This contribution discusses methodical challenges and results of a study in a teacher education project at the University of Bremen, Germany, that focuses on initiating reflective practice. Methodical challenges arise from the interplay of structural conditions of teacher education in Germany and the complexity of the construct of interest.

Teacher Education

Internationally, there is a great diversity of teacher education systems (cf. Darling-Hammond & Lieberman, 2012; Schwille, Ingvarson, & Holdgreve-Resendez, 2013). To some extent, this is true also within Germany, where education is regulated by the federal states.

In Germany, prospective secondary school teachers usually specialize in two subjects (including subject didactics), whereas in case of prospective primary school teachers the situation is much more diverse. In Bremen, those students also specialize in two major subjects (at least one of which must be mathematics or German). In Germany, all prospective teachers also attend courses in education, with topics including general pedagogy, pedagogical psychology, sociology, and diversity.

German universities generally “emphasize academic, theoretical study. This approach ensures a relatively advanced level of academic preparation for all future teachers […] This first phase also contains a great deal of required education coursework […], albeit with a heavy emphasis on theory.” (Schwille et al., 2013, p. 97)

During their studies in Bremen, future teachers complete three practical phases. The longest one, the practical study semester, for which university prepares students and during which it supports them, takes place in the second semester of Master’s courses and offers students broad teaching experiences. After completing their university studies, students spend 18-24 months as trainee teachers at a school. Caused by a current need for teachers in Germany, some students already work as assistant teachers during their university studies.

Reflective Practice

Since Schön (1983), there has been a surge in both programs initiating reflective practice in teacher education and scientific research on that topic. There have been various approaches to model the multifaceted construct reflective practice, see e.g. Beauchamp (2015) for an overview. Following Dewey (1933), some focus on sequential steps that characterize reflective thought (e.g. Korthagen, 1985), whereas others stress the diversity of depth, breadth, or contents of reflections (e.g. Hatton & Smith, 1995; Valli, 1997).

This study

The project consists of four subprojects that all address aspects of reflective practice. In an overarching mission statement, the project identified recontextualization, coherence, and collegiality as core ideas of reflection on which the project puts special emphasis. This study is part of the project’s evaluation, relating the project’s innovations to students’ self-perceived preparation for reflection.

In order to conceptualize a wide range of reflective thoughts and actions and to integrate the above core ideas, Valli’s (1997) classification proved adequately versatile and was therefore chosen as underlying model for generating questionnaire items.

As described above, students’ studies are split between two major subjects on the one hand and general education on the other. As a comprehensive assessment of students’ self-perceived preparation for reflection should account for these different parts, a central question is:

(RQ1) Do students in their perception of how university prepares them for reflection distinguish between subject-related and overarching facets?

Answering this question affirmatively would entail a more complex data modeling, using e.g. multilevel models in order to account for subject-relatedness of students’ answers.

With recontextualization being one of the project’s core ideas, many questionnaire items touched upon use of theories in practical contexts. Considering German universities’ general emphasis on theories, a further question is:

(RQ2) To what extent are teaching students’ attitudes towards theories related to their self-perceived preparation for reflection on the one hand, to what extent are practical experiences on the other?

Instruments and Design A questionnaire was constructed assessing how teacher students perceive how well university studies prepare them for reflective thinking and acting as teachers. Structurally, it was based on Valli’s (1997) classification, which distinguishes five facets of reflective thinking. In this contribution only technical reflection (TR) and deliberative reflection (DR) will be addressed. TR is characterized by a “focus on the narrow domain of teaching techniques” (Valli, 1997, pp. 74-75) and an emphasis on application of textbook theories. DR is distinguished by “decision making based on a variety of sources: research, experience, the advice of other teachers, personal beliefs and values […]” (Valli, 1997, p. 77). Content-wise, a special focus was on the key ideas outlined above. Items consisted of a description of some reflective teacher action and a seven-point Likert scale for assessing how well one feels prepared for this action. Items referring to subjects were answered once per subject. Sample items are: - “Analyzing specific classroom situations together with colleagues.” (DR) - „Applying subject-related theories in order to assess pupils’ learning progresses.” (TR) Along with several established instruments – among them scales on attitudes towards theories (Wagner, Klein, Klopp, & Stark, 2016) and empathic concern (Davis, 1980) –, which were considered constant background variables and were assessed only once, the above instrument was employed three times between autumn 2017 and summer 2018. Sample Across all three occasions, N(tot) = 145 (76% female, 64% master students, 70% prospective secondary school teachers) advanced undergraduates or master students of teaching participated, roughly mirroring the population of students. (RQ2) was answered based on students who participated at least twice (N(long) = 52). Analyses Analyses were conducted in steps: Firstly, TR and DR were analyzed using exploratory factor analyses (cf. e.g. Loehlin & Beaujean, 2017), leading to dimensionality results that answer (RQ1). Based on these findings, confirmatory factor analyses were applied to all scales and person parameters were generated. Secondly, missing background data were imputed using chained equations (van Buuren, 2018). Anticipating the answer to (RQ1), for the subject-related facets of TR and DR, the data consisted of two person parameters per person and occasion. Hence, thirdly, the final models specified in order to answer (RQ2) were multilevel regression models (cf. e.g. Snijders & Bosker, 2012), with one nesting structure pertaining to persons and, in case of subject-related facets, another nesting structure pertaining to subjects. Intra-class correlations provided additional evidence for the answer to (RQ1).

