Sources of Abstract

Self-Efficacy: Four This quasi-experimental study tested the po-

tency of different sources of self-efficacy beliefs.
Respondents were primary teachers (N ⫽ 93) in
Professional 9 schools who completed surveys of their self-
efficacy beliefs and level of implementation of a
Development Formats new teaching strategy for beginning readers be-
fore and after participating in 1 of 4 formats of
and Their professional development presenting the same
teaching strategy with increasing levels of
Relationship to efficacy-relevant input. Results indicated that
the professional development format that sup-
Self-Efficacy and ported mastery experiences through follow-up
coaching had the strongest effect on self-efficacy
Implementation of a beliefs for reading instruction as well as for im-
plementation of the new strategy. A substantial
proportion of the teachers who participated in
New Teaching formats that included a demonstration with lo-
cal students and a planning and practice session,
Strategy but no follow-up coaching, experienced a de-
crease in their self-efficacy for reading instruc-
tion.

Megan Tschannen-Moran Self-efficacy, the belief in one’s abilities to

College of William and Mary accomplish desired outcomes, powerfully
affects people’s behavior, motivation, and,
ultimately, their success or failure (Ban-
Peggy McMaster dura, 1997). Without self-efficacy, people
West Virginia ASCD
do not expend effort in endeavors because
they perceive their efforts will be futile.
Teacher self-efficacy is a teacher’s per-
ceived capability to impart knowledge and
to influence student behavior, even that of
unmotivated or challenging students (Gus-
key & Passaro, 1994). A growing body of
empirical evidence supports Bandura’s
(1977) theory that teachers’ self-efficacy be-
liefs are related to the effort they invest in
teaching, the goals they set, their persis-
The Elementary School Journal tence when things do not go smoothly, and
Volume 110, Number 2
© 2009 by The University of Chicago. All rights reserved.
their resilience in the face of setbacks.
0013-5984/2009/11002-0006$10.00 Teachers’ self-efficacy has been linked to
 SOURCES OF SELF-EFFICACY 229

their behavior in the classroom and the im- and practice teaching. These beliefs are
plementation of instructional change (Ash- specific to particular teaching contexts;
ton & Webb, 1986; Guskey, 1986; Haney, therefore, teachers form perceptions about
Wang, Keil, & Zoffel, 2007; McKinney, Sex- their personal capabilities in light of the
ton, & Meyerson, 1999; Timperley & Phil- requirements of a particular teaching task
lips, 2003). Teachers’ self-efficacy has also (Tschannen-Moran, Woolfolk Hoy, & Hoy,
been related to student outcomes such as 1998). As a motivational construct, self-
students’ self-efficacy beliefs and student efficacy influences teachers’ effort and per-
engagement, motivation, and achievement sistence that then affects performance, in
(Anderson, Greene, & Loewen, 1988; Midg- turn becoming a new source of efficacy in-
ley, Feldlaufer, & Eccles, 1989; Ross, 1992; formation. The cyclical nature of behavior
Shahid & Thompson, 2001). The present influencing self-efficacy, and thus new be-
study examined the relationship between haviors, forms a self-reinforcing cycle of
teacher self-efficacy and the format of pro- either success or failure that tends to be-
fessional development in a new skill and come quite stable unless a jarring experi-
the implementation of that skill in teaching ence provokes a reassessment. Professional
reading. Teacher self-efficacy was mea- development formats that emphasize vari-
sured before and after each of four different ous sources of teacher self-efficacy form the
workshop formats to reveal the extent to basis of this study.
which each format was related to increased
teacher self-efficacy. The study also exam-
ined program implementation in relation to Verbal Persuasion
the professional development format and Verbal persuasion involves verbal input
teacher self-efficacy. from others, such as colleagues, supervi-
sors, and administrators, that serves to
strengthen a person’s belief that he or she
The Development of Teacher possesses the capability to achieve a de-
Self-Efficacy sired level of performance. Bandura (1997)
Self-efficacy beliefs influence thought pat- noted that “it is easier to sustain a sense of
terns and emotions that enable goal- efficacy, especially in times of difficulty, if
directed actions in situations where people significant others express faith in one’s ca-
believe they can exercise some control. Self- pabilities than if they convey doubts” (p.
efficacy is a future-oriented belief about the 101). Verbal persuasion may be limited in
level of competence a person expects he or its power to create enduring increases in
she will display in a given situation. Given self-efficacy, but it can bolster self-change if
the pivotal role of self-efficacy beliefs in the positive appraisal promotes greater ef-
understanding human behavior, it is im- fort in the development of skills that sub-
portant to understand how these beliefs are sequently lead to a stronger sense of effi-
formed. Bandura (1997) postulated that cacy. In schools, teachers often receive
teachers make judgments of their self- verbal persuasion in the form of profes-
efficacy based on the verbal encouragement sional development workshops that pro-
of important others such as colleagues, su- vide knowledge of a new strategy as well
pervisors, and administrators (verbal per- as persuasive claims about its usefulness;
suasion), the success or failure of other unfortunately, this format allows little in-
teachers who serve as models (vicarious put from teachers (Stein & Wang, 1988).
experiences), perceptions of past experi- Teachers may also receive verbal persua-
ences of teaching (mastery experiences), sion in the form of specific feedback or en-
and the level of emotional and physiologi- couragement from a supervisor or col-
cal arousal experienced as they anticipate league designed to convince them that they
230 THE ELEMENTARY SCHOOL JOURNAL

can successfully implement a new teaching mance of the skill to be learned can provide
strategy. Verbal persuasion alone may not valuable information and insight.
be a powerful source of self-efficacy; how-
ever, in partnership with other sources of Mastery Experiences
efficacy, it may provide teachers the en- The most influential source of efficacy
couragement necessary to expend effort to- information is personal mastery experi-
ward realistic goals aimed at strengthening ences because they provide the most au-
their teaching skills. thentic evidence of whether one can master
whatever it takes to succeed in a particular
Vicarious Experience field or endeavor (Bandura, 1997). Suc-
A second source of self-efficacy is that cesses build a robust belief in one’s efficacy,
of observing another person successfully especially when success is achieved early in
perform the action that one is contemplat- learning with few setbacks. Self-efficacy be-
ing. Because teaching lacks absolute mea- liefs may be diminished when success is
achieved through extensive external assis-
sures of adequacy, teachers must appraise
tance, after considerable effort, or on a task
their capabilities in relation to the perfor-
perceived as easy or unimportant. Failures
mance of others (Bandura, 1997). The ob-
that cannot be attributed to a lack of effort
server has the opportunity to appraise his
or to external events are likely to have a
or her own capabilities because the model
deleterious effect on self-efficacy beliefs.
provides a standard and this can help the This has important implications for teacher
observer set goals for his or her own teach- professional development. The actual use
ing. The greater the assumed similarity be- of the new knowledge presented in a pro-
tween the observer and the model, the fessional development workshop has been
more persuasive will be the belief that one shown to contribute to changes in teacher
possesses the capabilities to master compa- self-efficacy, whereas simple exposure to
rable activities. When an observer watches the material did not (Ross, 1994). Teacher
a successful teaching exchange, he or she self-efficacy is a dynamic construct that is
is more likely to see the teaching task as cyclical in nature. The proficiency of a per-
manageable. Likewise, when the teaching formance creates a new mastery experience
model fails despite strong efforts, the ob- that serves as a new source of self-efficacy
server may judge the teaching task to be that either confirms or disrupts existing
out of reach. People actively seek proficient self-efficacy beliefs. Over time, the process
models who demonstrate the competencies stabilizes and a relatively enduring set of
to which they aspire. Competent models efficacy beliefs are established that tend to
transmit knowledge and teach observers be resistant to change (Bandura, 1997). The
effective skills and strategies for manag- preservice period, in particular, tends to be
ing task demands through their behavior a time of learning to teach that is marked by
and by revealing their thinking about the major changes in teacher self-efficacy (Ross,
task at hand. Some professional develop- 1994; Woolfolk Hoy & Burke-Spero, 2005).
ment models provide vicarious experiences
through videos of the skill or strategy in Physiological and Affective States
action. If the vicarious experience is limited When judging their own capabilities,
to watching the presenter, it may be only people rely partly on information conveyed
minimally effective at increasing teaching by physiological and emotional states (Ban-
skill (Joyce & Showers, 1988). However, as dura, 1997). A person’s level of arousal,
part of a comprehensive developmental ex- whether perceived positively as anticipa-
perience, observing a proficient perfor- tion or negatively as anxiety, can influence

