Educational Measurement: Issues and Practice xxxx 2019, Vol. 0, No. 0, pp. 1–9 Discrete Choice Modeling in Education: An Innovative Method to Assess Teaching Practices Vasilis Grammatikopoulos, University of Crete, Greece, Athanasios Gregoriadis, Aristotle University of Thessaloniki, Greece, and Michalis Linardakis, University of Crete, Greece Self-reported measures are an easy, time-efficient, and low-cost way to collect research data compared to other assessment methods. They are, however, characterized by several limitations regarding the quality and the clarity of the data they produce, especially when they are implemented in education. The main purpose of the current study was to use a method called discrete choice modeling (DCM) in education, in order to develop a self-reported instrument that will reduce the bias for socially desirable responses and to assess teachers’ practices in physical activity. DCM method has the ability to overcome the respondents’ tendency to answer in a socially desirable way. A nationally representative sample of 531 Greek early educators participated in this study and were administered a self-assessment questionnaire based on DCM. Results showed that the DCM based newly developed instrument manages to assess effectively educators’ practices and provided evidence of the applicability of the method in education. Further implications and future recommendations regarding the effective application of DCM in education are discussed. Keywords: assessing teaching practices, discrete choice modeling, early childhood education, self-assessment T here are several methods available for assessing teach- ing effectiveness in education. Apart from recent holis- tic and dynamic approaches to assess teaching effective- ness (e.g. Creemers & Kyriakides, 2008), value-added mod- els, classroom observations, principal evaluations, analysis of classroom artifacts, portfolios, and self-report measures are acknowledged as common methods to assess teaching effec- tiveness (Goe, Bell, & Little, 2008; Little, Goe, & Bell, 2009). A “common secret” in the teaching assessment field is that educators in general are reluctant to accept external assess- ments implemented in their classrooms, even in cases where anonymity is ensured and/or the data collection is being made only for research purposes. However, teachers’ self-reports are fast and easy to implement compared to other time- and cost-consuming methods. However, it is acknowledged that teacher self-reports are accompanied by several limitations regarding the quality and clarity of the data they collect. Para- phrasing a seminal comment by Nevo (2001, p. 104), it can be argued that everybody seems to hate external observations, whereas nobody trusts self-assessment reports. One way to address the credibility gap of the self-report methods may be to use discrete choice modeling (DCM). The research problem that this study attempts to address is to overcome the social desirability bias of traditional approaches by using a method that seems to be more effective. This method is the discrete choice modeling (DCM) method, a Vasilis Grammatikopoulos, University of Crete, Greece; gram- vas@bc.teithe.gr. Athanasios Gregoriadis, Aristotle University of Thessaloniki, Greece. Michalis Linardakis, University of Crete, Greece. statistical technique under the family of conjoint analysis. DCM has been implemented successfully in various areas such as econometrics, marketing, and management, but its use in education is very scarce. DCM in Education This study attempts to adapt and apply the method of DCM (McFadden, 1974) in education. More specifically, it employs DCM in order to assess early educators’ practices in teaching physical activity. DCM is based on the modeling of actual choices in realistic scenarios formed by mutually exclusive alternatives (i.e., teaching practices). It is used to estimate the utility or value the respondent places on attributes when making choices. The methodology and the advantages of using DCM over statistical methods of stated preference data have been thoroughly discussed in the literature (Hensher, Rose, & Greene, 2005; Louviere, Hensher, & Swait, 2000; Louviere & Islam, 2008; Train, 2003). To our knowledge, only a few recent studies in education have employed a similar methodology (Aubusson, Burke, Schuck, Kearney, & Frischknecht, 2014; Cunningham et al., 2014). Let us examine the hypothetical example where a study uses a questionnaire asking teachers to report (e.g., on a Lik- ert scale from 1 to 7) how they implement separately each of the important attributes on teaching physical activity in their daily practice. Important attributes on teaching phys- ical activity could be, for example, “how teachers evaluate children’s physical activity” or “how teachers engage in the activities with children.” The respondents tend, in that case, to overrate the relative implementation of the attributes, C  2019 by the National Council on Measurement in Education 1