Seven Questions about Creativity and Creative Thinking

6. Gender differences in creative thinking: are girls really better than boys at thinking creatively?

One of the most striking and consistent findings described in the PISA 2022 report was that boys did not perform better than girls, on average, in any of the 64 countries and economies that participated in the creative thinking assessment (OECD, 2024[1]). In fact, girls scored nearly 3 points higher than boys on average in the PISA creative thinking scale across OECD countries (a large performance difference). Even after accounting for students’ performance in reading, girls performed significantly better than boys with similar reading scores in around half of all countries and economies. These performance differences in favour of girls persisted across all types of task in the PISA test, including those targeting different ideation processes and applications in different domain contexts.

Gender differences in creative thinking may be rooted in broader cognitive, psychological and behavioural patterns. Prior research on creativity has found that girls tend to outperform boys in specific creative tasks, especially those involving associative thinking and elaboration (Baer and Kaufman, 2008[2]). While boys tend to perform better in divergent thinking tasks (Kazemian et al., 2024[3]; Baer and Kaufman, 2005[4]; Kim, 2006[5]), girls tend to score higher on originality and verbal creativity assessments. Socio-cultural norms and practices may contribute to these findings by influencing the development of boys’ and girls’ creative self-concept and reinforcing gendered cognitive styles and problem-solving strategies (Kim, 2006[5]; Bem, 1981[6]). Neuroimaging studies have also found gender differences in cognition: men and women activate different brain regions when solving creative tasks — particularly those requiring divergent thinking — suggesting the existence of gender-specific neural pathways (Abraham et al., 2013[7]). Other studies suggest that male advantage in creativity is influenced by personality traits such as openness to experience, assertiveness, and risk-taking (Feist, 1998[8]; Byrnes, Miller and Schafer, 1999[9]), although the PISA 2022 data found that 15-year-old girls generally reported higher levels of creative self-efficacy (especially in expressive or artistic tasks), and significantly stronger attitudes typically associated with supporting creative work including openness to experience and imagination and adventurousness (OECD, 2024[1]).

Overall, prior research on gender differences in creativity has found mixed results, with various factors including cognitive, motivational, affective and socio-cultural norms at play. One reason for these mixed findings in the research may be due to developmental differences between boys and girls across these factors, with periods of change and stability occurring at different ages (Slobodskaya, 2021[10]; Lau and Cheung, 2010[11]; He, 2018[12]). While the PISA 2022 results offered some hypotheses for why such strong gender differences in performance were observed in the PISA creative thinking test – namely differences in test engagement between boys and girls, and girls’ propensity to hold more positive attitudes and beliefs related to creativity – another contributing factor may have been the PISA scoring method itself. The PISA 2022 creative thinking scoring approach prioritised identifying sufficiently different or original responses rather than evaluating the holistic creative quality of students’ ideas. Will more granular scoring methods, such as those applied in the PISA CT Rescoring project (see Box 6.1), offer different insights into gender differences in creative thinking performance?

Box 6.1. Which score data are used in this chapter?

The analyses in this chapter draw upon two types of rescored data:

• Standardised holistic creativity scores: Each student response was scored based on judges’ evaluations of the overall creative quality of the response. The holistic creativity score reflects a relative ranking of creative quality compared to all other responses in that country-language-task group. To account for differences in the distribution of scores across country-language-task groups, the raw holistic creativity scores (range 1-7 score points) were standardised and transformed to a scale that ranges from 1-10 score points, with a mean of 5 score points.

• Criteria scores: Each idea within a student response was scored for its appropriateness, originality and value. A cumulative idea sum score was also produced by aggregating the scores across these three criteria. Tasks classified as ‘generate diverse ideas’ tasks required students to submit multiple ideas per response; for these tasks, each student response was also scored for flexibility (i.e. the number of different ideas in the response).

The analyses in this chapter focus on gender differences in scores resulting from these two methods. For more information on the scoring methods, see Annex A2.

6.1. Do gender differences in creative thinking persist across different scoring methods and across country-language contexts?

Across all country-language groups included in the PISA CT Rescoring study, girls achieved a higher mean standardised holistic creativity score than boys. Comparing the scores of boys and girls, girls scored 5.1 points on average whereas boys scored 4.9 points – a modest but statistically significant difference (0.2 points) in performance (Figure 6.1). The difference in mean score between boys and girls was also statistically significant in 13 of the 15 country-language groups included in the study, with the largest gaps observed in Slovenia (0.6 points), Greece (0.4 points), Saudi Arabia, Korea, Canada (both English and French provinces) and Hong Kong (China) (all 0.3 points). In Slovenia, this equates to over one third of a standard deviation in creative thinking performance between boys and girls – see Box 6.2. Only in Chile and Colombia was there no statistically significant difference in the performance of girls and boys in terms of their standardised holistic creativity scores. The PISA 2022 results also found no statistical differences in the performance of boys and girls in creative thinking in Chile, and a significant but below average difference in creative thinking performance in favour of girls in Colombia (OECD, 2024[1]).