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 231

his or her self-efficacy beliefs. Arousal, such space, and the attitudes of other teachers
as elevated heart and respiratory rate, in- and administrators, including the level of
creased perspiration, or trembling hands, trust, collective efficacy, and school climate.
may have enabling or debilitating effects, Self-efficacy beliefs can vary within teach-
depending upon whether the situation is ers depending upon the subject area, track
perceived as a challenge or a threat (Gre- level of students, and whether they are
goire, 2003). Moderate levels of arousal teaching outside their field of expertise
when perceived as a challenge can improve (Raudenbush, Rowan, & Cheong, 1992;
performance by focusing attention and en- Ross, Cousins, & Gadalla, 1996; Ross, Cous-
ergy on the task, whereas high levels of ins, Gadalla, & Hannay, 1999; Tschannen-
arousal perceived as a threat might inter- Moran & Johnson, 2004). In addition,
fere with making the best use of one’s skills Tschannen-Moran and Woolfolk Hoy (2007)
and capabilities. In the professional devel- found that contextual variables such as
opment context, exposure to new knowl- available resources, verbal persuasion in
edge and teaching strategies may evoke the form of collegial and community sup-
arousal in the form of interest and curios- port, and mastery experiences contributed
ity. Initial training experiences may cause more to the self-efficacy beliefs of novice
nervous anticipation for a teacher, espe- teachers than career teachers who had a
cially if the teacher is to be observed and wealth of mastery experiences on which to
the performance critiqued. But trying out a base their self-perceptions.
new strategy in a supportive workshop
setting where encouragement and assis-
tance are available can also help reduce the Teacher Self-Efficacy and the
fear of trying it with a room full of stu- Implementation of New Instructional
dents. With the ease that comes through Strategies
continued training and skill development, One of the most interesting and important
successfully implemented lessons create reasons for scholars and school leaders to
feelings of accomplishment, pride, and ex- pay attention to teachers’ self-efficacy is the
hilaration. role it plays in teachers’ implementation of
new teaching strategies presented through
professional development. In his model of
Analysis of the Teaching Task teacher change, Guskey (1986) hypothe-
In assessing their beliefs about their sized that the majority of instructional im-
teaching capability in a particular context, provement programs fail because they do
teachers will make two interrelated judg- not take into account what motivates teach-
ments: an assessment of their personal ers to engage in professional development
teaching competence and an assessment of and the process by which change in teach-
the assumed requirements of an antici- ers typically takes place. Researchers exam-
pated teaching task (Tschannen-Moran et ining teacher attitudes toward the imple-
al., 1998). Judgments of personal compe- mentation of new instructional practices
tence are those a teacher makes about his or have frequently found teachers’ self-efficacy
her capabilities based on an assessment of to be among the most powerful influences
internal strengths and deficits. The assess- on receptivity to change (Guskey, 1988;
ment of the teaching task may include stu- Poole & Okeafor, 1989; Shahid & Thomp-
dent factors such as their current level of son, 2001; Smylie, 1988). An efficacy-based
achievement, motivation, or socioeconomic change model of in-service teachers as they
status, and contextual factors such as the were introduced to whole-language teach-
instructional resources available, the prob- ing strategies was tested by McKinney et al.
lems or possibilities posed by the physical (1999). As participants moved through the
232 THE ELEMENTARY SCHOOL JOURNAL

three stages of initiation, implementation, involved new domain knowledge that re-
and refinement, a relationship between quired “redefining the reading task and
their self-efficacy and the concerns they ex- how to teach it” (p. 627). When teachers
pressed was observed. Participants with implemented the new methods and wit-
lower self-efficacy beliefs expressed con- nessed unanticipated changes in their
cerns characteristic of those in an early students’ achievement, their self-efficacy
stage of change, focusing on the impact it beliefs were enhanced. Timperley and Phil-
would have on them. Participants with lips (2003) proposed that “the change pro-
higher self-efficacy turned their attention to cess is likely to be an iterative rather than a
how the new strategies might affect their sequential one, where changes in beliefs,
students and their school and how they actions or outcomes are both shaped by,
might work to refine teaching practices and and built on, each other” (p. 630). The find-
relationships to better fit within their con- ings of the complex interplay of new
texts— concerns that are typical of later knowledge, changes in students’ achieve-
stages of change. Participants with the ment, and teachers’ feelings of self-efficacy
highest self-efficacy tended to view the in- led Timperley and Phillips to assert that
novation as both important and possible. teacher professional development needs to
Similarly, Scribner (1999) observed that simultaneously address teachers’ beliefs as
teachers’ level of efficacy influenced their well as the improvement in their practices.
response to professional development such Similarly, when Stein and Wang (1988) set
that high self-efficacy teachers were “op- out to identify factors related to teachers’
portunistic in their approach to profes- commitment to acquire and consistently
sional learning and they sought knowledge use the knowledge and skills necessary for
through their involvement in activities that the successful implementation of a new
often were not overtly professional devel- model of reading-skill development, they
opment opportunities” (p. 220) while low found a “pattern consisting of improve-
self-efficacy teachers were unable or un- ment in teachers’ actual expertise in pro-
willing to engage in the reforms because of gram implementation, followed by in-
a “perceived disconnection between the creases in their perceptions of self-efficacy
purposes of the efforts and their own needs for implementing the program” (p. 181).
as professionals” (p. 221). Interventions de- The cultivation of self-efficacy beliefs
signed specifically to address teachers’ self- among teachers has not been found to be
efficacy beliefs as they encountered new a straightforward process of incremental
instructional processes have been found to gains. A number of researchers have found
raise teachers’ self-efficacy and their imple- an “implementation dip” in self-efficacy as
mentation of problem-based instructional teachers began to implement a change ini-
practices in science (Haney et al., 2007). In tiative (Ross, 1994; Stein & Wang, 1988;
addition, such interventions have been Woolfolk Hoy & Burke-Spero, 2005). These
found to raise the self-efficacy for class- self-efficacy beliefs tended to rebound for
room management among mathematics teachers who were able to successfully im-
teachers when compared to a control group plement the change initiative. There is evi-
(Ross & Bruce, 2007). dence that teachers’ calibration of the level
There seems to be a complex interplay of their content knowledge is not especially
between teachers’ self-efficacy beliefs and accurate, with teachers tending to overesti-
their knowledge and skills to implement a mate their knowledge and skills (Cunning-
new instructional strategy. In a study of the ham, Perry, Stanovich, & Stanovich, 2004).
conditions required to achieve change in Thus, their self-efficacy beliefs may be
instructional practices in reading, Timper- based on an overestimation of their skills,
ley and Phillips (2003) found that change which may negatively affect self-efficacy

