at SciVerse ScienceDirect

Teaching and Teacher Education 30 (2013) 84e98

Contents lists available

Teaching and Teacher Education

journal homepage: www.elsevier .com/locate/ tate

Elementary and secondary teacher self-efficacy for teaching and pedagogicalconceptual change in a drama-based professional development program

Bridget Lee a,*,1, Stephanie Cawthon a,1, Kathryn Dawson b,1

a The University of Texas at Austin, Department of Educational Psychology, College of Education, 1 University Station D5800, Austin, TX 78712, USAbDepartment of Theatre and Dance, College of Fine Arts at The University of Texas at Austin, USA

h i g h l i g h t s

< Elementary teachers have higher self-efficacy for teaching than secondary teachers.< Elementary teachers have higher pedagogical conceptual change than secondary.< Self-efficacy is not predictive of pedagogical conceptual change.< Grade level taught moderates the effects of years teaching on self-efficacy.

a r t i c l e i n f o

Article history:Received 13 March 2012Received in revised form18 October 2012Accepted 24 October 2012

Keywords:Teacher self-efficacyConceptual changeProfessional developmentArts integration

* Corresponding author. Tel.: þ1 512 577 8621; faxE-mail addresses: bridgetlee@utexas.edu (B.

austin.utexas.edu (S. Cawthon), kathryndawson@aust1 All authors are from the University of Texas at Au

0742-051X/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.tate.2012.10.010

a b s t r a c t

This mixed-methods research study explores the potential relationship between the teacher self-efficacyand pedagogical conceptual change. The study context was a drama-based instruction professionaldevelopment model that specifically sought to facilitate pedagogical conceptual change. Significantdifferences were present between elementary and secondary teachers in self-efficacy for teaching and inpedagogical conceptual change. However, self-efficacy did not predict conceptual change. Theindependent variable (elementary and secondary teachers) was a significant moderator between yearsteaching experience and self-efficacy. We discuss the significance of these findings in light of teachertraining and teacher effectiveness.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Of the many malleable factors in education that can positivelyimpact student academic outcomes, teachers and their teaching areamong the most important. Interventions to improve studentsuccess often focus on changes in a teacher’s instructional practice,adaption of a new curriculum, or implementation of new strategiesand assessment tools. Yet a teacher’s contribution to studentlearning goes beyond the content of their instruction or their levelof training. Teachers’ perspectives and perceptions of their ownteaching, and by extension, of their students’ learning, are anintegral part of successful teaching practice.

For the last thirty years, many researchers have studied teacherself-efficacy for teachingdan educator’s beliefs about his or her

: þ1 512 471 1288.Lee), stephanie.cawthon@

in.utexas.edu (K. Dawson).stin.

All rights reserved.

capability to teach and affect student outcomes (Bandura, 1986).Despite differences in educational settings and systems, researchsuggests that this teacher belief is present in multiple cultures andcountries (Klassen et al., 2009; Scholz, Doña, Sud, & Schwarzer,2002). In addition, numerous empirical studies have suggesteda positive effect of high teacher self-efficacy on student achieve-ment outcomes (Armor et al., 1976; Ashton & Webb, 1986; Berman& McLaughlin, 1977; Caprara, Barbaranelli, Steca, & Malone, 2006;Ertmer, 2005; Roberts, Henson, Tharp, & Morena, 2001;Rosenshine, 1979; Ross, 1992; Tschannen-Moran & Johnson, 2011;among others). In short, when teachers have confidence in theirability to improve student learning, they are in fact better atdoing so.

However, teacher self-efficacy has proven to be a complexconstruct that varies throughout a teacher’s career and interactswith a teacher’s pedagogical practice (Alger, 2009; Brouwers &Tomic, 2000). A particular challenge in identifying the impact ofteacher self-efficacy on student outcomes is understanding (a) howself-efficacy may affect other changes in teacher behavior, and (b)what other teacher characteristics may influence teacher self-

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e98 85

efficacy. More specifically, efficacious teachers may be better able tochange their teaching approach in order to accommodate studentneeds. In other words, a teacher’s high level of self-efficacy mayfacilitate a teacher’s openness to new ideas about teaching and theability to undergo and act upon a pedagogical conceptual change.Consequently, professional development providers and teachereducators should address the relationship between teacher self-efficacy and pedagogical conceptual change as they developcurriculum and/or instructional interventions. If a teacher’s peda-gogy needs to change in order to incorporate new evidence-basedpractices, then an essential key to facilitating this change may beraising a teacher’s self-efficacy.

The purpose of this study was to investigate the potential rela-tionship between teacher self-efficacy and pedagogical conceptualchange. The guiding question was as follows: What is the rela-tionship between a teacher’s self-efficacy for teaching and theiropenness and/or ability to implement new instructional strategiesin the classroom? In this article, we first define and explore the twotheoretical constructs. Next, we describe the study context ingreater depth, providing examples of ways inwhich the structure ofthe professional developmentmodel and the focus on drama-basedinstruction together sought to facilitate pedagogical conceptualchange. We then present both quantitative and qualitative findingsfrom the study, including teacher demographics, self-efficacyscores, and measures of pedagogical conceptual change over thecourse of the professional development experience. We discuss thesignificance of these findings in light of teacher training, teachereffectiveness, and the use of teacher efficacy as a construct for in-service teacher research.

2. Theoretical frameworks

2.1. Teacher self-efficacy

As stated earlier, teacher self-efficacy is an educator’s beliefabout his or her capability to teach and affect student outcomes(Bandura, 1986). Teacher self-efficacy appears to be a stableunderlying construct throughout various contexts. In addition,research has shown that high teacher self-efficacy positively affectsstudent academic outcomes.

Within educational research, measuring teacher self-efficacyhas shifted throughout the last thirty years. In initial studies,the construct was measured by asking teachers to respond to twostatements (Rosenshine, 1979) which produced a general idea ofhow teachers thought they and their colleagues could have animpact on student learning. This early research suggested thatteacher self-efficacy was one of the few teacher characteristics tohave a significant influence on student achievement (Armoret al., 1976; Berman & McLaughlin, 1977; Rosenshine, 1979). Forthe next two decades of research, researchers developed andused an expanded measurement that included 20 statementsabout the teacher’s perceived effects of the effort to help herstudents. This new instrument drew a more direct link betweenbeliefs about one’s capacity to teach effectively and the effects oftheir teaching practice (Gibson & Dembo, 1984). Results of thisresearch suggested that teacher self-efficacy influencesa constellation of student outcomes related to achievement(Ashton & Webb, 1986; Ross, 1992), student beliefs (Midgley,Feldlaufer, & Eccles, 1989) and student motivation (Dembo &Gibson, 1985), among others.

More recently, work by Tscahnnen-Moran and Woolfolk-Hoyhas established the conceptualization and measurement ofteacher self-efficacy in even further depth. Their widely used andvalidated measure of self-efficacy, Teacher’s Sense of Efficacy Scale

(TSES), examines three subfactors: 1) efficacy for instructionalstrategies, 2) efficacy for classroom management, and 3) efficacyfor student engagement (Tschannen-Moran & Woolfolk Hoy,2001). These subscale divisions acknowledge that teaching hasmultiple components that are meaningful in a teacher’s percep-tion of effectiveness, and that a teacher’s efficacy can varybetween them. Findings from studies using this framework furthersupport the positive effects of teacher self-efficacy on variousteacher and student outcomes in multiple countries and contexts(e.g., Bryant, 2007; Caprara et al., 2006; Lumpe, Czerniak, Haney, &Beltyukova, 2012; Shidler, 2009; Yeo, Ang, Chong, Huan, & Quek,2008).

If research repeatedly suggests that highly efficacious teachersproduce stronger student outcomes, what do these teachers do intheir classrooms that may be different than teachers with lowerlevels of self-efficacy? Woolfolk, Rosoff, and Hoy (1990) found thatteachers with high self-efficacy were more supportive ofa student’s autonomy in learning, had well-managed classrooms,believed in their ability to make an impact on the students, andgave few extrinsic rewards (e.g., stars for good behavior), relyinginstead on intrinsic motivation in their students to support goodbehavior. More current research suggests that high self-efficacy isalso related to a teacher’s greater effort and persistence whenworking with students (Brouwers & Tomic, 2000) and their abilityto handle stress in the classroom (Chan, 2002). In sum, efficaciousteachers have high expectations for their students and are effec-tive in facilitating ways for their students to meet thoseexpectations.

