Analysis of discourse practices in statistical computing laboratory

Document Type

Conference Proceeding

Publication Date



Social interaction, Modelling, Questioning, Cognitive structuring


Classroom talk plays a significant role in a teaching and learning process. The author of this paper was concerned with the way student-student and teacher-student discourses promotes students’ statistical thinking in the computing laboratory. An observation study was thus conducted to examine how social interactions among students in small groups and the teacher’s intervention were organised in the laboratory. In the observation study, verbal speech and peer interaction were audiotaped. The various discourses were transcribed in full, with relevant excerpts being selected for analysis within a Vygotskian framework. The student-student discourse was analysed with the aid of Kumpulainen’s framework. It was found that the high use of the organizational, judgmental as well as external thinking functions in the student talk when presenting the chart title and axis labels; approving or disapproving group mate’s work; and monitoring data entry actively during keyboard input respectively.

The teacher-student discourse was analysed with the aid of Tharp and Gallimore’s framework. Questioning was used frequently by the teacher to discover what the students knew, understood, or misunderstood, and to offer direction towards deeper thinking at times when students could not otherwise make progress on tasks. The teacher also used modelling assistance to help students see how to assemble and organise pieces of knowledge they might have already grasped or possessed. Contingency management was another form of assistance exhibited in the form of praise that affirmed the quality of students’ work. The teacher offered cognitive structuring that assisted them to organise and justify various aspects of a regression modelling problem but left room for students to regulate their strategies based on their own creation and interpretation of a regression model they found was the best.

Source Publication

The 2nd 21st Century Academic Forum Conference, Harvard University, USA

Volume Number


Issue Number




First Page


Last Page


This document is currently not available here.