Growth mindset and academic outcomes: a comparison of US and Chinese students

This study investigates the seemingly paradoxical observation that while Chinese students often outperform their US counterparts academically, they tend to hold more fixed mindsets about intelligence compared to US students, who generally exhibit a growth mindset. Previous research, largely focused on Western contexts (WEIRD), has established a positive correlation between growth mindsets and academic success. However, this relationship may not be universal. Stevenson and colleagues' earlier work highlighted that Chinese and Japanese students attribute success more to effort than innate ability, a characteristic potentially linked to growth mindsets. This study challenges the assumption of a universal link between mindset and academic achievement by directly comparing the beliefs and academic performance of US and Chinese students. The researchers aim to understand how cultural differences in defining intelligence influence mindset beliefs and the relationship between mindsets, effort attributions, and academic performance. The authors hypothesize that while growth mindsets positively correlate with academic success in US students, this correlation might be weaker or even negative for Chinese students due to differing conceptualizations of intelligence and its relation to achievement. The study builds upon prior work by examining the nuanced ways culture shapes the understanding of intelligence (growth vs. fixed; fluid vs. crystallized) and how these conceptualizations affect educational outcomes. This research is crucial because it addresses the limitations of previous studies, which often neglected or misrepresented cross-cultural variations in intelligence mindsets and their impact on academic achievement.

Existing research reveals conflicting findings concerning the relationship between mindset and academic achievement, particularly across cultures. Studies have shown that Chinese and Japanese students tend to hold more fixed mindsets compared to their Western peers. However, this contradicts the observed academic success of these students, suggesting a cultural nuance in the interpretation of intelligence and its impact on motivation. Previous research has often focused on Western, Educated, Industrialized, Rich, and Democratic (WEIRD) societies and has not thoroughly examined how different cultural contexts shape the understanding of intelligence and its link to achievement. There's a need to move beyond a 'WEIRD' model and explore potential cultural variations in these relationships. Studies focusing on the cultural differences in how intelligence is defined – with Chinese individuals emphasizing aspects like reasoning and memory, while US individuals highlight problem-solving and social competence – suggest that the way intelligence is conceptualized might play a significant role in mindset development and academic outcomes. Prior research also indicates that students' beliefs about the malleability of intelligence are influenced by how intelligence is defined to them. Hence, understanding cultural differences in the definition of intelligence is critical to interpreting the relationship between mindset and academic performance. The current study directly addresses these gaps by exploring the cultural variations in the definition of intelligence and their relation to mindset beliefs and academic achievement across large samples of US and Chinese students.

The study employed a two-pronged approach, using both a large-scale dataset and a smaller, more in-depth investigation. **Study 1:** This study utilized data from the 2018 Programme for International Student Assessment (PISA) dataset, a large-scale international assessment of 15-year-old students' academic performance in mathematics, reading, and science. The sample included over 15,000 students from the US and China (specifically, Beijing, Jiangsu, Shanghai, and Guangdong provinces in China). Mathematics achievement was assessed using ten plausible values from the PISA mathematics test. Mindset of intelligence was measured using a single item from the questionnaire: "Your intelligence is something about you that you can't change very much." This item was rated on a four-point scale (1-Strongly disagree to 4-Strongly agree). Data analysis involved regression analyses using the R package 'intsvy', accounting for student demographics, school-level effects, and weight replicates. Effect sizes, confidence intervals, and visualizations were used to interpret the results. **Study 2:** This study involved a smaller sample of US (N=189) and Chinese (N=171) college students. Participants completed an online survey using Qualtrics. The survey comprised four blocks: 1) spontaneous definition of intelligence, 2) mindset of intelligence, 3) effort-ability attribution to academic achievement, and 4) demographic information. In Block 1, students provided their own definitions of intelligence, which were then coded into four categories (fluid only, crystallized only, both, neither). In Block 2, participants' malleability beliefs about intelligence were assessed with the Implicit Theories of Intelligence Questionnaire (ITIQ). In Block 3, effort-ability attributions for both expert and school-level mathematical achievements were assessed. Block 2 also included mindset items related to specifically defined fluid and crystallized intelligence, allowing researchers to explore how different conceptions of intelligence influence mindset beliefs. Analysis involved t-tests, correlations, ANOVA, and chi-square tests, comparing mindset beliefs, effort attributions, and spontaneous definitions of intelligence between the two cultural groups. Interrater reliability was assessed for the coding of open-ended responses, ensuring high levels of agreement among coders.

