ work environments predict intentions to

stay in the profession. Ebmeier (2003) has tied teacher commitment to teacher effort and

 

10 In the context of teaching, emotional exhaustion exists when a teacher feels like they do not have the
emotional resources left to continue their work. Depersonalization involves feeling detached from work
and as well as from interactions with their students. When teachers feel a sense of reduced personal
accomplishment, they do not feel like their efforts at work are worthwhile.

73

effectiveness, and his findings also suggest that commitment is predicted by relationships
with teacher colleagues, a finding supported in other research (Bryk & Schneider, 2002;
Desimone et al., 2002).

Theoretical Framework

Relevant to the study of early career teacher attrition is literature from the field of
organizational behavior. Researchers in this area have paid close attention to the
definition of job satisfaction, particularly its causes and consequences. Emerging from
this research is agreement that it is necessary to distinguish between individuals’
emotional responses to their work and their evaluations about the job. As Weiss and
Cropanzano (1996) suggest, “satisfaction is an evaluative judgment about one’s job that
partly, but not entirely, results from emotional experiences at work. It also partly results
from more abstract beliefs about one’s job” (p.2); similar definitions of job satisfaction
are offered elsewhere in the literature (Crites, Fabrigar, & Petty, I994; Eagly & Chaiken,
I993; Petty, Wegener, & Fabrigar, I997).

The distinction between emotional responses to one’s job and evaluative
judgments about work is made clear in Weiss and Cropanzano’s affective events theory,
where they focus directly on how individuals’ affective reactions to work events
influence job satisfaction, independent from the features of one’s work environment. In
defining “affective reactions,” Weiss and Cropanzano are careful to distinguish between
emotions and moods. Drawing on research by Frijda (1993) and Morris (1 989), the
authors suggest that emotions are responses to specific events, and are thus likely to be
intense but short-lived. Moods, in contrast, reflect more general affective states and are

more likely to be detached from specific work events. Both categories of affective

74

reactions are thought to consist of positive and negative dimensions, although other
categories of emotions are also presented: anger, fear, joy, love, sadness, and surprise
(p.21). Additionally, AET suggests that both emotions and moods are influenced by
experiences at work, and are thought to have direct consequences for work attitudes.

Weiss and Cropanzano (1996) acknowledge that work environments may have a
direct influence on attitudes about one’s job, yet they argue that more frequently the
process by which environmental features influence attitudes is indirect—the work
environment makes affect-producing events more or less likely. Research confirms the
important role of characteristics of the work environment. LeFevre (I 988) concluded that
emotional well-being at work varied by position, with managers and engineers reporting
higher levels of motivation, concentration, and creativity than clerical workers and
assembly line workers. Studies by Larson and Richards (1994) and Basch and Fisher
(1 998) have documented that emotional experiences at work depend on the activities in
which individuals engage. Further, Basch and Fisher found that the same types of events
do not always produce the same emotions across participants, suggesting that “event-
emotion connections” are individual-specific.

Finally, affective events theory suggests that emotional responses to work,
coupled with abstract beliefs about one’s job, influence overall judgments about job
satisfaction. Thus, individuals make evaluations based not only on beliefs about
characteristics of their jobs (e.g., the degree to which the job fosters independent
decision-making), but also based on their recall of past experiences. Weiss and
Cropanzano admit that the way in which individuals recall affective events is unclear, yet

they suggest that individuals are likely to rely on these events because they are concrete

75

and tangible (p.50). This viewpoint is supported by research comparing affect-based and
belief-based evaluations of work, which finds that both are important components of
attitudes (Breckler & Wiggins, 1989; Edwards, 1990).

Empirical data supports Weiss and Cropanzano’s assertion that affective
responses are associated with overall job satisfaction (Fisher, 2000, 2002; Weiss,
Nicholas, & Daus, 1999). Fisher (2000) considered the relationship between real-time
emotions during the work day and overall job satisfaction, using the experience sampling
method. She found that momentary reports of positive and negative emotions at work
were associated with overall job satisfaction, and she also reported that the frequency of
positive emotions was a better predictor of satisfaction than the intensity of emotions.
Weiss and colleagues (1999), meanwhile, collected data on 24 managerial workers 4
times a day over 16 days; their results suggest that average levels of pleasantness are
correlated with overall job satisfaction.

Purpose

Existing research in organizational behavior suggests that the organizational
characteristics of schools play an important role in determining how teachers evaluate
theirjobs, while affective events theory provides a framework for understanding how
individuals’ emotions mediate this relationship. Despite the prominence of this line of
research in organizational behavior, similar studies are missing in education. This study,
therefore, tests the usefulness of applying the affective events theory to an analysis of
factors predicting burnout and commitment among early career teachers. Particularly, I

focus on the predictors of affective responses at work, and investigate whether these

76

emotional experiences have consequences for early career teachers’ attitudes about their
work.

I also attempt to test two different conceptual definitions of how emotions might
influence attitudes about work. First, I use person-level means, representing participants’
average emotions at work during the school week; this definition of affective responses is
closely in line with Weiss and Cropanzano’s definition of mood, and is likely to reflect
participants’ overall affective state throughout the week—detached from any specific
events. A second aggregation strategy was to investigate the relationship between
participants’ “peak” emotions and their attitudes about work. Here, it was presumed that
rather than overall mood at work, it would instead be extreme emotional responses that
would influence teachers’ attitudes about their jobs.

Drawing on a sample of 42 early career teachers in Michigan and Indiana, I
analyze data on teachers’ experiences using the Experience Sampling Method, and
explore the associations between these experiences and global measures of job
attitudes—which are drawn from survey responses of the same teachers. Specifically, the
study was designed to address the following research questions:

1. Do affective responses to work (i.e., their reported levels of positive affect,
negative affect, fatigue, and skill) predict teachers’ levels of burnout,
controlling for teachers’ prior levels of burnout?

2. Do affective responses to work predict teachers’ levels of commitment,
controlling for teachers’ prior levels of commitment?

3. Which ESM-based measure of teachers’ emotions is more strongly associated

with global measures of teacher attitudes--aggregated means of participants’

77

emotions’ or aggregates based on teachers’ maximum reported levels of
emotions?
Method
Sample
The sample consisted of 42 first-, second-, and third-year K-8 teachers from three
school districts in Michigan and Indiana. The teachers in this sample were all participants
of a larger, three-year study of early career teachers in several urban districts in the two

states; the sample of the larger study is approximately 240 early career general education

and special education teachers.11 The three districts in the current analysis were recruited

because they are all medium to large in size and were hiring large numbers of early career

teachers.12 The three districts—Daus, Kaline, and Engram—ranged in enrollment from
9,448 to 19,055 students in grades K-12, with Daus being the largest of the three. The
districts were all similar in the percentage of their students eligible for free or reduced-
price lunch, ranging from 42% to 59%. Finally, the districts varied in the percentage of
their students who were racial minorities, ranging from 12% in Daus to 50% in Kaline.
Table 11 summarizes enrollment and demographic data from the three districts in 2006-
07.

[Insert Table 11 here]

 

11 The larger study is funded by the Carnegie Corporation of New York and examines how mentoring,
social networks, and district policies are associated with early career teachers’ commitment, retention, and

instructional practices, as well as student learning gains. This study began in 2006-2007 and continues
through the 2008-2009 school year.

Given the tight fiscal climate m both states, the number of districts hiring new teachers was limited. As a
result, the districts in the sample look different from one another in terms of size, percentage of minority
students, and percentage of low-income students.

78

In selecting teachers to participate in this study, we recruited general and special
education teachers who met the following criteria: they were teaching full-time, they had
earned a standard teaching certificate, and they had completed university-based teacher
preparation programs. Additionally, we only included teachers who provided instruction
in the core content areas in grades 1-8. For the general education teachers in our sample,
this meant that they either provided instruction at the elementary school level (grades 1—
5) or taught at the middle school level in the areas of language arts, history/social studies,
mathematics, and science. Of the special education teachers, we only included teachers
who provided academic instruction in grades 1-8 (i.e., we excluded individuals who did
not provide instruction, such as school psychologists and speech pathologists).

The sample included 26 general education teachers and 16 special education
teachers. The teachers varied in their years of experience, with a larger number of
second-year teachers (22) than either first-year (I I) or third-year teachers (9). Thirty nine
of the teachers in the sample were white and 35 of the 42 teachers were female, however,
both of these numbers are in line with national averages. Finally, the sample included a
disproportionate percentage of teachers (64%) from the Indiana district, Engram. The
demographic characteristics of the teachers in the ESM sample are summarized in Table
12; to allow for comparison with the teachers who participated in the larger study, the
table also includes demographic characteristics for all participants in the three districts.

[Insert Table 12 here]
Data and Measures
This study used data from two different sources, including surveys at two time

points--in Fall 2007 and again in Spring 2008--and data from a signal-contingent time

79

sampling method known as the Experience Sampling Method (ESM). The surveys
included questions related to the teachers’ backgrounds, professional preparation,
instructional practices, and the nature and frequency of their interactions with mentors
and colleagues, and experiences in their schools. Of particular interest to the current
study were the questions that asked early career teachers’ about the degree to which their
work had left them feeling burned out, as well as questions about their commitment to
their school, district, and to the profession more generally. These questions were asked on
both the fall and spring surveys, allowing us to track change across the school year.

The ESM is a week-long data collection activity in which participants respond to
signaling devices that are pre-programmed to beep eight times a day for seven days,
randomly within two-hour time intervals (Prescott, Csikszentmihalyi, & Graef, 1981;
Csikszentmihalyi & Larson, 1984; Csikszentmihalyi, 1997). The ESM differs from other
time diaries because it also measures participants’ primary and secondary activities, their
thoughts, their location, and their affective experiences at the time they were signaled.
Because the data is collected at random, and because participants provide multiple
responses to the same measures, the ESM can limit the recall error associated with other
survey methods; further, it has been shown to have high levels of reliability and validity
(Csikszentmihalyi & Larson, 1984; Hektner, Schmidt, & Csikszentmihalyi, 2007;
Robinson, 1985).

The ESM questionnaire and response procedures were adopted from Schneider
and Waite’s 500 Family Study (2005). All ESM participants participated in the ESM
process during the same school week in May of 2008, beginning on a Sunday and

continuing through the following Saturday. Upon entering the raw data into a data file,

80

two trained coders, using coding schemes adopted from previous ESM studies, coded

each of the open-ended questions about participants’ activities, locations, and thoughts; ‘3
inter-rater reliability for the coding of ESM activities ranged from .75 to .90. In order to
focus specifically on times when teachers were at work, two steps were taken to restrict
the total number of teachers’ responses. First, I excluded data from Saturday and Sunday,
and instead only focused on the 5-day work week (Monday-Friday). Second, I only
included teachers’ responses when they were coded as being at school, which resulted in
using 773 of the 1233 observations across all teachers.

Teachers’ Emotions at School. The ESM data provide a rich and informative
picture of individuals’ daily experiences, allowing for the measurement of a variety of
emotions across multiple contexts. For the current analysis, I focused on teachers’
emotions at work, employing a total of four emotion variables. In prior research testing
Affective events theory, positive affect and negative affect at work have been shown to
consistently predict job satisfaction (e.g., Weiss et al., 1999; Fisher, 2000, 2002). ESM
studies have made frequent use of these two constructs as well (Csikszentmihalyi &
Larson, 1984; Csikszentmihalyi, Rathunde, & Whalen, 1993; Koh, 2005). Thus, I follow
their strategies for constructing composite variables for these two measures.

The composite variable for positive affect was constructed from the mean of the
scale variables for happy, cheerful, friendly, and relaxed; negative affect was the mean of
the scale variables for angry, frustrated, irritated, lonely, nervous, worried. Both

composite variables had an internal consistency (Cronbach ’3 alpha) of higher than .80.

 

13 These categories (as well as the categories for instructional time use) were developed based on pilots of
the ESM in 2007-08 as well as in-depth interviews with pilot participants. See Jones & Youngs (2008) for
more details on the pilot.

8]

Two single scale measures of emotion were also included in the analysis. First, in
predicting teacher levels of burnout, I wanted to include a local measure of fatigue, an
emotion not represented in the negative affect measure. 1 therefore included the scale
variable that asked whether participants felt tired at the time they were beeped. I was also
interested in whether feeling skilled during activities at school was likely to lead to higher
levels of commitment (or conversely, lower levels of burnout). Therefore, each of my
final models also included the variable skilled. Table 15 in the appendix provides
information on the descriptive statistics for these variables, as well as information about
their construction.

To allow for regression analyses, it was necessary to develop person-level
aggregates of the four ESM emotion variables.l4 Two definitions of affect were
developed, a person-level mean and a person-level maximum. Person-level means and
maximums were created using only the beeps when teachers were at school. For the
person-level maximums, 1 first calculated beep-level z-scores, which represented how far
from their average emotional states did participants’ vary when they were beeped at a

given time point. Then, for every participant, I calculated their maximum z-score15 when

16
at work.

 

l4 . . . . . .
In correlations wrth job satisfaction, person aggregates of mood and emotion have been shown to
provide comparable results to multi-level models (see Fisher, 2000).
5

For the “maximum” variable, I used z-scores rather than raw scores because across the sample,
participants made full use of the response categories. As a result, when person maximums were calculated
using the raw scores, there was little variation between individuals. In contrast, the z-scores produce a
more accurate reflection of extreme deviation from person averages.

16 Although the activities that prompted the “maximum” emotions are not the focus of this analysis, Table
17.1-17.4 in the Appendix provides a list of teachers’ activities when they reported their highest levels of
positive and negative affect, relative to their mean levels.

82

Teachers’ Work Attitudes. To measure teachers’ evaluations of their jobs, I
created composite variables for two concepts frequently employed in literature on teacher
attrition: burnout and commitment; separate composites were made for the fall survey
and for the spring survey. The burnout variable was based on the Maslach Burnout
Inventory and consisted of the following items: I feel emotionally drained from my work,
I feel used up at the end of the workday, I feel fatigued when I have to get up in the
morning and face another day on the job, I feel burned out from my work, I feel
frustrated by my work, I feel I’m working too hard on my job. Responses on these single
measures ranged from I=strongly disagree to 4=strongly agree (a = .91); the fall
composite variable for burnout had a mean of 2.52 and a standard deviation of .74, while
the spring composite variable for burnout had a mean of 2.53 and a standard deviation of
.75. While it appears that average levels of burnout did not change from fall to spring,
when prior burnout was regressed on spring burnout, the residuals for the sample did not
violate assumptions of normality.

In order to reflect the various ways in which teachers might conceptualize their
future career plans, three measures of commitment were developed. This was an
important step because teachers may have based their evaluations of their career plans on
their satisfaction with their school or district specifically, or on their attitudes about their

subject area or grade more generally. Thus, I calculated variables reflecting a teacher’s
commitment to their school, to their grade or subject area,17 and to their district.
Composite variables for each definition of commitment were created by calculating the

mean values of teachers’ plans for the upcoming school year and their plans for five years

 

17 . . . . .
Elementary school teachers were asked about their commitment to their grade, while middle school
teachers were asked about their commitment to their subject area.

83

from now (see Table 15 in the Appendix for means and standard deviations for each
commitment measures).
Of the three measures of commitment, only commitment to grade/subject was

correlated with the emotion variables of interest, once controlling for prior levels of

commitment.18 Consequently, this was the only definition of commitment included in the
final analyses. One could argue that the commitment to grade/subject variable may not
fully encompass the ways that teachers think about their career plans. However, of the
three definitions, commitment to grade/subject most likely reflects a strong overall
measure of commitment to the profession, as it asks teachers to comment on their career
plans irrespective of their current school and district. Responses on the commitment to
grade/subject ranged from 1=strongly disagree to 5=strongly agree; the mean for fall
commitment was 4.14 with a standard deviation of 1.14, while the mean for spring
commitment was 4.08 with a standard deviation of 1.02.

Teacher Demographics. The models also included a series of variables indicating
characteristics about the teachers in the sample, including whether they taught general
education or special education, whether they were in middle school or in elementary
school, and how many years of teaching experience they had. Teaching experience was
classified as a binary variable, with second- and third-year teachers being collapsed into
one category; this step accounted for the unique factors facing first-year teachers,
including negotiating new relationships with colleagues and students, and implementing
curriculum for the first time. Finally, teacher race and gender were included in the

models, although the majority of the teachers in the sample were white females.

 

‘8 The correlations between each of the emotion variables and the study’s dependent variables can be
found in Table 16 in the Appendix.

84

Analytic Approach

In this study, I investigated the relationship between in-the-moment emotional
responses to teaching and attitudinal evaluations of work, in the form of teachers’
reported levels of burnout and commitment to their grade or subject area. Although this
paper argues that momentary affective responses (i.e., local measures of emotions and
mood) become aggregated into attitudinal evaluations of individuals’ work (i.e., global
measures of job satisfaction), an argument could also be made that equally important is

the impact of the global on the local, or howjob satisfaction influences an individual’s

daily affective states.19 To address this concern, the ordinary least squares regression

models in this analysis controlled for teachers’ prior levels of commitment and burnout

(as reported in the Fall 2008 survey).20 In addition, each model controlled for several
demographic characteristics (i.e., whether the teacher taught special education, whether
they were in elementary school or middle school, and whether they were a first-year
teacher). Race and gender were also included in the regression models.