Teacher efficacy is a construct that should be accessible toteachers across a range of settings and subject areas. On thestrength of literature supporting the positive relationship betweenteacher efficacy and student outcomes, building teacher efficacyhas been a focus of teacher preparation programs (for pre-serviceteachers) and professional development (for in-service teachers).Raising a teacher’s self-efficacy is now a desired outcome ofprofessional development, one that is complementary to what istraditionally the primary focus in teacher training, deepeningcontent knowledge (Ertmer, 2005; Roberts et al., 2001; Tschannen-Moran & Johnson, 2011). However, in a desire to create an inter-vention to increase teacher effectiveness with valid measures ofthese outcomes, “self-efficacy” persists as a somewhat vague andelusive construct (Klassen et al., 2011; Pajares, 1992; Tschannen-Moran & Woolfolk Hoy, 2001). One way to better develop thefield’s understanding of self-efficacy is to look at how self-efficacymay or may not predict a teacher’s ability to make changes to herpedagogy, particularly in the context of explicit training andprofessional development. In other words, does a teacher’s level ofself-efficacy predict a teacher’s openness and/or ability to undergoa pedagogical conceptual change?

2.2. Pedagogical conceptual change through professionaldevelopment

Pedagogical conceptual change (also referred to as accom-modative change in Hashweh, 2009 or pedagogical change inMaskit, 2011) employs the conceptual change framework fromthe field of cognitive psychology and applies it to a teacher’sconceptions about teaching (Thorley & Stofflett, 1996). In general,conceptual change is defined as “the alteration of conceptionsthat are in some way central and organizing in thought andlearning” (Strike & Posner, 1992, p. 148). More specifically then,when a teacher experiences a pedagogical conceptual change,a shift occurs in his or her underlying conceptions about teachingthat in turn affects their teaching practice. Pedagogical conceptualchange focuses on translating new ideas about teaching into

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e9886

classroom practice. In some cases, a teacher may have a general-ized belief that a particular instructional strategy best engagesstudents, but not yet know how to enact that belief. This is not anexample of pedagogical conceptual change because the changeremains an idea, not a practice. Pedagogical conceptual change isnot complete until teachers have acted on the new conception inmeaningful ways.

For example, imagine a teacher who was trained in an infor-mation processing approach to learning, but, through professionaldevelopment, is exposed to and embraces a new approach such asa constructivist view of learning. If a teacher has a true peda-gogical conceptual shift from an information processing toa constructivist view of learning, her/his teaching will likelyexhibit a more constructivist pedagogy (e.g., demonstratinga more intentional consideration of the students’ prior knowledgeand experience during instruction). If the teacher subscribes toa constructivist view of learning but still expects every student tobe filled with the same knowledge, then the pedagogicalconceptual change cycle is incomplete, focusing on the concept ofconstructivism but not the practice. This process of moving fromconceptual change to pedagogical conceptual change, in reality, ismore complex and reiterative than described here, and does notoccur in a linear and ordered progression. In practice, researchershave found that substantive changes in teaching practice happenover long periods of time and only after teachers have had ampleopportunities to “try out” the new conception (Guskey, 2002;Strike & Posner, 1992).

Thorley and Stofflett (1996) used the theory of conceptualchange in science learning to illustrate the necessary componentsfor pedagogical conceptual change (illustrated in Fig. 1). In order fora teacher to consider a pedagogical conceptual change, four thingsneed to be present: 1) a teacher needs to be dissatisfied with hercurrent pedagogical conception, 2) a teacher needs to see

Disscatisfaction with current concept

New concept is intelligible and intelligible

Experimnew co

1. Why won’t my students participate in class discussions?

2. My colleague tries to relate the topic to students’ lives.

4. The students were very interested in sharing their artifacts from home.

Fig. 1. Pedagogical co

a conceptual change as intelligible, 3) a teacher needs to seea change as plausible, and finally, 4) a teacher needs to seea conceptual change as fruitful. In addition to these four steps, it isessential that a teacher enact the new belief about teaching in orderfor pedagogical conceptual change to occur. In this action, teachersengage in two reiterative steps: 1) critical reflection with otherteachers or professionals and 2) experimental implementation inthe classroom (Keiny, 1994).

Fig.1 offers an example of this process. This reiterative process isthe fertile ground for pedagogical conceptual change.

When teachers are encouraged to reflect on their conceptions ofpersonal pedagogy, more experienced teachers state that they havechanged from a more teacher-centered pedagogy to a morestudent-centered pedagogy (Alger, 2009). This is to say thata teacher’s practicewill inevitably undergo changes over the courseof his or her teaching career. However, when teachers are offereda training to induce a pedagogical conceptual change, the resultswill likely vary in their ability to accommodate and/or incorporatethis change (Evers, Brouwers, & Tomic, 2002). The latter path topedagogical conceptual change is the focus of pre-service educa-tion and professional development and will likely be necessary forschool-wide reforms, changes in federal policy, among other things.How then might teacher educators be able to facilitate pedagogicalconceptual change?

2.3. Self-efficacy and conceptual change

Although the focus of this study is on teachers, most of theresearch on self-efficacy and conceptual change focuses onstudents and their learning. It is thus advantageous to consider thisadditional student-focused literature for insight into teacher self-efficacy and conceptual change about their teaching. Similar toteacher efficacy, student self-efficacy has been strongly linked to

entation with ncept

Critical reflection and discussion about experimentation

New concept is fruitful

3. I will try having students bring an artifact from home related to the topic.

5. I didn’t realize students could relate the topic to artifacts from home. They were really involved in the discussion, too.

nceptual change.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e98 87

increased achievement outcomes in specific subject areas (Pajares,2003). Student self-efficacy in learning appears to be linked toresilience and perseverance in a challenging task. Across variousdomains, researchers hypothesize that efficacious students havethe confidence in their abilities that allow them to persist whenfaced with difficult tasks or obstacles and thus open the potentialfor conceptual change (i.e., learning) (Pintrich, 1999). This is thedominant assumption about the relationship between efficacy andachievement in the student-focused literature.

How similar are the assumptions behind the factors thatcontribute to conceptual change in teachers to this model ofself-efficacy and learning in students? Recall that the model ofconceptual change for teachers included a “need to know” ora dissatisfaction with one’s current conceptual understanding(Fig. 1), an assumption not present in the student-focused literatureon the self-efficacy construct. It may be, in the case of teachers andpedagogical conceptual change, that high self-efficacy underminesconceptual change due to a teacher’s satisfaction with currentlevels of student understanding, which causes the teacher to havean increased resistance to change (Linnenbrink & Pintrich, 2003;Pintrich, Marx, & Boyle, 1993).

Thus, when testing the relationship between teacher self-efficacy and pedagogical conceptual change, we need to exploreboth possible effects, positive and negative, of high self-efficacy.However, students and teachers are developmentally different aslearners. Adult learners tend to be problem-centered rather thancontent-centered (Knowles, Holton, & Swanson, 2005). In otherwords, adult learners tend to seek new information in order to useit to resolve a situation that needs to change. Students, in contrast,have the task of learning information that is presented. Applyingthis to a teacher’s context, if a teacher has high self-efficacy forteaching when she attends a professional development workshop,she may not think there is a “problem”with the current conceptionof teaching. Therefore, even with evidence of effectiveness pre-sented in a workshop, this teacher may not be motivated to engagein a conceptual change. On the other hand, Knowles also assertsthat adult learners are more likely to be motivated to learn if thematerial can be applied directly to their situation or context. If thematerial presented in professional development is relevant, anefficacious teacher may be able to see appropriate applications ofnew concepts more easily and therefore, accommodate conceptualchange readily. Depending onwhat aspects of adult learning theoryare activated for a highly efficacious teacher, participation inprofessional development may or may not lead to pedagogicalconceptual change.

Looking at the limited number of studies that use self-efficacy asa predictorof teaching practice, efficacious teachers seem tobemoreopen to incorporating new ideas and aremorewilling to experimentwith new instructional strategies to serve their students better(Evers et al., 2002; Guskey, 1986; Stein &Wang, 1988). For example,a study in the Netherlands of teacher burnout, teacher self-efficacyand teacher reaction to innovative pedagogy, researchers found thatteachers with high self-efficacy had a more positive reaction andmore readily implemented innovative pedagogical changes thanteachers with low self-efficacy (Evers et al., 2002). Findings supportthe hypothesis that a teacher with high self-efficacy would be morelikely to experience a pedagogical conceptual change throughexperimentation and critical reflection. If, however, there is a posi-tive relationship between high self-efficacy and satisfaction withthe current conceptions (little dissatisfaction), then efficaciousteachers may not be willing to experiment in the classroom. Thismay lead to a negative relationship between level of self-efficacyand pedagogical conceptual change.

There are, of course, many factors that influence teacher percep-tions of teaching. Teachers vary in their perceptions of teaching

dependingon thenumberof yearsof experience, their school climate,andother factors (Alger, 2009). Relevant to this study, teacherefficacyalso varies by the grade level of their students. Interestingly, teachersin elementary grades (e.g., Kindergarten through fifth grade) aretypically more efficacious than teachers in secondary grades (e.g.,middle and high school grades) (Fuller & Izu, 1986; Guskey, 1986;Wolters & Daugherty, 2007) although some studies have found novariation (Bosma, Hessels, & Resing, 2012). If we are to betterunderstand howprofessional development and teacher trainingmaysupport a pedagogical conceptual change, we need to take a devel-opmental perspective and understand more fully how elementaryand secondary teachers respond similarly and/or differently toeducational needs of students.

