Self-Efficacy As A Predictor of Commitment To The Teaching Profession - A Meta-Analysis

Educational Research Review 15 (2015) 116
Contents lists available at ScienceDirect
Educational Research Review

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / e d u r e v
Review
Self-ecacy as a predictor of commitment to the teaching

profession: A meta-analysis
Steven Randall Chesnut *, Hansel Burley
Texas Tech University, Lubbock, Texas 79409-1071, USA
A R T I C L E
I N F O
Article history:
Received 8 May 2014
Received in revised form 14 January 2015
Accepted 14 February 2015
Available online 28 February 2015
Keywords:
Self-ecacy
Commitment to teaching
Meta-analysis
A B S T R A C T
This meta-analysis examined research on the effects of preservice and inservice teachers
self-ecacy beliefs on commitment to the teaching profession. Unlike previous studies on
self-ecacy and commitment, this review systematically examines the effects found within
the literature and highlights important theoretical and methodological issues. A total of
33 qualied studies were included in the nal analysis, including 16,122 preservice and
inservice teachers. Findings suggest that preservice and inservice teachers self-ecacy beliefs
inuence their commitment to the teaching profession (ES = +0.32). However, these effects
vary based upon the conceptual accuracy of the self-ecacy measure and the origin of data.
Conceptually accurate self-ecacy measures resulted in signicantly higher effect sizes.
Additionally, the specicity of questionnaire items and conceptual accuracy of the selfecacy measure positively predicted the relationships between self-ecacy beliefs and
commitment to teaching. Implications for the measurement of self-ecacy and interpretation of preservice and inservice teacher self-ecacy beliefs are presented.
2015 Elsevier Ltd. All rights reserved.
Contents
1.
2.
3.
4.
Introduction .............................................................................................................................................................................................................................. 2
1.1.
Teacher self-ecacy .................................................................................................................................................................................................. 2
1.2.
Role of self-ecacy in career decisions ............................................................................................................................................................. 3
1.3.
Commitment to the teaching profession .......................................................................................................................................................... 4
1.4.
Measuring self-ecacy beliefs ............................................................................................................................................................................. 5
1.5.
Rationale for current study .................................................................................................................................................................................... 5
Methods ..................................................................................................................................................................................................................................... 6
2.1.
The search .................................................................................................................................................................................................................... 6
2.2.
Inclusion ....................................................................................................................................................................................................................... 6
2.3.
Coding ........................................................................................................................................................................................................................... 7
2.4.
Effect size calculations and statistical analyses .............................................................................................................................................. 8
2.5.
Limitations of current meta-analysis ................................................................................................................................................................. 8
Findings ...................................................................................................................................................................................................................................... 8
Discussion ............................................................................................................................................................................................................................... 12
References ............................................................................................................................................................................................................................... 14
Authors note: This research, conducted by Steven Randall Chesnut and Hansel Burley, was completed in the Department of Educational Psychology at
Texas Tech University: 3008 18th Street, Lubbock, Texas, 79409, United States of America.
* Corresponding author. Tel.: +1 405 928 8310.
E-mail address: steven.chesnut@ttu.edu (S.R. Chesnut).
http://dx.doi.org/10.1016/j.edurev.2015.02.001
1747-938X/ 2015 Elsevier Ltd. All rights reserved.
S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116
1. Introduction
The commitment that preservice and inservice teachers have to enter and remain in the teaching profession has been a
construct of great interest among teacher educators (Chesnut & Cullen, 2014; Klassen & Chiu, 2011; Schlechty & Vance, 1981).
Teacher education programs and professional development training prepare future and current educators to meet the needs
of their students through a variety of mastery and vicarious experiences (e.g., Tschannen-Moran & McMaster, 2009). While
much of this effort on the part of teacher educators and professional development specialists is rewarded with improved
teacher performance, over half of those who become teachers end up leaving the profession (Ingersoll, 2003; Tait, 2008;
Tynjl & Heikkinen, 2011). This attrition is especially prevalent in urban schools, characterized by a high poverty, high minority student population (DeAngelis & Presley, 2011).
Throughout the years, researchers have attempted to quantify (e.g., track, measure) teacher attrition and explore the underlying reasons for teachers decisions to remain in or leave the profession. The decisions teachers make regarding entrance
into the profession, remaining in the profession, and leaving the profession have been suggested to originate from the direct
and indirect inuence of occupation-related beliefs (e.g., self-ecacy, outcome expectations), interests, and distal and proximal choice goals (Brown & Lent, 2006; Lent, Brown, & Hackett, 1994; Siwatu & Chesnut, 2014). One of the most controversial
factors mentioned has been self-ecacy. While most teacher self-ecacy researchers believe that self-ecacy beliefs can
predict, to a large extent, an individuals behaviors and performances (Bandura, 1977, 1986, 1997, 2006; Klassen, 2010; Klassen
& Chiu, 2010, 2011; Schunk & Pajares, 2009; Schunk & Usher, 2011; Usher, 2009; Usher & Pajares, 2006a, 2006b, 2008), some
have failed to observe that connection (Chapman, 1984; Friedman, 2003). Bandura (1997, 2006) and Bong (2006) have suggested that this lack of connection is due to weaknesses in the accuracy of the self-ecacy measures, the specicity of the
items, and the alignment of the prompted behaviors with the actual outcome measures.
Prior studies that have investigated the psychometric properties of the relationships between self-ecacy beliefs and
commitment have suggested that increased variability in the self-ecacy scale provides greater explanatory potential for
the variation in commitment responses (e.g, Pajares, Hartley, & Valiante, 2001). Additionally, the level of specicity of the
self-ecacy items has also been found to increase the explained variability in certain outcomes. While the eld has grown
substantially and we as researchers have been able to learn quite a bit about the measurement of self-ecacy beliefs, some
of the arguments posed about the characteristics of self-ecacy measures and their problems remain theoretical. These
theoretical arguments provide false comfort when attempting to interpret self-ecacy measures aimed at improving teacher
education and development.
1.1. Teacher self-ecacy
Teacher self-ecacy has come to be one of the most commonly examined factors believed to inuence preservice and
inservice teacher commitment, burnout, student achievement, and willingness to adopt and implement reform efforts (Chesnut
& Cullen, 2014; Tschannen-Moran & Woolfolk Hoy, 2001, 2007; Wheatley, 2000, 2002; Woolfolk Hoy & Davis, 2006). In his
seminal work, Bandura dened self-ecacy as beliefs in ones capabilities to organize and execute the courses of action
required to produce given attainments (Bandura, 1997, p. 3). In essence, self-ecacy is an individuals belief about what
he or she can do successfully (e.g., Bong, 2006). In regard to teaching, Dellinger, Bobbett, Olivier, and Ellett dened teacher
self-ecacy as individuals beliefs in their capabilities to perform specic teaching tasks at a specied level of quality in a
specied situation (Dellinger, Bobbett, Olivier, & Ellett, 2008, p. 752). Capturing these beliefs can prove benecial to researchers and teacher educators because they represent the underlying self-beliefs of teachers regarding what can be successfully
done in the classroom.
The perspectives on ecacy and how it has been conceptualized and operationalized has evolved since the earliest recorded studies in the 1960s. While Tschannen-Moran and Woolfolk Hoy (2001) provide a comprehensive history of teacher
self-ecacy, our purpose is to highlight the milestones and major events that lead up to our contemporary conundrum.
Based upon Rotters (1966) work in teacher beliefs, the Rand Corporation developed two items that sought to measure teachers beliefs as they concern their ability to inuence student achievement. One of these items focused on the teacher as the
inuential factor in student achievement, and the second focused on environmental factors as inuential in student achievement. Responses to these two items proved to be very powerful predictors of student achievement. More specically, teachers
that believed they could inuence student achievement were more likely to have students with higher achievement scores.
With this view of teacher ecacy, researchers pressed forward examining the inuence of teachers perceptions of their
inuence in student learning and student outcomes. Subsequent studies continued to show strong predictive relationships
between teachers ecacy beliefs and student outcomes (Armor et al., 1976; Guskey, 1981, 1982, 1988). The problem with
measures that looked teacher ecacy from this perspective is that they focused on the teachers beliefs that students changes
were based upon things that they can do, in its most general sense, juxtaposed to things that were outside of their control.
While subsequent research went on to examine teacher ecacy from the perspective of locus of control, and even receiving distinguishing titles such as personal and general teaching ecacy (Gibson & Dembo, 1984; Hoy & Woolfolk, 1990), they
ignored a crucial aspect of Banduras (1977) interpretation: the actual teaching behaviors that lead to changes in student
outcomes.
Before continuing, we want to clarify that the line of research focusing on teaching ecacy as locus of control has been
benecial to research on teacher education. It is benecial to know that teachers who believe that they are responsible for
student gains and that, on a global level, they can inuence student development make for more effective instructors. This