Analyses were conducted separately for each emotional predictor of interest. I
could have included multiple emotions in the same model; however, I decided to run
models with one emotion at a time, given the moderate correlation of many emotional

variables to one another. Thus, for every emotion, I investigated its relationship with

 

19 Fisher (2002) has attempted to address this issue by testing all components of AET using structural
equation analyses, a methodological strategy which allows one to test the direction of relationships. Her
results confirm the role of positive and negative emotions in predicting job satisfaction, but she also
suggests that disposition and characteristics of the work environment are predictive of affective responses.

20 As described by Allison (1990), an alternative approach for controlling for the dependent variable at a
prior time point is a change model where the dependent variable is represented as Y2 — Y.. I analde each
of the models in this analysis using change scores as outcomes. In most cases, the coefficients for the
significant predictors were similar across models. However, unlike in the models that controlled for prior
commitment, the models using change scores produced higher coefficients for “skilled” in predicting spring
commitment.

85

burnout and commitment respectively; I also compare the two aggregations of emotions
(“mean” and “max”) to determine which better predicted teacher attitudes.
Results

Comparing Mean and Max Constructs of Teacher Emotions. One of the goals of
this study was to determine whether teacher emotions were better estimated using mean
variables (the average of teachers’ responses across the ESM data collection) or max
variables (the maximum value that teachers reported during the week, relative to their
overall means). Thus, for each of the four emotional predictors (positive affect, negative
affect, fatigue, and skill), I ran separate regression models using both mean and the max
constructs. For the purpose of comparison, I have included the coefficients for both sets
of aggregate measures in Tables 17.1 — 17.4 in the Appendix.

In all models, the “mean” aggregates were stronger predictors of both burnout and
commitment than the “maximum” aggregates; as these comparison tables indicate, no
“maximum” predictor was significant at p S .10 for burnout or commitment. Further, in
some cases, the coefficients changed direction when using maximum values, often in
ways that did not make intuitive sense. For example, while mean negative affect was
positively associated with spring burnout, the association between maximum negative
affect and spring burnout was negative but was not significant. These results suggest that
average emotional states across the week were more salient for teachers’ job attitudes
than peak emotions at single time points. Also, given the relatively weak correlations
between the mean and maximum aggregates, this would seem to indicate that estimates of

emotions at any one time point were unlikely to represent teachers’ overall moods.

86

Consequently, the final models in Table III (Burnout) and Table IV (Commitment
to Grade/Subject) only include the mean constructs of positive affect, negative affect,
skill, and fatigue.

Teacher Emotions and Burnout. As shown in Table 13, the results of this analysis
indicate that even when controlling for prior levels of burnout, there is a moderate
association between two of the four emotional variables and burnout. Negative affect (the
composite of the degree to which participants felt angry, frustrated, irritated, lonely,
nervous, stressed, and worried) had a standardized coefficient of .46, and was significant
at p S .01, while “tired” was statistically significant at p S .05 and had a standardized
coefficient of .35; the R2 for the two models were .53 and .47 respectively. Further,

although they were not included in the final models, interaction terms between each of

the emotions and prior burnout were not associated with spring levels of bumout.” This

suggests that regardless of whether teachers felt burned out in the fall, negative emotions

during the year were likely to lead to higher levels of burnout at the end of the year.
[Insert Table 13 here]

The remaining two emotion variables, positive affect (the mean of teachers’ levels
of feeling happy, cheerful, friendly, and relaxed) and “skilled”—had negative coefficients of
-.12 and -.13 respectively, but neither was significant at p S .10; each model had an R2 of
.38. Despite the relatively small coefficients in each model, the direction of these
relationships, however, is suggestive of a potential mediating role of positive emotional

experiences in reducing teacher levels of burnout. Overall, the data indicate that the

 

2' Tables 18.1-18.2 show the final models with interaction terms.

87

negative emotions contributed to burnout more than either positive affect or feeling
skilled.

Of the variables representing teacher characteristics, only the dummy variable for
teaching special education was associated with teacher burnout independent of the
emotional variables and prior burnout. Only in the model that included the variable
“skilled” was special education status significant at p S .10; yet in all five models, the
standardized coefficient for teaching special education was at least .20. Interaction terms
between special education and each of the four emotion measures were non-significant
when included in each of the models, however, suggesting that the emotions did not have
a differential impact on burnout across the two groups of teachers. These results are in
line with previous research by researchers in special education who have indicated that a
high level of burnout is a primary cause of special education teacher attrition (Miller,
Brownell, & Smith, 1999; Singer, 1993). Neither year of teaching nor the elementary
teacher/middle school teacher distinction was associated with burnout once controlling
for emotions and prior burnout.

Teacher Emotions and Commitment to Subject/Grade. The results from the
regressions predicting teacher commitment to subject/grade are shown in Table 14. In
each of the four models, prior commitment had standardized coefficients ranging from
.47 to .51, all of which were significant at p S .01. Of the four emotion variables, positive
affect and feeling skilled both had moderate associations with teachers’ levels of spring
commitment, with standardized coefficients of .34 and .30; both relationships were
significant at p S .05. The R2 for positive affect (.44) was slightly higher than the R2 of

skilled (.42)—yet both values were smaller than the R2 values for the relationships

88

between emotions and burnout. Finally, the relationship between negative affect and
spring commitment was significant at p S .10, with a standardized coefficient of -.l6,
suggesting that when teachers experience feelings of negative affect, it may reduce their
commitment throughout the school year.

[Insert Table 14 here]

As in the models predicting burnout, there was a negative relationship between
teaching special education and commitment to grade/subject—independent of teacher
emotions and prior commitment—with standardized coefficients in the four models
ranging from -.29 to -.3 1. Also, interaction terms were again created to test the joint
effects of teaching special education and having high average values for each of four
emotion variables, yet none of these variables were significant. The results are also
suggestive of a negative relationship between first-year teachers and level of
commitment, as well as a small positive association between elementary school teachers
and level of commitment; however, given the size of the standard error of these variables
(relative to special education), these findings should be interpreted with caution.

Discussion

The purpose of this study was to investigate the influence of teachers’ in-the-
moment responses to their work on overall job attitudes—namely their stated levels of
burnout and their future career plans with respect to teaching in their grade or subject. In
doing so, I tested a framework developed in organizational behavior known as the
Affective events theory (AET), which proposes that emotional responses to work,
coupled with abstract beliefs about one’s job, emotional influence overall judgments

about job satisfaction. Further, AET posits that emotions provide a theoretical link

89

between workplace conditions and work attitudes; it is by influencing the kinds of events
that individuals experience (and respond emotionally to) that work context affects job
satisfaction. Studies in organizational behavior have empirically tested these relationships
and shown that emotional responses play an important role in evaluations of work
(Fisher, 2000, 2002; Weiss, Nicholas, & Daus, 1999). However, despite the potential
contribution of this theory in predicting teachers attitudes about their jobs (as well as their
future career decisions), no studies have made use of this framework in educational
research. This study therefore represented an initial application of Affective events theory
in predicting early career teachers’ levels of commitment and burnout.

Three questions were proposed to investigate these relationships. First, I analyzed
the degree to which four categories of emotional responses to work—teachers’ levels of
positive affect, negative affect, skilled, and tired—predicted burnout in Spring 2008,
controlling for prior burnout and several teacher demographic characteristics. Second, I
analyzed the same set of emotional items to determine how well they predicted teacher
commitment. Finally, 1 was interested in comparing two different conceptualizations of
teacher affect—one based on peak levels of emotion at school and the other based on
average levels of emotion at school over the work week

Related to the first research question, the results of my regression analyses
provide evidence of a moderate association between teachers’ average levels of emotions
at work—namely, their levels of “negative affect” and “fatigue”—and their stated levels
of burnout, a measure which I based on the Maslach Burnout Inventory. This finding is
noteworthy for multiple reasons. It lends support to Weiss and Cropanzano’s argument

regarding the importance of considering the impact of one’s emotional experiences on

90

their attitudes about their job. Additionally, it appears that negative affect and fatigue can
lead to higher levels of burnout, regardless of whether teachers felt burned out at the
beginning of the school year. Finally, the magnitude of the relationship between
momentary emotions and global measures of attitudes is noteworthy given the relatively
small sample size of the analysis (n=39).

Negative affect was also shown to be negatively associated with teachers’ career
plans, although the coefficient for this relationship was only significant at p S .10. In
addition to supporting the theoretical argument advanced in AET, this finding also
supports previous research in special education by Miller, Brownell, and Smith (1999),
who suggested that stress is one of the strongest predictors of attrition in special
education, even when controlling for a multitude of personal, background, classroom, and
school-level factors. The results also indicate that both positive affect and skill are
associated with higher levels of commitment, suggesting that positive emotional
experiences at school can impact teachers’ desire to stay in teaching. And, although the
data in the current study do not speak to teachers’ actual career decisions (only their
stated career plans) the association between teacher emotions and commitment is
important given studies that have linked commitment to retention (Weisberg & Sagie,
1999; Weiss, 1999).

The study’s last research question was posed in an effort to distinguish between
emotional responses to specific events and general categories of mood that might emerge
over time. Based on comparisons of models that separately tested the association of these
variables with teacher attitudinal outcomes, it appears that the mean variables more

strongly predicted both burnout and commitment. This suggests that when making

91

assessments about their work, teachers rely not on specific moments but on average
emotions over time. It is likely true that, at times, teachers do consider particularly salient
experiences when determining their future career plans (or assessing the degree to which
they feel burned out); however, attitudes are more likely to be based on one’s average
emotional state while at work.

Although these findings do not speak directly to how specific activities or
contexts influence average emotional states, prior analyses of these same data (Author,
2009a) have looked at how teachers’ emotions vary across contexts. The teachers in the
sample experienced significantly higher than average levels of skill, challenge,
engagement, self-esteem, and activation when engaged in instruction. Meanwhile, when
doing activities such as planning, paperwork, or grading, the teachers in this sample
reported lower than average levels of positive affect, enjoyment, and challenge. Given the
results of the current analysis, which suggest that average emotions are associated with
teachers’ work-related attitudes, further investigations into how teacher emotions in
specific situations shape overall affect are warranted.

Finally, of the teacher characteristics that were controlled for in the analysis, the
distinction between special education and general education teachers appeared to be the
most important (when predicting commitment), and the results indicate that average
affective responses did not differ significantly between general education and special
education teachers. However, the reasons for these differences in commitment were not
the focus of this study. In an analysis of the full MlECT sample, Author (2009b) has
investigated how informal support from one’s mentor and colleagues influences teachers’

career plans; their results indicate that informal support from colleagues is critical for

92

early career special education teachers’ commitment, as is a supportive school context
more generally.
Limitations

A primary shortcoming of these data was the limited number of teachers included
in the analysis. While I can be relatively confident about the general direction of the
associations measured (i.e., whether a predictor was positively or negatively associated
with one of the two outcomes of interest), more fine-grained interpretations were more
difficult to detect. For example, other than teaching special education, neither year of
teaching nor level of teaching (i.e., elementary vs. middle school) was significantly
associated with burnout or commitment. While this speaks to the predictive strength of
the special education/ general education distinction—even with a sample that only
included 16 special education teachers—it also suggests that the sample was too small to
distinguish between other differences in teacher characteristics. For example, while the
results for teaching special education were rather straightforward (i.e., special education
teachers felt more burned out and less committed than general education teachers), the
association between being a first-year teacher and the two outcomes was less easily
interpretable: the first-year teacher variable was consistently negative in predicting
commitment to grade/subject, but was also consistently associated with lower levels of
burnout. Perhaps with a larger sample, the nature of these relationships would become
more evident.

A second shortcoming of the current analysis was that it only focused on one
relationship within Weiss and Cropanzano’s theory. Although the authors’ main

argument is that emotions at work aggregate into attitudes about work, they also use

93

affective responses as a way to bridge the relationship between work conditions andjob
satisfaction. As I discussed in the introduction to this paper, it is frequently assumed in
educational research that teaching environments have a direct influence on teachers’
attitudes about work. Operating under this assumption, researchers and policymakers
have often attempted to increase teachers’ commitment levels by increasing their access
to resources, information, and support through mentoring and induction programs. While
the current study provides evidence suggesting that teachers’ emotional experiences are
also important predictors of their levels of burnout and future career plans, it does not,
however, directly assess the relationship between mentoring and induction policies and
teacher affect.

Similarly, Weiss and Cropanzano suggest thatjob satisfaction predicts whether
workers leave their jobs. Given that this study’s data were collected only over the course
of one year, it was not possible to determine whether participants followed through on
their career plans. Thus, I could not directly test how emotional responses to teaching
affects (through job attitudes) teacher attrition. An example that would be useful to the
future study of teachers’ affective responses to work is Fisher’s (2002) analysis of AET,
in which she used structural equation modeling to look at both antecedents and
consequences of positive and negative affective experiences at work. Structural equation
modeling has the additional advantage of allowing the researcher to investigate the
direction of relationships, which would address questions about the direction in which
workplace emotions and workplace attitudes operate.

A final limitation is the study’s measurement of teacher commitment. Although 1

base my measures on previous research on teachers’ levels of commitment, it will be

94

important in future studies to consider employ alternative measures of teacher
commitment to the profession—rather than relying on the proxy of commitment to
grade/subject. Additionally, although the distribution of the residuals when regressing
spring commitment on fall commitment did not violate the assumption of normality,
participants did not vary considerably in their levels of commitment from fall to spring.
While this may reflect that teachers’ levels of commitment are for the most part stable
across the school year, it does raise questions about the sensitivity of the measures to
more subtle changes in commitment. Despite these limitations, the moderate association
between teacher emotions and attitudes about the job suggests that affective variables are an
important (and underutilized) predictor of teachers’ plans to either stay or leave in the
profession.
Implications

Significant attention has been paid by researchers and policymakers to improving the
work conditions facing early career teachers. Districts, for example, have increasingly looked
to induction programs to assist elementary and secondary teachers in their transition from
preparation programs to their firstjobs.22 Further, the majority of new teachers are
provided with some sort of mentor to support them as they encounter the many
uncertainties of their jobs. The argument underlying many of these programs is based, at
least partly, on assumptions about teachers’ emotional responses to work. It is presumed
that in the face of undesirable work conditions, teachers will feel less attached to their
jobs, and will be more inclined to leave their position. And research confirms this to be

the case; teachers in challenging positions, such as teaching or teaching in an urban

 

22 These induction programs may take a variety of forms, such as workshops, collaborations, support
systems, orientation seminars, and mentoring (Smith & Ingersoll, 2004, p. 683).

95

school, are more likely to leave their positions (Billingsley, Carlson, & Klein, 2004;
Miller, Brownell, & Smith, 1999; Hamilton, Lankford, & Loeb 2002).

While much of the research on early career teachers has studied the effects of
teacher induction programs, few have rigorously collected data on how teachers’ actual
experiences impact their career plans. By connecting the fine-grained emotion data of the
ESM to longitudinal survey responses on teachers’ attitudes about theirjobs, this study
presents evidence that teachers’ affective responses to their work play an important role
in teachers’ judgment about their jobs. While the size of my sample precluded me from
focusing specifically on teachers’ emotional responses to aspects of induction (e.g.,
professional development, formal mentor meetings, observations with principals), the
findings on teachers’ general emotional states are significant. By taking account of
teachers’ emotional reactions to their work (in addition to features of their work
environments), researchers, policymakers, and district administrators will be provided

with ways of better supporting special educators in their early years of teaching.

96

Appendix A

Table 11: Demographic Characteristics of the Districts in the Sample

 

District Total Enrollment % F rec/Reduced Lunch % Non-white
Daus (MI) 19,055 51% 12%
Kaline (MI) 9,448 42% 50%
Engram (IN) 13.726 59% 47%

 

Table 12: Demographics of the ESM Sub-Sample vs. Full Sample (in 3 Districts)

 

Demographic Characteristics of ESM Sample

Demographics of Full Sample in 3 Districts

 

Total

Total n=42 n=92
Gender Gender

Female 3 5 8 3% Female 75

Male 7 17% Male 17
Grade Level Grade Level

Elementary School 27 64% Elementary School 5 3

Middle School 15 36% Middle School 34
Special Ed. Status Special Ed. Status

General Education 26 62% General Education 67

Special Education 16 3 8% Special Education 25
Year of Teaching Year of Teaching

First 1 1 26% First 30

Second 22 52% Second 38

Third 9 21% Third 24
Race/Ethnicity Race/Ethnicity

African-American 2 5% African-American 6

Asian 1 2% Asian 1

White 39 93% White 85
District District

Daus (Michigan) 7 17% Daus (Michigan) 17

Kaline (Michigan) 8 19% Kaline (Michigan) 20

Engram (Indiana) 27 64% Engram (Indiana) 55

81%
19%

63%
37%

73%
27%

34%
41%
25%

7%
1%
92%

1 8%
22%
60%

 

97

$25-3 S. w a... ”8. w a... n2. w a... ”202

 

 

 

2.. an. m. am. seesaw-z
33 SS :3 95.: .3235
on. a: a: $.—
33 $3 83 GB 225
8.- 2.- N..- R. 3.- S... 8- E -
Am... C . 2C . QC . Sc 22>»
S- 3- 3 : S 8 3 2
$3 :2 $3 63
.5 one .30 m:
S.- Sc 2.- 2.- 2.- 2.. Sr 2.- g H t .m
ANS $3 :3 $3 .
3 one «83
mm. em. 2. ...-9.. mm. 2... cm. 2.. s o. 3:. m
23 $3 :3 $3 .
0 one 00 0 Eu
:.- 2.. oo. 8. 8. S.- . No. 3. 2. E s m .m
Snatmtoaofiu .35ka
SS
0.:
mm. fav- 6 ..H
83
o _
8.- m..- n :.xm
$3
8 ozmwo
3. .....B._ a E. .. z
a:
00 mo
2.- 2.- a 5.2;: a
«fie-£05m LmQUCm-N
. E ”V . $5 . $5 . ac =£|5oE=m
_m aaamm mm aaamm hm aaahm av aaaom
n5 3% «we «we
MHMWm-Wocwm cofimocmom ~%M_o~_%v8:ow game-aux ”Nona-...... New 563.31 .Huomoh-Wocww cot-323% czar—ax, «savanna—u:—
. e So e .e m So u as. m So u .e o m So
at: 3:3. as? scam-2 new... ~32an

 

awn—5 ”fianmluaofiam League-F $5369..— ..8 ...-.252 ”3 033—.

