One Step, Two Step, Three Step, Four: Evaluating a Cognitive Model of Integration

• In Event: 45.024 - Integration: A Critical Competency for the Digital Age

Abstract

Objectives
The ability to draw on diverse sources of information to form coherent understanding is referred to as integration. Given its importance for learning in the digital age (Goldman & Scardamalia, 2013), a variety of studies have examined integration, including considerations of its manifestation during think-alouds (Wolfe & Goldman, 2005) and in the written responses that students compose (Wiley & Voss, 1999). Nevertheless, missing in these analyses has been a consideration of the cognitive processes underlying integration.

Theoretical Frame
Recently, List and Alexander (in press) proposed a four-step model of integration. Specifically, they suggest that integration occurs through students’: (1) becoming aware of a possible relation across texts (Step 1: Identification), (2) separately conceptualizing relation-relevant information within each text (Step 2: Separate Representation), (3) connecting both texts via a single statement (Step 3: Simultaneous Relation), and (4) specifying and explaining the relation among them (e.g., agreement/disagreement; Step 4: Relational Designation). In this study, we provide empirical evidence for this theorized four-step process.

Methods
Undergraduates (N=33) were asked to complete two multiple text tasks. First, students were asked to form an argument about why Stonehenge was constructed, based on conflicting explanations presented across two texts. Following reading, participants were asked to complete an evidence sort task, or to classify evidence introduced uniquely within one or across both texts, according to its origin. Second, participants were asked to form an argument about whether or not sex work should be legalized, based on conflicting points of view introduced across eight texts. Following reading, participants were asked to create a diagram to visually represent all eight texts and the relations among them. See Figure 1.
During reading, students were asked to highlight any relations they identified across texts and to explain them. Students’ relational explanations were coded according to the step in List and Alexander’s four-step process they reflected. See Table 1.

Results
Participants attempted to formulate an average of 4.61 (SD=4.21) relations across the Stonehenge texts and 12.06 (SD=8.50) relations across the sex work texts. The number of relations formed was significantly associated across the two text sets, r(33)=0.41, p<0.01. Between 34.42% and 57.14% of the relations that students attempted to form, across the sex work and Stonehenge texts, respectively, only reflected the first step of the four-step process. As a contrast, between 8.44% (Stonehenge texts) and 22.08% (sex work texts) of relations formed reflected full integration. Students’ formation of complete, four-step relations was associated with both their performance on the evidence sort task and with the number of cross-textual connections they formed when constructing relational diagrams based on multiple texts.

Scholarly Significance
This study provides empirical evidence for a proposed four-step process of integration. Consistent with prior work, students were found to be limited in both the quantity and quality of integration. At the same time, when fully integrating two or more texts, they were found to do so according to the four-step process proposed. This suggests that this process may be used to scaffold students’ multiple text integration in the classroom.

Authors

• Alexandra List, The Pennsylvania State University

• Hongcui Du, Florida State University

• Hye Yeon Lee, The Pennsylvania State University