2.4. Professional development in drama-based instruction

Two of the ways to increase a teacher’s self-efficacy are toprovide opportunities for positive vicarious and mastery experi-ences (Bandura, 1986). Through enacting or “trying out” instruc-tional strategies, teachers have a greater sense of self-efficacy andmore positive attitudes toward changes in pedagogy (Cawthon &Dawson, 2009; Evers et al., 2002). Professional development indrama-based instruction (DBI) offers a way to “try out” newinstructional strategies and apply them to a classroom context. DBIis a collection of applied theater strategies including role playing,improvisation, and theater games (Cawthon & Dawson, 2009). Thetheoretical underpinnings of DBI invite teachers to 1) create a dia-logic community of learners among teachers and students(Vygotsky, 1978); and, 2) shift from passive and/or active instruc-tional strategies toward constructive and/or interactive instruc-tional strategies (Fonseca & Chi, 2008).

A great deal of research has examined the effects of DBI onimmediate student achievement across a variety of subject areasincluding language arts (Moore & Caldwell, 1990; Podlozny, 2000;Wagner, 1986, 1990), science (Dorion, 2009; Francis, 2007; Sloman& Thompson, 2010), foreign language acquisition (Bournot-Trites,Belliveau, Spiliotopoulos, & Seror, 2007; Erdman, 1991), andmath achievement (Fleming, Merrell, & Tymms, 2004; Walker,Tabone, & Weltsek, 2011) among others. In addition, researchalso suggests a positive effect of DBI on student attitudes towardacademics (Bournot-Trites et al., 2007; Walsh-Bowers & Basso,1999).

More than a set of discrete instructional strategies, DBI canalso inform a teacher’s pedagogical beliefs about teaching. Thispotential change, however, is greatly under-researched in artsintegration and is part of the aim of this article. From previousresearch conducted by the first author, this shift seems to includea teacher adopting an artistic way of viewing curriculum andinstructional goals with an eye toward concepts that can beembodied or acted out. This results in implementing kinestheticand reflective strategies that invite students to be activelyinvolved in their learning. By inviting teachers to experimentwith DBI strategies and reflect on new concepts of teachingduring the professional development sequence, previous studiesshow that teacher self-efficacy is increased (Cawthon & Dawson,2009).

The purpose of the study was to look specifically at elementaryand secondary teacher self-efficacy and its potential relationship topedagogical conceptual change through a drama-based instructionintervention. Does higher self-efficacy predict a teacher’s peda-gogical conceptual change? Do elementary teachers have higherself-efficacy for teaching than secondary teachers? If there isa difference in elementary and secondary teacher self-efficacy, isthere also a difference in a teacher’s willingness to change?

Table 1Demographics for the elementary schools.

Race/ethnicity School 1 School 2 School 3 School 4

White & non-Hispanic 89 72 86 200Hispanic 307 381 390 209Black & non-Hispanic 63 48 46 31Other race/ethnicities 10 3 7 5Free & reduced

lunch/total students387/469 432/504 425/529 230/445

Source: National Center for Education Statistics, 2009e2010 school year.

Table 2Demographics for the secondary schools in this study.

Race/ethnicity School 1 School 2 School 3

White & non-Hispanic 1156 75 194Hispanic 1766 157 606Black & non-Hispanic 304 19 103Other race/ethnicities 45 2 6Free & reduced lunch/total students N/A N/A N/A

Source: National Center for Education Statistics, 2009e2010 school year.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e9888

3. Method

3.1. Research hypotheses

This research study was guided by three hypotheses.

1) Elementary teachers have higher initial self-efficacy for teachingthan secondary teachers. Based on previous research literaturecomparing elementary and secondary teachers, we expect self-efficacy for teaching to be significantly higher in elementaryteachers than their secondary counterparts.

2) Elementary teachers experience greater conceptual change thansecondary teachers. If high self-efficacy is predictive ofconceptual change, then the efficacious elementary teacherswill be more open and willing to incorporate a shift inconceptual change and then act upon that shift.

3) Teachers with higher self-efficacy have greater shifts in conceptualchange than teachers with lower self-efficacy. In light of limitedprevious research, it was unclear whether or not self-efficacywould be predictive of a shift in conceptual change. Whenconsidering the possible effects of high self-efficacy on open-ness and ability to change, the researchers thought the teacherswith higher self-efficacy would be more willing to experimentwith the new concepts in the classroom. Therefore, theseteachers would be able to accommodate a conceptual changebased on their experimentation and reflection.

3.2. Study context: the university-school partnership

The context of this study was a longitudinal, intensive profes-sional development program in drama-based instruction for bothelementary (Kindergarten through fifth grade) and secondary(sixth through twelfth grade) teachers in a public school district inSouth Central Texas, U.S.A. The program for the both elementaryand secondary teachers was specifically designed to encouragepedagogical conceptual change as part of its long-term partnershipwith the teachers in this district. This study was approved by theInstitutional Review Board at the researchers’ university understudy number 2007-09-0146.

At the time of this study, the school district had participated inTheater for All Teachers (pseudonym) for four years. (For a morethorough reviewof this partnership, see Cawthon &Dawson, 2009.)Based at a large research university in the southern part of theUnited States, Theater for All Teachers (TAT) provides research-basedprofessional development in drama-based instruction to shift thelearning culture of the classroom. The partnering K-12 schooldistrict serves approximately 14,500 students. The school admin-istration supported the expansion of TAT professional developmentsuch that, in a two year time period, it served teachers at allsecondary campuses in all subject areas and teachers on four of theelementary campuses.

3.2.1. Sample demographics: elementaryA total of n ¼ 12 elementary teachers participated in the TAT

professional development program in the 2009 academic year. Theelementary teachers came from four campuses, with at least twoteachers from each participating campus. Elementary teachers hadan average teaching experience ofM¼ 8.75 years (SD¼ 8.28); someteachers were new to the field (i.e., with zero years of previousteaching experience) whereas others were long-term careereducators (i.e., up to 26 years of experience). Teachers had priorteaching experience ranging from Kindergarten through 8th grade,with an emphasis on second (19%), third (22%), and fourth (26%)grade classrooms. None of the elementary teachers had workedwith high school students in their previous experiences.

The demographics for three of the participating elementaryschools were very similar (Table 1). The fourth school was desig-nated as a gifted and talented magnet campus. This fourth schoolhad fewer students who qualify for the free and/or reduced lunchand had a more equal representation of White/Non-Hispanic andHispanic students than the first three schools.

3.2.2. Sample demographics: secondaryA total of n ¼ 18 secondary teachers participated in the TAT

professional development program in the 2008 academic year. Thesecondary teachers came from three different campuses includingamiddle school, the high school and the alternative learning center.Secondary participants had been teaching an average of M ¼ 8.03years (SD ¼ 9.58); the large standard deviation indicates that someteachers were new to the field (i.e., zero years previous experience)whereas others were long-term career educators (i.e., up to 31years in the field). Teachers had prior teaching experience rangingfrom 5th grade to 12th grade, with an emphasis on ninth (17%),tenth (18%), and eleventh (21%) grade classrooms. Six of theteachers had worked with fifth and sixth grade students, but onlyfor one year.

The race and ethnicity demographics in the participatingsecondary schools were very similar to the demographics of thethree main elementary schools (Table 2). Data was not available forthe free and reduced lunch program at the secondary level. The fourelementary schools in this study were located in very differentareas of the city and all fed into the secondary schools; thus, thedemographics were likely to be very similar to that represented inthe elementary schools. Of note for the secondary schools, the highschool completion rate was 83% in 2008 and 89% in 2009, higherthan average for this state.

3.3. Program structure

The TAT program provided sustained and intensive supportfor teachers and students, based on Thomas Guskey’s researchin effective professional development (2002). Through after-schoolsmall group training sessions and one-on-one mentorship,teachers learned drama-based teaching techniques to use in theclassroom. The initial meeting and a six-hour training formed thefoundation and trajectory for the monthly site visits to the districtthroughout the academic semester. A site visit is outlined in Table 3.(For a sample training agenda, see Appendix A.)

Table 3Program structure.

Beforevisit

� TAT staff constructed a training agenda based on theneeds assessment conducted with the campus LearningAdministrator and initial teacher training; teachersidentified a lesson/curricular topic to activate througha drama-based instructional strategy.

Duringvisit

� TAT staff facilitated an after-school training in drama-basedinstruction;

� Teachers and TAT staff co-created lesson plans at the training;teachers and TAT staff emailed lessons back and forth in theevening before the class visit;

� Teachers and TAT staff modeled and practiced strategiesduring full-day residency in teachers’ classes with theirstudents.