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 233

beliefs in the face of evidence that their sess that they are not implicated in the
knowledge and skills were not as strong as changes being proposed and will process
they believed. The detrimental effect on the new content superficially. Teachers
self-efficacy beliefs as teachers attempt to who do feel implicated by the reforms pre-
practice a new teaching strategy may be sented will experience stress and discom-
ameliorated by additional support during fort. Those with low self-efficacy are pre-
the crucial early stages of change. In exam- dicted to respond to the reform initiative as
ining the effect of various components of a threat, leading to an avoidance intention
professional development models, Joyce and superficial belief change. On the other
and Showers (1988) found a jump in effect hand, teachers with high self-efficacy who
sizes when practice feedback was added to perceive that they have the resources, time,
information, theory, and demonstration; and support necessary to implement the
there was further increase in effect size proposed changes would likely interpret
when coaching to support implementation the reform as a challenge and consequently
was added to the other components. Re- engage in more systematic (and thus effort-
searchers have suggested that professional ful) processing of the information pre-
development programs that aim to support sented. This model explicates the mecha-
teachers’ ongoing utilization of new knowl- nisms through which teachers’ self-efficacy
edge regarding effective practice need to mediates their response to instructional
develop a delivery system characterized by change.
the provision of continued support and The present study was designed to ex-
follow-up after initial training (Guskey, plore the impact that four professional de-
1989; Stein & Wang, 1988). velopment formats with varying levels of
In a challenge to the view that “more is self-efficacy-relevant input would have on
better” when it comes to self-efficacy be- teachers’ self-efficacy for teaching. It was
liefs, Wheatley (2002) has proposed that predicted that as the richness of the sources
doubts about one’s efficacy can sometimes of self-efficacy increased, there would be an
be beneficial in that uncertainty or doubt is increase in teacher self-efficacy and that the
crucial for the teacher reflection that leads level of implementation of the new reading
to new insights. Wheatley challenged Ban- strategy would consequently increase.
dura’s (1997) claim that it is difficult for a
person to achieve while fighting self-doubt, Method
stating instead that it is difficult for teach-
This quasi-experimental study examined
ers to learn and improve without experi-
the role that the format of professional de-
encing self-efficacy doubts. This disequilib-
velopment played in increasing teacher
rium and uncertainty may come about
self-efficacy and implementation of a new
from a challenge to teachers’ beliefs about
instructional strategy. Four professional
their existing practices. Wheatley sug-
development formats that emphasized var-
gested that factors such as follow-up coach-
ious sources of self-efficacy beliefs postu-
ing might moderate the debilitating influ-
lated by Bandura (1997) were utilized. Im-
ence that teacher self-efficacy doubts may
plementation of the newly taught strategy
have on teachers, resulting in improved
in relation to professional development for-
practice. Gregoire (2003) has proposed a
mat and teacher self-efficacy was also ex-
model of teacher conceptual change based
amined.
on whether teachers, when presented with
an instructional reform initiative, appraise
it as either a threat or a challenge. In this Design
model, teachers who believe that they are This study utilized a quasi-experimental
already implementing the reform will as- quantitative design. Using cluster sampling,
234 THE ELEMENTARY SCHOOL JOURNAL

nine schools from five different public school ferent formats, taking an additive approach
systems in varied settings were placed into to including new or deeper sources of
one of four treatment groups. The participat- efficacy-relevant input. All participating
ing schools were socioeconomically diverse, teachers were administered surveys assess-
representing all four quartiles of free/re- ing their self-efficacy beliefs as well as prior
duced lunch participation in the state, with use of the target instructional strategy at
one school in the highest quartile, three in the the beginning of the initial workshop.
second and third quartiles, and two in the Treatment 1 (information). All schools
lowest quartile as identified by free/reduced received a 3-hour workshop entitled the
lunch participation. There was also diversity Tucker Signing Strategies for Reading
of context, with four rural schools, four sub- structured in alignment with the Tucker
urban schools, and one urban school. Four training manual. The standard one-time
schools received the professional develop- workshop format was used as the baseline,
ment in the spring, and five the following fall. with verbal persuasion as the identified
The four schools that were available for treat- source of self-efficacy beliefs. Treatment 1
ment and data collection in the spring were schools received the workshop alone. Us-
randomly assigned to the four treatment ing a lecture format, each of the 44 hand
groups, and the five that were available in the gestures was presented and demonstrated
fall were likewise randomly assigned to each as participants followed in their own man-
of the four groups with the exception that uals. Any questions were answered and
two schools were selected for Treatment verbal descriptions were given of the pre-
Group 1 to balance small numbers in that senter’s previous use of the strategies. Par-
group from the spring. A cluster sampling ticipants were invited to use the gestures to
design was selected because the professional decode words as outlined in the manual.
development treatment was offered at the Treatment 2 (information ⴙ modeling).
school level. To have teachers who received Treatment 2 added vicarious experience, a
different treatment formats in the same second source of self-efficacy, as the pre-
schools and on the same grade-level teams senter modeled the use of the Tucker hand
would have introduced a potential cross- cues with local students selected by the
contamination of treatment effects. teachers in the participating school, typi-
The Tucker Signing Strategies for Read- cally second or third graders who were not
ing was selected as the teaching strategy to yet reading. In this treatment, approxi-
be taught because the strategy is simple in mately 20 minutes of the 3-hour workshop
scope and short-term in implementation were devoted to a demonstration in which
(Tucker, 2001). The strategies address a the presenter taught these struggling read-
specific area of beginning reading—that of ers to use the hand cues, and participants
matching letters to sounds and decoding watched as these students successfully de-
words using hand signals to provide a con- coded new words. It was noted that teach-
ceptual bridge between the written symbol ers acknowledged the success of the hand
and the sound associated with that letter. gestures with students for whom other
This method has demonstrated compelling methods had not been successful. Many ex-
results with struggling readers (Cole & pressed their intention to use the strategies
Majd, 2003; McMaster, 2003). All treatment in their own classrooms as a result of seeing
delivery was done by the same trainer who the demonstration.
was trained by the author of the Tucker Treatment 3 (information ⴙ model-
strategies. ing ⴙ practice). Treatment 3 added a pro-
In an attempt to differentiate between tected mastery experience with the in-
the various sources of self-efficacy, this clusion of a one-and-a-half-hour practice
study separated the training into four dif- session. During the practice session, partic-