98

.6253 S. w 9...... m3. w a... 5.. w a... ““.82

 

 

 

3.. S- «M. 3.. Essa-é
:3 89: ES 39.: aces...
ELAN-N va. snow N W»
. 86 . 3o . 8S . 2 .6 225m
8 8 8 - S- 8 N. 2 - a. -
. am. V . :3 . an. o . Gm. o 223
S- 3- 8 8 8- 8- 3- 2-
:3 $3 33 RS
.5 one .80 P:
8.- a; 2.- 2.- 8.- 2.- 2.- mm- ; p S .-._
£3 :3 £3 83 .
.5 one Goa
_m.- free.- omfi 3.35.- _m.- 1.00.- oNr ...—er ..— ..—. Um— E. m
63 :3 GS 33 28¢
_ _. E. :. mm. _ _ mm. 8. 2. 623 .805
ooze-Imhooxcfiv Stoop-N
83
o:
3.- 2.- u .e
83
o _
cm. :3. u :.ea
3:
8 gamma
3.- an- a E. . z
83
00 0250
X. :3. a 5. .s. A.
Eotosm amt-oofi
G: a: a: a:
a lavas—EEO
8. ....m. .m. rim. 2.. ......S. S. ......2. = ... . . o
«mm.» «mm.» «mm.» «mm.»
“WNW-W ”Wm cosmocwom “WNW—W 0:an :o_mm2mom HMS.“ “w coax-6&3. “HMS-“Woohnwm :omwmocwom oswtm> «sou—596: _
. e So .u e m So a. .u e m So a. So
we: 33m. ES. seemmi use? mega

 

ASHE watgmlaaofizfifioU accuse-F $59.10...— ..o.— ...—eve: u: 935—-

99

Appendix B: Supplemental Tables and Figures

Table 15: Description of Variables

 

 

 

 

 

 

 

Description Mean S.D.
I. Dependent
Variables
Burnout Composite measure taken from the survey Spring 2.53 .75
questionnaire, consisting of 6 questions: I
feel emotionally drained from my work, I
feel used up at the end of the workday, I feel Fall 2.52 .74
fatigued when I have to get up in the
morning and face another day on the job, I
feel burned out from my work, I feel
frustrated by my work, I feel I’m working
too hard on my job. Responses on single
measures ranged from l=strongly disagree to
4=strongly agree. (a = .91)
Commitment to Composite measure taken from the survey Spring 4.08 1.05
Grade/Subject questionnaire, consisting of 2 questions: I
would prefer to continue teaching in this
grade/subject next year, I could see myself Fall 4.14 1.1 4
teaching in this grade/subject in five years.
Responses on individual measures ranged
from l=strongly disagree to 5=strongly
agree. (a = .87)
Commitment to Composite measure taken from the survey Spring 3.81 1.31
School questionnaire, consisting of 2 questions: I
would prefer to continue teaching in this
school next year, I could see myself teaching
at this school in five years. Responses on Fall 3.95 122
individual measures ranged from 1=strongly
disagree to 5=strongly agree. (a = .84)
11. Independent
Variables
Positive Affect Taken from the ESM questionnaire. Beep- Aggregate 4.18 .75
level composite variable including the (person
following scale variables: strong, happy, raw-score
cheerful, relaxed. (Cronbach’s alpha ((1) = mean)
.84) Aggregate 1.48 .51
(person 2-
score max)
Negative Affect Taken from the ESM questionnaire. Beep- Aggregate 4.17 1.22
level composite variable including the (person
following scale variables: angry, frustrated, raw-score
irritated, lonely, nervous, stressed, worried. mean)
(a = .86) Aggregate 2.72 .83
(person 2-
score max)

 

100

Description of Variables (continued)

Skilled

Tired

 

 

Taken from the ESM questionnaire. Beep- Aggregate 2.17 .56
level response to the question: “As you were (person
beeped, were you feeling skilled?” raw-score
Responses ranged from 0=not at all to mean)
3=very mud“ Aggregate 1.12 .81
(person 2-
score max)
Taken from the ESM questionnaire. Beep- Aggregate 1.14 .55
level response to the question: “As you were (person
beeped, were you feeling tired?” Responses raw-score
ranged from O=not at all to 3=very much. mean)
Aggregate 1.28 .56
(person 2-
score max)

 

111. Control Variables

Special Education

Elementary
School

First Year Teacher

Gender

Race

 

Dummy variable where 0 = general
education teacher and l = special education
teacher

Dummy variable where 0 = middle school
teacher and 1 = elementary school teacher

Dummy variable where 0 = not a first year
teacher and 1 = first year teacher

Dummy variable where 0 = male and 1 =
female

Dummy variable where O = non-white and 1
= white

 

101

Avo=fiumv S. W use...

.8. w ...... m....w .... éoz

 

 

8.. 8...- ...o- ...-8...- ..N... 8... .8...- ...... 2...- 2...- 8...- ...... “MEN:
8.. .....- 8...- ...-2.- IS... 8...- R...- 8... 8... .....- to... ...:omwww...

8.. 8...- 8... 2...- N... .82.- 8... m... t...- 8...- ..o......w..m>...o..

8.. 8...- 8...- 8... :8... 8...- .....- R... 8...- “mums.

8.. 2....- .. . ...- S... : ...- .2.- .o...- 8...- be... gumwwuws.

8.. 8...- ..N...- w . ...- 8...- .8- 18... .8...- gmwmfixt

8.. ... .o- t... 8... ...... an..- 60...... gmwwxe.

...... m . ... 8...- ..em... .2... .8...< owmmm...m.

8.. :8... ...-8... ...-...- .cogmwfiwwf.

...... 1...... 8...- .aoemhwflwmo.£.

...... :8...- ofimwwwfiw.€

co. Soc-Em C.

a: .... 8.. a. .... E G. G. .... .m. a. ... 33%;

 

£22.23. 2.. E can: 833...; .8.— mcozsotou 6— 033—.

102

 

mm. 3.. m... ..N. 833%...

 

 

 

 

.2... 3...... .3... .8” .. 2.8.2:.
8. N 8 8 . 8 .
. .8... . 2 m... . .8... . .8... use...
8- ...- N.- N..- 8- 8- 8- ...-
. Sm. . Gm. . . ....N. . . ...-.... 2...?
8- e..- 2.. ..T ..- ..N- 8 m.
.8. .8. 38 SN. 3:23
8.- m _ .- ....- mN- ...- a..- ....- ....- 30> at...
.2. an. 38 ...N. .38.;
.m. .8.- oN- .. . e- ..N. 8. 8N. 2.. .2. .20.-am
.8. an. ...N. ...N. .2089
8. w .. 8. N... .... 8. 8. 8. 62.8 .52.“.
S...w.28o§3&b
kings
2 m. .NN. 32.
.8. N . .- ..N- .8»... 3....8.
SN. 8... .532.
8.- 3... :8. N..- N. .- out... 38.8..
...-totofim .5535
. G.M. . G.M. . .2”. . .2“. ......lsossm
mm ...:ow NV 52.21..V mm ......an av ...aaom
Em. 3m. «.8. Q8.
......me ...-...... ”...-3w 3.8 .....M.............0 ...-E... ”...-...wa ...-...... ......fi...»
t a. c m “Sivan—.33: v .v w m 88335333 o b v .m voices—ESE: o b u .m ENE-accaacb . v v _
8%... N .88.. use... . 38:. as... N 38:. Nae-...... . Es:
3v”... .00-353.80 o. EocEEEoU an”... w..._.aml§oE=m

 

to»: 95.8.— ..o 2954 .20.—93a. :o vow-am 8.5.32 353,—. 959.10...— -.o._ £0.32 ":.— 03:.

103

 

 

 

3. N... 2.. mm. 3339.-..
.N . H .. .3... .3... .3... .322...
... N ......m N ..N . ... .
. .8... . ...mn. . ...-H. . ...m... 225,-.
3.. - N. - 3.. N. 3.. B ... - 3 -
..m. .3. .3. .3. 353
.....- m... N...- a...- o...- .N.- .... 3... .
.3. .33.. GN. ..NN. .28“;
m...- 3..- a...- NN.- m..- .N; m...- a...- 30> E:
.3. .N3. .MN. ..N. 28.3
3.- ...... .5- 3.- :8.- Nm. ..3. MN. 3. ...m 38%
.33.. .Nm. .3N. ..N. 28$
N... N .. . .. 3N. .... N... 8. ....- .85m .805
h0-h~k2b§k~u£U
...-$985
.8
...N. .2 .. . W...
N... a . .- EN
.... .N.- 96302
50
E... .33.. . wow...
.- *Nm-u - franc.— w
o. $- 2.... oz
38.85% Nut-ugh
. .2”. . .2. . .2 H. . .3. ...-.lsoeam
om flow NV ilk-v Nm .3th hm ......i-m
20.2.: 80 «cam-Waco E20580 30%-Waco 2.0.0580 .co.mw..w.uou E0658”. EBMW-Noou 03ME>
nofiuavcmfi vofivumccsmcD uni—zagafi ...-36:65.33 BEEN—=85 236.3532: BEEN—28% cofiufincfimcD Eon—.0 0....—
35. N E8: .283. . 38: 3.3. N 3%: 33:. . s8:

 

...-u... 60,352.20 9 505.558

8m“... wEEmI'SoEzm

 

«our: «>5..qu .... ...—«>04 ...-35:3. .8 03am nee—£34..- ..oauaoh. 959:5...— ..o._ 20.32 "fl: 0...:-

104

 

 

 

33. N... 33. 33. 3.3.3-..
...... .39... ..3... ...... 38.2...
...-...... N .... N. . ...N...
. .33... . .2... . .3... . .33... 295“.
3..- .N- ....- ...- 3..- 3. N.- 3N-
. .3. . . .33.. . .33. . . .3.... 8...?
... - N3- .... .... .. - .3 - .... ..
.3. .3. .NN. .NN. .2383
....- 3N.- 3..- .3.- N..- .N.- ... .- :.- §> ......
. . 3.. .. 3.. ...N. ...N. .2383
N3.- :3..- ..3.- ...-....- 3N. ...3... 3N. .N.. ...m 38%
.N3.. . . 3.. ...N. .3. 558-3
3... ... ... 3N. .... .... .5. 8. .853 .523.
MUNNMTMBGKQN‘U
kmfibumpm
...N.. .3... .3...
3N.- .3.- 3...- 3...- 3.6.3
...N. ...N. .532.
..3. :33. 3...- 3..- ...-......3
322.65% ..mfiumh
. .2“. . .2“. . .2”. . .2“. ......lsoesm
mm ....imm .m 3.23% cm ......th mm ......mm
.33. 3.3. 3...... $3.
«Mme—BMWNMW EQBEOOU «MW—mu-HWOFMNW EOBEOOU “WM—momww EOBEUOU “HMBFCOGHW EBUEOOU Go “Mum—0.8“,
u b c m vofifizgcflmca v .... c m vofinaccaficD c ... v m vo~_vacfim=D v .23. m quEEUGSmcD a m. c _
3.3.. N 3...: 3:83.. . 38:. 3.2. N 38:. 3.8:. ...-.8:

 

8v"... .uofszcEO o. .=o§.EEoU

.33”... wccgmlsoeam

 

:5—W MG n—0>0\— anhvfioloh. HQ Enflfl «05:3sz hOflOflOh. Mflmwum—utnn .2: 3050—2 "Qb— 256,—.

105

 

..... 3. 33. 3... ...-.3...-..

 

 

 

.33.. .33.. ...... .33. 2.8.3...
......N-NN .2..va ...ON._ on
. ...... . .3... . .3... . .3... 22......
.... - .. - .... .... 3..- .: ... - 3. -
. ...3.. . .33. . . .33. . . .N.... 3......
N.. 3.. N..- 3..- 3.- 3N- ...- ....-
..3. .33.. ...N. .3N. .2383
3...- 3..- 3....- .N.- 3..- NN.- N...- 3...- 30> .3...
.3. .33.. .3N. .NN. .2383
. 3.- .3...- . 3.- :33.- ..3. .3... NN. 3. ...3. 3.02.3
.33.. .33.. ...N. .3N.. .2333
3... 3.. . .. ..N. 8. .... ...- 3..- .853 .523.
mohwtmtfikafiu
.NNN‘UQNR
...N. ..N. ......2.
N...- .....- 3...- . ..- ...-c;
...3. ...N.. .532.
3...- 3..- 33. :3... 8.:
3.330205% kmguam-N
. .... . .... . .3.. . ....H. =5.|.8E=3.
om 1:..om om .3...—m om ......R-m 3 ...}...NW
. $3. $3. .5. Q3.
...-...... ...-...... ...-......E... 3...... ...... :25... 3......“ ...-......
v ... v m o-€8n=83=3 u b v .m ENESUGSED H. b u m waitstaaficz u b c m unfit-.3533... . U c —
3...... N 3...: 3:83.. . $5: 3.8:. N 5...: «:83. .3...:
2..."... 60.35.2320 9 EoEzEfioU 8m"... w=_.aml.:oE:m

 

2.3:...— ._o 3.25.— .anuaoh .3 vow-am 32:533.. 35...:- ufiugvflm .8.— 3_e_x._>. "v...— 2...:-

106

Table 18.1: Models for Predicting Teacher Burnout (with Interaction Terms)

 

 

 

Positive Negative - .
Aflect Aflect Skilled Tlred
0L5 OLS OLS OLS R r 'on
Independent Variable RegFCSSiO“ RegreSSiO“ RengSiO" (gig; 685‘
(SE) (SE) (SE) )
1.18 .00 1.39“ .49
Bumout—Fall (.73) (.28) (.56) (~37)
Teacher Emotions
. . .30
Posmve Affect (. 50)
_ -.92
Negative Affect (1.23)
_ .84
Skilled (.66)
. .43
Tired (30)
Emotions x Prior
Positive x Prior -.18
(.20)
_ _ .76
Negative x Prior (_43)
. . -.41
Skilled x Prior (,27)
. ' .02
Tired x Pnor (.31)
Teacher Characteristics
_05 -,07 .10 -.17
Elem. Teacher (.24) (.22) (.24) (~24)
. .40 .42 .40* 34
Specral Ed. ('24) (.21) (.23) (33)
. -.15 -.07 -.11 --03
First Year (.27) (.22) (.26) (~26)
. .11 -.07 .12 --04
White (_43) (.37) (.42) (-41)
Female -'07 -I3] ”07 -.13
(42) (29) (36) 034)
Interce t .00 2.18 -.82 .86
p (2.08) (.80) (1-64) (1-07)
R—squared .39 .56 .42 .47

 

107

Table 18.2: Models for Predicting Teacher Commitment (with Interaction Terms)

 

Positive

Negative

 

 

Aflect A flect Skilled Tired
OLS OLS OLS OLS
Independent Variable RegfeSSiO" Reg'eSSiO" RegreSSiO" Regression
(SE) (SE) (SE) (SE)
. l.14* .94M 1.18" .50
Commitment—F all (.61) (.38) (.44) (.45)
Teacher Emotions
Positive Affect “5*
(.64)
_ 2.79
Negative Affect (252)
. 1.85
Skilled (.85)
. -.20
Tired (2.10)
Emotions x Prior
.. . «17
Posmve x Prior (.15)
. . -.78
Negative x Prior (.58)
. ' -.32
Skilled x Prior (.20)
. . .01
Tired x Pnor (.48)
Teacher Characteristics
.16 .31 .17 .24
Elem. Teacher (.30) (33) (24) (.35)
S ecial Ed -.64** -.67** -.71** -.66*
p - (.30) (.32) (.30) (.35)
First Year -23 -.18 -.26 -21
(.33) (.35) (.33) (.39)
. -.20 -.02 -.11 -.09
White (.56) (.59) (.56) (.62)
Female -56 '06 -.20 .00
(.52) (.49) (.47) (.52)
Interce t -1'70 '41 -1 .47 2.39
P (2.50) (1.83) (1.83) (2-16)
R-squared .39 .56 .46 .34

 

108

References

Achinstein, B., Ogawa, R.T., & Speiglman, A. (2004). Are we creating separate and
unequal tracks of teachers? The impact of state policy, local conditions, and teacher

characteristics on new teacher socialization. American Educational Research Journal,
41(3), 557-603.