type of perspective can enhance orientations toward teacher education. In fact, Jerald (2007) reviewed the research and
reported that teachers with a stronger sense of self-ecacy had better levels of planning and organization, were more resilient when classroom strategies did not go well, were more open to experimentation, were less critical of students, and
referred students to special education less often. However, the eld needs more research in this area, including the role of
mastery experiences, modeling by principals, and social persuasion, for example, that all students can succeed. When Bandura
(1977) began discussing the role of self-ecacy as perceptions of ability and its predictive role in human behavior, he did
not do so with intention of measuring an individuals beliefs that they can enact change. Instead, his arguments centered
on the existence of individuals beliefs in their ability to successfully engage and maintain a behavior that, without necessarily stating, would lead to them being successful agents in their environment.
Highlighted in his 1986 publication, Bandura discussed the role of human agency as an individuals beliefs in his or her
abilities to enact change in an environment. Broadly speaking, Bandura utilized agency as a term for locus of control. When
an individual believes he or she is able to enact change in an environment, the subsequently observed change is likely to
be attributed to something that the individual did successfully. However, before someone believes he/she can inuence the
environment, he/she must be condent in his/her ability to execute and sustain the independent and combined behaviors
necessary for that change (Bandura, 1997). This focus on the behaviors has not been forgotten since its conceptual and operational work in the late 1970s, but it has been frequently ignored or inappropriately cited to give credence to the other
line of research.
Since the late 1990s, following Banduras (1997) publication, teaching self-ecacy research has began to focus more of
its attention on the behaviors that teachers need to be able to successfully engage in order to be effective instructors. That
is, instruments developed since 1997 have focused substantially more attention on effective teaching behaviors, such as plan
activities that accommodate the range of individual differences among my students (Dellinger et al., 2008), use my students cultural background to help make learning meaningful (Siwatu, 2007), and assist families in helping their children
do well in school (Tschannen-Moran & Woolfolk Hoy, 2001). Scales that utilize conceptually accurate items, as envisioned
and dened by Bandura (1977, 1997, 2006), have frequently shown stronger relationships between teacher self-ecacy and
student performance and learning (Tschannen-Moran & Woolfolk Hoy, 2001), and occupational related outcomes such as
commitment (Chesnut & Cullen, 2014; Klassen et al., 2012). Additionally, scales that utilize the context-specic aspect of
self-ecacy by increasing the specicity of the situation in which teachers must be able to enact a behavior (e.g., such as
in a second language classroom), is an aspect of Banduras operationalization of self-ecacy that should enhance a scales
ability to predict an outcome, but is rarely given consideration. The problem with using the alternative lines of research,
under the guise of self-ecacy, is that it does not provide the appropriate orientation to learn about teachers beliefs to engage
in the behaviors needed to be successful, and the inferences drawn from these incorrect operational denitions are of limited
use in further research and teacher education (Klassen, Tze, Betts, & Gordon, 2011; Skaalvik & Skaalvik, 2008; Wyatt, 2012).
1.2. Role of self-ecacy in career decisions
Self-ecacy beliefs have a foundational role in social cognitive theory (Bandura, 1986, 1997). Self-ecacy also plays a
fundamental role in Lent et al.s (1994) social cognitive career theory (SCCT). In SCCT models, self-ecacy beliefs contribute to the explanation of the development of an individuals career interests, occupational and educational choices, and success
in academic and career engagements (viz., Brown & Lent, 2006; Lent & Brown, 1996). As previously discussed, self-ecacy
beliefs are context-specic appraisals that can uctuate based upon an individuals interpretations of everyday engagements, observations, and interactions (Bandura, 1977, 1997; Usher & Pajares, 2008). As SCCT models suggest, these beliefs
inuence preservice and inservice teachers decisions to pursue a career in teaching and remain in the profession (Brown
& Lent, 2006; Lent et al., 1994).
SCCTs choice model (Lent et al., 1994) depicts a process by which an individuals occupational-related goals inuence
his or her decision to pursue a career path. In particular, inuenced by occupation-related beliefs (i.e., self-ecacy and outcome
expectations), occupational interests develop, leading to occupational choice goals (Brown & Lent, 2006; Lent et al., 1994).
These choice goals, in turn, motivate behaviors that will help individuals achieve their career-related goals. For example, a
prospective teacher has high teaching self-ecacy and believes in the outcomes associated with being a teacher. Consequently, these beliefs inuence his or her interest in the teaching profession. This interest inuences career-related goals
of becoming a teacher. With the goal and the intentions of becoming a teacher, the prospective teacher will likely enroll in
a traditional or alternative teacher preparation program and engage in the appropriate actions that will help him or her
become a teacher.
Lent and colleagues (1994) used the choice model to explain how career-related goals can change as a result of positive
or negative experiences related to pursuing a particular career. For example, after declaring an elementary education major,
the prospective teacher may engage in a wide variety of experiences in the classroom and in the eld. When reecting on
these experiences, the prospective teacher may realize the complexity of teaching math in an urban school. Should these
experiences decrease self-ecacy beliefs, preservice teachers may modify their career-related goals (e.g., preference for teaching in a suburban school rather than an urban school, goal of becoming a math teacher) or alter their choice of occupation.
Ultimately, the choice model examines an individuals commitment to an occupation. Commitment is a complex, multifaceted concept that has come to be identied as the psychological bond and identication of an individual to an organization
or occupation that has special meaning (Chan, Lau, Nie, Lim, & Hogan, 2008; Coladarci, 1992; Firestone & Pennell, 1993;
Fresko, Kr, & Nasser, 1997). Given this denition, someone who is committed might show a strong aliation or connection to a career or organization. For teachers, this commitment might be evidenced by extended hours at school, a genuine
level of care toward students and colleagues, and searching for new ways to reach students through professional development experiences. A lack of commitment, on the other hand, might be evidenced by a weakened bond to an organization,
career, or colleagues that could give rise to a myriad of subsequent personal decisions and behaviors, such as leaving the
profession.
1.3. Commitment to the teaching profession
From an organizational psychology perspective, organizational commitment is an employees affective bond to an organization (Mowday, Porter, & Steers, 1982). A leading framework proposed by Mowday et al. (1982) asserts that organizational
commitment can be characterized by strong condence in the organizations goals, willingness to exert considerable effort,
and a desire to stay with the organization. Commitment to the teaching profession, much like any other occupational commitment, is the psychological bond that an individual has with teaching, as a role, as an occupation, and as an institution
(Chesnut & Cullen, 2014). Ware and Kitsantas (2011) described teacher commitment as the intention to stay in teaching.
They also reported that teachers were found to have a higher level of commitment when they have ecacy to garner the
support of their principals, inuence policies at the workplace, and control instruction in their classes (Ware & Kitsantas,
2007), with teachers who associate instructional outcomes to factors they control confronting new challenges with optimism (Rosenholtz, 1989). When teachers feel that they cannot control the terms of their work, the sense of powerlessness
this causes results in disassociation from the products of work, including educational outcomes for students and avoidance of new work challenges (Rosenholtz, 1989).
Commitment has been, and can continue to be, considered a form of motivation (Rosenholtz, 1989). Motivation is broadly
dened as the process by which goal-directed activities are undertaken and sustained (Schunk, 2012). While the factors
that inuence this goal-directed behavior can be further subtyped into intrinsic and extrinsic motivators (Ryan & Deci, 2000),
commitment should be interpreted as the goal-directed behavior that is inuenced by both intrinsic (e.g., self-ecacy, value)
and extrinsic motivators (e.g., salary, workload). Teacher commitment is related specically to internal motivation, with high
motivation related to performance (Rosenholtz, 1989). Positive performance becomes self-rewarding while poor performance may result in alienation and disengagement, leading to poor performance (Ware & Kitsantas, 2011).
In the literature surrounding preservice teacher commitment, researchers have focused primarily on the positive nature
of the term. More specically, when viewing commitment with a positive frame of reference, researchers focus on aspects
of increasing psychological attachment, longevity of teaching career, and promoting the entrance of preservice teachers into
the teaching profession. In measurement, some researchers utilize questions that prompt teachers to indicate whether they
plan to return to or enter teaching in the next year (e.g., Evans & Tribble, 1986). Others seek answers to how many years
teachers plan to stay in the profession (e.g., Bruinsma & Jansen, 2010). Most researchers, however, tend to focus on psychological self-report measures that seek to clarify how individuals perceive the value of the profession, attachment to the
job, and perceptions of t into the expectations that are held for those in the profession (e.g., Chesnut & Cullen, 2014; Klassen
et al., 2012).
In the literature surrounding inservice teacher commitment, researchers tend to focus primarily on the negative nature
of the term. More specically, when viewing commitment with a negative frame of reference, researchers focus on aspects
that inuence detachment, depersonalization, and ultimately withdrawal from the profession. While some researchers have
been known to look at intentions to leave the profession (e.g., Coladarci, 1992), many utilize psychological self-report measures that examine the amount of emotional exhaustion, depersonalization, accomplishment, and stress (e.g., Betoret, 2009;
Chan, 2008; Chwalisz, Altmaier, & Russell, 1992). Ultimately, the combination of these elements conceptually denes burnout.
Teacher burnout has been examined nearly as much as teacher commitment. While there are different subcategories of
burnout (e.g., depersonalization, reduced personal accomplishment; Huberman & Vandenberghe, 1999), the most commonly utilized is emotional burnout. Emotional burnout has been dened as the state of physical and emotional depletion
resulting from the conditions of work (Freudenberger, 1974, p. 160). This depletion is the process of demotivation in which
teachers reevaluate the intrinsic and extrinsic motivators, ultimately resulting in lower identication and psychological bond
with the teaching profession, institution, or role as an educator. Teachers experiencing emotional burnout are likely to exhibit
signs opposite to teachers who are highly committed to teaching. For example, teachers experiencing high levels of emotional burnout are more likely to depersonalize their students and colleagues, reduce their desire for personal accomplishment,