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 235

ipants worked in groups to make decisions fied random selection. In each of the par-
regarding how they would use the strate- ticipating schools, all kindergarten through
gies, plan appropriate lessons for their stu- second-grade teachers participated in the
dents, and practice implementation of the study. The unit of analysis was the individ-
strategies (i.e., lessons that they could use ual teacher. The 93 teachers included in the
with their students the following day). To- study completed both the initial and final
tal time for this treatment was 4 hours and surveys.
30 minutes.
Treatment 4 (information ⴙ model- Instrumentation
ing ⴙ practice ⴙ coaching). Treatment 4 Three measures were used in this study.
added a stronger mastery experience with These included the Teachers’ Sense of Effi-
the inclusion of follow-up coaching in the cacy Scale (Tschannen-Moran & Woolfolk
use of the new skill. Coaching took place in Hoy, 2001), an adaptation of the Teacher
the weeks following the workshops and in- Sense of Efficacy for Literacy Instruction
cluded three components: (1) A 30-minute (Tschannen-Moran & Johnson, 2004), and a
small-group review of hand gestures, (2) a measure of implementation of the Tucker
15-minute one-on-one coaching session in method. These are described below.
dialogue with the presenter, and (3) a 30- Teacher self-efficacy. Teachers’ self-
minute coaching session with the presenter efficacy beliefs were measured using the
in the teacher’s classroom. Not only did 12-item Teacher Sense of Efficacy Scale
this coaching provide a mastery experience ([TSES]; Tschannen-Moran & Woolfolk Hoy,
as teachers were supported in their imple- 2001). Instructions directed the respon-
mentation of a new strategy, it also pro- dents to “Please respond to each of the
vided a more individualized and specific questions by considering the combination
experience of verbal persuasion. For those of your current ability, resources, and op-
teachers who requested it, the coaching portunity to do each of the following in
may have also included a second vicarious your present position.” Teachers rated the
experience within their own classroom set- items on a 9-point scale with anchors at 1
ting as they observed the presenter use the (none at all), 3 (very little), 5 (some degree),
Tucker Signing Strategies with their stu- 7 (quite a bit), and 9 (a great deal). The scale
dents. Total time for this treatment was 5 included three four-item subscales: Efficacy
hours and 45 minutes. for Instructional Strategies, Efficacy for
Physiological arousal as a source of self- Classroom Management, and Efficacy for
efficacy was not directly tested in the four Student Engagement. The following are ex-
professional development formats used in amples of items within each subscale:
this study, although teachers were no
doubt experiencing some emotional re- Efficacy for Instructional Strategies:
sponse to what they were learning.
Whether these feelings were beneficial or • How well can you implement alterna-
tive strategies in your classroom?
detrimental to their self-efficacy beliefs was
not explored in this study. Efficacy for Classroom Management:

• How much can you do to calm a stu-

Participants dent who is disruptive or noisy?
The participants were 93 primary (K–2)
Efficacy for Student Engagement:
and resource teachers from nine schools
within five different school districts. Through • How much can you do to get students
cluster sampling, schools were assigned to to believe they can do well in school-
workshop treatment groups through strati- work?
236 THE ELEMENTARY SCHOOL JOURNAL

The alpha coefficient of internal consistency out unknown words when they are
of the TSES for the current study was .90 for reading?
• To what extent do you use the Tucker
both administrations of the instrument. hand cues to meet the needs of strug-
Teacher self-efficacy for reading in- gling readers?
struction. Seven items adapted from the
Tschannen-Moran and Johnson (2004) mea- The alpha coefficient for the six-item imple-
sure of teachers’ sense of efficacy for liter- mentation scale was .99, indicating that
acy instruction were included to determine when a teacher indicated a particular level
teachers’ self-efficacy for the teaching of of implementation, she tended to indicate
reading (TSERI). The instructions and re- the same level across the various aspects of
sponse anchors were the same as those for reading instruction tapped in the six items.
the TSES. Examples of the reading items
follow:
Data Collection
• To what extent can you help your stu- School districts were selected for partic-
dents monitor their own use of read- ipation on the basis of their willingness to
ing strategies? include all primary teachers in the school
• To what extent can you teach the building rather than offering the workshop
sound/letter relationship to your stu- to volunteers after hours in a central loca-
dents?
• How much can you do to meet the tion. This allowed for a more realistic pro-
needs of struggling readers? fessional development experience. Once
permission was granted to conduct work-
The obtained alpha coefficients for the shops and collect data, principals were con-
TSERI were .91 in the first administration tacted to set dates for the workshops. At the
and .88 in the second. beginning of each workshop, the researcher
Implementation. Implementation of the explained the purpose of the study, assured
target teaching strategy was assessed at confidentiality, and asked teachers to com-
two stages. A one-item preassessment was plete the survey honestly. Follow-up sur-
designed to determine the level of imple- veys were administered to all participants
mentation of the Tucker Signing Strategies approximately 1 month after the work-
for Reading on the survey given prior to shops. Surveys were coded numerically so
training. This item was used to identify that initial responses could be matched
participants who were already using the with final responses by individual.
Tucker strategy. Since the intent was to
study teacher participation in training with Data Analysis
a new strategy, teachers who responded Descriptive statistics were computed
with a 7 or higher were removed from the for each treatment group and for the
sample. The posttraining implementation full sample for pre- and posttests of TSES
measure included five additional items that and TSERI as well as implementation.
were used to determine the relationship be- Repeated-measures ANOVAs were con-
tween teacher self-efficacy and implemen- ducted to compare the four treatment for-
tation of the strategy presented in the train- mats across time. A follow-up analysis
ing. Examples of the implementation items compared changes in teacher self-efficacy
follow: in general and teacher self-efficacy for
reading instruction that were greater than
• To what extent do you use the Tucker a standard deviation. Finally, an ANOVA
Reading Strategies? (Used on both
pre- and postsurveys) was used to explore the effect of treat-
• To what extent do you use the Tucker ment format on implementation of the
hand cues to help your students figure Tucker Signing Strategies for Reading.