Author. (2009a).
Author. (2009b).

Breckler, S. J, & Wiggins, E. C. (1989). Affect versus evaluation in the structure of
attitudes. Journal of Experimental Social Psychology, 2, 253-271.

Bryk, A.S., & Schneider, BL. (2002). Trust in schools.“ A care resource for improvement.
New York: Russell Sage Foundation.

Chemiss, C. (1995). Beyond burnout: Helping teachers, nurses, therapists, & lawyers
recover from stress and disillusionment, Routledge, NY.

Crites, S. L., Jr., F abrigar, L. R., & Petty, R. E. (1994). Measuring the affective and
cognitive properties of attitudes: Conceptual and methodological issues. Personality
and Social Psychology Bulletin, 20, 619-634.

Csikszentmihalyi, M. (1997). F indingflow: The psychology of engagement with everyday
life. New York: Basic Books.

Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the Experience
Sampling Method. Journal of Nervous and Mental Disease, 1 75(9), 526-536.

Day, C., & Leitch, R. (2001). Teachers’ and teacher educators’ lives: The role of
emotion. Teaching and Teacher Education, 1 7(4), 403-415.

Desimone, L. M., Porter, A. C., Garet, M. 8., Yoon, K. S., & Birman, B. F. (2002).
Effects of professional development on teacher’s instruction: Results from a three
year longitudinal study. Educational Evaluation and Policy Analysis, 24(2), 81-112

Dinham, S., & Scott, C. (1998). An international comparative study of teacher
satisfaction, motivation, and health: Australia, England, and New Zealand. Paper
presented at the annual meeting of the American Educational Research Association,
San Diego.

Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. San Diego, CA:
Harcourt Brace Jovanovich.

109

Ebmeier, H. (2003). How supervision influences teacher efficacy and commitment: An
investigation of a path model. Journal of Curriculum and Supervision, 18(2), 110-
14].

Edwards, K. (1990). The interplay of affect and cognition in attitude formation and
change. Journal of Personality and Social Psychology, 59, 202-216.

Fisher, C. D. (2002). Antecedents and consequences of real-time affective reactions at
work. Motivation and Emotion, 26, 3—30.

Fisher, CD. (2000). Mood and emotions while working: Missing pieces of job
satisfaction? Journal of Organizational Behavior, 21 185-202.

Fletcher, 8., Strong, M. & Villar, A. (2008). An investigation of the effects of variations
in mentor-based induction on the performance of students in California. Teachers
College Record, I 10(10), 2271-2289.

Frijda, N. H. (1993). The place of appraisal in emotion. Cognition and Emotion, 7, 357-—
387

Glazerman, S., Dolfin, S., Bleeker, M., Johnson, A., Isenberg, E., Hugo-Gil, J. Gruder,
M., Britton, E., Britton, E., & Ali, M. (2008). Impacts of comprehensive teacher
induction: Results from the first year of a randomized controlled study. Washington,
DC: US. Department of Education, Institute of Education Sciences.

Goddard, R., & Goddard, M. (2006). Beginning teachers’ burnout in Queensland schools:
Associations with serious intentions to leave. The Australian Educational Researcher,
33(2), 61-75

Gold, Y., Roth, A., Wright, G, & Michael, W. (1991). The relationship of scores on the
Educators Survey, a modified version of the Maslach Burnout Inventory, to three
teaching-related variables for a sample of 132 beginning teachers. Educational and
Psychological Measurement, 51, 429-438.

Grossman, P.L., & Thompson, C. (2004). District policy and beginning teachers: A lens
on teacher learning. Educational Evaluation and Policy Analysis, 26(4), 281-301.

Guglielmi, R. S. & Tatrow, K. (1998). Occupational stress, burnout, and health in
teachers: A methodological and theoretical analysis. Review of Educational Research,
68(1), 61-99.

Hargeaves, A. (2000). Mixed emotions: Teachers’ perceptions of their interactions with
their students. Teaching and Teacher Education, 16(8), 81 1-826.

Hargreaves, A. (2001). The emotional geographies of teaching. Teachers College Record,
103(6), 1056-1080.

110

Hargreaves, A., Beatty, B., Lasky, S., Schmidt, M., & Wilson, SJ. (2006). The emotions
of teaching. San Francisco: Jossey-Bass.

Hektner, J .M., Schmidt, J .A., & Csikszentmihalyi, M. (2007). Experience sampling
method: Measuring the quality of everyday life. Thousand Oaks, CA: Sage .

Ingersoll, R.M. (2001a). Teacher turnover and teacher shortages: An organizational
analysis. American Educational Research Journal, 38(3), 499-534.

Ingersoll, R. (200 l b). Teacher turnover, teacher shortages and the organization of
schools. Center for the Study of Teaching and Policy, University of Washington.

Jeffrey, B., & Woods, P. (1996). Feeling deprofessionalised: The social construction of
emotions during an OFSTED Inspection. Cambridge Journal of Education, 26(3),
325-343. ~

Johnson, S.M., & Birkeland, SE. (2003). Pursuing a "sense of success": New teachers
explain their career decisions. American Educational Research Journal, 40(3), 58 l -
617.

Johnson, S.M., Kardos, S.M., Kauffman, D., Liu, E., & Donaldson, M.L. (2004). The
support gap: New teachers’ early experiences in high-income and low-income
schools. Education Policy Analysis Archives, 12(61). Retrieved from
http://epaa.asu.edu/epaa/v12n6l/.

Kapadia, K., Coca, V., & Easton, J .Q. (2007). Keeping new teachers: A first look at the
influences of induction in the Chicago Public Schools. Chicago: Consortium on
Chicago School Research, University of Chicago.

Kardos, S.M., Johnson, S.M., Peske, H.G., Kauffman, D., & Liu, E. (2001). Counting on
colleagues: New teachers encounter the professional cultures of their schools.
Educational Administration Quarterly, 3 7(2), 250-290.

Lasky, S. (2000). The cultural and emotional politics of teacher-parent interactions.
Teaching and Teacher Education, 16(8), 843-860.

Lasky, S. (2005). A socio-cultural approach to understanding teacher identity, agency,
and professional vulnerability in a context of secondary school reform. Teaching and

Teacher Education, 2] (8), 899-916.

Little, J .W. (1996). The emotional contours and career trajectories of (disappointed)
reform enthusiasts. Cambridge Journal of Education, 26(3), 345-359.

111

Miller, M.D., Brownell, M.T, & Smith, SW. (1999). Factors that predict teachers staying
in, leaving, or transferring from the special education classroom. Exceptional
Children, 65(2), 201 -2 l 8.

Morris, J. A. (1989). Mood: The frame of mind. New York: Springer-Verlag.

Peske, H. G., Liu, E., Johnson, S. M., Kauffman, D., & Kardos, S. M. (2001). The next
generation of teachers: Changing conceptions of a career in teaching. Phi Delta
Kappan, 83(4), 304-311.

Petty, R. E., Wegener, D. T., & F abrigar, L. R. (1997). Attitudes and attitude change.
Annual Review of Psychology, 48, 609-647

Prescott, S. Csikszentmihalyi, M., & Graef, R. (1981). Environmental effects on
cognitive and affective states: The experiential time sampling approach. Social
Behavior and Personality: An International Journal, 9(1), 23-32.

Robinson, J .P. (1985). The validity and reliability of diaries versus alternative time use
measures. In F. T. Juster & F. P. Stafford (Eds.), Time, goods, and well-being (pp. 33-
62). Ann Arbor, MI: University of Michigan, Survey Research Center, Institute for
Social Research

Schmidt, M. (2000). Role theory, emotions, and identity in the department headship of
secondary schooling. Teaching and Teacher Education, 16(8), 827-842.

Schneider, B., & Waite, L.J. (Eds.). (2005). Being together, working apart: Dual-career
families and the work-life balance. Cambridge, UK: Cambridge University Press.

Shemoff, D., Csikszentmihalyi, M. Schneider, B., & Steele-Shemoff, E. (2003). Student
engagement in high school classrooms from the perspective of flow theory. School
Psychology Quarterly, 18, 158-176.

Shemoff, D., & Schmidt, J .A. (2008). Further evidence of an engagement-achievement
paradox among US. high school students. Journal of Youth and Adolescence, 3 7(5),
564-580.

Shemoff, D., & Vandell, D.L. (2007). Engagement in after-school program activities:
Quality of experience from the perspective of participants. Journal of Youth and
Adolescence, 36(7), 891-903.

Schmidt, J .A., Shumow, L. & Kackar, H. (2007). Adolescents' participation in service

activities and its impact on academic, behavioral, and civic outcomes. Journal of
Youth and Ado1escence, 36, 127-140.

112

Smith, T.M., & Ingersoll, R.M. (2004). Reducing teacher turnover: What are the
components of effective induction? American Educational Research Journal, 41(3),
681-714.

Stein, M.K., & D’Amico, L. (2002). Inquiry at the crossroads of policy and learning: A
study of a district-wide literacy initiative. Teachers College Record, 104(7), 1313-
1344.

Taylor, S. E. (1991). Asymmetrical effects of positive and negative events: The
mobilization-minimization hypothesis. Psychological Bulletin, 110, 67-85.

Weisberg, J. & Sagie, A. (1999). Teachers’ physical, mental, and emotional burnout:
Impact on intention to quit. The Journal of Psychology, 133, 333-339.

Weiss, EM. (1999). Perceived workplace conditions and first-year teachers’ morale,
career choice commitment, and planned retention: A secondary analysis. Teaching
and Teacher Education, 15, 861-879.

Weiss, H.M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion
of the structure, causes and consequences of affective experiences at work. Research
in Organizational Behavior, 8, 1-74.

Weiss, H.M., Nicholas, J .P., & Daus, CS. (1999). An examination of the joint effects of
affective experiences and job beliefs on job satisfaction and variations in affective
experiences over time. Organizational Behavior and Human Decision Processes, 78,
1—24

Youngs, P. (2007). District induction policy and new teachers’ experiences: An
examination of local policy implementation in Connecticut. Teachers College
Record, 109(3), 797-837.

113

CHAPTER 4: A COMPARISON OF THE INFLUENCE OF SOCIAL CONTEXT ON
THE COMMITMENT AND BURNOUT OF EARLY CAREER SPECIAL
EDUCATION AND GENERAL EDUCATION TEACHERS
Introduction

The struggle to staff all classrooms with highly-qualified teachers is a struggle
across all teaching not unique to special education (Darling-Hammond, 1997, 1999),
although but the severity of the problem is. The scarcity of special education teachers is
widespread across all regions of the US, with 98% of the nation’s school districts facing
chronic shortages in special education (Boe, Cook, Bobbitt, & Terhanian, 1998; Carlson
et al., 2002; ERIC, 2001, McClesky, Tyler, & Flippin, 2004). Positions remain vacant
each year because, in many cases, suitable candidates cannot be found. Instead, vacant
positions are likely to be filled by personnel who a) are not certified, b) are assigned to
positions for which they do not have certification, or e) have no certification at all.23
Taken cumulatively, uncertified personnel filled more than 30,000 teaching positions per
year in the 19908 (US. Department of Education, 2000 in McClesky et al., 2004).

The extent of the shortage of special education teachers has also increased
because of the growing special education population. For students ages 6-21, the demand
increased by 42% from 1987-88 to 2002-03, from 284,300 teachers to 403,100 teachers
(Boe, 2006). During this same time, the supply of teachers grew as well, but it could not
match the steady, long-term growth in demand. This trend shows no signs of abating, and

schools are likely to continue to use unqualified replacements, as well as to increase the

demands on the fully-certified teachers already in their positions.

 

23 Boe (2006) calculated that 1.1% of these teaching positions remain vacant, staffed by long-term
substitutes.

114

Finally, the teacher shortage has intensified because special education teachers are
more likely to leave the profession than their general education peers. Strong disparities
between the two fields are reported by Boe, Bobbitt, and Cook (1997). Using a sample of
4,782 public school teachers, they determined that 5.8% of general education teachers
and 7.9% of special education teachers left teaching between 1987-88 and 1988-89. They
also found that 13% of special educators transferred to general education teaching
positions during the same time. More recent data suggests that, on the national level, the
percentage of special educators either leaving or transferring to general education
remains close to 20%, with attrition rates much higher in some districts and states
(National Commission on Teaching and America's Future, 2003). The high rate of
transfer from special education to general education is worth noting because it suggests
that there is something about the special education setting itself (rather than the
profession of teaching) that is driving teachers out of special education.

The high rates of attrition across the teaching profession is due in large part to the
loss of beginning teachers, or what Ingersoll refers to as the “revolving door” of teaching
(Ingersoll, 2001). Approximately 30% of new teachers leave teaching within the first
three years, while 50% are likely to leave within five years (Huling-Austin, 1990;
Ingersoll & Smith, 2003; Olson, 2000). In special education, the problem of attrition is
even more pronounced. In their analysis of the retention decisions of first-year teachers
using the Schools and Staffing Survey, Smith and Ingersoll (2004) found that first-year
special education teachers were 2.5 times as likely to leave the profession as teachers in

other disciplines. This finding is corroborated by other researchers as well (Boe, 2006).

115

While individual characteristics are a primary determinant of leaving teaching
(such as age, class, content area), beginning teachers often make retention decisions
based on organizational factors, such as whether they receive support from their
colleagues or administrators, whether they have adequate resources, and whether they
feel successful with their students (Johnson & Birkeland, 2003; Johnson et al., 2004;
Ingersoll, 2001). As a wide body of research confirms, the organizational norms of
schools have consequences for whether or not a beginning teacher feels attached to their
school and to the teaching profession more generally.

Several strategies for addressing the field’s shortage of teachers have been
proposed. Given the multiple trends affecting the supply and demand of special education
teachers, Boe (2006) advocates focusing our attention on the ways in which we can
increase the quality and quantity of the supply of the teaching labor force—the system’s
inputs. He recommends luring teachers who have left the profession back into the labor
force, finding ways to attract general education teachers into special education, and
increasing the number of individuals becoming certified to teach special education in
colleges and universities.

This paper argues that a more viable policy aim would be to target factors that are
likely to improve retention, bringing together an analysis of two of the most significant
predictors of leaving teaching—being an early career teacher and teaching special
education. If we do not address the challenges facing these populations, any efforts to
increase the pool of potential teachers will likely be undermined by the adverse
institutional conditions that they face in the beginning of their careers. And, if we can

improve the quality of the work environment for these teachers, we can also increase

116

their ability to increase students’ achievement levels in language arts and mathematics
(Desimone et al., 2004; Hill, Rowan, & Ball, 2005). By describing the ways in which
schools as social organizations shape the commitment of early career special education
teachers, this paper will provide direction for district and state level policies related to
special education teacher retention.

Literature Review

Research on beginning teachers has rarely focused on both general education and

special education teachers.24 Yet, it is becoming increasingly important to simultaneously
consider the experiences of both groups of teachers, as students with disabilities are more
frequently being integrated into general education classrooms for instruction. Similarly,
from a policy perspective, it is important to investigate whether induction policies that are
thought to be effective for all teachers may have differential impacts on general education
and special education teachers. Consequently, when reviewing literature of the
experiences of early career teachers, I explore differences in curricular expectations,
induction experiences, and outcomes for both groups of teachers.

For beginning general education teachers, districts often take on a proactive role
in defining curricular expectations. Stein & D’Amico (2002) studied district policy
surrounding literacy improvement in New York City’s District #2. They found that
district leaders relied on professional development to facilitate teacher learning of
curricular expectations; additionally, policies related to student assessment and teacher
evaluation provided clear messages about how the instruction of new general education

elementary teachers should look. Similarly, Grossman and Thompson (2004) reported

 

This IS In part because the work lives of general education and specral education teachers have
traditionally been viewed as separate.

117

that districts facilitated teacher learning and communicated teaching expectations through
“(t)he tasks they assign(ed) to new teachers, the resources they provide(d), the learning
environments they create(d), the assessments they design(ed), and the conversations they
provoke(d) (2004, p.298).” Finally, these findings are aligned with those of Achinstein,
Ogawa, and Speiglman (2004), who found that in two California districts, teachers’
instructional practices were strongly influenced by district curricular choices and the
professional development opportunities they provided.

In contrast, research suggests that the curricular and instructional expectations
placed on beginning special education teachers tend to be ambiguous (Author, under
review; Billingsley & Tomchin, 1992; Carter & Scruggs, 2001; Conderman & Stephens,
2000; Mastropieri, 2001; Zabel & Zabel, 2001). Special education teachers are often
responsible for designing curriculum for multiple subjects across multiple grade levels,
yet most do so without adequate resources for determining how this instruction should
look (Kilgore et al., 2003). And, while general education teachers are typically provided
with materials such as textbooks, pacing guides, and state teaching standards, special
education teachers often do not receive the same kind of direction. The materials they do
receive are often identical to those given to general education teachers, but they are
provided without guidance for adapting these materials to meet their students’ needs.