Aftervisit

� Teachers and TAT staff evaluated the strengths andweaknesses of the DBI lesson;

� TAT staff evaluated the strengths and weaknesses of thetraining agenda.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e98 89

This process aimed to facilitate pedagogical conceptual knowl-edge through each of the phases of the program. Before the visit,teachers selected a curricular topic and lesson that they would liketo activate through drama-based instructional strategies. Throughguided reflection on previous implementations of the lesson, theteachers identified potential areas of dissatisfaction with the orig-inal lesson. During the training with teachers, the facilitatorsemphasized ways to use the DBI strategies to engage the students.This addressed the intelligibility of DBI strategies. Next, after thetrainers modeled each strategy, they invited the teachers to think ofways that they could use the DBI strategy with their students intheir classrooms. This not only helped teachers connect to theirown experiences but also created buy-in for the plausibility phase ofconceptual change. Finally, the fruitful phase of conceptual changemay have happened through teacher conversations (e.g., mystudents got really involved when I tried out this DBI strategy) orthrough experimentation in the teacher’s classroom. The latter wasachieved through the full-day residency in the teacher’s classroomwith additional support.

3.4. Drama-based instruction sequence

For this study, teachers participated in a professional develop-ment sequence that was focused on a pedagogical conceptualchange through drama-based instruction (Cawthon & Dawson,2009). That being said, countless DBI strategies may facilitate thischange. Although little research has been conducted about theusefulness of specific strategies, the researchers chose a sequenceof learning that was informed by adult learning theory and researchon professional development.

Initial training sessions focused on highlighting the experienceof learners in DBI through low stakes active strategies such asValues Clarification (Rohd, 1998), Grafitti Alphabet, and Real/IdealImages (Boal, 1974). As described earlier, the teachers fully partic-ipated in each strategy and then reflected on the learning after-ward. These initial strategies attempted to highlight the potentialproblem of passive learning strategies by engaging the teachersdirectly in active learning strategies (Knowles et al., 2005). Duringthe middle of the professional development sequence, researchersfocused on strategies that allowed the participants to developquestioning skills that encourage using them in multiple waysthroughout the curriculum. These strategies mostly involveddifferent versions of Image Work (Boal, 1974; Spolin, 1986). Byfocusing on skills of instructional strategies rather than a “canned”lesson plan, teachersmay be able to use these relevant skills in theircontext immediately (Knowles et al., 2005). The latter training

sessions focused on fully embodying learning and opportunities to“try out” the strategies within the training session (Guskey, 2002).These DBI strategies included Hotseating, Teacher-in-Role, Student-in-Role, and Mantle of the Expert (Heathcote & Bolton, 1995). All ofthese strategies and their descriptions are in the unpublisheddocument Drama-Based Instruction: A Handbook for Using Drama asan Educational Tool (Dawson & Lee).

3.4.1. InstrumentsIn order to situate this research in the current work in the field,

a mixed-methods approach was used. Much of the literature hasfocused on quantitative measures of self-efficacy (e.g., Bryant,2007; Caprara et al., 2006; Shidler, 2009); however, we alsoincluded a qualitative measure to begin to better understand theteachers’ initial self-efficacy. Pedagogical conceptual change,however, has mostly been measured using qualitative measures(e.g., Alger, 2009; Morton, 2012). We wanted to better understandindividual teacher’s shifts throughout the program as well as theteachers’ experiences as a whole group; therefore, we includedtwo qualitative and one quantitative measure of pedagogicalconceptual change. Overall, this study used three main datacollection tools: Participant Intake Form, Self-Efficacy Scale, andLesson Plan Evaluations. A brief description of each of these toolsis provided below.

3.4.2. Participant intake formTeachers completed the participant intake form on the first long

training day at the start of the fall term. The form included fourdemographic questions and seven open ended questions. Teachersresponded to questions about the number of years they had beenteaching and the grades that they had taught or were currentlyteaching. This form also had open ended questions about teachers’perspectives on teaching, including questions about what types ofactivities most engaged their students, teachers’ most and leastfavorite things about teaching, their hopes for the TAT program, andtheir own prior experiences with professional development. Theopen-ended questions were used to measure initial pedagogicalconceptions.

3.4.3. Self-efficacy scaleFor this study, we used the self-efficacy measure based on

Bandura’s (1977) model of self-efficacy factors and developed byTschannen-Moran and Woolfolk Hoy (2001). The Teacher’s Sense ofEfficacy Scale twelve-item short form includes three subscalesaddressed through Likert-type questions: Student Engagement,Instructional Strategies, and Classroom Management. The firstsubscale, instructional strategies, focuses on the how of teaching, orwhat teachers do to help their students learn the material. Thesecond subscale, classroom management, is an essential compo-nent in working with a group of students, and is often one that canundermine the effectiveness of a teacher’s instructional strategies ifnot implemented in conjunction with a focus on academic content.The third subscale, student engagement, refers to how wella teacher can create a learning environment that the students aremotivated to be present for the material, both physically andpsychologically. The reliability for the scale is 0.90 and theconstruct validity is 0.87, 0.89, 0.84 for each of the subscales,p < 0.01. The reliability for this sample population was 0.92. Weadministered this scale at the beginning and end of the TATprogram.

3.4.4. Lesson plan pre- and post-evaluationsAt each monthly visit to the school district, TAT staff worked

with the teachers to introduce new strategies and to apply them toupcoming lesson plans. The lesson planning and evaluation process

Table 4Comparing pre self-efficacy means for elementary and secondary teachers.

N Mean SD

Pre self-efficacy total scoreElementary 11 7.34 1.10Secondary 18 6.74 0.87

Classroom management subscoreElementary 11 7.64 0.80Secondary 18 7.03 0.72

Student engagement subscoreElementary 11 7.16 1.19Secondary 18 6.31 1.06

Instructional strategies subscoreElementary 11 7.23 1.53Secondary 18 6.96 1.28

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e9890

was structured, in part, by two measures: a pre- and post-evaluation form. This form provided a space for teachers todescribe the content and intent of their lessons, describe theirstudents’ level of engagement, and to rate their lesson on a varietyof elements of authentic learning (drawn from Newmann &Wehlage, 1995). This measure had three open-ended questions,seven Likert-type questions and one question to provide specificexamples from their classroom. The reliability for the scale withthis sample populationwas 0.92. The open-ended questions servedas a second qualitative data point for pedagogical conceptualchange.

Teachers completed the pre-measure during the after schooltraining and the post-measure after they had implemented the newstrategy in their classrooms. The measure included the teacher’sperceptions of student learning during the lesson, such as howstudents construct new knowledge and communicate their under-standing. This pre-evaluation focused on the teacher’s perception ofher ability to engage students and teach the content based ona previous implementation of similar content. The post-evaluationfocused on the teacher’s perception of her ability to engage thestudents and teach the content based on the new DBI lessonimplementation. The change scores of the teacher’s perceptionswere used as themeasure for conceptual change. See Appendix B forthe pre-lesson evaluation measure.

4. Results

In the following section, we present findings from both ourquantitative and qualitative measures to address the three studyhypotheses. 1) Before the training, Elementary teachers had higherinitial self-efficacy for teaching than secondary teachers. 2)Elementary teachers experienced greater conceptual change thansecondary teachers. 3) Teachers with higher self-efficacy hadgreater shifts in conceptual change than teachers with lower self-efficacy.

Table 5Categories comments of teacher perceptions of teaching.

Elementary teachers Secondary teachers

Best way toengage students

� Hands-on� Kinesthetic� Cooperative learning� Competitions� Real-world activities

� Hands on activities� Working with others� Speed math� Rewarding efforts

(candy, tickets)Favorite thing

about teaching� Seeing the aha

moments actually occur� Prepare them for the

future and hopefully apassion for something

� Making learningexciting

� I love learning myself� Working with students

one on one� Experiencing the

students makingconnections

� Getting kids excitedabout my passion

Least favorite thingabout teaching

� Paperwork!� TAKS� Not enough time� When kids struggle

� I HATE paperwork� Excessive testing� The time I put into lessons� Classroom management� kids dropping out

4.1. Self-efficacy of elementary and secondary teachers

To address the first hypothesis, we compared the initial scores ofself-efficacy for teaching between elementary and secondaryteachers. The data produced a significant result (t27 ¼ 1.71, p < 0.05for equal variance and unequal sample sizes). This suggested thatthe pre self-efficacy scores for elementary teachers were signifi-cantly higher than the pre self-efficacy scores for the secondaryteachers (see Table 4). In addition, we compared the pre self-efficacy scores for the teachers at the subscale level. For bothclassroom management and student engagement scales the pre-subscores were significantly different between the groups(t25 ¼ 2.04, p < 0.05; t27 ¼ 2.01, p < 0.05) with elementary teachershaving the higher scores; however, self-efficacy subscores forinstructional strategies were not significantly different between thetwo groups (t27 ¼ 0.57, p ¼ 0.62).

We also analyzed post self-efficacy scores, but had only fourelementary teachers and ten secondary teachers who completedthe final self-efficacy scale survey. The data did produce a signifi-cant difference in means at t13 ¼ 2.25 with p < 0.05, but again, thismay not be a stable and accurate depiction of the measure with sofew data points. The low number of post self-efficacy scores wasdue to the change in data collection procedures at the end of theprogram. All other measures were gathered in person, but wewere not on the school site for this final measure of self-efficacyand online attempts to complete the survey were not verysuccessful.