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 237

TABLE 1. Mean Self-Efficacy and Implementation Scores

Treatment 1 Treatment 2 Treatment 3 Treatment 4 Overall

(N ⫽ 28) (N ⫽ 21) (N ⫽ 20) (N ⫽ 24) (N ⫽ 93)
M SD M SD M SD M SD M SD

Initial TSES 7.19 (1.00) 6.91 (.68) 7.00 (.98) 6.98 (.86) 7.03 (.89)
Final TSES 7.52 (.84) 7.24 (.65) 7.27 (1.04) 7.69 (.78) 7.45 (.84)
Initial TSERI 7.17 (1.07) 7.20 (1.03) 7.08 (1.20) 6.96 (.99) 7.10 (1.06)
Final TSERI 7.78 (.77) 7.22 (.90) 7.11 (1.17) 7.99 (.81) 7.56 (.96)
Implementation 3.45 (1.82) 2.71 (1.99) 3.73 (2.43) 6.78 (1.67) 4.20 (2.49)

Limitations self-efficacy-relevant input in additive fash-

A limitation of this study design was ion would result in steady increases in self-
that as new sources of self-efficacy were efficacy was not confirmed. The results also
added in the third and fourth treatment raised questions regarding how treatment
formats, there was also an increase in conditions and self-efficacy beliefs were re-
time, presenting an untested confounding lated to implementation of a new teaching
variable. The amount of contact time with strategy.
the trainer ranged from 3 hours for Treat-
ments 1 and 2 to 53⁄4 hours for those in Descriptive Statistics
Treatment 4. In addition, all data were Descriptive statistics for each of the four
self-reported. This meant that the level of treatment groups for the initial and final
implementation was not corroborated teacher self-efficacy scores, as well as the
with classroom observations to confirm implementation scores, are displayed in Ta-
whether or not teachers were actually im- ble 1. Responses were on a 1–9 scale, rang-
plementing the new strategies at the level ing from “none at all” to “a great deal.” The
they reported, except for those teachers in mean scores for initial teacher self-efficacy
Treatment 4 during which the trainer (TSES) and initial teacher self-efficacy for
made classroom visits and discussed with reading instruction (TSERI) were quite sim-
teachers the implementation challenges ilar, with TSES scores ranging from 6.91 to
they were facing. In addition, teachers’ 7.19, and TSERI scores ranging from 6.96 to
self-reports of self-efficacy are indicators 7.20. These ranges are similar to those
of self-beliefs related to motivation and found by Tschannen-Moran and Woolfolk
are not necessarily well calibrated to their Hoy (2001). The posttest for final teacher
actual effectiveness in classroom instruc- self-efficacy ranged from 7.24 to 7.69 and
tion. Finally, the survey responses may final teacher self-efficacy for reading in-
have suffered from response bias, al- struction ranged from 7.11 to 7.99. All
though the generally moderate correla- showed sufficient variability, with stan-
tions between scales and across time dard deviations ranging from .65 to 1.17.
would suggest this was not a significant Mean implementation scores ranged from
problem. 2.71 to 6.78 indicating substantial variance
in implementation across treatment groups
Results and standard deviations ranging from 1.67
to 2.43 within groups.
The findings of this study demonstrated
some surprising relationships between
treatment formats that included various Changes in Self-Efficacy Beliefs
sources of self-efficacy and changes in self- A steady increase in self-efficacy was
efficacy. The expectation that increasing expected from Treatment 1 through Treat-
238 THE ELEMENTARY SCHOOL JOURNAL

TABLE 2. Repeated-Measures ANOVA for Teacher Self-Efficacy

Type III
Sum of Squares df Mean Square F Sig.

Between-subjects effects:
Treatment 2.782 3 .927 .730 .537
Error 113.024 89 1.270
Within-subjects effects:
Time 7.732 1 7.732 33.420 .000
Time ⫻ treatment 1.438 3 .479 2.072 .110
Error (time) 20.592 89 .231

ment 4 because of the additive approach and Treatment 4 —showed significant gains
of including additional sources of self- in TSERI, while the two middle treatment
efficacy-relevant input for each subsequent groups—Treatment 2 and Treatment 3—re-
treatment. A repeated-measures ANOVA mained unchanged.
did reveal significant increases in teachers’ Further exploration led to the discovery
self-efficacy (TSES) across time [F(3, 89) ⫽ that these measures of central tendency
33.42, p ⬍ .01], however, in this case the were masking a drop in self-efficacy among
treatment effect was not significant. This a substantial number of teachers. To ex-
meant that all teachers tended to express plore this dynamic, we recoded the posttest
increased general self-efficacy at the time of scores as falling within a standard devia-
the posttest (approximately 1 month later) tion of the initial scores, those with losses of
regardless of which treatment group they more than a standard deviation, and those
were in (see Table 2). with a gain of greater than a standard de-
For teachers’ self-efficacy for reading in- viation (based on the SD of initial self-
struction (TSERI), there was no main effect efficacy scores). Only 1% of the general
for treatment, but there was a significant TSES scores dropped more than a standard
effect for time, again demonstrating an deviation, and while 24% saw increases of
overall increase in self-efficacy for reading more than a standard deviation, the chi-
instruction at the time of the posttest [F(3, square was not significant. For TSERI, how-
89) ⫽ 19.69, p ⬍ .01]. In this case, there was ever, 14% and 20% of the teachers in Treat-
also a significant interaction effect between ment 2 and Treatment 3, respectively, saw
time and treatment [F(3, 89) ⫽ 6.49, p ⬍ .01] declines in their self-efficacy of greater than
(see Table 3). An examination of these data a standard deviation. On the other extreme,
revealed an unexpected pattern (see Fig. 1). 19% and 10%, respectively, of teachers in
While the TSERI scores were homogeneous these same two treatment groups saw in-
to begin with, two groups—Treatment 1 creases in their TSERI of more than a stan-

TABLE 3. Repeated-Measures ANOVA for Teacher Self-Efficacy for Reading Instruction

Type III
Sum of Squares df Mean Square F Sig.

Between-subjects effects:
Treatment 4.971 3 1.657 1.067 .367
Error 138.195 89 1.553
Within-subjects effects:
Time 8.235 1 8.235 19.687 .000
Time ⫻ treatment 8.144 3 2.715 6.490 .001
Error (time) 37.228 89 .418

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 239

dard deviation. Furthermore, no teachers in

Treatments 1 and 4 had TSERI declines of
more than a standard deviation, while 25%
of Treatment 1 teachers and 38% of Treat-
ment 4 teachers had increases of over a full
standard deviation (see Table 4).

Implementation of Reading Strategies,

Teacher Self-Efficacy, and Format
The final research question explored the
relationship of self-efficacy and treatment
FIG. 1.—Changes in teacher self-efficacy for read- format in relation to the implementation
ing instruction. of the teaching strategy introduced. An
ANOVA was conducted to explore the dif-
ferences in implementation of the Tucker

TABLE 4. Changes in Self-Efficacy Larger Than 1 Standard Deviation

Less Than 1 SD Within 1 SD Greater Than 1 SD

Changes in Self-Efficacya Below the Mean of the Mean Above the Mean Total

Treatment 1:
Count 0 23 5 28
% within treatment .0 82.1 17.9 100.0
Treatment 2:
Count 0 17 4 21
% within treatment .0 81.0 19.0 100.0
Treatment 3:
Count 1 16 3 20
% within treatment 5.0 80.0 15.0 100.0
Treatment 4:
Count 0 14 10 24
% within treatment .0 58.3 41.7 100.0
Total
Count 1 70 22 93
% within treatment 1.1 75.3 23.7 100.0

Changes in Self-Efficacy Less Than 1 SD Within 1 SD Greater Than 1 SD

for Reading Instructionb Below the Mean of the Mean Above the Mean Total

Treatment 1:
Count 0 21 7 28
% within treatment .0 75.0 25.0 100.0
Treatment 2:
Count 3 14 4 21
% within treatment 14.3 66.7 19.0 100.0
Treatment 3:
Count 4 14 2 20
% within treatment 20.0 70.0 10.0 100.0
Treatment 4:
Count 0 15 9 24
% within treatment .0 62.5 37.5 100.0
Total:
Count 7 64 22 93
% within treatment 7.5 68.8 23.7 100.0