and exhibit fewer emotional reactions to events in the learning environment (Fimian & Blanton, 1987; Fives, Hamman, &
Olivarez, 2007; Maslach, Jackson, & Schwab, 1986; Maslach & Pines, 1977). Research examining emotional burnout tends
to focus on inservice teachers. This bias makes conceptual sense, as preservice teachers are less likely to have experienced
enough negative events in the profession that would demotivate them from pursuing their career.
Conceptually similar to, but unique from, commitment, burnout describes the attachment and identication of an individual to the profession. For example, items from the commonly used Maslach Burnout Inventory (Maslach et al., 1986)
ask teachers if they feel that they perceive their daily tasks negatively, if they talk to others negatively about the job, whether
the job is draining, and whether they need increasing amounts of time to relax after work. On the other hand, items from
common commitment scales ask teachers whether they identify and take problems with the teaching profession as theirs
to solve, if teaching has personal meaning, and if teaching would be a good career even if money were no issue. Under the
umbrella of motivation to enter and remain in the teaching profession can be viewed on a loose continuum. At the negative end of the continuum, studies that investigate burnout are able to ll in information about lower and decreasing levels
of motivation to remain in the profession. At the positive end of the continuum, studies that investigate commitment are
able to ll in information about higher and increasing levels of motivation to enter and continue in the teaching profession. While the visualization and conceptualization of this continuum are easy, it remains unclear why researchers choose
to examine burnout over commitment and how this selection inuences the results that are reported.
With much of the focus on trying to increase teacher commitment and reduce teacher attrition (turnover), some researchers have investigated the uctuations of commitment during teaching. Skinner (1996) suggested that in many instances,
researchers could conclude the strength of commitment and the onset of emotional burnout by examining a teachers perception of the impact that they can make on their own environment. That is, teachers who believe that they can make a
difference in their environment through personal engagements tend to exhibit strong and adaptive self-ecacy beliefs. This
agency can greatly inuence a teachers commitment to the profession (Bandura, 1997). However, when weak and maladaptive, teacher self-ecacy beliefs can also be the source of decreased condence in perceived abilities and agency (Bandura,
2001). While weak and maladaptive self-ecacy beliefs can be detrimental to a teachers longevity (Klassen & Chiu, 2011),
these beliefs can be used to structure interventions to help prepare preservice and train inservice teachers (Siwatu & Chesnut,
2014). Being able to use self-ecacy beliefs to structure instruction and interventions is dependent upon knowledge of teachers self-ecacy beliefs, which could be biased if they are improperly measured.
1.4. Measuring self-ecacy beliefs
In the appraisal of self-ecacy beliefs, individuals tend to draw upon their self-knowledge, the prompted task, and the
strategies that might make engagement in the task a successful endeavor (Bandura, 1997; Knoblauch & Woolfolk Hoy, 2008).
In order to make an accurate appraisal, individuals must be able to compare what they can do to what is being asked of
them. While it is likely that self-ecacy appraisals will be slightly inaccurate (e.g., over- or underestimate abilities), interpreting and drawing implications from these appraisals is a greater concern (Bandura, 1997; Schunk & Pajares, 2009). It is
important to understand what exactly is being measured in a self-ecacy scale.
Bandura (1997) and Wheatley (2005) have suggested that self-ecacy instruments not only ignore the underlying factors
that inuence self-ecacy appraisals (e.g., sources of information, perspectives of task), but that the construction of many
famous instruments has been wrought with inaccurate conceptualizations and operational denitions (Klassen et al., 2011;
Wyatt, 2012). Basing decisions and interpretations of self-ecacy research upon studies that utilized inaccurate conceptualizations and operational denitions of self-ecacy beliefs might not only bias the outcomes of such processes, but may
further promulgate the use of inaccurate measurement methods. Instruments that claim to measure self-ecacy beliefs,
yet measure only locus of control, self-esteem, or some other self construct perpetuate these misconceptions (Bong, 2006).
Interpreting self-ecacy scores and ratings is no easy task, especially for conceptually inaccurate measures. When trying
to interpret high versus low ratings on individual items, it is important to determine what perspective the respondent took
and whether or not the items truly captured the nature of self-ecacy as Bandura (1977, 1997) envisioned. Additionally,
the spread of possible responses may bias outcomes. For example, a scale that measures self-ecacy on a 5-point Likerttype scale does not offer as much variability as a scale based upon a 0100 scale because respondents tend to avoid extreme
positions, so scales with just a few steps shrink quickly (Bandura, 2006). This lack of variability may lead to inaccurate determinations of who has higher self-ecacy (Pajares et al., 2001). Similarly, interpreting instrument composite scores, especially
of an instrument that poorly measures self-ecacy, may give way to inaccurate conclusions about the nature of selfecacy and its role in achievement, performance, or commitment. Without following up or interviewing participants within
a temporally acceptable time frame (e.g., Bong, 2006), sole interpretation of self-ecacy ratings may fail to portray accurately the self-ecacy beliefs and perceptions of the participants (Wheatley, 2005; Wyatt, 2012). In terms of research and
responsive interventions for preservice and inservice teachers, it is important to understand what is being measured in selfecacy scales and to be condent in their accuracy.
1.5. Rationale for current study
Our purpose in the present meta-analysis is to validate the above arguments made by self-ecacy researchers, as part
of an effort to encourage accurate and appropriate self-ecacy measures for research and intervention. To date, stringent
reviews of the relationship between teacher self-ecacy and commitment are absent in the literature. As self-ecacy grows
into an increasingly important factor of study in teacher commitment and burnout, it is important that we gain an overview of the eld and establish directions for future research. The present study will answer the following questions:
1. Are self-ecacy beliefs positively related to teacher commitment to the profession?
2. Does the relationship between self-ecacy and commitment differ for preservice and inservice teachers?
3. Does the relationship between self-ecacy and commitment differ when measuring commitment with a positive orientation (e.g., commitment, intention) compared to a negative orientation (e.g., burnout, attrition)?
4. Does the relationship between self-ecacy and commitment differ for teachers from diverse regions in the world?
5. Does the relationship between self-ecacy and commitment differ when measuring commitment with a positive orientation
(e.g., commitment, intention) compared to a negative orientation (e.g., burnout, attrition)?
6. Does the relationship between self-ecacy and commitment differ for studies that accurately measure self-ecacy (according to Bandura, 1997) compared to those that inaccurately measure self-ecacy?
7. Does the specicity of self-ecacy measures and accuracy of conceptualizations inuence the effects of the relationship between self-ecacy and teacher commitment?
2. Methods
The current meta-analysis employed techniques proposed by Lipsey and Wilson (2001). Multiple software suites were
utilized to carry out this meta-analysis. SPSS (v. 22) was utilized to facilitate data entry into databases and running the multiple regression analysis. Microsoft Excel and Comprehensive Meta Analysis (Borenstein, Rothstein, & Cohen, 2012) were
utilized to calculate effect sizes (e.g., correlation coecients to Fishers Z) and test Q and association parameters (presented as 2 estimates).
The procedures for this study followed the steps outlined by Lipsey and Wilson, (2001): (1) establishing a reason for the
study, (2) locating all studies that might t, (3) ltering the studies to keep only those that t within the pre-established
criteria, (4) coding these studies based upon a code sheet into a database, (5) calculating values that can be used in meta
analyses (e.g., Fisher Z, inverse variance weight), and (6) conducting the statistical analyses testing for homogeneity, signicance, and subgroup effects.
2.1. The search
All of the studies in this meta-analysis came from keyword-guided searches from online journal and dissertation databases and from searches guided by citations in relevant studies. Broad searches of published and unpublished literature on
teacher commitment and attrition were conducted using several strategies and databases. The following databases were
searched, focusing on manuscripts, theses, and dissertations from 1980 to 2013: ERIC, PsycINFO, ProQuest Dissertations and
Theses, and Google Scholar to ensure coverage of studies left out in previous searches. The following search terms were
utilized in the search for studies:
1. A mixture of teacher self-ecacy and terminology related to commitment (e.g., commitment, retention, intention, enter
teaching).
2. A mixture of teacher self-ecacy and terminology related to burnout (e.g., withdrawal, dropout, attrition, turnover).
3. A mixture of teacher self-ecacy and terminology related to commitment and burnout in studies that utilized popular
instruments to measure self-ecacy (e.g., Teacher Sense of Ecacy Scale, Teacher Ecacy Scale, Teachers Ecacy Beliefs
System; Dellinger et al., 2008; Gibson & Dembo, 1984; Tschannen-Moran & Woolfolk Hoy, 2001)).
Searches returned thousands of published articles, and nearly 300 dissertations and conference proceedings on the topics
of self-ecacy, teachers, and commitment. Of the studies that were returned in our searches, a majority of them were removed
on the premise of a qualitative framework (e.g., Hong, 2010). Those that utilized a quantitative framework were retained,
but many of them were a function of noise within the search algorithms. More specically, studies containing one of the
components (e.g., self-ecacy), but not the other (e.g., burnout), were returned. Additionally, studies were returned that
mentioned the relationship between self-ecacy and commitment in the manuscript, but focused the analysis on selfecacy and some correlate of commitment (e.g., job satisfaction). In the end, 33 independent studies from 27 manuscripts
were retained, and appropriately weighted, for analysis. Of these 33 studies, 31 were published manuscripts and two were
dissertation projects. Table 1 summarizes the studies retained for inclusion in the study.
2.2. Inclusion
To be included in this meta-analysis, the following inclusion criteria were established:
1. Studies involved preservice and inservice teachers perceptions and condence in ability (self-ecacy) as it relates to
their subsequent commitment to the teaching profession.
a. Commitment could either be interpreted positively as intention, retention, and commitment, or negatively as burnout,
attrition, and turnover.
2. Studies were written in English. While international studies were specically coded for nationality of sample, only those
written in English were retained.
3. Studies were completed between 1980 and the present.
4. Studies reported relationships between teacher self-ecacy and commitment that could be ultimately converted to effect
sizes meaningful to a multiple regression meta-analysis.
5. Both published and unpublished studies, including peer-reviewed journal articles, dissertations, and conference presentations.
Table 1
Studies included in the meta-analysis.
Author (year)
Type
Bruinsma and Jansen (2010)

