Conflicting evidence from personal epistemology surveys: When students' words contradict their reported beliefs

This paper examines alignment of students' beliefs and attitudes towards knowledge and learning in mathematics.  Personal epistemology survey instruments are more convenient to administer than time-consuming interview protocols.  This is particularly true for large groups of participants or over multiple occasions in a repeated-measures design.  However, there is still a concern among the research community about the construct validity of these survey instruments (Hofer & Pintrich, 1997). This paper reports findings from a survey administered to mathematics students at various levels in a community college in the Pacific Northwest region of the United States.  The survey measured self-reported epistemic beliefs and included open-ended questions assessing attitudes toward the nature of effective learning in a mathematics classroom.  The purpose of this report is (1) to determine if students' self-reported beliefs about the authoritarian nature of knowledge align or misalign with their attitudes regarding effective teaching and learning, (2) to provide examples of alignment and misalignment, and (3) to assess the frequency of misaligned beliefs and statements.  A mixed-methods analysis was employed.  Confirmatory factor analysis and hierarchical clustering was used to assign students to low-, medium-, and high-availing beliefs regarding the authoritarian nature of knowledge.  Students' statements about effective teaching were coded for evidence of availing & nonavailing attitudes toward the nature of authoritarian knowledge.  Preliminary results demonstrate both concurrence and discordance between students' reported personal epistemology beliefs and attitudes about effective learning in all three belief groupings.  This investigation is important for academic motivation researchers because different instructional methods may be appropriate for students with well-aligned reported beliefs and attitudes versus students with misaligned beliefs and attitudes.  Furthermore, different learning strategies (e.g., deep vs. shallow) may be more appropriate based on either a student's reported epistemic belief or a misalignment between belief and attitude.