Individual profiles in online reading : relationships between default psychological stances, goal orientations, strategies, and performance

Abstract

With the advent of the digital-explosion age, the prevalence of information technology and the ease of accessing online resources have propelled the emergence of human-driven digital media and textual production. This has transcended traditional conceptions of reading and encompasses the ability to acquire knowledge, comprehend information, and meaningfully interact with technology. In response to these demands, the Progress in International Reading Literacy Study (PIRLS), conducted every five years since 2001, developed the e-PIRLS assessment in 2016. This assessment seeks to evaluate students' online reading ability by simulating web browsing experiences and motivating students to read texts, interact with them, and answer comprehension questions. The above trends show that the skill of reading online effectively has emerged as a necessary competency for students in the present era. However, scholars have found that despite being somewhat computer literate, students born in the 21st century still encounter significant challenges in online reading. These difficulties include, but are not limited to, becoming lost in the abundance of information and struggling to establish connections between multimodal content. Consequently, it is imperative to employ appropriate strategies and provide psychological support to overcome these challenges. That said, previous research has neither adequately investigated the influence of psychological factors on online multimodal reading nor explored the effects of strategies for objective online reading assessments, and in particular it has not taken individual differences into consideration. Therefore, this study, guided by the Cognitive Affective Engagement Model (CAEM), employed a person-centered approach to investigate 280 fourth-grade students' online multimodal reading in Hong Kong. Using latent profile analysis, this study first identified the participants' psychological default stance shaped by their online reading self-efficacy and emotions. Three distinct profiles emerged: positive (G1), moderate (G2), and ambivalent (G3). Second, multinomial logistic regression analyses were conducted to examine whether goal orientations for online reading tasks influenced the memberships of these three profiles. The results indicated that goal orientations significantly affected the classification of students in the G1 and G2 groups, with higher goal orientations increasing the likelihood of students being classified into G1. However, no significant effects were observed for G3. Third, using the Wald chi-square test, this study further explored differences in online reading strategy use and performance among students with varying psychological profiles and found that G3 students exhibited the highest confidence in their strategy application but achieved the lowest reading scores. Students in G1 reported greater strategy use than those in G2 did, which was also reflected in their superior online reading performance. Finally, mixture path analyses were conducted to examine how different psychological profiles moderated the relationship between students’ online reading strategies and their reading performance, and the results revealed that strategies significantly impacted only students in the G2 and G3 groups. Specifically, mastering the ability to manage disorientation significantly improved online reading performance for both the G2 and G3 students. The strategy of identifying the main idea was effective only for G3 students, while a trial-and-error approach proved beneficial exclusively for G2 students. Interviews provided additional insights into the differences among the three groups and explained how goal orientation influenced the formation of their profiles. To strengthen the findings, as triangulation sources for the mixture path analyses the study also utilized backend data (e.g., question-answering duration, clicks on multimodal elements) that were extracted from the testing platform, combined with interview responses. This study not only validates and extends the Cognitive Affective Engagement Model for multiple sources from a theoretical perspective, but it also provides empirical support for the critical role of psychological mechanisms and reading strategies in the complex context of online reading. Importantly, it highlights how individual differences influence the interactions among these variables. The findings offer valuable pedagogical insights for enhancing the students’ online reading performance, including how educators can help students cultivate more positive psychological profiles and which strategies should be emphasized for students with different psychological characteristics.