Chesnut and Cullen (2014)
Rots, Aelterman, Vlerick, and Vermeulen
(2007)
Evans and Tribble (1986)
Klassen and Chiu (2011)
Rots and Aelterman (2009)
Rocca (2005)
Chan et al. (2008)
Coladarci (1992)
Fives et al. 2007
Betoret (2006)
Betoret (2009)
Brouwers and Tomic (2000)
Klassen et al. (2012) 1
Schwarzer and Hallum (2008) 1
Schwarzer and Hallum (2008) 2
Watt and Richardson (2007)
Chapman (1984)
Shen (2009)
Brouwers, Evers, and Tomic (2001)
Chan (2002)
Chan (2008)
Evers, Brouwers, and Tomic (2002)
Friedman (2003)
Skaalvik and Skaalvik (2007)
Brissie, Hoover-Dempsey, and Bassler
(1988)
Madden-Szeszko et al. (2000)
Punch and Tuettemann (1990)
Louis (1998)
Chwalisz et al. (1992)
Journal
Journal
Journal
Sample
Origin
Outcome
198
209
209
Preservice
Preservice
Preservice
Europe
N. America
Europe
Intent
Commit
Commit
ES (r)
0.210
0.350
0.350
Journal
Journal
Journal
Dissertation
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
Journal
179
379
301
215
3715
364
49
247
724
558
379
203
211
394
608
595
294
1282
530
277
83
159
490
322
244
1213
Preservice
Preservice
Preservice
Preservice
Inservice
Inservice
Preservice
Inservice
Inservice
Inservice
Preservice
Preservice
Preservice
Preservice
Inservice
Inservice
Preservice
Inservice
Inservice
Inservice
Preservice
Inservice
Inservice
Inservice
Inservice
Inservice
N. America
N. America
Europe
N. America
Asia
N. America
N. America
Europe
Europe
Europe
N. America
Europe
Asia
Asia
Europe
Europe
Australia
N. America
Asia
Europe
Asia
Asia
Europe
Europe
Europe
N. America
Commit
Commit
Commit
Intent
Commit
Commit
Burnout
Commit
Burnout
Burnout
Commit
Commit
Commit
Commit
Burnout
Burnout
Commit
Commit
Burnout
Burnout
Burnout
Burnout
Burnout
Burnout
Burnout
Burnout
0.230
0.310
0.310
0.421
0.330
0.280
0.570
0.403
0.322
0.450
0.350
0.300
0.160
0.310
0.170
0.480
0.160
0.070
0.168
0.340
0.330
0.290
0.610
0.130
0.310
0.430
Dissertation
Journal
Journal
Journal
73
574
528
316
Inservice
Inservice
Inservice
Inservice
N. America
Australia
N. America
N. America
Burnout
Burnout
Intent
Burnout
0.160
0.270
0.420
0.344
SE measure
Acc.
Spec.
TSES Modied
TSES Original
TSES Modied
Yes
Yes
Yes
3
3
3
TES
TSES Original
TSES Translated
TSES Original
OSTES Modied
TES
TSES
Author Created
TSES Original
TSES Original
TSES Original
TSES Original
TSE
TSE
FIT Choice
Not Described
GSES Shortened
Author Created
Schwarzer, 1999
Author Created
Author Created
Not Described
Author Created
Author Created
No
Yes
Yes
Yes
Yes
No
Yes
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
Yes
Yes
Yes
Yes
No
Yes
No
1
3
3
3
3
1
3
3
4
5
3
3
3
3
1
1
1
1
1
3
1
3
5
*
5
1
TES
Author Created
Author Created
Author Created
No
No
Yes
No
1
1
2
1
TSES Teacher Sense of Ecacy Scale (Tschannen-Moran & Woolfolk Hoy, 2001); OSTES Ohio State Teacher Ecacy Scale.
TES Teacher Ecacy Scale (Gibson & Dembo, 1984); TSE Teacher Self-Ecacy (Schwarzer, Schmitz, & Daytner, 1999).
GSES General Self-Ecacy Scale (Wang, 1999).
* Full scale could not be retrieved for coding.
2.3. Coding
To examine the relationship between teacher self-ecacy beliefs and their commitment to the profession, studies that
remained after the nal selection screenings were coded. Descriptive and substantive features, such as type of population,
nationality of teacher sample, accuracy of self-ecacy measures, breadth of self-ecacy measures, and the correlation (e.g.,
relationship) between self-ecacy and commitment, were placed in the database. The authors of this study trained with
the coding scheme before manuscripts were coded for this study. While disagreements initially arose, the iterative process
of meeting and realignment facilitated an interrater agreement of 93% on the codes for 10 random studies. Minute disagreements on the specicity of the self-ecacy were the most common. Given that specicity was coded on a 5-point scale,
the discrepancies tended to arise when one author coded the study a 4 and the other a 5. In situations where there was a
discrepancy in coding, the authors met to discuss and mutually decide upon a rating. The study features utilized in the analyses were coded in the following way:
1.
2.
3.
4.
Type of population: Preservice or inservice teachers.

