Dissecting a Many-Headed Beast: Exploring Students' Ability to Integrate Causal Strategies Into a Written Text

• In Event: International Perspectives on Writing in History: The Role of Language

Abstract

Background and significance. The explicit teaching of historical reasoning strategies has been the focus of many studies on historical reasoning (e.g. Nokes, Dole, & Hacker, 2007; Stoel, van Drie, & van Boxtel, 2016). In history, the ability to reason historically is assessed primarily through writing tasks based on the reading of multiple sources. However, a previous study (Stoel et al., 2016) on causal historical reasoning highlighted the complexity of such a writing task and raised the question whether students were able to integrate knowledge of causal strategies in a written account.

In the current qualitative study, we analyzed 11th-grade pre-university students’ revisions to a previously written essay, after a lesson-unit in which second-order concepts and causal strategies had been explicitly taught. We explored how the intervention influenced the argumentative quality of their texts―e.g. insertion of claims and conclusions, and working with evidence and historical sources (e.g. Van Drie & Van Boxtel, 2008). Furthermore, we explored how initial text-structure interacted with the types of revisions students made (Coffin, 2004).

Method. A sample of twenty essays was analyzed. The revised essays were compared to the original essays. Through open and axial coding, we defined three main revision categories: “causal elaboration,” “use of second-order concepts,” and “argumentation.” Each category was divided into several subcategories. Two raters categorized the text-structure of the essays as predominantly linear-chronological, or non-linear-thematic. We did not only categorize the revisions, but also explored the domain-specific quality of the changes.

Results. We found that students were able to revise their essays in ways that revealed an increase in conceptual knowledge about historical causation. Students were able to integrate descriptive and analytical second-order concepts that were part of the intervention (e.g. “trigger”, “catalyst”, “(in)direct”); elaborated the causal density by including concrete events and abstract phenomena, and insert argumentative elements that strengthened author-presence (e.g. causal claims). We also found that initial text-structure was related to the amount and type of revisions students made. On average, linear essays contained 14 revisions versus 23 revisions in non-linear essays. The non-linear texts integrated more analytical second-order concepts and causal claims. Furthermore, in the non-linear essays, conceptual elaborations were often developed as the central theme of the paragraph. This process of “nominalization” (e.g. Coffin, 2004) was witnessed less often in linear-chronological texts. Finally, results showed that although many revisions were coded as the insertion of a causal claim, these claims tended to lack argumentative support and almost no reference to evidence was made.

Conclusion. We concluded that knowledge of second-order concepts and strategies is an important building block of causal historical writing. However, if students are expected to progress in their ability to write causal explanations, explicit attention in history classes should also be given to knowledge of causal text-structures, strategies for working with historical evidence, providing arguments and counter-arguments.

Authors

• Gerhard Stoel, University of Amsterdam

• Carla Van Boxtel, University of Amsterdam