To further understand the context of the self-efficacy scorefindings, we analyzed the open-ended comments from theparticipant intake form. Teacher responses to their perspectiveson teaching were gathered at the start of the professionaldevelopment sequence alongside the self-efficacy scale to geta sense of their beliefs and thoughts about teaching as theyentered the TAT program. High frequency and/or uniqueresponses comparing answers from elementary and secondaryteachers to the three questions about teaching are provided inTable 5. The first open-ended question addresses both hypothesisone and two; we included their responses in this part of thediscussion to highlight the striking similarities betweenelementary and secondary teachers’ responses on all three qual-itative questions.

Elementary and secondary teachers’ perspectives on teachingwere collected in the same manner and time point in their expe-rience with the TAT program. We first asked teachers to identify thetypes of activities that they felt best engaged their students. Bothelementary and secondary teachers emphasized kinesthetic orphysically active strategies, games, and competition as a good fit fortheir students. A few secondary teachers offered unique comments,such as “videos”, “music” and “the internet”. This seems develop-mentally appropriate that technology may be more integrated andengaging for secondary students.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e98 91

The second question asked participants to list their favoritething about teaching. High frequency answers included “talkingwith the students, hearing what they’ve learned”, “seeing the a-hamoments actually occur”, and “when students ‘get it’”. Unique tothe secondary teachers, quite a few teachers mentioned theirenthusiasm for sharing their passion for the specific subject matterwith their students. One thing that is noticeable across theseanswers is that students figured prominently in teacher’s thoughtsabout the positive aspects of teaching: students are at the heart ofwhat these teachers find appealing about their work.

In contrast, the final question asked the teachers to identify theirleast favorite thing about teaching. By far, the most commonanswer to this question was paperwork. Other common responseswere time constraints, state testing, classroom management, andseeing kids struggle. In addition, two secondary teachers discussedtheir disappointment when students fail or drop out of school. Itseems that elementary and secondary teachers believed that theyneeded to offer active instructional strategies to engage theirstudents in the material. When they were able to engage theirstudents, the teachers most enjoyed their work. Teachers enjoyedtheir work least when administrative demands or poor behaviorhindered their ability to engage students. Table 5 offers a compar-ison of high frequency teacher comments by both the elementaryand secondary teachers. The researchers grouped all the commentsinto similar categories and then used a teacher comment torepresent each category.

4.2. Elementary and secondary teacher pedagogical conceptualchange

To address the second hypothesis, first we looked at thedescriptive statistics for the change from pre and post lesson planevaluations as the marker of pedagogical conceptual change. SeeTable 6 for the descriptive statistics on this measure.

Comparing themeans for the elementary and secondary groups,there was a difference in the pedagogical concept change scores,with elementary having a mean of 3.41 change (range: 0.71e7) andsecondary having a mean of 0.61 change (range:�1.07 to 4.03). Theindependent samples t-test was significant (t27 ¼ 4.860, p < 0.001with equal variances assumed). This suggests that there wasa significant difference for elementary versus secondary teachers intheir pre/post change scores for the lesson plan evaluations, withelementary teachers experiencing a larger increase in conceptualchange than secondary teachers.

As part of the pre/post evaluation survey for each lesson,teachers had the opportunity to provide comments about the lessonand their thoughts about its effectiveness. Very few teachers wroteanycomments on the pre lesson evaluation questionnaire; however,most teachers provided comments on at least one of the post lessonevaluations. Elementary teachers provided three to eight commentsat each time point; whereas, the secondary teachers provided fourto twelve comments at each time point. Many of the commentsabout the lessons seemed to be very similar between elementaryand secondary teachers. For example, one elementary teachercommented, “All the students were actively participating. Even the

Table 6Descriptive statistics for pedagogical conceptual change in elementary andsecondary teachers.

Pre-/post-lesson N Min change Max change Mean change SD

Elementary 12 0.71 7.00 3.41 1.75Secondary 17 �1.07 4.13 0.61 1.36

reluctant learners.” Similarly, a secondary teacher commented,“Students paid more attention than usual.”

Overall the elementary teachers’ comments were much morepositive than the secondary teachers’ comments. Only oneelementary teacher had a moderately negative comment, “I wasquite conscious of time and an [.] assessment I needed toadminister.” This teacher was concerned about the amount of timethat had been spent using the DBI instructional strategy asopposed to a more direct-instruction method. On the other hand,secondary teachers offered negative comments on a variety ofareas including: classroom management, retention of knowledgeand implementing the strategies. For example, one teacher com-mented that students had “off-topic, private discussions” andanother teacher stated, “I am not sure how much influence thisactivity will end up having on the students’ [understanding thecontent].” The other area of concern for the secondary teachersseemed to be implementing the strategies. For example, oneteacher commented, “The students need to establish some groundrules before beginning.”

The qualitative data support the quantitative significant differ-ence in their conceptual change scores. Secondary teachers weremuch more reluctant to implement the strategies and when theydid use the strategies, theyweremore likely to question the efficacyof the instructional strategy for teaching the material and engagingstudents compared to the perceived amount of additional time theDBI strategy required.

4.3. Elementary and secondary teacher self-efficacy andpedagogical conceptual change

To address the final hypothesis of self-efficacy predictingconceptual change, we analyzed the pre self-efficacy score ofteachers as a predictor of the change score in lesson plan evalua-tions using a linear regression analysis. The regression model wasnot significant (F1,25 ¼ 0.359, p ¼ 0.555) for the mean pre self-efficacy scores. When analyzing the data separately, for elemen-tary teachers, the pre self-efficacy subscale scores were nota significant predictor of the change score in lesson plan evaluations(F1,9 ¼ 0.99, p¼ 0.46). For secondary teachers, the regressionmodelwas significant (F1,14 ¼ 4.71, p ¼ 0.02). Looking further at thesecondary teacher subscores, classroom management was notsignificant but student engagement and instructional strategieswere significant. In sum, as a secondary teacher’s self-efficacy forengaging students increased, her conceptual change also scoreincreased (t ¼ 3.75, p < 0.01). In contrast, as a secondary teacher’sself-efficacy for instructional strategies decreased, her conceptualchange score increased (t ¼ �2.31, p ¼ 0.04).

As part of a post hoc analysis, we analyzed the data fora significant grouping variable (elementary or secondary teachers)as a moderator for self-efficacy predicting pedagogical conceptualchange. After centering the self-efficacy scores for the participants,we calculated the product of the centered self-efficacy scores andthe dummy-coded grouping variable (i.e., elementary ¼ 0,secondary ¼ 1). The centered self-efficacy score and the productvariable were regressed on the change scores for the teachers. Theinteraction was not significant (FD1,24 ¼ 1.42, p ¼ 0.25). Althoughnot significant, this result may warrant follow up future research tofurther understand this tendency for efficacious elementary andsecondary teachers to experience different levels of pedagogicalconceptual change. See Fig. 2 for a visual representation of thistrend in the data.

After we became aware of this tendency in the data, we con-ducted an additional post hoc analysis of the pre self-efficacy scoresas predicted by years teaching experience. Past research has sug-gested that years teaching experience may contribute to a teacher’s

Fig. 2. Pre self-efficacy and conceptual change by elementary and secondary teachers.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e9892

self-efficacy but with mixed results, some reports have showna significant effect while others have found no significant effect onteacher self-efficacy (e.g., Bosma et al., 2012; Bryant, 2007;Tschannen-Moran & Johnson, 2011; Wolters, & Daugherty, 2007;Yeo et al., 2008). First we conducted an independent sample t testand confirmed that there was no significant difference in thenumber of years teaching between the elementary and secondaryteachers (t23 ¼ 0.18, p ¼ 0.86). Next, we centered the number ofyears teaching variable and created a cross product between thenewcentered variable and the grouping variable (i.e., elementary orsecondary). We regressed the number of years teaching and thecross product on the pre self-efficacy scores. The change from thefirst model (with the centered years teaching) to the second model(adding the product variable) was significant with FD1,22 ¼ 10.56,p ¼ 0.004. This suggests that there is a significant interactionproduced by the grouping variable. As number of teaching yearsincrease, secondary teachers have a higher self-efficacy for teachingwhereas elementary teachers have a lower self-efficacy forteaching. For a visual representation of the significant interaction,see Fig. 3.

5. Discussion

In this section, we discuss the quantitative and qualitativeresults for each of the research hypotheses, discuss limitations tothis study, and offer possible future directions for research.