**Study 1:** The results revealed that US students endorsed a significantly more growth mindset than Chinese students (68.39% vs. 55.61% viewed intelligence as malleable). Regression analysis indicated that country and mindset significantly predicted mathematics scores, with a significant interaction effect. Pairwise comparisons showed a positive correlation between growth mindsets and mathematics scores for US students (medium-to-large effect sizes), but a slightly negative, albeit small, correlation for Chinese students. Visualizations (box plots) of the data reinforced these findings, clearly showing different score distributions across mindset levels for the two countries. **Study 2:** This study replicated Study 1's findings: Chinese students endorsed a less growth mindset for intelligence compared to US students (p<0.001). However, Chinese students showed significantly stronger effort-oriented attribution towards school-level achievement (p<0.001), but were more ability-oriented for expert-level achievement. Correlations between intelligence mindset and effort-ability attribution were significant in the US sample (r=0.22, p=0.003) but not in the Chinese sample (r=0.10, p=0.196). ANOVA revealed significant main effects for country and intelligence definition, and a significant interaction effect on mindset beliefs. Post-hoc analyses showed Chinese students consistently endorsing more fixed mindsets across all definitions (self-defined, fluid, and crystallized intelligence). Analysis of spontaneous intelligence definitions indicated cultural differences: Chinese students were more likely to include fluid intelligence, while US students favored crystallized intelligence in their descriptions. A chi-square test confirmed this difference. This suggests that the way students conceptualize intelligence is related to their mindset beliefs.

The findings challenge the assumption of a universal association between growth mindsets and academic success. The study demonstrates significant cross-cultural differences in how students conceptualize intelligence and how this influences their mindset beliefs and their relation to academic outcomes. US students’ mathematics performance is strongly associated with their beliefs about the malleability of intelligence, while this relationship is much weaker for Chinese students. This helps explain the apparent contradiction of Chinese students holding more fixed mindsets while still achieving high academic performance. The study proposes two key explanations: 1) Cultural differences in defining intelligence (US students tend to define intelligence as more crystallized, which might be perceived as more malleable; Chinese students emphasize fluid intelligence). 2) The link between growth mindset, motivational beliefs, and academic outcomes is stronger for US students than Chinese students. These findings suggest that interventions aiming to promote growth mindsets should consider cultural contexts and might benefit from targeting beliefs about the malleability of school performance rather than intelligence itself. The study highlights the importance of acknowledging cultural belief systems when studying motivation and academic achievement.

This research provides compelling evidence challenging the universality of the growth mindset theory. The study reveals significant cultural differences in the conceptualization of intelligence and its relationship with academic motivation and performance. The findings suggest that interventions should be culturally sensitive and focus on promoting effort-oriented attributions for school performance rather than solely targeting beliefs about the malleability of intelligence. Future research should explore other cultures and delve deeper into the nuances of how individuals conceptualize intelligence to gain a more complete understanding of the interplay between mindset, culture, and academic achievement. The limitations of focusing solely on the 'intelligence mindset' construct are highlighted, emphasizing the need for a more comprehensive approach that considers the specific aspects of mindset beliefs in relation to different cultural contexts.

While the study utilizes large sample sizes and employs rigorous statistical analyses, there are limitations to consider. Study 1 relies on a single item to measure mindset, limiting the depth of understanding. Study 2’s coding of open-ended responses, while achieving high inter-rater reliability, involves subjective interpretation. The college student sample in Study 2 may not fully represent the broader population, possibly limiting the generalizability of these findings. Future research could use more comprehensive measures of mindset, incorporate qualitative data to provide richer contextual understanding, and extend these analyses to more diverse cultural groups. The focus on mathematics might not fully capture the complexity of mindset and achievement across diverse academic domains.