Supporting Students' Ability to Argue From Evidence: Encountering Evidence in Text Versus Short Q&A Format

• In Event: Learning to Argue From Evidence

Abstract

The objective of the present work was to examine whether encountering evidence in the form of questions and answers (Q&A), versus text, is more effective in supporting students’ ability to construct evidence-based arguments. Cognitive load theory assumes working memory is limited, thus reducing cognitive load could provide an advantage and enhance motivation. Most important, the Q&A format makes salient the purpose of each piece of evidence, i.e., it potential function in the service of an argumentive claim.

Forty-one twelve-year old students (42 girls) from a public elementary school in Cyprus participated. Half of the students were assigned to the Q&A condition, and the other half to the text condition. Students in both conditions, working in pairs, engaged in a series of electronic dialogs, using instant messaging, with other pairs of peers and in some reflective activities based on transcriptions of the dialogues. The only difference between the two conditions is that students in the text condition encountered all the relevant information in the context of a traditional text, which they were asked to read in its entirety and was available to them thereafter, whereas students in the Q&A condition received the same information in the form of short pieces of Q&A text distributed to them across the sessions.

Students in both conditions were engaged in nine 90-minute-sessions following the curriculum implemented in earlier work (Kuhn, Hemberger, & Khait, 2016). The topic of the intervention was the best method for producing electricity, solar energy vs. natural gas. Students’ individual argument skills were assessed at both initial and final sessions by means of a written essay on the intervention topic.

Students’ essays were divided into idea units and coded based on their function. Students in the Q&A condition at the posttest used more functional evidence, that is the connection between evidence and claim was clear, in their essays, (M = 3.45, SD = 1.39) compared to students in the text condition (M = 2.05, SD = 1.49), t(39) = 3.097, p = .004. Significant differences were also observed regarding the function of evidence used. In particular, students in the Q&A condition employed more evidence both to support their own position (M = 1.95, SD = 1.05), t(39) = 2.237, p = .031, and to weaken the opposing position (M = 1.3, SD = .98), t(39) = 2.14, p = .038, compared with students in the text condition (M = 1.28, SD = .85 and M = .67, SD = .91, respectively).

Our results have important educational implications, suggesting that making available information in the Q&A format, which makes salient the purpose of each piece of evidence, distributed over time, compared to traditional text, is more effective in supporting students’ ability to construct evidence-based arguments.

Authors

• Kalypso Iordanou, University of Central Lancashire

• Deanna Kuhn, Columbia University