Type of study: Dissertation, thesis, conference, or publication.
Origin of data: Countries were coded and combined into continents.
Accuracy of self-ecacy items: Conceptually accurate or inaccurate (Bandura, 1997, 2006; Bong, 2006). Accurate questionnaire items will reect what a respondent can do. These items will not reect intentionality, self-esteem, locus of
control, or outcome expectancies (Bandura, 2006).
5. Type of motivation to enter or remain in the profession: Commitment or burnout.
6. Specicity of items: Scale of 15 indicating global (1), behaviorally specic (3), behaviorally and contextually specic
(5) self-ecacy items. Items at level 1 will have no clear domain or clear behavioral task, and may look like questions in
measures of self-concept or locus of control (e.g., I can affect students learning). Items at level 3 will lack a clear domain,
but will have a clear behavioral task (e.g., I can control disruptive behaviors of students in my classroom). Items at level
5 will have a clear and relevant domain of functioning (Bandura, 2006), with that domain being some aspect of teaching (e.g., I can provide realistic challenges in mixed ability classes). In situations where a measure has varying levels of
specicity, ratings were averaged and rounded, making possible the levels of 2 and 4.
2.4. Effect size calculations and statistical analyses
Since the relationship between self-ecacy and commitment is common among all studies in this review, the Pearson
correlation coecient (r) was used as the effect size statistic. Effect sizes were calculated using the Pearson r to Fisher z
transformation (Lipsey & Wilson, 2001). In cases were studies reported multiple effect sizes (e.g., r), it was rst determined
whether these effect sizes were independent of one another before utilizing both. For example, a study that utilized three
different preservice teacher samples from different countries reported three different correlation coecients between selfecacy and commitment. Given that each of these correlations was independent from the other, they were coded as three
different effect sizes. To answer the rst research question, xed and random main effects were calculated. The main effects
and subgroup analyses for the subsequent research questions were conducted rst using Microsoft Excel and then veried
by using comprehensive meta analysis (Borenstein et al., 2012). The multiple regression was calculated using a weighted
regression analysis in SPSS (v 22) and by then recalculating the standard errors and converting to z signicance values (DeCoster,
2004; Lipsey & Wilson, 2001).
Signicance of moderating variables was calculated by comparing the total (QT), within (QW), and between (QB) variances to a corresponding 2 threshold (DeCoster, 2004; Lipsey & Wilson, 2001). The 2 threshold was determined by calculating
degrees of freedom for each level of variance. Degrees of freedom for total variance (QT) were calculated by k 1, with k
being equal to the number of studies in the sample (e.g., 33). Total degrees of freedom for within variance (QWtotal) were
calculated by k j, with k being equal to the number of studies in the sample and j being equal to the number of categories
in moderating variable(s). Disaggregated degrees of freedom for each within category (QW) were calculated by kwithin 1, with
kwithin being equal to the number of studies in each category of the moderating variables. Summing the disaggregated degrees
of freedom for each group results in the same degrees of freedom for the QWtotal, but provides a foundation upon which group
level signicance testing for homogeneity of variance can be calculated. Degrees of freedom for between variance (QB) were
calculated by j 1, with j being equal to the number of categories in moderating variable(s). Using the appropriate degrees
of freedom, the Q statistics were compared against corresponding 2 thresholds for signicance at the p < 0.05 level.
2.5. Limitations of current meta-analysis
Because this study required quantitative effect sizes, qualitative studies were removed from the analysis. While qualitative analyses offer a great deal of knowledge to any eld of research, they could not be quantied to the extent needed
for inclusion. As many of the studies on teacher self-ecacy and commitment were conducted from a qualitative framework, the removal of this information could weigh heavily on the interpretations and directions of future research in this
eld. Future reviews should meta-analyze the results of these qualitative studies so that the voices of these researchers will
not be overlooked. Second, attempts were made to contact the authors of older studies that utilized measures of teacher
self-ecacy and commitment, yet failed to report correlational information. The non-response of some of these individuals undoubtedly limited the scope and power of this study.
Another limitation is that the researchers selected only those studies written in English. Unfortunately, at the time of
data collection, the researchers deemed translating the studies as cost and time prohibitive. The reader should be aware
that because of this limitation, the sample of included studies might be biased. However, according to Thornton and Lee
(2000), all meta-analyses that do not review all studies are similarly biased. Still, a retrospective review of 303 metaanalyses using the English language restriction found no systematic bias (Morrison et al., 2012). Therefore, despite this limitation,
some condence is warranted when evaluating this studys outcomes.
Meta-analyses are unique in that they offer a systematic investigation of the literature within eld. While this aspect of
a meta-analysis may be its strength, it is also its weakness. Due to the collection procedures in meta-analyses, studies that
remain unpublished are ultimately left out. Failing to include these studies in the analyses biases the results in favor of those
that were fortunate enough to be published (Rosenthal, 1979, 1991). The studies that fail to be published are stored away
and typically never see the light of day. In order to account for this le drawer problem, a fail safe N was calculated. The
fail safe N is a calculation of the number of studies with little or no effect size that would be needed to be included in the
analysis to reduce the overall correlation effect size to a non-signicant level. Considering studies with a small, but nonzero correlation effect size, it would require 54 studies to lower the estimated correlation to a level that is no longer signicant.
3. Findings
A total of 33 studies qualied for inclusion in the analysis. The 33 studies covered 16,122 preservice and inservice teachers from North America, Europe, Asia, and Australia. The overall weighted effect size (pooled correlation) was 0.318 for xed
effects and 0.317 for random effects (SE = 0.008 for xed effects and 0.026 for random effects). The large Q statistic value
indicated that this collection of studies was quite heterogeneous (Q = 295.67, df = 32, p < .001). Table 2 reports the xed and
Table 2
Overall effect sizes.
k
Fixed
Random
ES
33
33
.318
.317
SE
.008
.026
Variance
.000
.001
Test of null
95% CI
Lower
Upper
Test of homogeneity
Q-value
df (Q)
P-value
41.63
12.82
<.001
<.001
.304
.271
.331
.362
295.67
32
<.001
random effects models for the main effect. This suggests that the variance between cases within this study is larger than
could be explained by sampling error. In order to account for the variance in effect sizes, key methodological and sample
characteristics were used to model some of the variation.
With regard to the teacher level, the studies were divided into two major categories: (1) preservice teachers (n = 14) and
(2) inservice teachers (n = 19). Table 3 presents the summary results of the analysis by teacher level. Results suggested nonsignicance between group differences (QB = 1.65, df = 1, p = .2). This means that the difference in effect sizes between preservice
and inservice teachers was not signicant. The variance within preservice teachers indicated heterogeneity (QW = 25.81, df = 13,
p < .05), as did the variance within inservice teachers (QW = 268.21, df = 18, p < .001). While inservice teachers had a higher
pooled correlation between self-ecacy and commitment, that difference was not great enough to establish signicance.
Results from the random effects model corroborated this nding of a non-signicant difference between preservice and inservice
teacher groups.
With regard to the orientation in which researchers measured the motivation to enter and remain in the teaching profession, the studies were divided into two major categories: (1) positive orientation (e.g., commitment, intention, entrance),
and (2) negative orientation (e.g., burnout, attrition). Table 4 presents the summary results of the analysis by orientation.
Initial results from the xed analysis suggested signicance between group differences (QB = 21, df = 1, p < .001). This means
that the differences in effect sizes between orientations of measures were signicant. However, the results from the random
effects model do not corroborate this nding. Instead, results from the random effects model indicate no signicant difference in effect size as a function of the orientation of the commitment measure (QB = 1.06, df = 1, p = .3). Given that there
will always be more variance in the population than what can be accounted for in a sample, the results of the random effects
model provide a wider perspective. Both models, however, indicated heterogeneity within each of the two groups (Negative: QW = 164.16, df = 18, p < .001; Positive: QW = 110.17, df = 16, p < .001). The lack of signicant difference between groups
provides evidence upon which we can validate the decision to examine burnout and commitment as measuring a similar
construct, yet at distinct ends of the same continuum.
With regard to the conceptual accuracy of the self-ecacy scale being utilized, the studies were divided into two major
categories: (1) conceptually accurate (n = 21) and (2) conceptually inaccurate (n = 12). Table 5 presents the summary results
of the analysis by conceptual accuracy. The determination of conceptual self-ecacy was based upon Banduras (1997, 2006)
denition of self-ecacy items and manner in which items should be phrased. Results suggested signicance between group
differences (QB = 43.62, df = 1, p < 0.001), with accurate self-ecacy measures reporting an effect size of +0.354 and inaccurate
Table 3
By teacher level.
Model
Fixed
Random
Group
Preservice
Inservice
QB
Preservice
Inservice
QB
ES
SE
Test of null
95% CI
Lower
Upper
Test of homogeneity
Q
df(Q)
25.81
268.21
1.65
25.81
268.21
0.32
13
18
1
13
18
1
<.05
<.001
=.2
<.05
<.001
=.6
14
19
.300
.322
.018
.009
16.67
35.78
<.001
<.001
.268
.307
.330
.338
14
19
.301
.324
.023
.033
13.09
9.82
<.001
<.001
.255
.259
.346
.386
ES
SE
Table 4
By orientation.
Model
Fixed
Random
Group
Positive
Negative
QB
Positive
Negative
QB
Test of null
95% CI
Lower
Upper
Test of homogeneity
Q
df(Q)
110.17
164.16
21
110.17
164.16
1.06
16
18
1
16
18
1
<.001
<.001
<.001
<.001
<.001
=.3
17
19
.289
.356
.010
.011
29.82
31.73
<.001
<.001
.270
.334
.308
.376
17
19
.294
.342
.029
.038
10.29
8.94
<.001
<.001
.238
.267
.348
.413
10
Table 5
By conceptual accuracy.
Model
Fixed
Random
Group
Accurate
Inaccurate
QB
Accurate
Inaccurate
QB
ES
SE
Test of null
95% CI
Lower
Upper
Test of homogeneity
Q
df(Q)
130.38
121.67
43.62
130.38
121.67
4.44
20
11
1
20
11
1
<.001
<.001
<.001
<.001
<.001
<.05
21
12
.354
.256
.009
.012
38.56
21.33
<.001
<.001
.336
.233
.371
.280
21
12
.354
.253
.026
.043
13.88
5.903
<.001
<.001
.304
.169
.402
.333
self-ecacy measures reporting an effect size of +0.256. This indicates that scales employing an accurate measure of selfecacy had a signicantly higher correlation with commitment to the teaching profession than did studies that employed
scales that inaccurately measured self-ecacy beliefs. This is consistent with Banduras (1997) claim that self-ecacy beliefs
can better predict outcomes when the beliefs are accurately measured. The variance within the accurate studies indicated