Results Exploratory factor analyses and two-dimensional confirmatory factor analyses clearly implied that subject-related facets of TR and DR must be modeled as factors of their own, answering (RQ1). Hence, four outcome variables (TR and DR general and subject-related, respectively) were investigated in the following. CFI- and TLI-values between 0.94 and 0.99 indicated good model fit for the TR scales. Only for the subject-related DR scale, model fit was mediocre (CFI = 0.94, TLI = 0.9). In null models, 25.5% (subject-related TR) and 13.9% (subject-related DR) of the variance were explained by the data being nested in subjects, supporting the affirmative answer to (RQ1). Final multilevel regressions showed, among other things, a significant positive relation between subject-related scales and favorable attitudes towards theories as well a significant negative relation between extramural teaching experiences and general DR. Discussion The answer to (RQ1) suggests that including subject-related aspects in an instrument assessing teacher students’ preparation for reflective thinking might necessitate more complex modeling approaches, as it did in this study. However, it does not settle the questions to what extent this holds true in teacher education systems putting less emphasis on subjects or for experienced teachers with some distance to university studies. The two results of the regression analyses mentioned above raise questions of validity (Are favorable attitudes towards theories just a proxy for a general inclination towards university studies?) and interpretation / causality (Are teaching students more critical of their university studies because they know what is required in practice?). Again, further research is needed in order to understand whether these results can be translated to teacher education systems focusing less heavily on theory. There being no evidence of the practical study semester having similar negative relations with self-perceived preparation for reflection might suggest differences between supported and relatively unsupported practical experiences.

Beauchamp, C. (2015). Reflection in teacher education: issues emerging from a review of current literature. Reflective Practice, 16(1), 123-141. Darling-Hammond, L. & Lieberman, A. (Eds.). (2012). Teacher Education around the World. London: Routledge. Davis, M. H. (1980). A Multidimensional Approach to Individual Differences in Empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85. Dewey, J. (1933). How we think. Buffalo: Prometheus Books. Hatton, N. & Smith, D. (1995). Reflection in teacher education: Towards definition and implementation. Teaching & Teacher Education, 11(1), 33-49. Korthagen, F (1985). Reflective Teaching and Preservice Teacher Education in the Netherlands. Journal of Teacher Education, 36(1), 11-15. Loehlin, J. C. & Beaujean, A. A. (2017). Latent Variable Models (5th ed.). New York: Routledge. Schön, D. A. (1983). The reflective practitioner: How professionals think in action. New York: Basic books. Schwille, J., Ingvarson, L., & Holdgreve-Resendez, R. (Eds.). (2013). TEDS-M 2008 Encyclopedia: A Guide to Teacher Education Context, Structure, and Quality Assurance in the Seventeen TEDS-M 2008 Countries. Amsterdam: IEA. Snijders, T. A. B. & Bosker, R. J. (2012). Multilevel Analysis (2nd ed.). Los Angeles: Sage. Valli, L. (1997). Listening to other voices: A description of teacher reflection in the United States, Peabody Journal of Education, 72(1), 67-88. van Buuren, S. (2018). Flexible Imputation of Missing Data (2nd ed.). Boca Raton: CRC Press. Wagner, K., Klein, M., Klopp, E., & Stark, R. (2016). Förderung anwendbaren bildungswissenschaftlichen Wissens anhand kollaborativem Lernen aus Fehlern [Facilitation of Applicable Knowledge in Teacher Education by Collaborative Learning from Instructional Errors]. Unterrichtswissenschaft, 44(4), 373-390.