A new perspective on memorization practices among East Asian students based on PISA 2012

Abstract

This study examined learning strategy use in mathematics among East Asian students in East Asian educational systems. By employing latent class analysis on the Programme for International Student Assessment (PISA) 2012 data, we found four classes of learning strategy types, namely memorization with metacognitive strategies (17.49%), metacognitive strategies with memorization (50.70%), elaboration only (10.33%), and metacognitive strategies with elaboration (16.47%). The results showed that the majority of the students in all seven East Asian educational systems belonged to the ‘metacognitive strategies with memorization’ class, and most students adopted more than one type of learning strategy when learning mathematics. Additionally, students who reported the use of metacognitive strategies along with either memorization or elaboration showed higher mathematics achievement. We conclude that the cognitive processes employed by students of East Asian backgrounds are more complex and nuanced than the previous perception that they relied heavily on memorization.

Keywords:

• Learning strategies
• mathematics
• PISA
• latent class analysis

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

Notes

1 The model selection was based on BIC, a_BIC, LMR as well as theoretical interpretations. For Japan and Macau, BIC, a_BIC and LMR had consistent results, indicating a 3-class model as the most appropriate solution. Additionally, based on interpretations, the 3-class model was the final model in Japan, Macau. In Singapore, BIC and a_BIC were lowest in the 3-class model; however, LMR had the significant p-value at the level .05 in the 4-class model, suggesting that the 4-class was favoured. Given that two other statistical criteria preferred the 3-class model and the interpretations were meaningful in the 3-class model, the 3-class model was the final model in Singapore. In Taiwan, BIC and a_BIC were lowest in the 4-class model; LMR had the significant p-value at the level .05 in the 2-class model, which indicated the 2-class was favoured. Given that two other statistical criteria preferred the 4-class model and the interpretations were meaningful in the 4-class model, the 4-class model was the final model in Taiwan. In Hong Kong, BIC was lowest in the 2-class model; a_BIC was lowest in the 3-class model; LMR had the significant p-value at the level .05 in the 2-class model, which indicated the 2-class was favoured. Given that two other statistical criteria preferred the 2-class model and the interpretations were meaningful in the 2-class model, the 2-class model was the final model in Hong Kong. In South Korea, BIC was lowest in the 2-class model; a_BIC was lowest in the 3-class model; LMR had the significant p-value at the level .05 in the 3-class model, which indicated the 3-class was favoured. Given that two other statistical criteria preferred the 3-class model and the interpretations were meaningful in the 3-class model, the 3-class model was the final model in South Korea. In Shanghai, BIC was lowest in the 3-class model; a_BIC was lowest in the 4-class model; LMR had the significant p-value at the level .05 in the 2-class model, which indicated the 2-class was favoured. Although there were inconsistent statistical results, the 3-class model was the final model in Shanghai due to the interpretations of classes.