heterogeneity (QW =130.38, df = 20, p < 0.001), similar to the variance within inaccurate studies (QW = 121.67, df = 11, p < 0.001).
Results from the random effects model corroborate this signicant nding (QB = 4.44, df = 1, p < 0.05). Figure 1 demonstrates this heterogeneity. Overall, the conceptual accuracy of scales can inuence the magnitude of the correlations found
between self-ecacy and teacher commitment; however, signicant variance within each of these categories requires further
examination.
With regard to sample location, the studies were divided into four major categories, each describing the continent in
which the participants were located: (1) North America (n = 12), (2) Europe (n = 11), (3) Asia (n = 8), and (4) Australia (n = 2).
Fig. 1. Comparison between conceptually accurate and inaccurate measures.
11
Table 6
By location.
Model
Fixed
Random
Group
N. Amer.
Europe
Asia
Australia
QB
N. Amer.
Europe
Asia
Australia
QB
ES
SE
Test of null
95% CI
Lower
Upper
Test of homogeneity
Q
df(Q)
125.23
72.28
38.60
2.57
56.99
125.23
72.28
38.60
2.57
10.18
11
10
7
1
3
11
10
7
1
3
12
11
8
2
.302
.404
.282
.234
.013
.014
.012
.033
23.68
29.33
24.03
7.17
<.001
<.001
<.001
<.001
.277
.377
.259
.170
.327
.429
.305
.296
12
11
8
2
.328
.381
.234
.223
.046
.038
.038
.056
7.06
9.96
6.20
4.01
<.001
<.001
<.001
<.001
.237
.306
.160
.114
.413
.452
.305
.326
<.001
<.01
Table 6 presents the summary of results of the analysis by location. Results suggested signicance between group differences (QB = 56.99, df = 3, p < 0.001). Within group differences suggested heterogeneous variance for the North America
(QW = 125.23, df = 11, p < 0.001), Europe (QW = 72.28, df = 10, p < 0.001), and Asia (QW = 38.60, df = 7, p < 0.05) groups. Results
for the Australia group suggest homogeneous variance between the cases (QW = 2.57, df = 1, p > 0.05); however, because it
was a small sample, this value should be interpreted with caution. Results from the random effects model corroborate these
ndings. Overall, results indicate that there is signicant between-group effect sizes with much within-group variance.
Meta-analyses are used to nd not just between group differences, as one might hope for with an ANOVA, but also homogeneity within groups indicating a goodness of t (Lipsey & Wilson, 2001). As there was much variance within the groups
in both the conceptually accurate and by location analyses, further investigation was warranted. A 4 2 comparison analysis was conducted by dividing location effect sizes by the accuracy of self-ecacy conceptualization. Table 7 presents the
summary of the results of the analysis. Results suggest signicance between group differences (QB = 106.53, df = 6, p < 0.001).
Contrary to the previous analyses, the further disaggregated groups began to demonstrate levels of within group homogeneity of variance. Signicant within group variance emerged in the Accurate by Europe group (QW = 64.38, df = 8, p < 0.001),
in the Accurate by Asia group (QW = 20.29, df = 5, p < 0.001), and in the Inaccurate by North America group (QW = 98.10, df = 5,
p < 0.001). While it is logically understandable to have within group variance in studies that fail to accurately conceptualize self-ecacy as it relates to teacher commitment, studies that accurately conceptualize and measure self-ecacy should
be more consistent.
As there was much variance within the groups in both the conceptually accurate and by orientation analyses, further
investigation was warranted. A 2 2 comparison analysis was conducted by dividing orientation of measure (e.g., positive,
Table 7
Self-ecacy moderated by accuracy of conceptualization origin of population.
Country of origin
North America
Europe
Asia
Australia
QB
# Cases (sample)
Effect size (SE)
95% CI
QW (df)
P-value
# Cases (sample)
Effect size (SE)
95% CI
QW (df)
P-value
# Cases (sample)
Effect size (SE)
95% CI
QW (df)
P-value
# Cases (sample)
Effect size (SE)
95% CI
QW (df)
P-value
Accurate conceptualization
Inaccurate conceptualization
6 (1759)
.3980 (.024)
[.351.445]
7.56 (5)
P > 0.05
9 (3075)
.4503 (.018)
[.414.486]
64.38 (8)
P < 0.001
6 (5170)
.3123 (.014)
[.285.340]
20.29 (5)
p < 0.001
0 (0)
N/A
N/A
N/A
N/A
Test of homogeneity
Q-value
106.53
6 (3427)
.2679 (.017)
[.234.301]
98.10 (5)
P < 0.001
2 (971)
.3575 (.032
[.294.421]
1.54 (1)
P > 0.05
2 (852)
.1562 (.034)
[.089.223]
0.33 (1)
P > 0.05
2 (868)
.2379 (.034)
[.171.305]
2.57 (1)
P > 0.05
df (Q)
6
P-value
<0.001
Cells are not optimal sizes for the analysis; however, this information does make a strong argument
about the forces behind the observed relationship between self-ecacy and teacher commitment.
12
Table 8
Self-ecacy moderated by accuracy of conceptualization measure orientation.
Measure of orientation
Accurate conceptualization
Inaccurate conceptualization
Positive measure
12 (6941)
.330 (.013)
[.309.351]
19.44 (11)
p > .05
9 (3063)
.405 (.015)
[.375.435]
94.99 (8)
P < .001
5 (2366)
.163 (.020)
[.123.202]
34.57 (4)
p < .001
7 (3752)
.313 (.015)
[.284.342]
50.30 (6)
P < .001
Test of homogeneity
Q-value
df (Q)
P-value
96.38
3
<.001
Negative measure
# Cases (sample)
Effect size (SE)
95% CI
QW (df)
p-value
# Cases (sample)
Effect size (SE)
95% CI
QW (df)
P-value
QB
negative) effect sizes by the accuracy of self-ecacy conceptualization. Table 8 presents the summary of the results of the
analysis. Results suggest signicance between group differences (QB = 96.38, df = 6, p < 0.001). As a moderator, commitment
scales that utilized a negative orientation (e.g., burnout) had higher relationships with teacher self-ecacy than positive
measures of commitment. Additionally, when accurate measures of self-ecacy were utilized, these relationships were the
highest. While these reect the prior analyses, further questions are raised. Why are the relationships between negatively
oriented measures of commitment and inaccurately measured self-ecacy beliefs nearly as large as positively oriented measures of commitment and accurately measured self-ecacy beliefs? Are there characteristics of negatively oriented measures
of commitment that better reference the identication and orientation toward the teaching profession? While the groups
show decreased levels of within group variances as indicated by lower QW values, they are still signicant. This signicant
within group variance indicates that more moderators should be able to account for the variance. However, theoretically
and pragmatically, these potential moderators are currently unknown.
The specicity of self-ecacy measures has been a topic of major discussion within the eld (Bandura, 1997, 2006; Bong,
2006; Dellinger et al., 2008). The specicity of a self-ecacy measure tends to span a global to specic continuum. On the
global end of the spectrum, self-ecacy items tend to reference what outcomes an individual believes he/she can enact.
For example, a global item might read, I can affect student learning. This item represents something that an individual
believes he/she can accomplish, but ignores the small details of the actual behaviors that lead to student learning and the
context in which student learning must occur. On the specic end of the spectrum, self-ecacy items tend to not mention
the accomplishment of an outcome, but focus more on the behaviors and contexts of an individuals engagement. For example,
a specic item might read, I can provide realistic challenges in mixed ability classes. Bandura (1997, 2006) suggests that
global self-ecacy measures will suffer from predictive power when the outcome is specied or contextualized. Selfecacy measures that incorporate highly specic items, however, will provide only predictive power when the outcome is
in the same context or domain.
Given the claims that contextually specic and conceptually accurate self-ecacy measures should more strongly predict
the relationship with a contextualized outcome (in this case teacher commitment), we utilized a multiple regression. Results
suggest that instrument specicity and conceptual accuracy were both statistically signicant predictors of the relationships between teacher self-ecacy and commitment to the teaching profession. The predictors accounted for 22.9% of the
variance in the relationship between self-ecacy and commitment. Table 9 summarizes the regression information.
4. Discussion
The ndings of this meta-analysis indicate that preservice and inservice teachers self-ecacy beliefs are positively
related to their commitment to the teaching profession (ES = +0.32). This nding is statistically and practically signicant.
Cohen (1977, 1988) would classify this as a moderate effect size (Low: r = 0.1; Moderate: r = 0.3; Large: r = 0.5). By squaring
this value, we are able to attain the amount of variance that self-ecacy explains in commitment. Alone, self-ecacy can
account for 10% of the variance in the decision that teachers make to enter, remain, and leave the profession. The values of
the ES and r2 represent strong effects. Using the binomial effect size display (BESD), if half the population were categorized
Table 9
Instrument specicity and conceptual accuracy as predictors of the relationship between self-ecacy
and commitment.
Variable
Adjusted SE of B
Z ( replacement)
Instrument specicity
Conceptual accuracy
0.034
0.040
0.008
0.020
4.25
2.00
<0.001
<0.05
13
as low commitment and the other as high commitment (with each new variable split at the median) and high self-ecacy
and low self-ecacy, an ES of .32 represents an improvement rate of 34% to 66% (Rosenthal, Rosnow, & Rubin, 2000).
Therefore, the ES between self-ecacy and the extent to which teachers are committed to the profession hypothetically
represents a gain of 32%, another indication that the relationship between self-ecacy and teacher commitment is a
strong one.
Consistent with the theoretical perspectives surrounding self-ecacy beliefs (Bandura, 1997), this study conrms and
corroborates the research suggesting that preservice and inservice teacher self-ecacy beliefs can be used to predict commitment to the teaching profession (e.g., Chesnut & Cullen, 2014; Klassen & Chiu, 2011). Unlike the ndings in previous
self-ecacy studies that downplay the link it might have with performance and choice outcomes, this study implements
and examines the fundamental characteristics of self-ecacy measures (e.g., accuracy and specicity) in understanding the
relationships that self-ecacy beliefs share with commitment to the teaching profession.
Previous studies on preservice and inservice teacher self-ecacy beliefs suggest that inservice teachers are likely to have
more accurate appraisals than preservice teachers (e.g., Woolfolk Hoy & Spero, 2005). While the results from this metaanalysis cannot corroborate or refute these claims, they do suggest that there is a non-signicant difference between the
role that these beliefs play in commitment to the teaching profession between these two groups. In other words, while the
pooled correlation between inservice teachers self-ecacy beliefs and commitment to teaching was larger than preservice
teachers, it was not signicantly different. Because of this, we suggest that researchers not overlook the value of selfecacy beliefs in preservice teachers for engaging in teaching practices, even if they seem to be something beyond the
respondents repertoires. The condence that preservice teachers have toward their ability to be successful in teaching practices signicantly inuences their commitment to the teaching profession.
A critical interest of educational researchers is trying to understand how their ideas, theories, and constructs transfer to
international populations with different cultures and socialization trends. According to the ndings from this study, the correlations between self-ecacy beliefs and commitment to teaching were largest in Europe and smaller in magnitude for
North America, Asia, and Australia. When we examined the role of conceptually accurate and inaccurate self-ecacy measures by country, we found that regardless of the country, conceptually accurate self-ecacy measures had signicantly
higher correlations with commitment than conceptually inaccurate measures. Further studies should examine the reason
for these international discrepancies. Are the differences related to the domain of this study (e.g., teacher commitment) or
