Investigating the validity of a survey intended to gauge mathematics problem-solving disposition along the impulsive-analytic dimension

Miguel Mendoza, University of Texas at El Paso

Abstract

This study investigates the validity of the Likelihood-to-Act (LtA) survey developed to assess respondents' problem-solving disposition along the impulsive-analytic dimension. Sixteen pre-service teachers were strategically selected from a pool of 495 LtA survey respondents to participate in a semi-structured task-based interview approximately one hour in length. The interview consisted of three main parts: a problem solving section, a consistency section, and a classification section. (1) Part 1 of the interview was developed to get interviewees to solve five open-response problems meant to elicit an impulsive response. Interviewees were also asked to verbalize their thought process in order for the researchers to gain some insight into their problem-solving disposition. Four of the five problems were extracted directly from the LtA survey and one additional unrelated problem was added because it cleverly elicits an impulsive response. (2) Part 2 of the interview involved having interviewees retake 10 of the LtA items and also had them explain any interview responses that varied from the original response by two or more points. (3) In Part 3, each interviewee classified the problem solving act described in each of 10 LtA items (different from those previously seen in Part 2) as being representative of either an impulsive or analytic disposition. A moderately high correlation of 0.693 (p = 0.004) was found between the problem solving data in Part 1 and the original LtA survey scores. In addition, two-scale analysis showed that 60.2% of the corresponding original survey responses matched the problem solving disposition scores assigned by the rubric-scoring team. Reliability issues arose in the consistency section where 63 out of 150 responses to LtA items during the interview were inconsistent with classroom responses to the same LtA items. Survey-taking context issues and other hidden variables surfaced as participants were prompted to explain inconsistent and consistent responses.

Subject Area

Mathematics education|Educational tests & measurements|Mathematics

Recommended Citation

Mendoza, Miguel, "Investigating the validity of a survey intended to gauge mathematics problem-solving disposition along the impulsive-analytic dimension" (2013). ETD Collection for University of Texas, El Paso. AAI1551237.
https://scholarworks.utep.edu/dissertations/AAI1551237

Share

COinS