5.1. Elementary teachers had higher self-efficacy than secondaryteachers

Elementary teachers had a significantly higher initial self-efficacy score than secondary teachers. This finding was in line

with previous research that found a difference in elementary andsecondary teachers’ self-efficacy for teaching (Fuller & Izu, 1986;Guskey, 1986; Wolters & Daugherty, 2007). More specifically,elementary teachers had significantly higher self-efficacy thansecondary teachers on two subscales: classroom managementand engaging students. In the qualitative data, classroommanagement was one of the main frustrations expressed bysecondary teachers with implementing drama-based instructionin their classrooms. In previous studies on self-efficacy forclassroom management and emotional exhaustion, researchersfound a direct relationship (Brouwers & Tomic, 2000) suggestingthat emotional exhaustion reduced the number of masteryexperiences and, therefore, lowered the teacher’s self-efficacy. Isit possible that the secondary teachers in our study were expe-riencing emotional exhaustion above and beyond what theelementary teachers were experiencing? It may be useful to addthis measure to future studies.

Upon initial analysis, the different outcomes in efficacy forengaging students and instructional strategies seemed odd. Ifa teacher is efficacious in instructional strategies, it seems that theywould also be efficacious in engaging students. For the qualitativedata point, we asked, “What activities best engage your students?”This question invites participants to list strategiesdnot necessarilya gauge of a teacher’s ability to actually engage the students throughthese strategies. Secondary and elementary teachers know theinstructional strategies that they believe should best engagestudents, however, they feel differently efficacious in engaging theirstudents.

Taking all of these results together, secondary teachers perceivethat they struggle more with managing and engaging students intheir classrooms than elementary teachers. Both the qualitative andthe quantitative data suggests that secondary teachers know how

Fig. 3. Centered pre self-efficacy scores as predicted by number of years teaching with sub-groups.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e98 93

to teach the material (i.e., instructional strategies), but are notequally confident in their knowledge and ability to engage andmanage students. This result implies that differences in teacherefficacy for student engagement may not correspond with differ-ences in how teachers express their ideas about strategies toencourage student participation; in other words, efficacy andconceptions about teaching may not follow the same pattern ortrajectory.

Teachers, in general, may have a common belief in theinstructional strategies that best engage students but they havea diverse ability and/or confidence to enact and follow through onthose beliefs in the classroom. If, in general, teachers believe thathands-on activities are the best way to engage students, activestrategies (like drama-based instruction) might be more readilyimplemented in the elementary classroom. In the secondaryclassroom, the content is more challenging to teach through activestrategies and the students may be more reluctant to participatedue to their developmental stage. As students advance tosecondary school, they may begin to focus less on play and moreon productive “work” (Erikson, 1959). Thus, active learning strat-egies may be seen as simple play and not as beneficial for olderstudents. This classroom culture coupled with the counter beliefthat kinesthetic activities facilitate learning may influencea secondary teacher’s self-efficacy.

5.2. Elementary teachers experience greater pedagogicalconceptual change

In this study, elementary teachers experienced a greater peda-gogical conceptual change than did secondary teachers. Yet resultsfrom this study cannot statistically support the hypothesis that self-efficacy significantly contributed to this change. Although theregression model was not statistically significant, it did suggest

a tendency for efficacious elementary teachers to have less changethan lower self-efficacy elementary teachers. The teachers’ lowerself-efficacy may be indicative of their dissatisfaction with theircurrent conception of teaching (Thorley & Stofflett, 1996); thisallowed them to be problem-focused to shift their conception(Knowles et al., 2005). In contrast, efficacious elementary teachersmay not have felt a need to change their current conceptions. Thissatisfaction may have undermined their desire and/or ability toundergo a pedagogical conceptual change (Linnenbrink & Pintrich,2003; Pintrich et al., 1993). Efficacious secondary teachers, on theother hand, experienced the greatest change in their pedagogicalconceptions. It seems we may have a case of “it depends”. If anelementary teacher is satisfied with her current conception ofteaching, she will be less open to change; whereas, if a secondaryteacher is satisfied with her current conception of teaching, shewillbe more open to change.

As suggested earlier, raising teachers’ self-efficacy is the focus ofnumerous professional development programs (including the onediscussed here). Depending on the teachers’ context, this focus onself-efficacy may or may not be a productive route toward peda-gogical conceptual change. For elementary teachers, the content ofthis particular professional development program was in DBI,focusing on active strategies that align well with their expectationsfor how to best engage students in the classroom. An elementaryteacher who is dissatisfied with the current status of her classroom(as represented by a low self-efficacy score) may see this type ofprofessional development as a relevant, intelligible and plausiblesolution (Thorley & Stofflett,1996; as implied by Fig.1). She thenmayput forth more of an effort to change her teaching practice to seea fruitful change.

For secondary teachers, it may be that the kinesthetic natureof this professional development program does not seem plau-sible for their current conception of what is most effective in their

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e9894

specific classroom. Although the secondary teachers commentedthat the best way to engage students is through kinestheticstrategies, this may not be the expectation (personal and/oradministrative) for how they interact with their students. Activelearning comes with many risks, particularly in the areas ofclassroom managementdanother area of low self-efficacy forsecondary teachers. Because of the higher risk and developmentalstage of the students, secondary teachers, may need moremastery experiences in drama-based instruction to see DBI asrelevant, intelligible and plausible instructional strategies and toincrease self-efficacy.

5.3. Years teaching experience negatively predicts self-efficacy forelementary teachers

In our post-hoc analysis, we expected to see a similar patternin the data for both elementary and secondary teachers for yearsteaching experience and self-efficacy based on previous research(e.g., Bosma et al., 2012; Bryant, 2007; Tschannen-Moran &Johnson, 2011; Wolters, & Daugherty, 2007; Yeo et al., 2008). Asteachers gain more experiences, it seems like they would bemore efficacious. However, if the experiences they are gaining arenot positive or mastery experiences, then this may lower theirefficacy. Our data did not suggest that years teaching experiencepromoted self-efficacy for elementary teachers. As elementaryteachers had more years of teaching experience, they hadsignificantly lower self-efficacy scores (and were more likely toexperience pedagogical conceptual change). One possible reasonfor this interaction of elementary and secondary teachers’ yearsteaching experience may be the focus and consistency of thecurriculum. For a secondary teacher, she is usually expected to“master” the content and the content changes very little over thecourse of a teacher’s career. Pedagogical approach is not as muchof a focus in curricular reforms at the secondary level. For anelementary teacher, she is usually expected to “master” thepedagogy to teach the content (McPartland, Coldiron, &Braddock, 1987). Again, the content changes very little over theteachers’ career; however, the pedagogy may change signifi-cantly. Elementary schools adapt new curriculum periodicallyand typically this implies a change in pedagogy as well. As anelementary teacher tries to keep up-to-date on the latest methodfor teaching a specific content, this may impact her self-efficacyfor teaching.

5.4. Limitations

There are several limitations to the generalizability of thefindings in this study. The first is the content (i.e., focus on DBI)and structure (i.e., long-term, embedded in classroom) of the TATprogram. This program is unique and represents a relatively non-traditional approach to teaching and professional development.We believe that this focus on DBI is a strength to the study but itis a very specific instance of professional development. It may notbe appropriate to generalize these findings to other types ofprofessional development that are not directly addressinginstructional strategies (e.g., a training for implementing newcontent or for addressing social/emotional development ofstudents).

Second, the sample size for this study was extremely small(N ¼ 30 teachers total), but is quite large for the intensive, longi-tudinal partnership model of this professional developmentprogram. It is very difficult to conduct research on professionaldevelopment for many reasons, including long-term requirementsfor conceptual change to occur, the complex context in whichteachers work, and thewide range of variables involved (e.g., school

programs, professional development days, changes in schedules)(Kucsera & Svinicki, 2010). One of the additional variables in thisstate is the standardized assessment schedule, where studentsparticipate in annual assessments acrossmultiple content areas. It isdifficult to conduct research in the spring semester because of theshift toward preparing students for the high-stakes, standardizedtests.

5.5. Conclusions and future directions

Teacher self-efficacy continues to be a highly researched theo-retical construct with great import. The fact that efficacious teacherspositively impact student outcomes across multiple disciplines andcontexts is reason alone to continue to pursue this somewhat elusiveconstruct. More specifically, researchers need to continue to inves-tigate ways to better understand how teacher self-efficacy affectsother changes in teacher behavior and what other teacher charac-teristics influence teacher self-efficacy. This article attempts toexplain one of the influences by researching the potential relation-ship between teacher self-efficacy and pedagogical conceptualchange.

In line with previous research, elementary teachers havea higher self-efficacy for teaching than secondary teachers;however, this difference is present in only two of the measuredsubscales: classroom management and engaging students. Boththe qualitative and the quantitative data suggests that secondaryteachers know how to teach the material (i.e., self-efficacy forinstructional strategies), but are not equally confident in theirknowledge and ability to engage and manage students. Thissuggests that professional development for secondary teachersneeds to focus on raising self-efficacy for classroom managementand engaging students. This could be done through on-goingmodeling of strategies in the classroom with debriefingbetween the teachers; thus, allowing more vicarious positiveexperiences.

