PRE-SERVICE TEACHERS SELF-EFFICACY BELIEFS FOR TEACHING PHYSICS

PRE-SERVICE TEACHERS’ SELF-EFFICACY BELIEFS FOR TEACHING PHYSICS Claudia Meinhardt, Thorid Rabe and Olaf Krey University of Potsdam, Germany Abstract:...
Author: Brian Baker
3 downloads 0 Views 295KB Size
PRE-SERVICE TEACHERS’ SELF-EFFICACY BELIEFS FOR TEACHING PHYSICS Claudia Meinhardt, Thorid Rabe and Olaf Krey University of Potsdam, Germany Abstract: Science teachers’ self-efficacy is considered to be a central variable in order to predict science teachers’ classroom management, students’ self-efficacy beliefs and their achievement. Nevertheless there is an ongoing discussion about the appropriate sense of specificity of current instruments and also about their validity and reliability. A theoretically well grounded test instrument has been designed to measure physics teachers’ self-efficacy beliefs in the fields of “experimenting”, “analyzing and preparing physics contents” and “dealing with students’ conceptions”. For each of these fields two scales in the dimensions “planning” and “conducting” (physics lessons) have been developed. Results of pilot studies show by means of statistical analysis (CFA, correlation analysis) and qualitative methods (interviews, expert rating), that it is possible to develop an instrument for this level of specificity and indicate how items can be improved for the final test instrument. Keywords: self-efficacy beliefs, pre-service physics teachers, test developing, test validation

BACKGROUND, FRAMEWORK, PURPOSE Teachers’ self-efficacy can be defined as “the teacher's belief in his or her capability to organize and execute courses of action required to successfully accomplish a specific teaching task in a particular context” (Tschannen-Moran, Woolfolk Hoy & Hoy, 1998, p. 233). For our work we have used an even more specific definition describing self-efficacy as the perceived certainty to resolve a new or difficult issue successfully, even if some kind of resistance is encountered (Schmitz & Schwarzer, 2000, p. 13). Relevant characteristics of the construct come to the fore by the given definitions: teachers’ self-efficacy beliefs 

describe a judgment of their own abilities,



concern difficult actions and



are task- and context-specific.

Teachers’ self-efficacy has been studied in combination with many other variables. On the teacher’s side correlations between self-efficacy and a variety of constructs were found, including persistence, resilience, effort, goal setting, enthusiasm, commitment, instructional behavior, levels of planning and organization, open mindedness with regard to specific teaching tasks. On the learners’ side teachers’ self-efficacy correlates with achievement, motivation and students’ self-efficacy (Woolfolk Hoy, Hoy, & Davis, 2009). Mastery experience (i.e. own teaching experience) and vicarious experience (i.e. observing other teachers) are considered to be important sources for the development of teachers’ self-efficacy (Tschannen-Moran et al., 1998). In some cases first experiences in teaching seem to cause a “reality shock” indicated by decreasing self-efficacy beliefs (Woolfolk Hoy et al., 2009).

In the field of science teaching the Science Teaching Efficacy Beliefs Instrument (STEBI, Riggs & Enochs, 1990) is widely spread and used. However, „researchers have questioned the validity and reliability of existing measures“ (Tschannen-Moran & Woolfolk Hoy, 2001, p. 784), including the STEBI and the SETAKIST-R (Pruski et al., 2013). Furthermore, “there are questions about the extent to which teacher efficacy is specific to given contexts and to what extent efficacy beliefs are transferable across contexts. In addition, the appropriate level of specificity in the measure of teacher efficacy has been difficult to discern“ (Tschannen-Moran & Woolfolk Hoy, 2001, p. 784). To the best of the authors’ knowledge an instrument that allows measurement of self-efficacy for teaching science on a more specific level is not available so far. However, such an instrument is a prerequisite for studying science teachers’ self-efficacy beliefs and the influence of these beliefs on their science teaching. Therefore Cakiroglu et al. (2012, p. 458) “echo the need for new or revised measure(s) that would reliably assess science teaching efficacy and its components.” The study at hand aims at developing a theory-driven, valid and reliable instrument that allows measuring efficacy for specific fields of planning and conducting physics lessons. Because of the difficulties mentioned above the main research question is the following: Is it possible to develop an instrument that allows measuring physics teachers’ self-efficacy beliefs for teaching physics while meeting the criteria of both reliability and validity?