␹ (6, 93) ⫽ 9.40, ns.

a 2

␹ (6, 93) ⫽ 13.37, p ⬍ .05.

b 2
240 THE ELEMENTARY SCHOOL JOURNAL

Signing Strategies for Reading based on tially a second vicarious experience with
treatment format. The results revealed that the teacher’s own students). Through the
when controlling for initial self-efficacy, addition of follow-up coaching sessions,
40% of the variance in implementation teachers increased their self-efficacy for
could be explained by treatment format reading instruction and also their imple-
[F(3, 89) ⫽ 19.57, p ⬍ .01]. A Tukey post-hoc mentation of the Tucker Signing Strategies
test identified Treatment 4 (information, for Reading.
demo, practice, and coaching) as the only As in previous studies of teacher self-
training format that varied significantly from efficacy using the TSES, the mean scores
each of the other three groups in terms of were above the midpoint on the response
implementation of the instructional strategy scale, with means around 7 on a 9-point
introduced. Follow-up coaching was the ele- scale (Tschannen-Moran & Woolfolk Hoy,
ment that distinguished Treatment 4 from 2001), but with standard deviations be-
the other training models. tween .6 and 1.1 there should not be a con-
cern of ceiling effects. These strong means
reveal that career teachers tend to have a
Discussion fairly strong sense of efficacy for teaching.
This study explored the relationship be- Closer inspection of the results revealed
tween professional development formats that few practicing teachers selected re-
and changes in teacher self-efficacy beliefs sponse options 1–3 (“none at all” to “very
and teachers’ self-reports of implementa- little”) concerning their capability as teach-
tion of a new teaching strategy. Four pro- ers. Studies that segregate novice teachers
fessional development formats for the pre- have found their means to be significantly
sentation of the Tucker Signing Strategies lower than those of career teachers (al-
for Reading were designed to include, in though still above the midpoint), suggest-
additive fashion, three of the four sources ing that teachers with low self-efficacy tend
of self-efficacy identified by Bandura (1997). to either increase their self-efficacy or exit
The first three formats included verbal per- the field within the first few years of teach-
suasion in the form of information about ing (Tschannen-Moran & Woolfolk Hoy,
the teaching strategy, vicarious experience 2007).
in the form of a demonstration of the strat- For the primary-grade teachers in this
egy with local students, and a protected study, general sense of efficacy for teaching
mastery experience in the form of a plan- and self-efficacy for reading instruction
ning and practice session with colleagues. were highly correlated, sharing 55% of the
These all were related to modest gains in variance in these constructs. This shared
teacher self-efficacy, and the first was re- variance was not surprising because pri-
lated to gains in teacher self-efficacy for mary teachers who consider themselves
reading instruction. They were not, how- skillful at teaching reading are also likely to
ever, related to increases in implementation consider themselves capable teachers in
of the target teaching strategy. general because the teaching of reading is
In keeping with Bandura’s (1997) asser- so central to the task of teaching in the
tions, the most powerful professional de- primary grades. Although self-efficacy is a
velopment format included an authentic context- and task-specific construct (Ban-
mastery experience embedded in the teach- dura, 1997) that has been shown to vary
er’s regular teaching context. This mastery within teachers across subject and grade-
experience included the use of the strategy level classes at the secondary level (Rau-
in the teacher’s own classroom with the denbush et al., 1992; Ross et al., 1996, 1999),
support of a coach providing specific ver- for these primary teachers, the group of
bal persuasion and feedback (and poten- students they were referencing in their as-

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 241

sessments of general self-efficacy and self- rather than the proposed sources of self-
efficacy for reading instruction were the efficacy. The pattern of gains and dips in
same. Nonetheless, there was sufficient dis- self-efficacy, however, offers evidence that
crepancy to justify analyzing general self- contact time was not the only salient vari-
efficacy for teaching and self-efficacy for able, although it may have played a role.
reading instruction as separate constructs. The contact times for Treatment 1 and
The Tucker Signing Strategies for Read- Treatment 2 were the same, and yet there
ing workshop was targeted specifically to was a .61 increase in self-efficacy for read-
the skill of teaching beginning reading and ing instruction in Treatment 1 and almost
had a significant effect on teacher self- no mean change in Treatment 2, in part
efficacy for reading instruction in the first because of declines in self-efficacy for read-
and fourth treatment formats. The mean ing instruction in Treatment 2. Treatment 3
of general teachers’ self-efficacy also in- extended the workshop from 3 hours to 41⁄2
creased during the course of the study, hours, and yet again the mean change in
however, even though the instructional self-efficacy for reading instruction was
strategy taught only targeted reading in- negligible because over half of teachers ev-
struction for beginning readers. This find- idenced declines in their self-efficacy for
ing bears further exploration in future stud- teaching reading, with 20% evidencing
ies because it suggests that providing losses of more than a standard deviation.
teachers with instructional strategies and Treatment 4 extended the total contact time
support in one content area may transfer to to 53⁄4 hours, and over 90% of the teachers
a more general sense of self-efficacy. in this treatment experienced increases in
With exposure to a new teaching strat- both general self-efficacy and self-efficacy
egy, a substantial proportion of teachers in for reading instruction.
two of the four treatment groups saw sub- Because self-efficacy has been associ-
stantial declines in self-efficacy for reading ated with willingness to try new strategies
instruction. This drop indicates that the de- in previous research, the relationship be-
velopment of self-efficacy is more complex tween initial self-efficacy and implementa-
than expected. It seems that the awareness tion of the new strategy was tested. The
of a new instructional strategy that is lack of a relationship between initial self-
shown to have an impact on struggling efficacy and implementation raises a chal-
readers caused some teachers to reassess lenge to studies that have found that the
their definition of good teaching and to level of implementation of a new method
provoke a recalibration of their own self- depends on the self-efficacy of teachers
efficacy beliefs against this new standard. (Poole & Okeafor, 1989; Scribner, 1999). In
Without coaching to assist teachers in the the present study, variance in the level of
implementation of the new skill, a signifi- implementation of the new instructional
cant proportion of teachers were left feeling strategy was related to the addition of the
more inadequate than they had before. In follow-up coaching components of train-
Treatment 4, the group that received coach- ing, not to initial self-efficacy. There is,
ing, virtually all of the teachers experienced however, some alignment with earlier re-
increases in both general teacher self- search in the finding of a relationship be-
efficacy and teacher self-efficacy for read- tween teachers’ final self-efficacy for read-
ing instruction, and nearly 4 in 10 experi- ing instruction and implementation.
enced gains of more than a standard
deviation in both types of self-efficacy.
One might argue that the findings of Implications
this study have more to do with the contact The findings from this study suggest that
time involved in the various treatments the process of influencing the self-efficacy
242 THE ELEMENTARY SCHOOL JOURNAL

beliefs of practicing teachers is not a gies introduced will be implemented in the

