Development of Relational Reasoning Ability Among Chinese Elementary and Middle School Students

• In Event: 43.012 - Pushing the Boundaries of Relational Reasoning in Research and Practice: Cross-National, Cross-Domain, and Cross-Age Explorations

Abstract

Objective

This study aims to picture the developmental trajectory of relational reasoning (RR) in Chinese children and adolescents, a less studied population, and to investigate its relations with their mathematical ability development.

Theoretical Framework

Relational reasoning is a fundamental cognitive ability underlying human learning (Krawczyk, 2012) and has been empirically associated with academic performance in areas including mathematics (Richland et al., 2007) and science (Murphy et al., 2017). While prior research has shown developmental differences in children’s RR (Crone et al., 2009; Jablansky et al., 2015), such differences remain to be confirmed using psychometrically-sound measures. Further, RR development has been studied among western European populations (Jablansky, Alexander, Eilam, Aharon, & Sun, 2017). Little is known about RR development in Chinese children. Moreover, it is important to understand how children’s RR development corresponds to changes in their mathematical abilities over time.
In this study, the Test of Relational Reasoning-Junior (TORRjr), a figural measure of children’s ability to reason with four forms of relations (analogy, anomaly, antinomy, antithesis), was translated into Chinese and administered to 3rd through 7th graders. Their RR performance was then examined in relation to their demonstrated mathematical abilities.

Method

Data were collected from over 900 third to seventh graders from two regions of China. The TORRjr results for 38 classes in south China (n=740) have been analyzed so far. Four of the third to sixth grade classes (n=173) also completed the Mathematics Ability Test (MAT), which consists of 12 released items from the TIMSS 2011 eighth-grade Mathematics Concepts and Mathematics Items covering of number, algebra, and geometry. The specific items were chosen based on item difficulties (one easy, two moderately difficult, one difficult) for Hong Kong students. All students completed the two tasks on iPads in their classrooms during normal school hours under the supervision of their teacher and a trained research assistant.

Results

When examined by grade level and by scale (Table 1), a series of ANOVAs indicated that RR as measured by TORRjr increased rather linearly over the grades (Figure 1). Also, while there was significant overall improvement for each form, their developmental trajectories varied to some degree (Figure 2). Specifically, both analogy and anomaly appeared to develop rather early, whereas antinomy and, especially antithesis appeared later. This was reinforced by the performance data indicating that students overall tended to perform best on the analogy and anomaly scales, and worst on the antithesis scale (Figure 3). Moreover, multiple group regression analysis revealed that RR significantly predicted mathematics scores for all grades (Table 2). However, RR was more closely related with mathematical ability at the higher grade level (Table 3).

Significance

This study represents the first attempt to plot the development of RR of Chinese children using a formal assessment, TORRjr. In addition, RR’s developmental trajectory was linked to changes in children’s mathematical ability.

Authors

• Hongyang Zhao, University of Maryland - College Park

• Yuting Sun, University of Maryland - College Park