ITEM CONSTRUCTION, STUDY DESIGN, RESEARCH QUESTIONS In the scientific community there is a lively debate about the construct of pedagogical content knowledge in academic settings. One of the questions discussed is, how to operationalize this construct. Actually there is no consensus existing that concretely points out knowledge and skills students have to learn in science education courses. However, to identify relevant fields of physics teaching self-efficacy, standards for teacher preparation courses in Germany (KMK, 2010) and a document created by the science teaching research community (Korneck, Lamprecht, Wodzinski & Schecker, 2010) have been analyzed. Three fields of interest were chosen for item operationalization because they seem to represent a common denominator. Those fields are experimenting (ex), analyzing and preparing physics contents (pc) and dealing with students’ conceptions (sc). Each of these fields has been subdivided into the dimensions “planning” (p) and “conducting” (c). In order to represent the chosen self-efficacy definition (see above) we developed items presenting a first person point of view, describing actual behavior/ability that requires some effort or persistence and stating some kind of resistance or missing resource (barriers of action). These characteristics can be found in the following two item examples. •

I have no difficulties to prepare appropriate experiments for my lessons although the school lab is not well equipped. (experimenting – planning (ex-p))



I can stage an experiment comprehensively even if the experimental setup is rather complex. (experimenting – conducting (ex-c))

Obviously the operationalization is representing a certain level of specificity that is not intended to consider physics contents (such as mechanics e.g.). A questionnaire was developed and given to 84 German pre-service teachers. Each field was represented by 7 items in each of the two dimensions using a 1 to 4 Likert scale format (disagree/somewhat agree/strongly agree/fully agree). The questionnaire also included instruments to assess general self-efficacy (Schwarzer & Jerusalem, 1999), general

teachers’ self-efficacy (Schmitz & Schwarzer, 2000) and the subject related self-concept (adopted from Hoffmann, Häussler & Lehrke, 1998). Several pilot studies for validation and revision purposes have been arranged using quantitative and qualitative approaches. For pilot study 1 the newly developed scales have been statistically analyzed (CFA, ML estimation with Amos 18.0) and correlations to the other constructs have been examined in order to validate the construct at hand. In a second pilot study 18 experts for physics education have been asked to rate the adequacy and content validity of the scales (see questions below). For this a special questionnaire was compiled and experts’ answers were summarized by means of content analysis. In a third pilot study we conducted 21 interviews using the think aloud method to make sure that pre-service and beginning in-service teachers understand the items and do not consider them to be artificial. The interviews were recorded and partly transcribed. In summary the following questions are relevant for validation purposes in our studies: 

Are the items understood the way they are laid out?



Did the item-operationalization succeed? Are the barriers of action perceived as authentic? Do the operationalizations distinguish selectively between the dimensions planning and conducting?



Are the chosen fields of teaching physics relevant and do the developed items fit to the fields addressed? Are all relevant aspects of these fields of interest covered by the developed items?

RESULTS Pilot study 1 For the confirmatory factor analysis (CFA) factor loadings >.5 and the following values for fit indices are considered acceptable: χ2/df.95, TLI>.95, RMSEA.8) all of our scales meet our standards (see Table 1). Correlation analysis (Kendall’s Tau B is used here) shows that there are no relevant correlations between our physics teaching self-efficacy scales and the self-concept scale, which supports the idea of two different psychological constructs. Small correlations can be found between our physics teaching self-efficacy scales and the general self-efficacy scale (τ>.225, ρ>.295, α.404, ρ>.521, α

Suggest Documents