TEACHER BELIEFS AND BELIEF REPORTS: WHY THE DIFFERENCE REALLY MATTERS Alfredo Gaete Presented at EDULEARN13, Barcelona, July 1-3, 2013 Abstract The concept of belief has become one of the most important concepts in contemporary education. No wonder, since people’s beliefs are normally taken to be highly influential upon their behavior and, consequently, the last two or three decades have witnessed a wealth of studies concerned with identifying what teachers, student teachers, and teacher educators believe about a wide variety of issues. Most of these studies assume that it is relatively unproblematic to determine what subjects believe in virtue of what they say they believe. On this understanding, the procedures and instruments used to identify beliefs are based upon belief reports: interviews, Likert-type scales, questionnaires, etc. In this work I argue that the assumption in question is mistaken and, therefore, we better re- interpret recent educational research on beliefs as revealing information not necessarily about the beliefs of teachers and other educational agents but about the stories they tell about what they believe. I also suggest some alternative procedures for identifying beliefs that are not based upon belief reports. My point is that if we are interested in what educational agents really believe rather than in what they claim to believe, these alternative procedures seem far more appropriate or, in fact, valid. Keywords: Beliefs, teacher beliefs, belief reports. 1 INTRODUCTION The concept of belief has become one of the most important concepts in contemporary education (see [1], [2], [3], [4], [5], [6], [7]). There is nothing surprising about this. After all, people’s beliefs are normally taken to be highly influential upon their behavior. What the teacher does to teach, what the student does and is ready to learn, what the teacher educator talks about during her lecture, her very giving of a lecture rather than her doing of something else – all this depends, to some extent, on what these people believe about such things as teaching, learning, knowledge, the subject-matter in question, one’s own and others’ personal efficacy, the goals of education, and a myriad of other issues. Consequently, the last three decades have witnessed a wealth of studies concerned with identifying beliefs like these. Now in most of these studies it is assumed, often implicitly, that it is relatively unproblematic for one to get to know what a person believes by means of asking her to tell one what she believes. In words of Alexander and Docky [1], who have made this assumption explicit: “[W]e assumed that the responses that participants shared would be accurate reflections of their thoughts and views” (p. 416). On this understanding, the procedures and instruments that the great majority of the studies use to identify beliefs are based upon different sorts of belief reports: interviews, Likert-type scales, questionnaires, and so on (see e.g., [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18]). In this work I argue that the assumption in question is mistaken; more specifically, that under certain conditions – which commonly obtain in educational research – to ask people to say what their beliefs are is not the best way to know what they actually believe. Even though this strategy might work, there are, I contend, many (relevant) contexts in which it is a particularly feeble procedure. I also suggest where to look in order to find alternative procedures.