straightforward one. Because self-efficacy classroom. Previous researchers have also
is a motivational construct associated with found that experiences limited to telling
the willingness to try new practices, persis- (verbal persuasion) and watching others
tence, and resilience in the face of setbacks, (vicarious experiences) were minimally ef-
the expectation of the researchers was that fective in leading to instructional change
the inclusion of additional sources of self- and have emphasized the need for provid-
efficacy in the four formats would lead ing teachers with feedback and support in
to incremental increases in teacher self- their implementation efforts (Guskey, 1989;
efficacy that would result in incremental Joyce & Showers, 1988; Stein & Wang,
increases in implementation of the new in- 1988). In the present study, professional de-
structional strategy. A more complex pat- velopment training that included follow-up
tern emerged, however, of both increases coaching, in which participants received
and decreases in self-efficacy beliefs. support as they developed the new teach-
The results of this study demonstrated ing skill, was related to increased imple-
that verbal persuasion, vicarious experi- mentation. The lack of relationship be-
ences, and even a limited mastery experi- tween implementation and the first three
ence did not prove to be particularly pow- training formats underscores the impor-
erful in creating the conditions to support tance of these follow-up coaching experi-
implementation of a new instructional ences in providing teachers needed assis-
strategy when they took place in a large tance as they attempt to implement a new
group setting. Furthermore, the results skill.
lend support to the importance of an au- The primary-grade teachers in this
thentic task-specific mastery experience study tended to view themselves as doing a
and of individualized verbal persuasion in good job of teaching in general, and specif-
raising self-efficacy beliefs and supporting ically of teaching reading, as evidenced by
implementation of a new teaching strategy the fairly high mean scores of both teacher
(Bandura, 1997). Only in the real setting can self-efficacy in general and teacher self-
a teacher experience a true test of his or her efficacy for reading instruction in the initial
capabilities. Thus, as teachers’ bolstered data collection. It is also true that in each of
self-efficacy supports their successful im- the nine schools, second- and third-grade
plementation of the new strategy, the im- students were found for the demonstra-
plementation experience becomes a mas- tions that had not yet started reading. In
tery experience that contributes to future most cases, those children had been the
self-efficacy assessments. The findings of students of several of the teachers present.
this study support professional develop- Prior to the Tucker workshop, it may have
ment models that mediate the effect of been that when these teachers had a stu-
the reassessment of participants’ analysis dent in their class who did not learn to
of the teaching task and their personal read, it did not disrupt their self-efficacy as
competency as they participate in skill- teachers because they consoled themselves
development programs (Tschannen-Moran, with the belief that a few students simply
Woolfolk Hoy, & Hoy, 1998). can’t be taught to read during the primary
The relationship evidenced between years. Receiving as passive observers infor-
professional development format and im- mation about a teaching strategy that most
plementation of a new instructional strat- were unfamiliar with in Treatment 1 may
egy in this study aligns with previous have challenged some of the teachers’ as-
findings that the form that professional de- sumptions about the requirements of read-
velopment takes is an important variable in ing instruction, but if, as Gregoire (2003)
the likelihood that the instructional strate- has proposed, they did not feel personally

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 243

implicated, they may not have engaged in reassessment in the face of evidence of a
the systematic processing of the informa- teaching strategy that is shown to be suc-
tion. When the teachers in Treatment 2 wit- cessful with students with whom they have
nessed a demonstration in which strug- failed. Wheatley (2002) challenged initia-
gling readers who had been students in tives for promoting increases in teacher self-
their classes and had not yet learned to efficacy beliefs as too simplistic and has pro-
read were taught to decode complex words posed that entertaining doubts about one’s
within a 20-minute demonstration, many of efficacy might actually provoke greater
them may have felt personally implicated. openness to new ideas. Thus if doubts,
Witnessing the trainer achieving success as reflected in decreases in teacher self-
where they had failed may have caused efficacy beliefs, are part of an expected
some teachers to adjust their evaluation of growth pattern as teachers learn to imple-
their own capability to teach reading by ment a new skill, they should be viewed as
creating new, higher standards for accept- natural and desirable reflections that help
able practice. The added component of create conditions of readiness to benefit
practice in Treatment 3 allowed an oppor- from professional learning. To overcome
tunity for limited skill development but these doubts, however, teachers may need
also revealed the difficulties inherent in assistance such as follow-up coaching in
putting it into practice, such that even more confronting and resolving the anxiety and
teachers began to interpret the intervention uncertainty inherent in changing profes-
as a threat as opposed to a challenge (Gre- sional practice (Guskey, 1988; Ross, 1992).
goire, 2003), resulting in a drop in their There is a need for more research to explore
self-efficacy for teaching reading. The the effects of teacher self-efficacy doubts on
follow-up conversations and assistance re- teacher development as they attempt to in-
ceived during coaching may have allowed crease their skill and stabilize their imple-
for deeper processing of the information mentation of new strategies.
presented and helped bring teachers’ skill That implementation of the reading
levels in line with their revised standards of strategy introduced was related to final
good practice, resulting in strengthened self-efficacy for reading instruction and not
self-efficacy. to general self-efficacy is of note because it
A substantial proportion of the partici- demonstrates the importance of assessing
pants in this study experienced a decrease self-efficacy as a task-specific construct
in their self-efficacy for reading instruction (Bandura, 1997). Although teachers who
with exposure to this new instructional participated in this professional develop-
strategy. This dip in self-efficacy beliefs is ment program generally came away with a
comparable with previous studies that stronger sense of their capability as teach-
have found that in the initial stages of im- ers, this was not necessarily related to their
plementation of a new skill, self-efficacy willingness or ability to implement the in-
beliefs may suffer (Guskey, 1984; Ross, structional strategy introduced. Increases
1994; Stein & Wang, 1988; Woolfolk Hoy & in self-efficacy for reading instruction may
Burke-Spero, 2005). It is interesting that the have led teachers to take the risks involved
greatest “dips” in self-efficacy came among in attempting a new instructional strategy,
treatment groups with only a small level of or conversely it may have been the result of
implementation. These decreases in self- successfully having implemented the new
efficacy for reading instruction may pro- strategy and having witnessed improved
vide further evidence that teachers’ calibra- student learning as a result. Scholars have
tion of the level of their content knowledge proposed that teacher self-efficacy is a cy-
is not especially accurate (Cunningham et clical process in which self-efficacy be-
al., 2004), such that it is vulnerable to a comes both a product and a constructor of
244 THE ELEMENTARY SCHOOL JOURNAL

experiences (Tschannen-Moran, Woolfolk achievement. Alberta Journal of Educational