are they the result of some underlying social or cultural aspect that has yet to be reported?
Many years after introducing self-ecacy, Bandura began to specically give attention to the measurement and
interpretation of self-ecacy instruments (Bandura, 1997, 2006). The primary issue surrounding the measurement and
interpretation of self-ecacy beliefs began with the development of popular self-ecacy questionnaires created by Rotter
(1966) and Gibson and Dembo (1984). While questionable self-ecacy measures were created before and have been
created since, the problem that self-ecacy researchers have expressed is that the measure of self-ecacy is unique and
not the same construct as locus of control, self-esteem, or self-concept, which many of these instruments unintentionally
measure. Bandura (1997, 2006) and Bong (2006), among others, have suggested that the inaccuracies within self-ecacy
measures can be the reason that these purported measures of the construct fail to predict outcomes. The ndings in this
study indicate that conceptually accurate self-ecacy measures report signicantly larger correlations with commitment
to teaching than conceptually inaccurate measures. When measuring self-ecacy beliefs, it should be a primary concern
for the researchers that they adhere to self-ecacy instruments as the construct is dened and exemplied by Bandura
(1997, 2006) and Bong (2006). Failing to utilize conceptually accurate self-ecacy measures will not only inuence
observed relationships with outcome variables, but it may also inuence the likelihood of publication and the effectiveness of responsive interventions.
Alternatively, given the ndings of this study, researchers should further investigate the differences in conceptualization and operationalization of commitment measures as positive and negative measures of commitment do not share
common relationships with self-ecacy. In our study, the ndings (from xed effects) suggest that studies utilizing negatively oriented measures of commitment will report signicantly larger effect sizes. While the random effects model did
not corroborate this nding, the trend still poses an interesting question about how we interpret measures of commitment
and burnout. Additionally, when examining the 2 2 matrix of accuracy by orientation, this question is expanded. From
the results, we can conclude that measures that employ accurate self-ecacy conceptualizations will have a larger correlation with commitment than inaccurate conceptualizations. We can additionally conclude that negatively oriented measures
of commitment will have a larger correlation with self-ecacy. Given that the correlation between negatively oriented
measures of commitment and inaccurate conceptualizations of self-ecacy was nearly as strong as between positively
oriented measures of commitment and accurate conceptualizations of self-ecacy, what is it about negatively oriented
measures of commitment that enhance the relationship? Are measures of burnout and emotionality more closely aligned
with the processes of leaving the profession than psychological identication with the process of entering and remaining
in the profession? Researchers would serve the eld well by further analyzing the negative and positive orientations of
commitment measures.
In addition to the need for adherence to conceptual accuracy, Bandura (1997, 2006) also suggested that self-ecacy instruments be as specic as necessary for the study. Other researchers have offered similar thoughts (e.g., Schunk & Usher,
2011), suggesting that if self-ecacy measures are too general, they will fail to account for much of the variation in the
outcome; however, if they are too specic, they will not provide much for generalization of results. Building upon these
14
perspectives, the ndings from this study suggest that the accuracy of self-ecacy measures and the specicity of the items
will predict the magnitude of the correlation between self-ecacy beliefs and commitment to teaching. That is, conceptually accurate and specic (e.g., contextual, behavioral) self-ecacy measures are more likely to culminate in higher
relationships than those that are global and/or inaccurate. Mirroring Banduras and other researchers recommendations,
self-ecacy instruments should be conceptually accurate and as contextually and behaviorally specic to the intended outcome
measure as possible. Using global self-ecacy scales will result in less impressive predictability and may negatively impact
ones analyses and study results.
The construct known as self-ecacy has come to be widely studied in many domains. While it has a stronghold in education, many have inaccurately utilized the construct and misinterpreted the inuence and the value of self-ecacy in
predicting peoples behaviors. Additionally, inappropriate usage of self-ecacy has inuenced the development and implementation of incorrect interventions (Klassen et al., 2011; Wyatt, 2012). The knowledge of self-ecacy beliefs can provide
much information to researchers and teacher educators. For example, these beliefs can be used to structure teacher education and professional development experiences. Utilizing accurate and specic self-ecacy belief measures can enhance
the ability for researchers and teacher educators to meet the needs of preservice and inservice teachers.
References
*Indicates references used in the analysis.
Armor, D., Conroy-Oseguera, P., Cox, M., King, N., McDonnell, L., Pascal, A., et al. (1976). Analysis of the school preferred reading programs in selected Los
Angeles minority schools. (Report no. R-2007-LAUSD). Santa Monica, CA: Rand Corporation. ERIC Document Reproduction Service No. 130 243.
Bandura, A. (1977). Self-ecacy: toward a unifying theory of behavioral change. Psychological Review, 84(2), 191215. doi:10.1037/0033-295X.84.2.191.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1997). Self-ecacy: The exercise of control. New York: W.H. Freeman and Company.
Bandura, A. (2001). Social cognitive theory: an agentic perspective. Annual Review of Psychology, 52, 126. doi:10.1146/annurev.psych.52.1.1.
Bandura, A. (2006). Guide for creating self-ecacy scales. In F. Pajares & T. Urdan (Eds.), Self-ecacy beliefs of adolescents (pp. 307337). Greenwich, CT:
Information Age Publishing.
*Betoret, F. D. (2006). Stressors, self-ecacy, coping resources, and burnout among secondary school teachers in Spain. Educational Psychology, 26(4), 519539.
*Betoret, F. D. (2009). Self-ecacy, school resources, job stressors and burnout among Spanish primary and secondary school teachers: a structural equation
approach. Educational Psychology, 29(1), 4568.
Bong, M. (2006). Asking the right question: How condent are you that you could successfully perform these tasks? In F. Pajares & T. Urdan (Eds.), Self-ecacy
beliefs of adolescents (pp. 287305). Greenwich, CT: Information Age.
Borenstein, M., Rothstein, H., & Cohen, J. (2012). Comprehensive meta-analysis: A computer program for research synthesis [computer software]. Englewood,
NJ: Biostat.
*Brissie, J. S., Hoover-Dempsey, K. V., & Bassler, O. C. (1988). Individual, situational contributors to teacher burnout. The Journal of Educational Research,
82(2), 106112. doi:10.2307/40539578.
*Brouwers, A., Evers, W. J. G., & Tomic, W. (2001). Self-ecacy in eliciting social support and burnout among secondary-school teachers. Journal of Applied
Social Psychology, 31(7), 14741491. doi:10.1111/j.1559-1816.2001.tb02683.x.
*Brouwers, A., & Tomic, W. (2000). A longitudinal study of teacher burnout and perceived self-ecacy in classroom management. Teaching and Teacher
Education, 16(2), 239253.
Brown, S. D., & Lent, R. W. (2006). Preparing adolescents to make career decisions. In F. Pajares & T. Urdan (Eds.), Self-ecacy beliefs of adolescents (pp.
201223).
*Bruinsma, M., & Jansen, E. P. (2010). Is the motivation to become a teacher related to preservice teachers intentions to remain in the profession? European
Journal of Teacher Education, 33(2), 185200.
*Chan, D. W. (2002). Stress, self-ecacy, social support, and psychological distress among prospective Chinese teachers in Hong Kong. Educational Psychology,
22(5), 557569. doi:10.1080/0144341022000023635.
*Chan, D. W. (2008). Dimensions of teacher self-ecacy among Chinese secondary school teachers in Hong Kong. Educational Psychology, 28(2), 181194.
doi:10.1080/01443410701491833.
*Chan, W. Y., Lau, S., Nie, Y., Lim, S., & Hogan, D. (2008). Organizational and personal predictors of teacher commitment: the mediating role of teacher
ecacy and identication with school. American Educational Research Journal, 45(3), 597630. doi:10.3102/0002831208318259.
*Chapman, D. W. (1984). Teacher retention: the test of a model. American Educational Research Journal, 21(3), 645658. doi:10.3102/00028312021003645.
Chesnut, S. R., & Cullen, T. A. (2014). Effects of self-ecacy, emotional intelligence, and perceptions of future work environment on preservice teacher
commitment. The Teacher Educator, 49(2), 116132. doi:10.1080/08878730.2014.887168.
*Chwalisz, K., Altmaier, E. M., & Russell, D. W. (1992). Causal attributions, self-ecacy cognitions, and coping with stress. Journal of Social and Clinical
Psychology, 11(4), 377400.
Cohen, J. (1977). Statistical power analysis for the behavioral sciences (Rev. ed). New York: Academic Press.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsday, NJ: Erlbaum.
*Coladarci, T. (1992). Teachers sense of ecacy and commitment to teaching. The Journal of Experimental Education, 60(4), 323337. doi:10.1080/
00220973.1992.9943869.
DeAngelis, K. J., & Presley, J. B. (2011). Toward a more nuanced understanding of new teacher attrition. Education and Urban Society, 43(5), 598626.
doi:10.1177/0013124510380724.
DeCoster, J. (2004). Meta-analysis notes. <http://www.stat-help.com/notes.html> Retrieved 28.09.13.
Dellinger, A. B., Bobbett, J. J., Olivier, D. F., & Ellett, C. D. (2008). Measuring teachers self-ecacy beliefs: development and use of the TEBS-self. Teaching
and Teacher Education, 24(3), 751766. doi:10.1016/j.tate.2007.02.010.
*Evans, E. D., & Tribble, M. (1986). Perceived teaching problems, self-ecacy, and commitment to teaching among preservice teachers. The Journal of Educational
Research, 80, 8185.
*Evers, W. J. G., Brouwers, A., & Tomic, W. (2002). Burnout and self-ecacy: A study on teachers beliefs when implementing an innovative educational
system in the Netherlands. British Journal of Educational Psychology, 72(2), 227243. doi:10.1348/000709902158865.
Fimian, M. J., & Blanton, L. P. (1987). Stress, burnout, and role problems among teacher trainees and rst-year teachers. Journal of Occupational Behavior,
8(2), 157165.
Firestone, W. A., & Pennell, J. R. (1993). Teacher commitment, working conditions, and differential policies. Review of Educational Research, 63(4), 489525.
*Fives, H., Hamman, D., & Olivarez, A. (2007). Does burnout begin with student-teaching? Analyzing ecacy, burnout, and support during the student-teaching
semester. Teaching and Teacher Education, 23(6), 916934. doi:10.1016/j.tate.2006.03.013.
Fresko, B., Kr, D., & Nasser, F. (1997). Predicting teacher commitment. Teaching and Teacher Education, 13(4), 429438. doi:10.1016/s0742-051x(96)00037-6.
Freudenberger, H. J. (1974). Staff burn-out. Journal of Social Issues, 30(1), 159165.
15
*Friedman, I. A. (2003). Self-ecacy and burnout in teaching: the importance of interpersonal-relations ecacy. Social Psychology of Education, 6(3), 191215.
doi:10.1023/A:1024723124467.
Gibson, S., & Dembo, M. (1984). Teacher ecacy: a construct validation. Journal of Educational Psychology, 76, 569582. doi:10.1037/0022-0663.76.4.569.
Guskey, T. R. (1981). Measurement of responsibility teachers assume for academic successes and failures in the classrooms. Journal of Teacher Education,
32, 4451.
Guskey, T. R. (1982). Differences in teachers perceptions of personal control of positive versus negative student learning outcomes. Contemporary Educational
Psychology, 7, 7080.
Guskey, T. R. (1988). Teacher ecacy, self-concept, and attitudes toward the implementation of instructional innovation. Teaching and Teacher Education,
4, 6369.