The organization of schools in many instances reinforces the distinction between
the work lives of general educators and special educators. With the exception of co-
teaching arrangements, there are few formal avenues for collaboration between special
education teachers and their general education peers. Further, even though students with

disabilities may be co-taught by general education and special education teachers, these

118

students are frequently thought to be the primary responsibility of special education
teachers (Kilgore et al., 2003). Additionally, with teachers often grouped according to
team or grade level, there can be physical distance between special education and general
education teachers, limiting opportunities for informal interactions. These circumstances
may heighten the isolation experienced by early career special education teachers, and
there is evidence that such isolation has negative consequences for special educators’
intentions to remain in teaching as well as their actual retention decisions (Billingsley &
Cross, 1992; Billingsley & Tomchin, 1992; Crane & Iwanicki, 1991; Mastropieri, 2001;
Miller et al, 1999).

The procedural demands and paperwork facing special education teachers also
limit them from engaging in conversations with colleagues about instructional issues. In
particular, along with their teaching responsibilities, special education teachers must
complete individualized education plans (IEPs), conduct initial and follow-up evaluations
of students, and track special education goals and objectives. Billingsley, Carlson, and
Klein (2004) conducted a study of a nationally representative sample of special education
teachers with less than five years of experience and found that 72% of the teachers in
their sample indicated that routine duties and paperwork interfered with their teaching. In
her review of existing literature, Billingsley (2004) found a similar pattern, noting that
paperwork was a dominant factor influencing decisions to leave the classroom
(Billingsley, 2004).

With regard to mentoring and induction, research from both general and special
education suggests that mentors can play an important role for early career teachers

(Holloway, 2001; Miller, Brownell, & Smith, 1999; Smith & Ingersoll, 2004; Strong &

119

St. John, 2001; Whitaker, 2000). In general education, for example, Youngs (2007a)
reported that when mentors were knowledgeable about novice teachers’ curricula and
state assessments, they were more likely to help them plan instruction and analyze
student learning. In their analysis of the Schools and Staffing Survey, a nationally
representative sample of teachers, Smith and Ingersoll (2004) found that having a mentor
in one’s field reduced the risk of leaving teaching by approximately 30%.

Similar evidence has been found in special education research; when relationships
with mentors are informal and more personal, new special educators are more likely to
intend to stay in teaching (Whitaker, 2000). In addition, it seems that for beginning
special educators, mentors in special education can offer unique pedagogical and
assessment skills, knowledge of students, and knowledge of subject matter that general
education teachers cannot (Lane & Canosa, 1995; White, 1996). Further, such mentors
are well positioned to provide emotional support to early career special educators
(Whitaker, 2000; Kueker & Haensly, 1991). At the same time, many new special
education teachers are not assigned to mentors who teach special education in their
schools. And, even when a beginning special educator is matched with a mentor at their
school who teaches special education, the mentor may have a different job description
(i.e., they may be responsible for working with disabilities that differ from those to which
the novice is assigned) and, thus, may not be familiar with the novice’s curriculum or
caseload.

Research also suggests that induction programs that rely on formal mentoring
may not capture the important role that other colleagues within schools play in the

socialization of teachers. The degree to which a teacher feels supported by colleagues

120

appears to have a significant influence on whether or not novices stay in teaching
(Johnson & Birkeland, 2003; Johnson et al., 2004; Ingersoll, 2001; Kardos etal., 2001).
A recent study conducted by the Consortium on Chicago School Research reported that
informal support from colleagues, rather than formal mentoring programs, appeared to
have a more significant influence on whether or not novices stayed in teaching (Kapadia,
Coka, & Easton, 2007). Smith and Ingersoll (2004) found that having a common
planning period with other colleagues or collaborating with other teachers on instruction)
increased the rate of retention of new teachers by more than 43%.

Similarly, Billingsley, Carlson, and Klein (2004) foundthat when special
education teachers had access to support from colleagues and administrators, they were
more committed to staying in teaching. Although relatively few empirical studies have
directly examined the informal support available to new special education teachers,
research indicates that when new special education teachers do not interact with
colleagues, they are likely to struggle to meet the needs of their diverse students and to
experience high levels of stress. Further, the lack of support from colleagues also appears
to have consequences for their intent to remain in teaching, as well as actual retention
decisions (Billingsley, Carlson, & Klein, 2004; Billingsley & Tomchin, 1992;
Mastropieri, 2001; Miller, Brownell, & Smith, 1999). The degree to which administrators
support early career special education teachers also appears to have important
consequences; Billingsley, Carlson, and Klein (2004) found that when teachers reported
that their principals did not understand the responsibilities of a special education teacher,

they were most susceptible to feeling dissatisfied with their jobs. Lack of administrative

121

support was also cited as a strong determinant of stress for first year teachers by
Billingsley and Tomchin (1992) and Kilgore and Griffin (1998).

Research on mentoring and induction in general education has produced
comparable findings. For example, Kardos et al. (2001) found that new teachers were
well served in schools where veterans supported novices through mentoring,
collaborative work in grade teams and departments, and professional development. In
both studies, principals played a key role in successful induction experiences by
addressing issues related to instruction, collaboration, and student learning (Kardos et al.,
2001; Youngs, 2007b).

In research on retention, Smith and Ingersoll (2004) used the 1999-2000 SASS to
investigate the influence of induction activities on first-year teacher attrition and
migration while controlling for other factors. They reported that having a mentor in one’s
field reduced the risk of leaving teaching by about 30%, but had little effect on teacher
migration. In addition, collaborating regularly with colleagues on instruction reduced the
risk of leaving teaching by 43% and lowered the risk of migration by 25% (Smith &
Ingersoll, 2004). In contrast, recent research by Glazerman and colleagues (under review)
found no significant differences between schools implementing comprehensive induction
programs and those that did not, raising doubts about the potential impact of formal
induction programs.

In sum, research indicates that the school experiences of early career general
education teachers and special education teachers vary considerably. In particular,
compared to their counterparts in general education, new special education teachers

receive less guidance regarding curricular expectations and experience higher levels of

122

isolation at work. Studies in both general and special education have shown that, through
mentoring and induction programs, districts can play a proactive role in providing new
teachers with opportunities to gain resources, instructional skills, and support from their
colleagues. However, few of these studies have incorporated beginning general education
and special education teachers. As a consequence, it is unclear how the two groups of
teachers respond to the same set of district induction policies, as well as how their
experiences differ as they negotiate key relationships with mentors, special and general
education colleagues, administrators, and other individuals in their schools and districts.
Theoretical Framework

Social Capital Theory. To explain why the experiences of early career special
education and general education teachers differ, this paper advances a conceptual
framework of new teacher support based on social capital theory. This theory pays
specific attention to the process by which individuals access relationships with
colleagues, as well as to the consequences of these interactions on a teacher’s burnout
and commitment.

For a new teacher to be successful in their teaching, they can take steps to
increase their productivity by increasing their human capital (their value in the
marketplace based on their skills, intelligence and experience) and/or by increasing their
access to social capital. Social capital theory suggests that an individual can also invest in
their social relations (thereby gaining access to the resources of others) in order to
produce increased returns. While human capital is an individual-level attribute, social
capital can be thought of the resources embedded in an individual’s social context (Burt,

1992, 2000; Coleman, 1990; Lin, 1999; Putnam, 1993). Through a variety of avenues,

123

social capital provides advantages to individuals that would otherwise not be available to
them.

This paper adopts a definition of social capital based on that of Lin (1999), who
defines social capital as: “Investment in social relations by individuals through which

they gain access to embedded resources to enhance expected returns of instrumental or

expressive actions” (39-40).25 Lin’s theory remains rooted in individual interactions; he
suggests that there are group-level profits made available through social capital, but only
in the aggregate. Thus, if we return to the question of how relationships with colleagues
influence new special education teachers, it is important to locate the resources available
to a novice teacher through their interactions with colleagues, determine how early career
teachers incorporate these resources into their own thinking about their teaching, and
investigate how this social capital in turn impacts their perceptions of their work
environment and their affective response to these conditions.

To clearly measure social capital, Lin suggests a network approach in which one
locates the resources available to an individual based on who they interact with. Frank,
Zhao, and Borman (2004), for example, have done work in the area of the diffusion of
information about technology among networks of teachers. One of their central questions
was how teachers access expertise about technology (i.e., using computers in the
classroom) from their colleagues. Frank et al. calculated the social network’s influence on

an individual by summing the resources they received across all of the individuals they

 

5 This is in contrast to Coleman’s theory, in which social capital is defined as any aspect of a social
structure that is useful to the individuals in an organization. This is useful in how widely it can be applied,
but it also introduces problems with differentiating between the causes of social capital and its effects,
something that Lin discusses directly.

124

named. Their findings suggest that schools as organizations provide a context for sharing
resources that are present within the social network of teachers.

In this study, I conceptualize social capital as the help and expertise that teachers
access for their personal benefit. In an organization such as schools, individuals will not
interact with all available actors. Instead, individuals (“egos”) seek out certain specific
others (“alters”), exchanging resources with one another (Wellman & Frank, 2001). The
social capital available to individuals varies according to the resources embedded in the
network as well as their location within this network. Both of these factors are critical in
understanding the social capital available to early career teachers, as well as why special
educators and general educators may differ in their ability to access resources in their
school networks.

Collective Assets of the School Organization. Lin’s theory of social capital also
takes the important step of distinguishing social capital from other forms of collective
goods, such as trust, collective efficacy, and teachers’ perceptions of professional fit.
Separate from the social capital directly available to teachers through their interactions
with colleagues, more general characteristics of schools as organizations may have a less
direct, though still important, influence on the socialization of early career teachers. As
Lin argues, the process through which these organizational characteristics influence
individuals is by either constraining or enabling individuals to access the social capital
(the resources) that are embedded in the system.

Relational trust, as suggested by Bryk and Schneider (2002), produces advantages
at the organizational level in terms of how schools operate. When individuals in an

organization trust one another, efficiency in decision-making can be increased, because

125

individuals will be more likely to trust one another’s intentions. Also, high levels of
relational trust allow organizations to “coordinate meaningful collective action,” by
providing individuals with clear expectations of their role expectations. However, while
organizational-level benefits of relational trust are noteworthy, the contribution of
relational trust to the social capital literature described here is that its presence increases
the likelihood that individuals can make use of the social capital existent in the
organization. In organizations with high levels of relational trust, actors will be more
likely to mobilize the resources that exist in their relationships. In short, as described by
Hoy and Tschannen-Moran (1999), relational trust ensures that individuals have access to
cohesive and productive relationships.

A second, related asset of school organizations is collective responsibility, or the
degree to which the faculty share common goals for improving student learning (Lee &
Loeb, 2000; Lee & Smith, 1996; Newman & Associates, 1996). When a teaching faculty
views the success or failure of their students as a shared responsibility, it has been shown
to have a positive impact on student achievement gains (Lee & Loeb, 2000; Lee & Smith,
1996). And, although collective responsibility is an emergent feature of schools, it can
provide personal benefits to individuals. For early career teachers, having common goals
with colleagues may increase the likelihood that they feel connected to the school
organization. When individuals perceive that their colleagues are capable of producing
high levels of student learning, this will alter the ways in which they access and make use
of resources from their colleagues.

Finally, an important characteristic of schools as organizations is the degree to

which teachers feel that they “fit” into the overall school community (Kristof, 1996;

126

Kardos et al., 2001). If teachers feel that their beliefs and practices are aligned with their
mentor and other colleagues, it may increase the degree to which induction programs are
effective (Youngs, Qian, & Holdgreve-Resendez, in progress). Kardos and her colleagues
(2001) explain how the school culture—and whether or not it is supportive of new
teachers—can influence new teacher’s retention decisions. Outlining three different
forms of professional cultures (veteran-oriented cultures, novice-oriented cultures, and
integrated professional cultures), they found that in integrated cultures—where beginning
teachers received frequent support and interacted with colleagues across experience
levels—novice teachers were best served (Kardos et al., 2001).
On their own, each of these theories regarding the role of collective assets is useful in
examining the performance of individuals, but taken as part of a larger model of social
capital, they become powerful predictors of a series of individual-level returns.
Purpose

By studying the social capital available to new teachers (both directly through
their interactions with their mentors and colleagues as well as indirectly through their
perceptions of the school organization), I improve upon previous models of how mentors
and other colleagues influence early career teachers’ beliefs and career plans.
Analytically, this study consists of three separate research questions. The first research
goal is exploratory in nature; I investigate the degree to which early career special
education and general education teachers differ in their access to formal and informal
channels of support from mentors colleagues. Specifically, I look at how frequently early
career teachers interact with colleagues in regard to instructional issues, as well as the

content of these interactions.

127

Second, I investigate the association between early career teachers’ interactions
with colleagues and two relevant teacher outcomes: teachers’ stated career plans
(expressed as their commitment to their schools, district, and grade/subject) and their
levels of burnout. I quantify the importance of support novices receive from mentors and
colleagues, and also incorporate into these models teachers’ perceptions of the school
organization more generally (i.e., their perceptions of trust and collective responsibility,
as well as their perceptions of organizational fit). These models control for teachers’
personal and professional backgrounds, producing a more powerful model for estimating
the effect of the theoretical constructs on commitment and burnout. I also include
interaction terms for teaching special education and each of the theoretical variables of
interest, to determine, for example, whether high levels of relational trust have a
differential effect for special education teachers (as compared to general education
teachers).

Finally, while the previous research question indirectly measures the importance
of social capital for beginning special education and general education teachers, the final
goal of this study is to directly test the flow of resources and information from colleagues
to new teachers, using social network analyses. I incorporate survey data from mentors
and key colleagues into the early career teacher data, in order to ask whether social
network members’ own levels of commitment—as well as their perceptions of the
school—influence early career teachers’ levels of commitment and perceptions of their
schools. I also investigate whether teaching special education is associated with teacher

outcomes, once controlling for influence from key colleagues.

128

Method

Sample

The sample for this study included first-, second-, and third-year K-8 teachers
from eight school districts in Michigan and Indiana.26 The districts in the current analysis
were recruited because they were all medium to large in size and were hiring large
numbers of early career teachers (ECTs). Although the eight districts were somewhat
similar with respect to size and student composition, there were some important
differences, which are outlined in Table 19.

[Insert Table 19 here]

The two smallest districts in the sample, Kaline and Wagner, had total K-12
student populations of less than 10,000 students, while the largest district in the sample,
Greenberg, had over 21,000 students. Similarly, the districts varied with respect to the

percentages of their students who were racial/ethnic minorities, although all but one of

the districts (Daus) were over 40% minority.27 The districts were somewhat similar with
respect to the percentage of their students who were from low-income backgrounds. The
largest district in the sample, Greenberg, had the highest number of low-income students,
with 65% eligible for free or reduced lunch, while Wagner--the smallest district in the
sample--had the fewest low-income students (36%). Given the tight fiscal climate in both
states, we were limited in the pool of districts who were hiring new teachers; thus, the

districts differ in size and student composition

 

26 . . . . . .
Teachers who were new to their district but who had more than three years of prior teaching experience
were not included in this study.

27 Daus had a high percentage of Arab-American students (i.e., more than 50%), many of whom were
identified as non-minority. This helps to explain the low percentage of non-white students in the district.

1.29

Early Career Teacher Sample. In selecting general education teachers to
participate in the study, we required that teachers teach in a core content area (math,
science, social studies, English/language arts, and elementary general education) in
grades 1-8, and we also required that they had standard teaching certificates and had
completed university-based teacher preparation programs. For special education teachers,
we also required that they be providing academic instruction in grades 1-8 (i.e., we
excluded individuals who did not provide instruction, such as school psychologists and
speech pathologists). The response rate for the two groups of teachers was comparable,
with 68% of general education teachers and 67% of special education teachers
completing the fall 2007 survey; we also had a spring retention rate of approximately
75%. In total, the sample for this study included 185 teachers: 47 special education
teachers and 135 general education teachers. The characteristics of the early career
teacher sample are provided in Table 20.

[Insert Table 20 here]
Seventy percent of the teachers in the sample were elementary school teachers (n=129),
and the number of teachers in their first, second, and third year of teaching were
distributed fairly evenly (61 first-year teachers, 80 second-year teachers, 43 third-year
teachers). Additionally, the overwhelming majority of the teachers were white (90%) and
female (83%).

Mentor/Colleague Sample. To allow for social network analysis, we asked each
early career teacher in the sample to provide the name of their mentor and up to eight

colleagues at their schools with whom they engaged in professional interactions (e.g.,

130

interactions about curriculum, instruction, students, and school policies).28 We then

contacted up to 4 of the classroom teachers nominated by the ECTs; colleagues who were

not full-time classroom teachers or who were instructional support providers29 were not
asked to participate in the study. Of the teachers who were contacted, 72 teachers named
as mentors by ECTs and 325 named as colleagues completed surveys in winter 2008. The
overall response rate for mentors and colleagues was approximately 61%.
Data

All participating ECTs completed surveys at two time points, in the fall of 2007
and spring of 2008. The surveys asked questions related to the teachers’ backgrounds,
professional preparation, interactions with others, experiences in their schools, and their
future career plans. In order to understand the content of the conversations of ECTs with
their mentors and colleagues (e.g., curriculum, instruction, classroom management,
psychological support), as well as how frequently they had professional interactions,
novices were also asked to provide information on their relationships with their mentor
and, separately, colleagues. In addition to describing the content and frequency of their
interactions, the novices also rated the usefulness of each of these relationships. Finally,
to account for changes in teachers’ levels of burnout and commitment from fall to spring,
teachers were asked the same questions at both time points. This allowed me to more
accurately model the impact of mentors, colleagues, and school-level characteristics on

early career teacher outcomes.