Hoy, & Hoy, 1998). Although evidence of Research, 34(2), 148 –165.
increased student learning was not in- Ashton, P., & Webb, R. (1986). Making a differ-
ence: Teachers’ sense of efficacy and student
cluded in the study, Guskey (1988, 1989) achievement. New York: Longman.
has likewise noted that teacher self-efficacy Bandura, A. (1977). Self-efficacy: Toward a uni-
and willingness to innovate are both a fying theory of behavioral change. Psycho-
cause and an outcome of witnessing im- logical Review, 84, 191–215.
provements in student achievement. Bandura, A. (1997). Self-efficacy: The exercise of
The findings of this study revealed the control. New York: W. H. Freeman.
Cole, C., & Majd, M. (2003). Tucker Signing Strat-
complexity of the development of self-
egies for Reading national study. Blooming-
efficacy and the kinds of support needed ton: Indiana University. Retrieved June 25,
for teachers to implement a new instruc- 2009, from http://www.ahaprocess.com/
tional strategy. Future studies are needed files/Tuckernationalstudy1.pdf
to probe these complexities further. How Cunningham, A. E., Perry, K. E., Stanovich,
much self-efficacy is enough to sustain the K. E., & Stanovich, P. J. (2004). Disciplinary
knowledge of K–3 teachers and their knowl-
self-confidence needed to take on the risks
edge calibration in the domain of early liter-
inherent in change, and how much is too acy. Annals of Dyslexia, 54, 139 –167.
much, leading to complacency and an im- Gregoire, M. (2003). Is it a challenge or a threat?
munity to change? If decreases in self- A dual process model of teachers’ cognition
efficacy are to be expected, and to some and appraisal processes during conceptual
extent are even helpful to fostering a moti- change. Educational Psychology Review, 15,
vation for change, at what point do they 147–179.
Guskey, T. (1984). The influence of change in
become debilitating? Are lecture-based instructional effectiveness upon the affective
workshops that raise self-efficacy but are characteristics of teachers. American Educa-
not powerful enough to lead to implemen- tional Research Journal, 21, 245–259.
tation to be abandoned, or do they have a Guskey, T. (1986). Staff development and the
role to play in setting the stage for future process of teacher change. Educational Re-
change? Is it counterproductive to present searcher, 15(5), 5–12.
Guskey, T. (1988). Teacher efficacy, self-concept,
teachers with compelling evidence of new and attitudes toward the implementation
instructional strategies that raise their stan- of instructional innovation. Teaching and
dards of professional practice without pro- Teacher Education, 4(1), 63– 69.
viding the kinds of supports needed to Guskey, T. (1989). Attitude and perceptual
meet those higher standards themselves? change in teachers. International Journal of
How much coaching is needed to sustain Educational Research, 13(4), 439 – 454.
Guskey, T., & Passaro, P. (1994). Teacher effi-
deep conceptual change and lasting in-
cacy: A study of construct dimensions.
structional changes? Which components of American Educational Research Journal, 31,
coaching are the most powerful in support- 627– 643.
ing self-efficacy beliefs and implementa- Haney, J. J., Wang, J., Keil, C., & Zoffel, J. (2007).
tion? These questions offer rich new ave- Enhancing teachers’ beliefs and practices
nues for future investigation into the role of through problem-based learning focused on
pertinent issues of environmental health sci-
self-efficacy in the professional develop-
ence. Journal of Environmental Education,
ment of teachers. 38(4), 25–33.
Joyce, B., & Showers, B. (1988). Student achieve-
References ment through staff development. White Plains,
NY: Longman.
McKinney, M., Sexton, T., & Meyerson, M.
Anderson, R., Greene, M., & Loewen, P. (1988). (1999). Validating the efficacy-based change
Relationships among teachers’ and students’ model. Teaching and Teacher Education, 15,
thinking skills, sense of efficacy, and student 471– 485.

DECEMBER 2009
 SOURCES OF SELF-EFFICACY 245

McMaster, P. (2003). Learning through gestur- tional Research Association, Seattle (ERIC
ing: The Tucker reading strategy. Virginia Document Reproduction No. ED 453 170).
Educational Leadership, 2(1), 39 – 43. Smylie, M. (1988). The enhancement function of
Midgley, D., Feldlaufer, H., & Eccles, J. (1989). staff development: Organizational and psy-
Change in teacher efficacy and student self- chological antecedents to individual teacher
and task-related beliefs in mathematics dur- change. American Education Research Journal,
ing the transition to junior high school. Jour- 25, 1–30.
nal of Educational Psychology, 81(2), 247–258. Stein, M., & Wang, M. (1988). Teacher develop-
Poole, M., & Okeafor, K. (1989). The effects of ment and school improvement: The process
teacher efficacy and interactions among ed- of teacher change. Teaching and Teacher Edu-
ucators on curriculum implementation. Jour- cation, 4, 171–187.
nal of Curriculum and Supervision, 4(2), 146 – Timperley, H., & Phillips, G. (2003). Chang-
161. ing and sustaining teachers’ expectations
Raudenbush, S. W., Rowan, B., & Cheong, Y. F. through professional development in liter-
(1992). Contextual effects on the self- acy. Teaching and Teacher Education, 19,
perceived efficacy of high school teachers. 627– 641.
Sociology of Education, 65, 150 –167. Tschannen-Moran, M., & Johnson, D. (2004,
Ross, J. (1992). Teacher efficacy and the effects of April). Teachers’ sense of efficacy for literacy
coaching on student achievement. Canadian instruction. Paper presented at the annual
meeting of the American Educational Re-
Journal of Education, 17(1), 51– 65.
search Association, San Diego.
Ross, J. (1994). The impact of an inservice to
Tschannen-Moran, M., & Woolfolk Hoy, A.
promote cooperative learning on the stabil-
(2001). Teacher efficacy: Capturing an elu-
ity of teacher efficacy. Teaching and Teacher
sive construct. Teaching and Teacher Educa-
Education, 10(4), 381–394.
tion, 17, 783– 805.
Ross, J., & Bruce, C. (2007). Professional devel- Tschannen-Moran, M., & Woolfolk Hoy, A.
opment effects on teacher efficacy: Results of (2007). The differential antecedents of self-
randomized field trial. Journal of Research in efficacy beliefs of novice and experienced
Education, 101, 50 – 60. teachers. Teaching and Teacher Education, 23,
Ross, J. A., Cousins, J. B., & Gadalla, T. (1996). 994 –956.
Within-teacher predictors of teacher effi- Tschannen-Moran, M., Woolfolk Hoy, A., &
cacy. Teaching and Teacher Education, 12, 385– Hoy, W. K. (1998). Teacher efficacy: Its
400. meaning and measure. Review of Educational
Ross, J. A., Cousins, J. B., Gadalla, T., & Hannay, Research, 68, 202–248.
L. (1999). Administrative assignment of Tucker, B. (2001). Tucker Signing Strategies for
teachers in restructuring secondary schools: Reading. Highland, TX: Aha! Process, Inc.
The effect of out-of-field course responsibil- Wheatley, K. (2002). The potential benefits of
ity on teacher efficacy. Educational Adminis- teacher efficacy doubts for educational re-
tration Quarterly, 35, 782– 804. form. Teaching and Teacher Education, 18,
Scribner, J. (1999). Teacher efficacy and teacher 5–22.
professional learning: Implications for school Woolfolk Hoy, A., & Burke-Spero, R. (2005).
leaders. Journal of School Leadership, 9, 209 –234. Changes in teacher efficacy during the early
Shahid, J., & Thompson, D. (2001). Teacher effi- years of teaching: A comparison of four
cacy: A research synthesis. Paper presented at measures. Teaching and Teacher Education, 21,
the annual meeting of the American Educa- 343–356.