| |
what was studied
This study examined the impact of paired discussion of a topic (here, capital punishment) on the quality of reasoning about that topic. Two age levels were examined: early adolescents and young adults. Paired discussion for roughly 10 minutes per week for five weeks (with different peers who held a range of different opinions) was compared to single-occasion paired discussion and a control condition limited to repeated elicitation of the participant's own opinions and arguments.
how effectiveness was measured
Students' pretest and posttest opinions were assessed using a 13-point opinion scale that was designed to minimize extreme responses (i.e., ceiling effects). The quality of students' argumentative essays was assessed using a topic-dependent hierarchical coding scheme that distinguished between functional arguments (arguments that addressed the function or purpose of capital punishment), nonfunctional arguments (arguments that focused in the conditions that made capital punishment justified without consideration of its function), and nonjustificatory arguments (such as appeals to sentiment or authority). Functional arguments were broken down into those that addressed alternatives to capital punishment and those that did not. The number of references to evidence (or acknowledgement of its necessity) and metacognitive statements (reflective statements) contained in each essay were also coded and counted. Each student's pretest arguments (constructed before the instructional intervention) was then compared to his or her posttest argument (constructed after the intervention).
|
|
| |
what the findings were
Twice as many adolescents exhibited opinion change than adults; yet, for both age groups, paired discussion significantly enhanced quality of reasoning relative to the two other conditions. The range of different arguments increased from pretest to posttest, suggesting that a process of social transmission of new knowledge occurred; however, ten different types of improvement in reasoning also appeared, including a shift from 1-sided non-comparative arguments to 2-sided comparative arguments and the appearance of metacognitive statements indicating a greater awareness of multiple perspectives. Analysis of discussion transcripts evidenced two surprising results. First, in roughly half of the cases in which arguments improved, new elements incorporated into the posttest argument first appeared during discussion between individuals who held congruent rather than discrepant opinions. Second, in roughly half the cases, these new elements were first introduced into the discussion by the individual who later incorporated them into his or her posttest essay rather than by his or her discussion partner. |
|