Elementary teachers experience greater pedagogical conceptualchange; however, the greatest change occurred for low efficaciouselementary teachers and high efficacious secondary teachers. Inaddition, years teaching experience negatively predicts anelementary teacher’s self-efficacy but positively predictsa secondary teacher’s self-efficacy. When planning and recruitingfor a training session in active learning strategies, we may need todirect efforts toward novice elementary teachers who are strug-gling and more experienced secondary teachers who are excelling.This may seem counterintuitive, but this population seems themost ripe for change in their classroom practice.

Further research is needed to continue to clarify the self-efficacyconstruct and its possible relationship to pedagogical conceptualchange. The researchers look toward a qualitatively study that offersa few single case studies to further understand what may behappening with both elementary and secondary teachers as theydevelop through the DBI professional development sequence. Inaddition, the elementary and secondary teacher interaction on self-efficacy and years of teaching experience needs to be addressed. Ifelementary and secondary teachers have significantly different self-efficacies for teaching, what might be influencing and/or causingthis difference?

As demands of high student outcomes increase, teachers need tobe supported in the classroom to facilitate effective curriculumdelivery. This may require a shift in their pedagogical conceptualunderstanding of how to best impact students’ learning and anability to act upon this new conceptualization. If so, research needsto tease out the process to best facilitate this change. According tothis study, focusing on raising self-efficacy may not be the primaryroute for all teachers to undergo a pedagogical conceptual change.

Appendix A. Training agenda

Activating CSCOPE through drama-based instruction

ACTIVELY EXPLORING THROUGH ACTIVE DISCUSSION STARTERS

4:00 Poster Dialogue: One thing I want to learn today is . . .

My students learn best when . . .

I learn best when . . .

One challenge I have with CSCOPE is . . .

One opportunity I have with CSCOPE is . . .

4:15 Introduction:

Name, Affiliation (e.g. Debbie, Fifth grade teacher)

Part of what we will be doing today is learning new active learning strategies or drama based instructional strategies. Looking at the poster “I learn best when”, I see . . . We will be doing a lot of active strategies today—we invite you to jump in and participate. We will provide handouts of the strategies so no need to take notes.

4:20 Goals

Reference Post-It dialogue

o What is drama-based instruction (DBI)?o How can I use DBI with my curriculum, specifically CSCOPE?

Process Poster Dialogue: Application Discussion

1. Describe—What were we doing during these activities?2. Analyze—Where in the curriculum do you need to do these things?3. Relate—How can you use this in your classroom?

ACTIVELY EXPLORING THROUGH THEATRE GAMES

4:25 This is not a roll of tape

"This is not a triangle, it is a piece of pizza." Next person names something different, and it continues till everyone in circle has a chance to name items that are triangular shape. Play may continue with other shapes, including 3-Dimensional.

Process Theatre Game: Application Discussion

Describe—What were we doing during these activities?Analyze—Where in the curriculum do you need to do these things?Relate—How can you use this in your classroom?

ACTIVELY EXPLORING THROUGH MOVEMENT

4:45 Cover the Space

Math: Unit 9: Fourth Grade---Performance Indicator state that students are to identify and describe lines and angles when given a tangram. Name a polygon when given its attributes and identify a figure that coincides with the given attributes of the polygon. Describe the attribute of a named geometric solid and identify a figure that coincides with the given attributes of the geometric solid.

Cover the Space The whole group forms an attribute of a shape when instructed by facilitator.

Ex. Form a line; parallel lines; intersecting line; perpendicular line; obtuse angle; acute angle; right angle; polygon; non-polygon.

Process Movement: Application discussion

Describe—What were we doing during these activities?Analyze—Where in the curriculum do you need to do these things?Relate—How can you use this in your classroom?

ACTIVELY EXPLORING THROUGH IMAGE WORK

5:00 Frozen Image

What makes a good picture? Ex. Colors, See people’s faces, Lots of levels, Focus

Circle Sculpt—Vocabulary (in circle)

1. Take on image by yourself2. Ask for volunteer to share one image

a. Describeb. Analyzec. Relate

3. Get into inner and outer circle4. Mirror image for another person5. Walk around—call out what you see

a. Describeb. Analyzec. Relate

5:30 Lesson Planning with TAT Staff

All teachers meet with assigned TAT Staff and write lesson plan applying what we’ve learned in drama-based instruction to a curriculum topic.

Appendix B. Pre-lesson plan evaluation

1. Your name: ______________________ 2. Date: ____________________

3. Name of Lesson/Topic/Unit: ____________________________

4. Please give a brief description of the goals for this lesson/topic/unit:

5. When you have previously taught this lesson (i.e. without TAT strategies), what percentage of your class actively participated? Please circle one of the options below. Active participation includes contributing in a written, verbal, or physical way, either individually or part of a group. If you have not taught this lesson before, circle "New Lesson" and skip to #7.

Less than 20% 20% 40% 60% 80% 100% New lesson.

6. If you have taught this lesson before, can you give examples of how students were engaged in this lesson? In other words, how did you know if students were participating?

7. Sharing your thoughts about this lesson/topic/unit:

Directions: Please indicate your opinion about each of the questions below by circling any one of the nine responses in the columns on the right side, ranging from (1) “None at all” to (9) “A Great Deal” as each represents a degree on the continuum. If this is a new lesson/topic/or unit, think about how these questions apply to similar lessons you have taught without TATstrategies.

Without TAT strategies, this lesson, topic, or unit…

Non

e at

all

Ver

y L

ittl

Som

e D

egr

Qui

te

Bit

A

Gre

at

1. Is something I enjoy doing with the class. 1 2 3 4 5 6 7 8 9

2. Keeps students’ attention. 1 2 3 4 5 6 7 8 9

3. Reaches students with a range of abilities. 1 2 3 4 5 6 7 8 9

4. Showcases my strengths as a teacher. 1 2 3 4 5 6 7 8 9

5. Leads to a variety of assessment strategies. 1 2 3 4 5 6 7 8 9

6. Is one I would recommend to a colleague. 1 2 3 4 5 6 7 8 9

7. Appears to be effective in helping students 1 2 3 4 5 6 7 8 9

learn.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e98 97

References

Alger, C. L. (2009). Secondary teachers’ conceptual metaphors of teaching andlearning: changes over the career span. Teaching and Teacher Education, 25,743e751.

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 minorityschools (Report No. R-2007-LAUSD). Santa Monica, CA: Rand Corporation.

Ashton, P. T., & Webb, R. B. (1986). Making a difference: Teachers’ sense of efficacy andstudent achievement. New York: Longman.

Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change.Psychological Review, 84, 191e215.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.Englewood Cliffs, NJ: Prentice Hall.

Berman, P., & McLaughlin, M. (1977). Factors affecting implementation and continu-ation (Report No R-1589/7-HEW)In Federal programs supporting educationalchange, Vol. II. Santa Monica, CA: Rand Corporation.

Boal, A. (1974). Theate of the oppressed (C. A. Maria-Odilia & L. McBride, Trans.). NewYork City: Theatre Communications Group.

Bosma, T., Hessels, M., & Resing, W. (2012). Teachers’ preferences for educationalplanning: dynamic testing, teaching experience and teachers’ sense of efficacy.Teaching and Teacher Education, 28, 560e567.

Bournot-Trites, M., Belliveau, G., Spiliotopoulos, V., & Seror, J. (2007). The role ofdrama on cultural sensitivity, motivation and literacy in a second languagecontext. Journal for Learning Through the Arts, 3(1), 1e35.

Brouwers, A., & Tomic, W. (2000). A longitudinal study of teacher burnout andperceived self-efficacy in classroom management. Teaching and TeacherEducation, 16, 239e253.

Bryant, L. R. (2007). The relationship between urban teachers’ self-efficacy beliefs andadequate yearly progress. 68, ProQuest Information & Learning.

Caprara, G. V., Barbaranelli, C., Steca, P., & Malone, P. S. (2006). Teachers’ self-efficacybeliefs as determinants of job satisfaction and students’ academic achievement:a study at the school level. Journal of School Psychology, 44, 473e490.

Cawthon, S., & Dawson, K. (2009). Drama for schools: impact of a drama-basedprofessional development program on teacher self-efficacy and authenticinstruction. Youth Theatre Journal.

Chan, D. W. (2002). Stress, self-efficacy, social support, and psychologicaldistress among prospective Chinese teachers in Hong Kong. EducationalPsychology: An International Journal of Experimental Educational Psychology,22(5), 557e569.

Dembo, M. H., & Gibson, S. (1985). Teachers’ sense of efficacy: an important factor inschool improvement. The Elementary School Journal, 86(2), 173e184.

Dorion, K. R. (2009). Science through drama: a multiple case exploration of thecharacteristics of drama activities used in secondary science lessons. Interna-tional Journal of Science Education, 31(16), 2247e2270.

Erdman, H. (1991). Conflicts of interest: bringing drama into the elementary foreignlanguage classroom. Youth Theatre Journal, 5(3), 12e14.

Erikson, E. (1959). Identity and the life cycle. NewYork: International Universities Press.Ertmer, P. A. (2005). Teacher pedagogical beliefs: the final frontier in our quest for

technology integration? Educational Technology Research and Development,53(4), 25e39.