 

28 ECTs were asked to nominate colleagues in fall 2007.

29 . . . . .
Examples of Instructional support provrders included literacy or math coaches, as well as mentors who
have been released from teaching.

131

Similarly, to compare ECTs to their mentors and colleagues, identical items were
asked in both the ECT and the mentor/colleague surveys. Specifically, colleagues were
asked about the school organization (i.e., their perceptions of relational trust, collective
responsibility, and their fit within the organization), as well as their own future career
plans. Data was also collected on the experienced teachers’ educational backgrounds,
demographic characteristics, and whether they had prior experiences as mentors.
Measures

Teacher Outcomes. This analysis focused on two teacher outcomes. In the special
education literature, burnout has been an important concept for understanding the
struggles of early career special education teachers. For special educators, a common
response to overwhelming or negative job circumstances is a high level of stress, which
has been shown to influence commitment (Billingsley & Cross, 1992; Gersten et al.,
2001; Wisniewski & Gargiulo, 1997) and attrition (Miller, Brownell, & Smith, 1999;
Singer, 1993). When this stress goes unaddressed over a long period of time, and
negative experiences accumulate, the teacher is likely to find their work increasingly
unsatisfying. The term burnout has been used to describe the accumulation of this stress.
In this study, I measure burnout through a battery of items taken from the Maslach
Burnout Inventory (Maslach & Jackson, 1981), such as “I feel used up at the end of the
day” and “I feel that I’m working too hard at my job.” Responses for each scale variable
ranged from 0 = “strongly disagree” to 4 = “strongly agree,” and the composite variable
was the mean of these six items (a = .91). The mean level of burnout for general

education teachers in the sample was 2.39 (standard deviation (SD) = .67); the mean for

132

special education teachers was 2.46 (SD = .71). Table 29 in the Appendix provides
descriptive statistics for all variables included in these analyses.

The second outcome used in this analysis is commitment, which as Ingersoll
(2001b) has argued, is believed to be a more relevant outcome variable for policymakers
than retention itself, because commitment more closely reflects attitudes about one’s job.
Retention decisions, on the other hand, may result from circumstances not having to do
with work itself (e.g., the desire to move out of the area, start a family, or pursue
additional schooling) or due to declines in student enrollment and/or district finances.
Commitment has been a useful construct for understanding how characteristics of
individuals’ work environments predict intentions to stay in the profession. Ebmeier
(2003) has found teacher commitment to be associated with teacher effort and
effectiveness, and he also suggests that commitment is predicted by relationships with
teacher colleagues, a finding supported in other research (Bryk & Schneider, 2002;
Desimone et al., 2002). Finally, commitment has been to shown to be predictive of career
decisions (Weisberg & Sagie, 1999; Weiss, 1999).

To measure commitment, I considered three sets of items that addressed teachers’
commitment to their school, grade or subject area, and district. For each category of
commitment, the survey included two questions, one asking the teacher about their
commitment to school, subject/grade, and district for the next school year and the second
asking about their commitment in five years. Responses to these items ranged from 0 =
“strongly disagree” to 4 = “strongly agree.” Comparisons of teachers’ responses to these
items revealed little variation (i.e., teachers differed very little in how they

conceptualized commitment to school, grade/subject, and district). Consequently, I

133

collapsed the total of six items into one composite measure of commitment, representing
teachers’ career plans overall. The final composite variable had an internal consistency,
as measured by Cronbach’s alpha ((1 ), of .90. The mean of this commitment variable for
general education teachers was 3.36 (SD = .69); the mean for special education teachers
was 3.06 (SD = .81).

Support fiom Mentors/Colleagues. For the descriptive analysis (in which I
compared the nature and frequency of interactions with mentors/colleagues for special
education and general education ECTs), I looked at how frequently ECTs interacted with
their colleagues around six categories of professional support: curriculum, teaching
strategies, student behavior, assessments, standardized testing, and psychological support.
Responses were coded as the number of school days per month that ECTs interacted with
colleagues about a given topic. For the regression analyses, 1 used an additional measure
of colleague support, based on how much the novice teachers valued the support they
received from mentors and colleagues. I treated “importance of mentor support” and
“importance of colleague support” as two different dummy variables, where 0 = support
is “not important at all” or “somewhat important,” and 1 = “very important” or
“extremely important.”

Collective Assets of the Social Structure. Variables were created to represent the
theoretical constructs of relational trust, collective responsibility, and perception of fit—
all of which were considered to be additional sources of support for early career teachers.
For each variable, I created a composite measure by taking the mean of single scale items
from the ECT survey. Relational trust consisted of items such as “teachers in this school

trust each other” and “teachers in this school respect those colleagues who are experts in

134

their craft;” responses ranged from 0 = “strongly disagree” to 4 = “strongly agree.” This
composite variable (a=.89) had a mean of 3.16 for general education teachers and 3.02
for special educators. Perception of fit consisted of 6 survey items, and the resulting
composite variable (a=.89) had a mean of 3.34 for general educators and 3.26 for special
educators. Lastly, the items making up collective responsibility asked teachers about the
proportion of their school-based colleagues who took responsibility for such things as
student discipline and student learning, with responses ranging from 1 = none to 5 = all.
The mean for the composite variable for collective responsibility was 3.61 for general
education ECTs and 3.58 for special education ECTs (a = .90).

Person- and School- Demographic Characteristics. In addition to special
education status, the models included other teacher demographic variables, including
whether they taught in middle school or in elementary school, and whether they were in
their first, second, or third year of teaching. Teaching experience was classified as a
binary variable, with second- and third-year teachers being collapsed into one category;
this step allowed me to account for the unique factors facing first-year teachers, including
negotiating new relationships with colleagues and students, and learning a new
curriculum-Although the majority of the teachers in the sample were white females, 1
also included race and gender in all models.

Analytic Approach

Descriptive Analysis. Given that few studies have jointly investigated the
experiences of beginning general education teachers and special education teachers, this
analysis began with a descriptive comparison of the formal and informal support

available to the two groups of teachers. Characteristics of mentors were determined based

135

on answers in the colleague surveys, while ECT responses were used to calculate mean
levels of interaction between ECTs and their mentors, as well as ECTs and their
colleagues. The last step in the descriptive analysis was to measure the degree to which
novices saw the support from mentors and colleagues as important; as with the previous
comparisons, means were calculated separately for general education and special
education teachers.

Regression Analysis. Following the descriptive analysis, I estimated the
relationship between teachers’ perceptions of support from colleagues (as well as their
perceptions of the resources embedded in the broader school context) and their levels of
burnout and commitment. The models predicting burnout were framed as an exploration
of school factors that may mediate the relationship between burnout in the fall and
burnout in the spring, while the models predicting commitment investigated the factors
that might increase the likelihood that teachers were committed in the spring.

The models for this analysis were run in three stages. First, I ran separate ordinary

least squares (OLS) regression models30 predicting burnout and commitment, focusing
specifically on the measures of mentor and colleague support. I then included the
predictors for the three measures of collective assets (i.e., relational trust, collective
responsibility, and perceptions of fit) to determine if the expanded models better
predicted the teacher outcomes. Finally, I investigated the degree to which there was an
interaction between teaching special education and each of the predictor variables of

interest. For the purpose of interpretation, these interaction terms were considered one at

 

30 . . . . .
I had consndered usmg HLM to account for the nesting of teachers Wltl'lln schools. However, the average

number of cases within schools was two, suggesting that multilevel models would add not contribute
greatly to my understanding of the relationship between support and teacher outcomes. Additionally, an
unconditional model (i.e., a multilevel model with no predictors) indicated that for spring burnout, there
was no school-level variation.

136

a time. At each stage of the analysis, I paid specific attention to the coefficient for the
variable indicating whether a teacher was in special education. Additionally, each model
included as a control the teacher’s prior measure of commitment or burnout, in order to
more closely estimate the effect of the predictors of interest. These models controlled for
the personal and professional backgrounds of each novice teacher, producing a more
powerful model for estimating the effect of the theoretical constructs on commitment.

Social Network Analyses. In addition, I used social network analysis to measure
the influence of colleagues’ levels of commitment and perception of the school
environment on early career teachers’ own commitment and beliefs. This analysis
provided a more direct measure of social capital, taking into account both the “resources”
that were being exchanged (e.g., levels of commitment) and the frequency of ECT-
colleague interactions. It was hypothesized that colleagues’ commitment and perceptions
of the school organization (i.e., relational trust, perception of fit, and collective
responsibility) would influence early career teachers’ levels of these measures. The
model below represents how I have conceptualized and measured the influence of

colleagues on new teachers, using the example of teacher commitment:

commitit = (commititq) + [31 [2 Xip Yim] + [32(flagi) + [33(special education teacheri) + e;
ni‘

In this model, the outcome variable of teacher commitment is represented for

novice teacher i at time t (spring of 2008), and the influence of the network is captured by

137

B]. The network, 2 Xii' Yi-H, represents the mean31 influence of the ECT-colleague

relationship (Xii) and the colleague’s commitment at time t-l (Yim). This total figure

captures the expertise of the others with whom one talks as well as the frequency of a
novice’s interactions with these individuals. In order to measure change over time as a

result of interactions with colleagues and school-level factors, the model also controls for
teachers’ commitment at time t-l (fall of 2007). The model also includes a flag (flag,) for

whether the teacher had data from their colleagues; this ensured that participants without
data from their colleagues were not dropped from the model. Finally, 1 developed similar
models for predicting the influence of colleagues’ perceptions of three school context
variables (relational trust, perception of fit, and collective responsibility) on ECTs’ own
perceptions of these measures.
Results

Descriptive Results

Support from Mentors. Table 19 provides information about the characteristics of
mentor relationships for special education and general education teachers. Seventy-seven
percent of special education teachers and 69% of general education teachers in the
sample had formal mentors.32 A surprising difference was the percent of teachers whose
mentors were full-time teachers in their schools; in special education, 62% had full-time
teachers in their schools as mentors, compared to 91% in general education. With fewer

special education teachers per school, the number of eligible “same-subject” mentors is

 

31 The mean is derived by dividing the sum of the network influence and dividing it by the number of
colleagues with which the novice interacted (ny).

32 This is partly attributable to the fact that more general education teachers were in their third year of
teaching than special education teachers (25% versus 17%).

138

much smaller, which likely explains the difference between the groups in their frequency
of interactions with mentors.
[Insert Table 19 here]

As shown in Figure l, a comparison of the frequency of interaction with mentors
shows important differences between special educators and general educators. The
majority of special education ECTs reported interacting with their mentors between “1-3
times per month” and “1-2” times per week. The frequency of interaction between
general education teachers and their colleagues was more broadly distributed; forty
percent interacted with their colleagues between “3-4 times per week” and “every day,”
versus 21% in special education.

[Insert Figure 1 here]

Figure 2 presents the frequency of interaction between ECTs and their mentors
around specific content areas, expressed in days per month. The largest disparities
between ECT-mentor interactions for special education and general education teachers
was in the area of curriculum, where special education teachers spent approximately 1.2
days per month interacting with mentors and general education teachers spent 5.2 days
per month; a Wilcoxon rank-sum test revealed this mean difference to be significant at p
< .01. Similarly, the difference in means for interacting about teaching strategies and
assessments were each significant at p <.05. The most frequent topic of interaction
between special education teachers and their mentors was student behavior, at
approximately 3.2 days on average per month.

[Insert Figure 2 here]

139

Support from Colleagues. When comparing special education and general
education teachers’ frequency of interaction with their colleagues around specific content
areas, there were fewer significant differences. Both groups of teachers interacted with
their colleagues on a more frequent basis than their mentors, as reflected in Figure 3.
Student behavior was the most frequent area of conversation, with general education
teachers talking with their colleagues about student behavior on average 8.9 days per
month and special education teachers interacting with their colleagues 10 days per month.
Curriculum, teaching strategies, and psychological support were frequent topics of
conversation for both groups of teachers; the only significant difference (as measured by
a Wilcoxon rank-sum test) was that special education teachers spent significantly more
time per month talking with their colleagues about psychological support. Assessments
and standardized testing were less frequent topics of conversation for both groups.

[Insert Figure 3 here]

Lastly, teachers were asked to assess the importance of the support they received
from their mentors and colleagues, with response categories ranging from “not at all
important” to “extremely important.” In line with the results reported in the above
figures, ECTs placed a higher value on the support they received from colleagues (special
education (SE) = 3.53; general education (GE) = 3.42), than they did from mentors (SE =
2.95; GB = 3.05). Given the high average levels of importance of support from
colleagues, in the next section’s regression analyses, I treated colleague support as a

dummy variable where 0 is “not important at all” or “somewhat important” and 1 = “very

140

important” or “extremely important.”33 Overall, these analyses suggest that the mentor-
mentee relationship looks different for special education teachers, with general education
teachers having closer professional relationships with their mentors. In contrast, both
groups more heavily rely on support from colleagues.
Regression Results

Predicting Teacher Burnout. The results from the regression analyses predicting
spring levels of teacher burnout (once controlling for fall burnout, teacher characteristics,
and the theoretical predictors of interest) are presented in Table 22. Not surprisingly, the
most important predictor of spring burnout was burnout in the fall, with a standardized
coefficient of .59 (p301). Teachers who reported that support from colleagues was “very
important” or “extremely important” were significantly less likely to have high levels of
burnout in the spring. The standardized coefficient for colleague support, -.29, was
negative and significant at p S .05. Also important was the lack of significance for the
special education variable, suggesting that, controlling for all other variables, general
education and special education teachers had common levels of burnout.

[Insert Table 22 here]

When the collective assets variables (i.e., perception of fit, relational trust, and

collective responsibility) were included in the model, none were significantly associated

with burnout, which suggests that the presence or absence of these forms of support for

early career teachers is likely not important for teachers’ levels of burnout.34 The

variance explained in the model predicting burnout increased from (R2 = .41) to (R2 =

 

33 . .
Because only 73% of teachers had mentors, and because the relative importance of support from
mentors was less than support from colleagues, this variable was not included in the regression analyses.

34 . . . . . . . . . .
The standardized coeffiCient for collective responSibility (-.16), however, is suggestive of an association
with burnout.

141

.53) with the addition of the collective asset variables; however, once these variables
were included, the coefficient for colleague support decreased from -.13 to -.09. To test
for multicollinearity in my predictors, I ran a test of variance inflation and found that

none of the included variables presented serious threats of multicollinearity—no variables

had variance inflation factors over 2.00.35

Predicting Teacher Commitment. When predicting teacher’s levels of spring
commitment, the first model only included teacher characteristics and the variable for the
importance of support from colleagues. As in the model predicting burnout, both prior
commitment and colleague support were positively associated with spring commitment.
The coefficient for prior commitment was .51, while colleague support had a
standardized coefficient of .13, and both of these values were significant at p S .01. In
addition, two of the teacher characteristic variables were significant once controlling for
prior commitment and support from colleagues; the coefficients for teaching special
education (-.14) and teaching middle school (-.23) were both significant at p S .01.

[Insert Table 23 here]

Many of the significant correlations in the first model of commitment decreased
once the three collective asset variables were included in the analysis. The coefficients
for prior commitment (.42), special education (-.1 l), and middle school (-.20) decreased
in size but remained significant at p S .01; in contrast, colleague support decreased to .07
and was no longer significant. Of the three collective asset measures, perception of fit and
collective responsibility were both positively associated with spring commitment, with

coefficients of .22 and .17 respectively; these values were significant at p S .01. Also, the

 

35 . . . . . . . .
Tests for multicollinearity were also non-Significant in each of the models predicting teacher
commitment.

142

inclusion of the collective asset measures resulted in a change in R2 of .09, suggesting
that the expanded model did a better job of explaining the variance in teachers’ levels of
commitment. Finally, the size of the coefficient for teaching special education—one of
the measures of theoretical importance to this analysis—indicated that testing for
interaction with other important predictors was warranted.

Four interaction terms were tested in separate models in order to determine
whether the relationship between teaching special education and commitment was
dependent on higher levels of colleague support (as measured by the “importance of
colleague support” variable and by the three collective asset variables) As outlined in
Table 24, the interaction terms for support from colleagues, perception of fit, and
collective responsibility were significant at p S .05, while relational trust was significant
at p S .10.

[Insert Table 24 here]
These interaction terms indicate that the relationship between colleague support variables
and teacher commitment functions differently for the two groups of teachers. Given that

special education teachers have lower levels of commitment in the spring to begin with,

the positive coefficients36 for the interaction coefficients suggest that the importance of
colleague support and a positive school environment is greater for novice special
education teachers than novice general educators. This is especially the case for the
variables of collective responsibility and perception of fit, both of which were
significantly correlated with commitment. An analysis of the scatter plots of each of the

collective assets on spring commitment confirms what the findings from the regression

 

36 . . . . . . . . . . .
The standardized coefficrents for the interaction variables were derived by standardizmg the indiVidual

variables before calculating their product.

143

models suggest. In each plot, we see that higher values for the collective asset measures
are more strongly correlated with commitment for special education teachers than they
are for general education teachers.