Hong, J. Y. (2010). Pre-service and beginning teachers professional identity and its relation to dropping out of the profession. Teaching and Teacher Education,
26(8), 15301543.
Hoy, W. K., & Woolfolk, A. E. (1990). Socialization of student teachers. American Educational Research Journal, 27, 279300.
Huberman, M., & Vandenberghe, R. (1999). Introduction: Burnout and the teaching profession. In R. Vandenberghe & M. Huberman (Eds.), Understanding
and preventing teacher burnout: A sourcebook of international research and practice. Cambridge, UK: Cambridge UP.
Ingersoll, R. (2003). Is there really a teacher shortage? A report co-sponsored by the center for the study of teaching and policy and the center for policy
research in education. Seattle: University of Washington, Center for the Study of Teaching and Policy.
Jerald, C. D. (2007). Believing and achieving (issue brief). Washington, DC: Center for Comprehensive School Reform and Improvement.
Klassen, R. M. (2010). Teacher stress: the mediating role of collective ecacy beliefs. The Journal of Educational Research, 103(5), 342350. doi:10.1080/
00220670903383069.
Klassen, R. M., & Chiu, M. M. (2010). Effects on teachers self-ecacy and job satisfaction: teacher gender, years of experience, and job stress. Journal of
Educational Psychology, 102(3), 741756. doi:10.1037/a0019237.
*Klassen, R. M., & Chiu, M. M. (2011). The occupational commitment and intention to quit of practicing and pre-service teachers: inuence of self-ecacy,
job stress, and teaching context. Contemporary Educational Psychology, 36(2), 114129. doi:10.1016/j.cedpsych.2011.01.002.
Klassen, R. M., Tze, V. M. C., Betts, S. M., & Gordon, K. A. (2011). Teacher ecacy research 19982009: signs of progress or unfullled promise? Educational
Psychology Review, 23(1), 2143. doi:10.1007/s10648-010-9141-8.
*Klassen, R., Wilson, E., Siu, A. F. Y., Hannok, W., Wong, M. W., Wongsri, N., et al. (2012). Preservice teachers work stress, self-ecacy, and occupational
commitment in four countries. European Journal of Psychology of Education, 28, 12891309. doi:10.1007/s10212-012-0166-x.
Knoblauch, D., & Woolfolk Hoy, A. (2008). Maybe I can teach those kids. The inuence of contextual factors on student teachers ecacy beliefs. Teaching
and Teacher Education, 24, 166179.
Lent, R. W., & Brown, S. D. (1996). Social cognitive approach to career development: an overview. Career Development Quarterly, 44(4), 310
322.
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory of career and academic interest, choice, and performance. Journal
of Vocational Behavior, 45(1), 79122. doi:10.1006/jvbe.1994.1027.
Lipsey, M., & Wilson, D. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.
*Louis, K. S. (1998). Effects of teacher quality of work life in secondary schools on commitment and sense of ecacy. School Effectiveness and School
Improvement, 9(1), 127.
Madden-Szeszko, G. (2000). Variables contributing to teacher ecacy: An examination of burnout, affect, demographic variables and general self-ecacy
(unpublished doctoral dissertation). Hempstead, NY: Hofstra University.
Maslach, C., Jackson, S., & Schwab, R. (1986). Maslach Burnout Inventory: Educator survey.
Maslach, C., & Pines, A. (1977). The burnout syndrome in the day care setting. Child and Youth Care Forum, 6(2), 100113.
Morrison, A., Polisena, J., Husereau, D., Moulton, K., Clark, M., Fiander, M., et al. (2012). The effect of English-language restriction on systematic review-based
meta-analyses: a systematic review of empirical studies. International Journal of Technology Assessment in Health Care, 28(2), 138144. doi:10.1017/
S0266462312000086.
Mowday, R. T., Porter, L. W., & Steers, R. (1982). Organizational linkages: the psychology of commitment, absenteeism, and turnover. turnover. San Diego,
CA: Academic Press.
Pajares, F., Hartley, J., & Valiante, G. (2001). Response format in writing self-ecacy assessment: greater discrimination increases prediction. Measurement
and Evaluation in Counseling and Development, 33(4), 214221.
*Punch, K. F., & Tuettemann, E. (1990). Correlates of psychological distress among secondary school teachers. British Educational Research Journal, 16(4),
369382. doi:10.1080/0141192900160405.
*Rocca, S. J. (2005). Predicting preservice agriculture teachers intentions to teach utilizing person inputs, contextual inuences, teacher ecacy, and outcome
expectations (Doctoral dissertation). University of Florida.
Rosenholtz, S. J. (1989). Teachers workplace: The social organization of schools. White Plains, NY: Longman.
Rosenthal, R. (1979). The le drawer problem and tolerance for null results. Psychological Bulletin, 86(3), 638641.
Rosenthal, R. (1991). Meta-analytic procedures for social research. Newbury Park, CA: Sage.
Rosenthal, R., Rosnow, R. L., & Rubin, D. B. (2000). Contrasts and effect sizes in behavioral research: A correlational approach. Cambridge: Cambridge University
Press.
*Rots, I., & Aelterman, A. (2009). Teacher education graduates entrance into the teaching profession: development and test of a model. European Journal
of Psychology of Education, 24(4), 453471.
*Rots, I., Aelterman, A., Vlerick, P., & Vermeulen, K. (2007). Teacher education, graduates teaching commitment and entrance into the teaching profession.
Teaching and Teacher Education, 23(5), 543556.
Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80(1), 128.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist,
55(1), 6878.
Schlechty, P. C., & Vance, V. S. (1981). Do academically able teachers leave education? The North Carolina case. The Phi Delta Kappan, 63(2), 106112.
Schunk, D. (2012). Learning theories: An educational perspective (6th ed.). Boston: Pearson.
Schunk, D. H., & Pajares, F. (2009). Self-ecacy theory. In K. R. Wentzel & A. Wigeld (Eds.), Handbook of motivation at school (pp. 3554). New York, NY:
Routledge.
Schunk, D. H., & Usher, E. L. (2011). Assessing self-ecacy for self-regulated learning. In B. J. Zimmerman & D. H. Schunk (Eds.), Handbook of self-regulation
of learning and performance (pp. 282297). New York: Routledge.
*Schwarzer, R., & Hallum, S. (2008). Perceived teacher self-ecacy as a predictor of job stress and burnout: mediation analyses. Applied Psychology, 57,
152171. doi:10.1111/j.1464-0597.2008.00359.x.
Schwarzer, R., Schmitz, G. S., & Daytner, G. T. (1999). Teacher self ecacy. Retrieved December 1, 2013, from http://userpage.fu-berlin.de/%7Ehealth/
teacher_se.htm
*Shen, Y. E. (2009). Relationships between self-ecacy, social support and stress coping strategies in Chinese primary and secondary school teachers. Stress
and Health, 25(2), 129138. doi:10.1002/smi.1229.
Siwatu, K. O. (2007). Preservice teachers culturally responsive teaching self-ecacy and outcome expectancy beliefs. Teaching and Teacher Education,
23(7), 10861101.
Siwatu, K. O., & Chesnut, S. R. (2014). The career development of preservice and inservice teachers: Why teachers self-ecacy beliefs matter. In H. Fives
& M. Gill (Eds.), International handbook of research on teachers beliefs (pp. 212229). New York: Routledge.
16
*Skaalvik, E. M., & Skaalvik, S. (2007). Dimensions of teacher self-ecacy and relations with strain factors, perceived collective teacher ecacy, and teacher
burnout. Journal of Educational Psychology, 99(3), 611625. doi:10.1037/0022-0663.99.3.611.
Skaalvik, E. M., & Skaalvik, S. (2008). Teacher self-ecacy: conceptual analysis and relations with teacher burnout and perceived school context. In R. Craven,
H. W. Marsh, & D. McInerney (Eds.), Self-processes, learning, and enabling human potential (pp. 223247). Connecticut: Information Age Publishing.
Skinner, E. A. (1996). A guide to constructs of control. Journal of Personality and Social Psychology, 71(3), 549.
Tait, M. (2008). Resilience as a contributor to novice teacher success, commitment, and retention. Teacher Education Quarterly, 35(4), 5775.
Thornton, A., & Lee, P. (2000). Publication bias in meta-analysis: it causes and consequences. Journal of Clinical Epidemiology, 53, 207216.
Tschannen-Moran, M., & McMaster, P. (2009). Self-ecacy: four professional development formats and their relationship to self-ecacy and implementation
of a new teaching strategy. The Elementary School Journal, 110(2), 228245.
Tschannen-Moran, M., & Woolfolk Hoy, A. W. (2001). Teacher ecacy: capturing an elusive construct. Teaching and Teacher Education, 17, 783805.
Tschannen-Moran, M., & Woolfolk Hoy, A. W. (2007). The differential antecedents of self-ecacy beliefs of novice and experienced teachers. Teaching and
Teacher Education, 23(6), 944956. doi:10.1016/j.tate.2006.05.003.
Tynjl, P., & Heikkinen, H. L. T. (2011). Beginning teachers transition from pre-service education to working life. Zeitschrift fr Erziehungswissenschaft,
14(1), 1133. doi:10.1007/s11618-011-0175-6.
Usher, E. L. (2009). Sources of middle school students self-ecacy in mathematics: a qualitative investigation. American Educational Research Journal,
46(1), 275314. doi:10.3102/0002831208324517.
Usher, E. L., & Pajares, F. (2006a). Sources of academic and self-regulatory ecacy beliefs of entering middle school students. Contemporary Educational
Psychology, 31(2), 125141. doi:10.1016/j.cedpsych.2005.03.002.
Usher, E. L., & Pajares, F. (2006b). Inviting condence in school: invitations as a critical source of the academic self-ecacy beliefs of entering middle school
students. Journal of Invitational Theory & Practice, 12, 716.
Usher, E. L., & Pajares, F. (2008). Self-ecacy for self-regulated learning. Educational and Psychological Measurement, 68(3), 443463. doi:10.1177/
0013164407308475.
Wang, L. (1999). The structure and measurement of self-ecacy. Acta Scientiarum Naturalium, Universitatis Pekinensis, 35(3), 414420.
Ware, H. W., & Kitsantas, A. (2007). Teacher and collective ecacy beliefs as predictors of professional commitment. Journal of Educational Research, 100,
303310.
Ware, H. W., & Kitsantas, A. (2011). Predicting teacher commitment using principal and teacher ecacy variables: an HLM approach. The Journal of Educational
Research, 104, 183193.
*Watt, H. M. G., & Richardson, P. W. (2007). Motivational factors inuencing teaching as a career choice: development and validation of the FIT-Choice
scale. The Journal of Experimental Education, 75(3), 167202. doi:10.3200/JEXE.75.3.167-202.
Wheatley, K. F. (2000). Positive teacher ecacy as an obstacle to educational reform. Journal of Research and Development in Education, 34(1), 1427.
Wheatley, K. F. (2002). The potential benets of teacher ecacy doubts for educational reform. Teaching and Teacher Education, 18, 522.
Wheatley, K. F. (2005). The case for reconceptualizing teacher ecacy research. Teaching and Teacher Education, 21(7), 747766.
Woolfolk Hoy, A., & Davis, H. A. (2006). Teacher self-ecacy and its inuence on the achievement of adolescents. In F. Pajares & T. Urdan (Eds.), Self-ecacy
beliefs of adolescents (pp. 117137). Greenwich, CT: Information Age.
Woolfolk Hoy, A., & Spero, R. B. (2005). Changes in teacher ecacy during the early years of teaching: a comparison of four measures. Teaching and Teacher
Education, 21, 343356.
Wyatt, M. (2012). Towards a re-conceptualization of teachers self-ecacy beliefs: tackling enduring problems with the quantitative research and moving
on. International Journal of Research & Method in Education, 37, 166189. doi:10.1080/1743727X.2012.74205.

Self-Efficacy As A Predictor of Commitment To The Teaching Profession - A Meta-Analysis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Self-Efficacy As A Predictor of Commitment To The Teaching Profession - A Meta-Analysis

Uploaded by

Copyright:

Available Formats

Educational Research Review 15 (2015) 116

Contents lists available at ScienceDirect

Educational Research Review

Self-ecacy as a predictor of commitment to the teaching

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

Bruinsma and Jansen (2010)

Type of population: Preservice or inservice teachers.

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

Fig. 1. Comparison between conceptually accurate and inaccurate measures.

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

S.R. Chesnut, H. Burley / Educational Research Review 15 (2015) 116

You might also like