Evers, W., Brouwers, A., & Tomic, W. (2002). Burnout and self-efficacy: a study onteachers’ beliefs when implementing an innovative educational system in theNetherlands. British Journal of Educational Psychology, 72, 227e243.

Fleming, M., Merrell, C., & Tymms, P. (2004). The impact of drama on pupils’language, mathematics, and attitude in two primary schools. Research in Dramaand Education, 9(2), 177e197.

Fonseca, B., & Chi, M. T. H. (2008). The self-explanation effect: a constructivelearning activity. In R. Mayer, & P. Alexander (Eds.), The handbook of research onlearning and instruction. Routledge Press.

Francis, M. (2007). The impact of drama on pupils’ learning in science. SchoolScience Review, 89(327), 91e102.

Fuller, B., & Izu, J. A. (1986). Explaining school cohesion: what shapes theorganizational beliefs of teachers? American Journal of Education, 94(4),501e535.

Gibson, S., & Dembo, M. (1984). Teacher efficacy: a construct validation. Journal ofEducational Psychology, 76, 569e582.

Guskey, T. R. (1986). Staff development and the process of teacher change. Educa-tional Researcher, 75(5), 5e12.

Guskey, T. R. (2002). Professional development and teacher change. Teachers andTeaching: Theory and Practice, 8(3/4).

Hashweh, M. Z. (2009). Teacher accomodative change. Teaching and TeacherEducation, 19, 421e434.

Heathcote, D., & Bolton, G. (1995). Drama for learning: Dorothy Heathcote’s mantle ofthe expert approach to education. Portsmouth: Heinemann.

Keiny, S. (1994). Teacher’s professional development as a process of conceptualchange. In I. Calgren, G. Handal, & S. Vaage (Eds.), Teachers’ minds and actions:Research on teachers’ thinking and practice. Bristol, PA: Taylor & Francis, Inc.

Klassen, R. M., Bong, M., Usher, E. L., Chong, W. H., Huan, V. S., Wong, I. Y. F., et al.(2009). Exploring the validity of a teachers’ self-efficacy scale in five countries.Contemporary Educational Psychology, 34, 67e76.

Klassen, R. M., Tze, V. M., Betts, S. M., & Gordon, K. A. (2011). Teacher efficacyresearch 1998e2009: Signs of progress or unfulfilled promise? EducationalPsychological Review, 23, 21e43.

Knowles, M., Holton, E. F., III, & Swanson, R. A. (2005). The adult learner: Thedefinitive classic in adult education and human resource development (6th ed.).Burlington, MA: Elsevier.

Kucsera, J., & Svinicki, M. (2010). Rigorous evaluations of faculty developmentprograms. Journal for Faculty Development, 24(2), 5e18.

Linnenbrink, E. A., & Pintrich, P. R. (2003). The role of self-efficacy beliefs in studentengagement and learning in the classroom. Reading and Writing Quarterly, 19(2),119e137.

Lumpe, A., Czerniak, C., Haney, J., & Beltyukova, S. (2012). Beliefs about teachingscience: the relationship between elementary teachers’ participation inprofessional development and student achievement. International Journal ofScience Education, 34(2), 153e166.

McPartland, J., Coldiron, J. R., & Braddock, J. (1987). School structures and classroompractices in elementary, middle and secondary schools. Baltimore, MD: Office ofEducational Research and Improvement, 14.

Maskit, D. (2011). Teachers’ attitudes toward pedagogical changes duringvarious stages of professional development. Teaching and Teacher Education,27, 851e860.

Midgley, C., Feldlaufer, H., & Eccles, J. S. (1989). Change in teacher efficacy andstudent self- and task-related beliefs in mathematics during the transition tojunior high school. Journal of Educational Psychology, 81(2), 247e258.

Moore, B. H., & Caldwell, H. (1990). The art of planning: drama as rehearsal forwriting in the primary grades. Youth Theatre Journal, 4(3), 13e20.

Morton, T. (2012). Classroom talk, conceptual change and teacher reflection inbilingual science teaching. Teaching and Teacher Education, 28, 101e110.

Newmann, F. M., & Wehlage, G. G. (1995). Successful school restructuring: A report tothe public and educators. Madison, WI: Wisconsin Center on Education Research.

Pajares, M. F. (1992). Teachers’ beliefs and educational research: cleaning upa messy construct. Review of Educational Research, 62(3), 301e332.

Pajares, F. (2003). William James: our father who begat us. In B. J. Zimmerman, &D. H. Schunk (Eds.), Educational psychology: A century of contributions (pp. 41e64). Mahwah: Erlbaum.

Pintrich, P. R. (1999). The role of motivation in promoting and sustaining self-regulated learning. International Journal of Educational Research, 31, 459e470.

Pintrich, P. R., Marx, R. W., & Boyle, R. A. (1993). Beyond cold conceptualchange: the role of motivational beliefs and classroom contextual factors inthe process of conceptual change. Review of Educational Research Summer,63(2), 167e199.

Podlozny, A. (2000). Strengthening verbal skills through the use of classroomdrama: a clear link. Journal of Aesthetic Education, 34(3/4), 239e275.

Roberts, J. K., Henson, R. K., Tharp, B. Z., & Morena, N. P. (2001). An examination ofchange in teacher self-efficacy beliefs in science education based on theduration of in-service activities. Journal of Science Teacher Education, 12(3),199e213.

Rohd, M. (1998). Theatre for community, conflict, and dialogue: The hope is vitaltraining manual. Portsmouth, NH: Heinemann.

Rosenshine, B. (1979). Content, time, and direct instruction. In P. L. Peterson, &H. J. Walberg (Eds.), Research on teaching: Concepts, findings, and implications.Berkeley, CA: McCutchan.

Ross, J. A. (1992). Teacher efficacy and the effects of coaching on student achieve-ment. Canadian Journal of Education, 17(1), 51e65.

Scholz, U., Doña, B. G., Sud, S., & Schwarzer, R. (2002). Is general self-efficacya universal construct? Psychometric findings from 25 countries. EuropeanJournal of Psychological Assessment, 18(3), 242e251.

Shidler, L. (2009). The impact of time spent coaching for teacher efficacy on studentachievement. Early Childhood Education Journal, 36(5), 453e460.

Sloman, K., & Thompson, R. (2010). An example of large-group drama and cross-year peer assessment for teaching science in higher education. InternationalJournal of Science Education, 32(14), 1877e1893.

Spolin, V. (1986). Theater games for the classroom: A teacher’s handbook. Ebanston,IL: Northwestern University Press.

Stein, M. K., & Wang, M. C. (1988). Teacher development and school improvement:the process of teacher change. Teaching and Teacher Education, 4, 171e187.

Strike, K., & Posner, G. (1992). A revisionist theory of conceptual change. InR. A. Duschl, & R. J. Hamilton (Eds.), Philosophy of science, cognitive psychology,and educational theory and practice.

Thorley, N. R., & Stofflett, R. T. (1996). Representation of the conceptual changemodel in science teacher education. Science Education, 80(3), 317e339.

Tschannen-Moran, M., & Johnson, D. (2011). Exploring literacy teachers’ self-efficacy beliefs: potential sources at play. Teaching and Teacher Education, 27,751e761.

Tschannen-Moran, M., & Woolfolk Hoy, A. (2001). Teacher efficacy: capturing anelusive construct. Teaching and Teacher Education, 17, 783e805.

Vygotsky, L. S. (1978). In M. Cole, V. John-Steiner, S. Scrbner, & E. Souberman (Eds.),Mind in Society: The development of higher psychological processes. HarvardUniversity Press.

Wagner, B. J. (1986). The effects of role playing on written persuasion: An age andchannel comparison of fourth and eighth graders. PhD, University of Illinois atChicago, Chicago.

Wagner, B. J. (1990). Dramatic improvisation in the classroom. In D. I. Rubin, &S. Hynds (Eds.), Perspective on talk and learning. Urbana, IL: National Council ofTeachers of English.

Walker, E., Tabone, C., & Weltsek, G. (2011). When achievement data meet dramaand arts integration. Language Arts, 88(5), 365e372.

B. Lee et al. / Teaching and Teacher Education 30 (2013) 84e9898

Walsh-Bowers, R., & Basso, R. (1999). Improving early adolescents’ peer relationsthrough classroom creative drama: an integrated approach. Social Work inEducation, 21(1), 23e32.

Wolters, C. A., & Daugherty, S. G. (2007). Goal structures and teachers’ sense ofefficacy: their relation and association to teaching experience and academiclevel. Journal of Educational Psychology, 99(11), 181e193.

Woolfolk, A. E., Rosoff, B., & Hoy, W. K. (1990). Teachers’ sense of efficacy andtheir beliefs about managing students. Teaching and Teacher Education, 6,137e148.

Yeo, L. S., Ang, R. P., Chong, W. H., Huan, H. S., & Quek, C. L. (2008). Teacher efficacyin the context of teaching low achieving students. Current Psychology, 27(3),192e204.