[Insert Figure 4.1 — 4.3 here]
Social Network Analysis Results

In the previous step, the regression analyses used indirect measures of the

influence of social capital on teacher commitment and teacher burnout. However, these
variables do not capture how interactions with colleagues actually influence early career
teachers’ attitudes about their work. Therefore, the advantage of using social network
analysis is that it directly measures the manifestation of social capital by accounting both
for the frequency of interactions between ECTs and their colleagues and for the resources
embedded in the teacher’s network (in this case, colleagues’ own levels of commitment,
as well as their perception of collective responsibility, fit, and relational trust). The
question underlying this analysis, then, is the degree to which colleagues’ attitudes about
work influence the attitudes of early career teachers. Tables 25 — 28 show the results of
these social network analyses.

[Insert Tables 25 - 28 here].

While the focus of each of the four models was on the influence variable (which
represented the mean influence measured across all colleagues for a given ECT), each
model also included a prior-level measure of the attitudinal measure of interest, a flag
variable for whether the teacher had influence data available, and a control for whether
the teacher taught special education. My results show that the influence variable was

significantly and positively associated with ECT spring measures of attitudes. For

144

predicting ECT commitment, the influence variable for colleagues’ level of commitment
had a standardized coefficient of .29. The size of this coefficient is similar to those
calculated in the other three influence models--the coefficient for influence in the
perception of fit model was .30, the coefficient in the perception of trust model was .25,
and in the collective responsibility model, it was .34. On average, a standard deviation in
influence was associated with between .25-.34 standard units in early career teachers’
own attitudes. Lastly, the coefficients for the special education variable were significant
in all but the collective responsibility model, indicating that even once controlling for
prior attitudes and controlling for the influence of colleague attitudes, teaching special
education was still associated with teachers’ attitudes (i.e., their commitment, their
perception of fit in the school organization, and their perceptions of relational trust).
Discussion

The primary purpose of this study was to compare the experiences of early career
special education and general education teachers as they became socialized into their
schools. I explored differences in their interactions with their mentor and informal
colleagues—in terms of the frequency and content of these interactions. Additionally, I
investigated the association between the support available to new teachers and two
relevant teacher outcomes: their overall commitment to their jobs and their levels of
burnout. These models included predictor variables for the importance of support novices
received from colleagues, as well as teachers’ perceptions of the school organization
more generally (i.e., their perceptions of trust and collective responsibility, as well as
their perceptions of organizational fit). I also included interaction terms to see if the

sources of support in the school organization had a differential impact on general

145

educators and special educators. Finally, I used social network analyses to investigate the
direct influence of colleagues’ commitment and perceptions of the school organization on
early career teachers own attitudes, incorporating survey data from ECTs’ key
colleagues.

Overall, my findings point to important differences in the induction experiences
of early career special education and general education teachers, and I provide evidence
to suggest that these differences are critical for teachers’ career plans. My descriptive
analysis suggests that even though special education teachers and general education
teachers are equally likely to have a mentor, fewer special education teachers have
mentors who are teachers in their own schools. This is a likely reason why general
education teachers engage in professional interactions much more frequently with their
mentors, and do so across a wide variety of topics. The two groups of teachers seem to
place relatively equal importance in the support they receive from informal colleagues in
their schools; ECTs across the sample most frequently engaged with their colleagues
about curriculum, teaching strategies, and psychological support, and student behavior.
These results suggest that the support early career teachers receive from formal mentors
is likely less valuable than the support from informal relationships with their colleagues.
And, overall, the special education teachers in this sample do not appear more isolated
from their colleagues than general education teachers.

The regression analyses predicting teacher burnout and commitment support this
conclusion. For both groups of teachers, support from colleagues was negatively
associated with teacher burnout, suggesting that colleagues can, in some cases, help

novice teachers cope with the stress and turmoil of learning to teach. More important was

146

the role of colleagues in predicting higher levels of teacher commitment, even when
controlling for prior commitment. The findings indicate that when early career teachers
feel that they fit into the school environment, and when they perceive high levels of
collective responsibility among the faculty, they are likely to be more committed. This is
particularly true for special education teachers; while they have lower average levels of
commitment, results from analyses including interaction terms for special education show
that, for special educators, higher levels of support are more closely associated with high
levels of commitment.

When I measured the influence of colleagues’ attitudes on the attitudes of early
career teachers, I found a direct relationship between colleagues’ beliefs about their
school organization and ECTs’ own beliefs; similar results were found for ECTs’ future
career plans. These results are in line with previous research by Frank and colleagues
regarding how teachers’ beliefs are influenced by others in their social network. Further,
given the positive role of collective assets-- such as collective responsibility and
professional fit--in predicting teacher commitment, these findings suggest that the role of
informal relationships with colleagues should not be overlooked.

Theoretically, this study expands on research suggesting that school organizations
may provide an important source of social capital for new teachers. When a new teacher
feels that they fit in professionally with their colleagues, it will likely increase the
frequency with which they reach out to their peers for guidance and assistance. Similarly,
if individuals perceive that there are high levels of collective responsibility in their
schools (i.e., their colleagues are capable of producing high levels of student learning),

this may alter the ways in which they access and make use of resources from colleagues.

147

When novice teachers feel that support is available to them—and when they value this
support—it is no wonder that they are likely to feel more committed to their schools
(Kardos et al., 2001).

The results from my influence models suggest that teachers do not develop their
attitudes in isolation; rather, their judgments about their school are derived both from
their own previous beliefs but also the beliefs of others around them. As Frank, Zhao, and
Borman (2004) write: “The function of social capital in intraorganizational diffusion
helps us understand the transitions between the macrolevel social entity of the
organization and the microlevel action of independent individuals” (p. 162). My findings
support-this claim, suggesting that social capital can serve as a bridge between the
broader school context and teachers’ beliefs. Although my analysis is limited to
collective assets and teachers’ career plans, it is likely that this process is replicated with
other resources embedded in novices’ social networks. For example, just as Frank and
colleagues show that teachers access help and expertise through their interactions with
colleagues, a similar argument could be made about the influence colleagues have on the
instructional beliefs and practices of new teachers. If this is the case, school- and district-
Ieaders would do well to take a more strategic role in structuring early career teachers’
interactions with colleagues.

Limitations

Several limitations of this study are worth noting. There are likely factors of
teachers’ induction experiences that I have not adequately accounted for. For example,
research has suggested that principal support is a key component of a successful

induction program (Kardos et al., 2001; Youngs, 2007b). Particularly for special

148

educators, principals can foster collaboration with general education colleagues, provide
curricular and instructional support directly aimed at special educators, and may shape
the overall climate of schools. Principals also often take an active role in assigning early
career teachers to mentors; given the proportionately low percentage of special education
teachers with mentors in their schools, this could be critical for their access to a
supportive mentoring relationship.

A second source of support not measured in this analysis was the role of external
resources—family, friends, and colleagues who were not in the same schools as the
participants. Even if a new teacher feels isolated from their colleagues, or lacks support
around certain areas of instruction, they very well may gain access to support from
external sources. This support could ultimately serve as a buffer against high levels of
burnout or plans to leave the profession. These individuals were excluded from this

analysis because of my interest in the role of schools as organizations; also, it was

presumed that policy can only leverage resources available to teachers in their schools.37

Another potential source of error is that I did not collect data on teachers’
personal characteristics (including their disposition) that may have shaped how
participants responded to their school environments. Data was gathered on teachers’
educational background and basic demographic characteristics (e.g., race, gender,
undergraduate institution), but there are likely other factors that shape teachers’ responses
to challenging work conditions. Examples of these characteristics are teachers’ reasons

for entering the profession and the degree to which they teachers were introverted or

 

37 Although the surveys in the study asked teachers how important they rate their nonschool-based
channels, systematic data was not collected on these individuals.

149

extroverted. However, the fact that I collected data on prior levels of commitment and
burnout likely lessens the potential impact of personal characteristics to some degree, yet
future research should pay closer attention to these differences.

Finally, researchers in special education would likely argue that l have not
adequately explored the variation within the special education teacher sample. For
example, teachers of students with learning disabilities or mild cognitive impairment are
more likely to spend their time interacting with general education teachers because they
may share responsibility for mainstreamed students. On the other hand, this experience
may contrast greatly from teachers of students with severe multiple impairments; there
will likely be fewer opportunities for these teachers to collaborate on instruction with
other teachers in their schools. To make claims about these teachers as a group overlooks
the varying needs of special educators, as well as potential differences in whether these
needs are being met.

Finally, the analyses of colleague data that I have conducted are relatively simple
applications of social network analysis. Possible extensions of this study would be to look
at colleagues’ influence on commitment for other (and potentially multiple) predictors. If,
for example, experienced teachers complained to novices about the negative aspects of
the school organization (and perceived that there were low levels of trust or collective
responsibility) this could very well influence early career teachers’ commitment to their
schools. The current study, however, did not consider these relationships. Additionally, I
have employed egocentric social network analyses—Le, I collected data on colleagues
based on who the ECT nominated. As a result, my estimations of teachers’ working

conditions are based on early career teachers’ perceptions. If I had collected sociocentric

150

data by surveying entire schools, 1 would have had a more complete picture of the school
as an organization.
Implications

Previous studies on induction in special education have focused primarily on the
relationship between beginning teachers and their mentors, but have not provided a
comprehensive analysis of how novice special educators become socialized into their
schools. The purpose of this study was to analyze the ways in which beginning special
education teachers’ mentors and peers (collectively thought of as their social network)
influence their experiences in their early careers, and to compare these results to a sample
of novice general educators. By focusing on social networks rather than on mentor-
mentee relationships alone, I am able to provide a more comprehensive picture of the
qualities of the induction experience that increase the likelihood that new special
education teachers feel competent in their profession, feel committed to their schools, and
intend to stay in teaching.

This study provides researchers, policymakers, and district administrators with
ways of better supporting special educators and general education teachers in their early
years of teaching. Researchers are becoming increasingly mindful of the ways in which
relationships in schools impact the work lives and career decisions of teachers, yet few
existing studies address patterns unique to special education. The findings of this study
demonstrate that informal relationships are critical for the experiences of new teachers, as
are the broader school conditions in which these new teachers work. Strong relationships
with colleagues—and positive perceptions of the school environment—are likely to

reduce the attrition of both special educators and general educators. As evidence from

151

these analyses suggests, for new special education teachers, these channels of support are

especially salient.

152

Appendix A

Table 19: Demographic Characteristics of Districts in the Sample

 

# of Schools in Total K-12 % Eligible for
District . . Student % Minority Free or Reduced
District P l .
opu ation Lunch
Daus (MI) 36 18386 0.12 0.51
Greenberg (MI) 80 21448 0.80 0.65
Kaline (MI) 18 9139 0.50 0.42
Wagner (MI) 12 7994 0.46 0.36
Engram (IN) 20 13666 0.48 0.62
Luckman (IN) ’1 8 16138 0.57 0.44
Payton (IN) 13 10662 0.84 0.50
Wilson (IN) 19 12483 0.59 0.58

 

Table 20: Demographic Characteristics of Teachers in the Sample

 

Total n=185 %
Special Ed. Status

General Education 138 75%

Special Education 47 25%
Grade Level

Elementary School 129 70%

Middle School 56 30%
Year of Teaching

First 61 34%

Second 80 43%

Third 43 23%
Race/Ethnicity

White 167 90%

African-Ameri can 1 3 7%

Other 5 3%
Gender

Male 31 17%

Female 154 83%

 

153

Table 21: Characteristics of Mentoring Relationships

 

Special Ed.

Percent w/ mentor 77%
Percent of mentors who are full-time

. 62%
teachers in same school
Percent w/ mentor match 59%
Frequency of Interaction w/Mentors (in 7 00
days/month) '

General Ed.

69%
91 %
49%
9.42

 

154

Figure 1: Frequency of Interactions Between ECTs and their Mentors

 

35
30
25 —
20 '*i

I Special Ed.

'/o of Teachers

D General Ed.

  

 

 

 

Never Less than 1-3 times per 1-2 times per 3-4 times per Every day
once a month month week week

Frequency of Interaction

Figure 2: Content of Interactions Between ECTs and their Mentors

 

 

I Special Ed.
D General Ed.

 

Days per Month

 

 

 

 

 

 

 

Curriculum Teaching Student Assessrmnts Standardized Psychological
Strategies Behavior Testing Suppon

155

Figure 3: Content of Interactions Between ECTs and Colleagues

Days per Month

 

 

 

 

 

 

Cm'riculum Teaching Student Assessrmnts Standardized Psyclnbgical
Strategies Behavior Testing Support

156

 

I Special Ed.

T DGeneral Ed.

 

Table 22: Models for Predicting Teacher Burnout-—Spring (n=164)

 

 

 

Model 1 Model [I
_ Unstandardized Unstandardized
Independent Var iable Coefficient Standardized Coefficient Standardized
(SE) Coefficient (SE) Coefficient
63*“ .59 .58*** .54
Bumout—Fall ('07) (.07)
-.29* -. l 3 -.22 -.09
Colleague Support (.16) (.16)
Collective Assets
. . -.04 -.03
Perception of Fit (.10)
. -.02 -.02
Relational Trust (.09)
. . . . -.15 -.16
Collective Responsrbility (.07)
Teacher Characteristics
. .08 .06 .07 .05
SpeCial Ed. Teacher (.09) (.09)
Middle School Teacher '06 ‘ 02 '01 00
(.09) ' (.09) '
. .04 .03 .07 .05
First Year Teacher (.09) (.09)
. .05 .02 -.03 -.01
mm" (.14) (.14)
Female -.18 -.10 -.15 -.09
(.11) (.11)
Interce t 1.26 2.12
P (.29) (.43)
R-squared .4 l .5 3

 

Note: *p S .10; "p S .05; "*p S .01 (2-tailed).

157

Table 23: Models for Predicting Teacher Commitment—Spring (n=l67)

 

 

 

Model 1 Model 11
Independent Variable urginflhfiglfd Standardized UhCSctilant‘lifighied Standardized
(SE) C 06 fficient (SE) Coefficient
Commitment—F all (07)” 51 (37)“ .42
Colleague Support (g; * * -l3 (i i g) .07
Collective Assets
Perception of Fit (3133);" .22
Relational Trust (:33) .03
Collective Responsibility (2:379; * * ~17
Teacher Characteristics
Special Ed. Teacher 2213);" ‘14 Thai)” -.11
Middle School Teacher I33; * it "23 2329'; * "‘ -.20
First Year Teacher (73):) '-05 (“399) --06
White (if) --05 ((1)2) .00
Female 0(1)?) .00 (ill?) -.02
l .28 * "' * -.09
Intercept (.32) (‘37)
R-squared .47 ' 5 6

 

Note: *p S .10; "p S .05; ***p S .01 (2-tailed).

