The effects of concept mapping and summarisation on L2 readers’ comprehension monitoring and metacognitive accuracy: A mixed-methods study

Toumi, Nour Elhouda and Kormos, Judit (2020) The effects of concept mapping and summarisation on L2 readers’ comprehension monitoring and metacognitive accuracy: A mixed-methods study. PhD thesis, Lancaster University.

Text (2020ToumiPhd)
2020toumiphd.pdf - Published Version
Restricted to Repository staff only until 20 May 2025.
Available under License Creative Commons Attribution-NonCommercial-NoDerivs.
Download (3MB)

Abstract

Successful reading comprehension plays a substantial role in the effectiveness of English Medium Instruction (EMI) in tertiary education. Students’ ability to accurately evaluate and regulate their comprehension during reading is central to their academic success. Evidence from the existing literature showed that some readers’ comprehension monitoring performance is inadequate when reading texts written in their second language (L2). Despite the growing interest in studying the meta-level processes in adult readers, there is a lack of converging empirical evidence on the most effective instructional programmes that could promote L2 readers’ comprehension monitoring and metacognitive accuracy. Using a mixed-methods intervention research design, this study examined the effects of two types of instructional approaches, concept-mapping and summarisation, on the comprehension and monitoring skills of adult readers enrolled in an EMI course in Algeria. This research also investigated how working memory (WM) capacity and vocabulary knowledge mediate the effects of these two interventions on learners’ reading comprehension, comprehension monitoring and metacognitive accuracy performances. Sixty-three EMI undergraduate students in Algeria were recruited for this study. Using a pre-test post-test non-equivalent group design, the participants were divided into three groups: the summary and the concept mapping-intervention groups and a control group. Participants’ reading comprehension, comprehension monitoring and metacognitive accuracy were assessed before and after the intervention. Vocabulary and WM measures were also obtained during the study. A combination of offline and online measures was used in the present research. An error detection task was used to assess participants’ comprehension monitoring performance. This task involved three measures of error detection: lexical/micro errors, textual contradictions (internal consistency) and contradictions between textual information and background knowledge (external consistency). The internal and the external consistency items together form the macro measure. Participants’ reading comprehension and vocabulary knowledge were measured with the Gates-MacGinitie reading comprehension (GMRT) and vocabulary tests. Based on the GMRT, participants were asked to provide metacognitive judgments on their comprehension performance in this test. The metacognitive accuracy measure involved two indices: absolute accuracy which was operationalised by calculating the absolute difference between participants’ actual comprehension performance and their judgments, and bias index which shows whether participants were under- or over-confident in judging their comprehension performance. WM capacity was measured with the use of an operation digit span task. Moreover, a technology-supported online instrument (eye tracker) was applied to stimulate participants to recall their thoughts during reading. The aim of this procedure was to investigate the mental processes involved during reading informational texts. The obtained verbal protocols were manually coded and analysed. The obtained quantitative data revealed a significant pre-test to post-test increase in comprehension monitoring, micro- and macro-error scores of the summary group. Participants who received the summary intervention were significantly more successful at detecting the lexical and semantic errors in the post-test compared to the pre-test. However, no statistically significant difference was found between this group and the concept mapping and the control groups in any of the aforementioned measures. Meanwhile, the concept mapping group showed significant improvement in reading comprehension, comprehension monitoring, detecting semantic errors and evaluating external consistency in the post-test compared to the pre-test. The results of the between groups ANOVA test showed a significant difference between concept mapping and the control group in the external consistency and bias index gain scores. Participants of the concept mapping group were significantly more successful at detecting the external errors and less biased in judging their comprehension performance than the control group. The effect of concept mapping on participants’ comprehension and monitoring skills was not mediated by their vocabulary knowledge or by their WM capacity. However, a significant correlation was found between the bias index gain scores and vocabulary knowledge of the participants in the summary condition. High vocabulary knowledge was associated with decreased bias in the post-test within this group. Participants’ verbal protocols reflected the use of 13 mental processes during reading, including careful local reading, careful global reading, skimming, search reading, word recognition, establishing propositional meaning, building a mental model, guessing word meanings from context, rereading, using background knowledge, evaluating spelling, evaluating internal consistency and evaluating the external consistency of the text. These processes were organised into the Local Global Situation Cognitive Metacognitive (LGSCM)-framework. This framework involves 13 reading individual processes and eight composite processes including, local comprehension, global comprehension, situation model, cognitive processes, metacognitive processes, cognitive local comprehension (CLC), metacognitive local comprehension (MLC) and metacognitive global comprehension (MGC) composite processes. The number of participants who engaged in establishing propositional meaning, building a mental model, global comprehension, cognitive and MGC processes has increased significantly from the pre-test to the post-test for the concept mapping group. Similarly, this group showed a significant increase in the use of evaluating internal consistency in the post-test. Nevertheless, the number of participants who used careful local reading in the post-test has decreased significantly against an increase in global reading for this group. However, participants who received the summary intervention showed increased use of evaluating external consistency in the post-test. The number of participants who engaged in situation model composite processes has also increased in this group. Unexpectedly, the number of participants using global comprehension composite processes has significantly increased from the pre-test to the post-test in the control condition. No statistically significant difference was found between the summary, concept mapping and the control groups in the use of these processes. The findings of this research have confirmed the effectiveness of using graphic organisers and self-evaluation tasks in enhancing L2 readers’ comprehension and monitoring skills.