Statistical Discourse Analysis: Contextual Effects of Algebra Students’ Actions on Microcreativity Processes

• In Event: Connecting the Moving Dots: Comparing Approaches to Coordinating Temporal Analyses of Groups and Individuals

Abstract

This study tests ecological hypotheses about processes over time via statistical discourse analysis (SDA, Chiu, 2008, Chiu & Khoo, 2005). Sequences of actions may affect the likelihoods of multiple processes, even after accounting for explanatory variables at multiple levels (conversation turn, time period, individual, group, classroom, school …). Specifically, this study tests whether social metacognition (especially disagreements) might influence micro-creative processes. See figure 1.

Theory
Group members' micro-creative processes (creating new ideas and assessing their utility via justifications) might serve as the building blocks of successful outcomes (Chiu, 2008a). Social metacognition is group members' monitoring and control of one another's knowledge, emotions, and actions (Chiu & Kuo, 2009). Social metacognition can aid group problem solving through evaluation of one another's ideas and identification of problems, especially through disagreements (Chiu & Kuo, 2009).

According to socio-cognitive conflict theory, group members can recognize problems, express them through disagreements, which stimulate justifications that support creation of new ideas (Buchs, et al., 2004). However, politeness theory suggests that rude and polite disagreements have different effects (Brown & Levinson, 1987). Rude disagreements (“You’re wrong. It’s fifty”) can threaten face, escalate interpersonal conflict, and often hinder justifications and creation of new ideas. On the other hand, a polite disagreement (“If we multiply five by ten, don’t we get fifty?”) reduces blame and creates common ground to support interpersonal relationships. In response to a polite disagreement, a person is less likely to retaliate, more likely to try to understand the criticism and more likely to justify his original suggestion or recognize the flaw and correct it with a new idea (Chiu & Khoo, 2003).

Method
84 students answered a survey and worked in groups of four on an algebra problem during class. Informed by students’ written work, video transcripts were coded along the dimensions in table 1.

SDA addresses the analytical difficulties of modeling group processes over time. See table 2. When the effect of a group (or individual) process on another process(es) (or outcome[s]) differs across contexts (micro-time contexts, time periods, people, groups, classrooms, etc.), SDA enables computation of each effect in each specific context.

Results and Discussion
Statistical discourse analysis (SDA) of students’ 3,296 conversation turns showed that when group members agreed, rudely disagreed or had lower peer status, they created fewer new ideas. After a command, new ideas and justifications were less likely. After actions by members with higher mathematics grades, justifications were more likely, though this effect differed significantly across time periods and across groups.
The effects of rude disagreements differed across groups and across micro-time contexts. In groups that did not solve the problem, rude disagreements showed much stronger negative effects on new ideas. After a wrong idea, a rude disagreement sharply increased the likelihood of a new idea. By combining the regression coefficients, the effect of rude disagreement in each specific context can be computed. Future work can extend SDA to mixed category outcomes (continuous, ordered or binary).

Author

• Ming M. Chiu, University at Buffalo - SUNY