158

 

 

 

 

an. a. 3. am. Essa--2
e ...m. e a...
8.- 2mm» 8.- ”NW-C 8.- Mww 3.- Maw-e .205 30> E:
.- E: .- 33 .- as 53 .
a {ea—MM 2 3.....wa 2 align“ ......amm... .550on Begum BE
.- 6.. .- 6... .- am. .- :2 .
: ...-ofi- : :8.- 2 ...-mm..- 2 ...-8.7 essay 8 Exam
2. Ammo 230:8 .. 8%
a...
3. MW is .. Guam
2. a: .. E .. Guam
*emm .
E. 3C 533:8 .. 8%
new»
. £3 . E: . E: . E: ezeeaem
VN ...o_. 2 .3:- w_ tin—- mm .30—- 0382—00
33 33 £3 33
S. . 8. . S. . 8. . as; .2233.
S 8 - 8 8 .
6: 8: 2 E 33
. . . . Q0 0
MN «amvm.~ _N ........Nm.V _N ....Mmmu.V NN tinny urn—mo :0: o.— m
3. :2. 2. 3:.
no 8. S 8. 8 cm. 2 3. e am 2. 8:8
. E: . E: . E: E:
. n i=0 .EEO
:. ...-e... z. ...-.3. a. 2:3. 3. ...-3.. = ... . .5. 0
Ga are m5 m6
nfihflfleflfim 568%; “thmfiegcmwm commocwom “WNW-“35W cognac-m uflMW-M-"HMW scamewom 033:; ”concave:—
. So . who . So . So
s.=.e=o..o-Qm% est-3% 5.3% .temmsmzoo-Qmmm

 

33".: metamlaeofizfifieu ..oguaoh “53:5...— ué £3.32 "en 03.:-

159

Figure 4.1: Scatter Plot of the Association Between Perceptions of Fit and Commitment (by Special
Education Status)

 

 

 

 

v .-
3
.g ,., _
G.
‘1’,
E
a)
E
z
E
E N -
O
o
I— - O O
I I I If
2 2.5 3 4
Perception of Fit (Spring)
0 Non-Sped x Sped
— Non-Sped Fitted Values ----------------- Sped Fitted Values

 

160

Figure 4.2: Scatter Plot of the Association Between Perceptions of Relational Trust and Commitment
(by Special Education Status)

 

 

 

 

3
.E
a
‘3
E
o
.E
E
E
O
o
X X
o o
o
1- - 0 O
I I I I
1 2 3 4
Relational Trust (Spring)
0 Non-Sped x Sped
—-—— Non-Sped Fitted Values ----------------- Sped Fitted Values

 

161

Figure 4.3: Scatter Plot of the Association Between Perceptions of Collective Responsibility and
Commitment (by Special Education Status)

 

 

 

 

q _
’5
C
Em d
‘1’,
E
o
E
.2
E
g..-
0
x 8
X X
o
O
‘- 4 O 0
1 3 5
Collective Responsibility (Spring)
0 Non-Sped s Sped
—— Non-Sped Fitted Values ----------------- Sped Fitted Values

 

 

 

162

Table 25: Influence Models Predicting ECT Commitment

 

Unstandardized

 

 

Coefficient Standardized
. (SE) Coefficrent
Independent Variable
. .66“ .57
Commitment—Fall ('07)
Mean Influence .O3*** .29
(Colleagues’ Commitment) (.01)
. . -.31*** -.18
Spec1al Education Teacher ( 10)
.47*** .27
Flag (.18)
.64***
Intercept (.25)
R-squared ,43
N 178

 

Table 26: Influence Models Predicting ECT Perceptions of Fit

 

 

 

Unstandardized Standardized
Coeffiment .
_ CoeffiCient
Independent Variable (SE)
. .55*** .49
Fit—Fall (.07)
Mean Influence 02*" .30
(Colleagues’ Fit) (.01)
S . lEd t' T h -.14"”" -.12
pec1a uca ion eac er (.07)
.31" .26
Flag (.13)
1.25***
Intercept (.23)
R-squared .33
N 182

 

163

Table 27: Influence Models Predicting ECT Perceptions of

 

 

 

Trust
”“5‘andafd'zed Standardized
CoeffiCient ff .
_ (SE) Coe iCient
Independent Variable
.65*** .58
Trust—Fall (.07)
Mean Influence .03 ** .25
(Trust) (.01)
S 'lEd t' T h -.18** -.12
peeia uca ion eac er (.08)
.45*** .28
Flag (.15)
.7611”!!!
Intercept ('22)
R-squared .42
N 181

 

Table 28: Influence Models Predicting ECT Perceptions of

 

 

 

Collective Responsibility
”“5‘”da’.d‘zed Standardized
CoeffiCIent .
. (SE) Coefficrent
Independent Variable
Collect. Responsibility— .40*** .43
Fall (.06)
Mean Influence .03“ ** .34
(Collective Responsibility) (.01)
. . .01 .01

SpeCial Education Teacher (.09)

.64*** .43
Flag (.16)

.85
Intercept ( 1 _ 0 8)
R-squared .31
N 171

 

Appendix B

Table 29: Description of Variables

 

General Special
Education Education

 

 

 

 

 

 

 

 

Mean Mean
(SD) (SD)

1. Dependent Variables

Burnout Composite measure taken from the ECT survey F07 2.36 2.36
questionnaire, consisting of 6 questions: I feel (.64) (.64)
emotionally drained from my work, I feel used up
at the end of the workday, I feel fatigued when I 808 (2:79) (2;?)
have to get up in the morning and face another ' ' '
day on the job, I feel burned out from my work, I
feel frustrated by my work, I feel I’m working too
hard on my job. Responses on single measures
ranged from 1=strongly disagree to 4=strongly

jgree. (a = .91)

Commitment Composite measure taken from the ECT survey F07 3.47 3.45
questionnaire, consisting of 6 questions: I would (.63) (.63)
prefer to continue teaching in this grade/subject
next year, I could see myself teaching in this 808 3'36 3 '06
grade/subject in five years, I would prefer to ('69) ('81)
continue teaching in this school next year, I could
see myself teaching at this school in five years, I
would prefer to continue teaching in this district
next year, I could see myself teaching in this
district in five years.

Responses on individual measures ranged from
1=strongly disagree to 4=strongly agree. (a = .90)

II. Independent Variables

Importance of Dummy variable created from the ECT survey .71 .70

Mentor questionnaire, where 0 = mentor support is “not (.46) (.46)

Support important at all” or “somewhat important” and 1 =
“very important” or “extremely important”

Importance of Dummy variable created from the ECT survey .90 .96

Colleague questionnaire, where 0 = colleague support is (.30) (.20)

Support “not important at all” or “somewhat important”
and 1 = “very important” or “extremely
important”

Perception of Composite measure taken from the ECT survey 3.16 3.02

School-Level questionnaire, consisting of 4 questions: It’s OK (.63) (.67)

Trust in this school to discuss feelings, frustrations, and

worries with other teachers, Teachers in this
school trust each other, Teachers in this school
respect other teachers who take the lead in school
improvement efforts, Teachers in this school
respect those colleagues who are experts in their
craft. Responses on individual measures ranged
from 1=strongly disagree to 4=strongly agree. (a
= .89)

165

Description of Variables (continued)

Perception of
School-Level
Collective
Responsibility

 

Composite measure taken from the ECT survey 3.61 3.58
questionnaire; participants are asked the (.74) (.69)
proportion of teachers in their school who: Help

maintain discipline in the entire school, not just

their classrooms, Take responsibility for helping

one another do well, Take responsibility for

improving the overall quality of teaching in the

school, F eel responsible for helping students

develop self-control, Set high expectations for

academic work, Feel responsible for ensuring that

all students learn. Responses on individual

measures ranged from 1 = none to 5 = all. (a =

 

.90)
Perception of Composite measure taken from the ECT survey 334 3,26
Fit in School questionnaire, consisting of 6 questions: My ' (.48) (.40)

approach to teaching fits in throughout this
school, My professional interests are the same as
those of other teachers throughout this school, I
identify with other teachers throughout this
school, My professional goals are the same as
those of other teachers throughout this school, I
matter to other teachers throughout this school,
Other teachers throughout this school matter to
me. Responses on individual measures ranged
from 1=strongly disagree to 4=strongly agree. (a
= .89)

III. Teacher Characteristic Variables

 

Dummy variable where 0 = general education

Special Education teacher and

Middle School

I = special education teacher

Dummy variable where 0 = elementary school
teacher and

1 = middle school teacher

Dummy variable where 0 = not a first year teacher

First Year Teacher and

Gender

Race

1 = first year teacher

Dummy variable where 0 = male and 1 = female

Dummy variable where 0 = non-white and 1 = white

166

References

Achinstein, B., Ogawa, R., & Speiglman, A. (2004). Are we creating separate and
unequal tracks of teachers? The impact of state policy, local conditions, and teacher
characteristics on new teacher socialization. American Educational Research Journal,
41(3), 557-603.

Billingsley, B. (2004). Special education teacher retention and attrition: A critical
analysis of the research literature. Journal of Special Education, 33(1), 39-55.

Billingsley, B., Carlson, E., & Klein, S. (2004). The working conditions and induction
support of early career special educators. Exceptional Children, 70(3), 333-347.

Billingsley, B.S., & Cross, L. (1992). Predictors of commitment, job satisfaction, and
intent to stay in teaching: A comparison of general and special education teachers.
Journal of Special Education, 25, 453-471.

Billingsley, B. S., & Tomchin, E. M. (1992). Four beginning LD teachers: What their
experiences suggest for trainers and employers. Learning Disabilities Research and
Practice, 7, 104-112.

Boe, E. (2006). Long-terrn trends in the national demand, supply, and shortage of special
education teachers. Journal of Special Education. 40(3), 138-150.

Boe, E.E., Bobbit, S.A., & Cook, L.H. (1997). Whither didst thou go? Retention,
reassignment, migration, and attrition of special education and general education
teachers from a national perspective. The Journal of Special Education, 30, 371-389.

Boe, E.E., Cook, L.H., Bobbitt, S.A., & Weber, AL. (1998). Retention and attrition of
teachers at the district level: National trends in special and general education.
Reston, VA: National Clearinghouse for Professions in Special Education.

Bourdieu, P. (1985). The forms of capital. in J.G. Richardson (Ed.), International
Handbook of Theory & Research for Sociology of Education (pp. 241-258). New
York: Greenwood.

Bryk, A., & Schneider, B. (2002). Trust in schools: A core resource for improvement.
New York: Russell Sage Foundation.

Burt, R. (1992). Structural Holes. Cambridge, MA: Harvard University Press.

Burt, R. (2000). The network structure of social capital. In (Eds) R. Sutton and B. Staw,
Research in Organizational Behavior. Greenwich, CT: JAI Press.

Carlson, E., Brauen, M., Klein, 8., Schroll, K., & Willig, S. (2002). SPeNSE: Key
findings. Retrieved June 1, 2009 from http://www.spense.org/Resultshtml.

167

Carter, K.B. & Scruggs, TE. (2001). Thirty-one students: Reflections of a first-year
teacher of students with mental retardation. Journal of Special Education, 35, 100-
104.

Coburn, C .E., & Russell, J. (2008). District policy and teachers’ social networks.
Educational Evaluation and Policy Analysis, 30(3), 203-235.

Coleman, J .S. (1988). Social capital in the creation of human capital. American Journal
of Sociology, 94, (Supplement 1988): 895-8120.

Conderman, F., & Stephens, J .T. (2000). Reflections from beginning special educators.
Teaching Exceptional Children, 33(1), 16-21.

Crane, S. J ., & Iwanicki, E. F. (1986): Perceived role conflict, role ambiguity, and
burnout among special education teachers. Remedial and Special Education, 7(2),
24-31.

Darling-Hammond, L. (1997). Doing what matters most: Investing in quality teaching.
New York: National Commission on Teaching and America’s Future.

Darling-Hammond, L., Wise, A., & Klein, S. (1999). A license to teach: Raising
standards for teaching. San Francisco: Jossey-Bass.

Desimone, L., Porter, A.C., Garet, M., Suk Yoon, K., & Birman, B. (2002).
Effects of professional development on teachers’ instruction: Results from a
three-year study. Educational Evaluation and Policy Analysis, 24 (2), 81-112.

ERIC Clearinghouse on Disabilities and Gifted Education. (2001, April). Educating
exceptional children: A statistical profile. Arlington, VA: The Council for
Exceptional Children.

Frank, K. & Yasumoto, J. Linking action to social structure within a system: Social
capital within and between subgroups. The American Journal of Sociology, 104(3),
642-686.

Frank, K.A., Zhao, Y., & Borman, K. (2004). Social capital and the diffusion of
innovations within organizations: The case of computer technology in schools.
Sociology of Education, 77(2), 148-171.

Gersten, R., Keating, T., Yovanoff, P., & Hamiss, M.K. (2001). Working in special
education: Factors that enhance special educators’ intent to stay. Exceptional
Children, 67(4), 549-567.

Glazerman, S., Dolfin, S., Bleeker, M., Johnson, A., Grider, M., Lugo-Gil, J ., & Isenberg,
E. (under review). Impacts of high intensity teacher induction on student and teacher

168

outcomes. In T. M. Smith, L. M. Desimone, & A.C. Porter (Eds.), Organization and
Eflectiveness of High-Intensity Induction Programs for New Teachers.

Goddard, R.D. & Goddard, Y. (2001). A multilevel analysis of the relationship between
teacher and collective efficacy in urban schools. Teaching and Teacher Education,
17, 807-818.

Goddard, R.D., Hoy, W.K.., & Hoy, A.W. (2000). Collective teacher efficacy: Its
meaning, measure, and effect on student achievement. American Educational
Research Journal, 3 7(2), 479-507.

Goddard, R.D., Hoy, W.K., & Hoy, A.W. (2004). Collective efficacy beliefs: Theoretical
developments, empirical evidence, and future directions. Educational Researcher,
33(3), 3-13.

Goldstein, J. (2004). Making sense of distributed leadership: The case of Peer Assistance
and Review. Educational Evaluation and Policy Analysis, 26(2), 173-197.

Granovetter, M. (1973). The strength of weak ties. The American Journal of
Sociology, 78(6), 1360-1380.

Grossman, P.L., & Thompson, C. (2004). District policy and beginning teachers: A lens
on teacher learning. Educational Evaluation and Policy Analysis, 26(4), 281-301.

Hill, H. C., Rowan, B., & Ball, D. L. (2005). Effects of teachers’ mathematical
knowledge for teaching on student achievement. American Educational Research
Journal, 42, 371—406.

Holloway, J. (2001). Benefits of mentoring. Educational Leadership, 58(8), 85-86.

Hoy, W. K., & Tschannen-Moran, M. (1999). Five faces of trust: An empirical
confirmation in urban elementary schools. Journal of School Leadership, 9, 184-208.

Huling-Austin, L. (1990). Teacher induction programs and internships. In W. R. Houston
(Ed.), Handbook of research on teacher education (pp. 535-548). New York:
Macmillan Publishing Company.

Ingersoll, R. & Smith, T. (2003). “The wrong solution to the teacher shortage”
Educational Leadership 60:8:30-33.

Johnson, S. M., & Birkeland, S. E. (2003). Pursuing a "sense of success": New teachers

explain their career decisions. American Educational Research Journal, 40(3), 581-
617.

169

Johnson, S. M., Kardos, S. M., Kauffman, D., Liu, E., & Donaldson, M. L. (2004). The
support gap: New teachers’ early experiences in high-income and low-income
schools. Education Policy Analysis Archives, 12(61).

Kardos, S.M., Johnson, S.M., Peske, H.G., Kauffman, D., & Liu, E. (2001). Counting on
colleagues: New teachers encounter the professional cultures of their schools.
Educational Administration Quarterly, 37(2), 250-290.

Kilgore, K. L., & Griffin, C. C. (1998). Beginning special educators: Problems of
practice and the influence of school context. Teacher Education and Special
Education, 21, 155-173.

Kilgore, K., Griffin, C., Otis-Wilbom, A., & Winn, J. (2003). The problems of beginning
special education teachers: Exploring the contextual factors influencing their work.
Action in Teacher Education, 25(1).

Kristof, A. L. 1996. Person-organization fit: An integrative review of its
conceptualizations, measurement and implications. Personnel Psychology, 49, 1-49

Kueker, J., & Haensly, P. (1991). Developing mentor/induction year teacher dyads in a
generic special education teacher-training program. Teacher Education and Special
Education, 14(4), 257-262.

Lane, G., & Canosa R. (1995). A mentoring program for beginning and veteran teachers
of students with severe disabilities. Teacher Education and Special Education, 18(4),
230-239.

Lin, N. (1999). Building a network theory of social capital. Connections, 22(1), 28-51.

Marsden, P. (1987). Core discussion networks of Americans. American Review of
Sociology, 52(1), 122-131.

Mastropieri, M.A. (2001). Introduction to the special issue: Is the glass half full or half
empty? Challenges encountered by first-year special education teachers. Journal of
Special Education, 35, 66-74.

McLeskey, J ., Tyler, N.C., & F lippin, SS. (2004). The supply and demand for special
education teachers: A review of research regarding the chronic shortage of special
education teachers. The Journal of Special Education, 38(1), 5-21.

McPherson, J. M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily

in social networks. In J. Hagan & K. S. Cook (Eds.), Annual review of sociology, vol.
27: 415—444. Palo Alto, CA: Annual Reviews.

170

Miller, M.D., Brownell, M.T, & Smith, SW. (1999). Factors that predict teachers staying
in, leaving, or transferring from the special education classroom. Exceptional
Children, 65 (2), 201 -2 I 8.

National Commission on Teaching and America’s Future. (2003). No dream denied: A
pledge to America ’s children—Summary report. New York: Author.

Olson, L. (2000). Finding and keeping competent teachers. Quality counts: Who should
teach. Education Week Special Issue, XLX(18), 12-17.

Portes, A. (1998). Social capital: Its origins and applications in modern sociology. Annual
Review of Sociology, 24, 1-24.

Smith, T.M., & Ingersoll, R.M. (2004). Reducing teacher turnover: What are the
components of effective induction? American Educational Research Journal, 41(3),
681-714.

Stein, M.K., & D’Amico, L. (2002). Inquiry at the crossroads of policy and learning: A
study of a district-wide literacy initiative. Teachers College Record, 104(7), 1313-
1344.

Strong, M. & St. John, L. (2001). A study of teacher retention: The eflects of mentoring
for beginning teachers. Research Working Paper, #3. Santa Cruz, CA: New Teacher
Center, UC Santa Cruz.

Wellman, B., & Frank, K. (2001). Getting support from personal comunities. In N. Lin,
R. Burt, & K. Cook (Eds.), Social capital: Theory and research . Chicago, IL: Aldine
De Gruytere

Whitaker, S.D. (2000). Mentoring beginning special education teachers and the
relationship to attrition. Exceptional Children, 66(4), 546-566.

White, M. (1996). Factors contributing to special education teacher attrition: How a
one-year internship affects the attrition rates of special education teachers in
Kentucky. Unpublished doctoral dissertation, Vanderbilt University, Nashville, TN.

Youngs, P. (2007a). District induction policy and new teachers’ experiences: An
examination of local policy implementation in Connecticut. Teachers College
Record, 109(3), 797-837.

Youngs, P. (2007b). How elementary principals’ beliefs and actions influence new
teachers’ induction experiences. Educational Administration Quarterly, 44(2), 101-

137.

Youngs, R, Qian, H., & Holdgreve-Resendez. (in progress). Teacher induction for
diverse, urban contexts.

171

Zabel, R.H., & Zabel, M.K. (2001). Revisiting burnout among special education teachers:

Do age, experience, and preparation still matter? Teacher Education and Special
Education, 24(2), 128-139.

172