Box 6.2. How big of a score difference?

The standardised holistic creativity scores in each country-language-task group were centred around a mean of 5 score points, with a standard deviation of 1.5 score points. This means that a standardised holistic score difference of 0.5 score points between boys and girls equates to a third of a standard deviation in performance (Table B6.2).

As for the criteria-based scores used in this chapter, the standard deviation in students’ appropriateness scores was 0.8 score points on average across country-language groups. For the originality and value criteria, the standard deviation in students’ scores were 0.9 score points, respectively, on average across country-language groups. For all three criteria, a score difference of 0.3 score points equates to around a third of standard deviation in performance, on average – although scores in some country-language groups varied more than others (Table B6.3).

Comparing gender differences in the PISA CT Rescoring project with gender differences in PISA 2022

In general, gender differences in performance observed in the PISA CT Rescoring project are similar to or smaller than those observed in the PISA 2022 results for both creative thinking and reading (OECD, 2023[13]; OECD, 2024[1]). Table 6.1 summarises the gender effect size in each test context.

Table 6.1. Gender effect sizes in creative thinking and reading

Gender differences expressed as a proportion of the standard deviation in performance (all students)

	PISA CT Rescoring project	PISA 2022 Creative Thinking	PISA 2022 Reading
Belgium (French community)	0.1	0.2*	0.3*
Brazil	0.1	0.2	0.2
Canada (English provinces)	0.2	0.2*	0.2*
Canada (French provinces)	0.2
Chile	0.0	0.1	0.1
Colombia	0.0	0.1	0.1
Germany	0.2	0.3	0.2
France	0.1	0.2	0.2
Greece	0.2	0.3	0.3
Hong Kong (China)	0.2	0.4	0.2
Italy	0.1	0.2	0.2
Korea	0.2	0.3	0.3
Saudi Arabia	0.2	0.5	0.4
Portugal	0.2	0.2	0.2
Slovenia	0.4	0.4	0.5

Note: *Gender effect sizes for the PISA 2022 creative thinking assessment and reading assessment are computed at the country level for Canada and Belgium, not at the country-language group level. Effect sizes are rounded to the nearest decimal point.

Source: Table B6.4

Figure 6.1. In no country-language group do boys significantly outscore girls on average

Mean standardised holistic creativity scores of boys and girls (all tasks), by country-language group

Note: Significant differences between boys and girls are shown by filled markers, and values indicating the mean score difference have been rounded to the nearest decimal point. Standardised holistic creativity scores have a mean of 5 score points for each country-language-task group. Country-language groups are ranked in descending order of the difference in mean standardised holistic creativity scores between boys and girls. Mean standardised holistic creativity scores are computed within each country-language group on the pooled responses across all tasks.

Source: Table B6.1.

Looking at the top 25% and bottom 25% of performers across country-language-task groups also reveals significant gender differences. Figure 6.2 shows that, on average, 54% of students within the top quartile of performers are girls compared to 46% boys. In Brazil, over 20 percentage point more girls than boys are top performers in the tasks included in the study, and in Greece, Canada (French provinces), France, Portugal and Belgium (French community), the gender difference is over 10 percentage points in favour of girls (Table B6.5). However, in nine country-language groups, there are no significant differences between boys and girls within the top 25% of performers across all tasks, suggesting that the performance of boys and girls is more variable towards the lower end of the score scale. Indeed in the bottom quartile, the gender differences are much starker: close to 56% of students in the bottom 25% of performers are boys, on average across country-language groups, compared to just 44% of girls. In Slovenia, nearly 65% of the lowest quartile of performers are boys (a gender difference of around 30 percentage points), and in Saudi Arabia, the figure is 63% (a difference of 26 percentage points).

Figure 6.2. There are relatively more girls than boys amongst top performers, and relatively many more boys than girls amongst low performers

Percentage of boys and girls within the top/bottom 25% of standardised holistic creativity scores per task (pooled responses across tasks), on average across country-language groups

Note: Differences in the share of boys and girls amongst responses within the bottom/top quartile per task (pooled across tasks) are statistically significant. The standardised holistic creativity scores were jittered, i.e. a random small digit (1e-6) was added to each score, so that equal quartiles in each country-language-task group could be produced. Values in the figure have been rounded to the nearest integer.

Source: Table B6.5

A second scoring method applied in the PISA CT Rescoring study involved scoring different dimensions of students’ ideas separately (see Box 6.2). These dimensions included idea appropriateness, originality and value. Additionally, judges scored each response to ‘generate diverse ideas’ tasks for idea flexibility (i.e. how different were the multiple ideas provided by students). Our results show that girls scored higher than boys, on average, in all three idea score criteria (appropriateness, originality and value) to a small but significant degree, representing a difference of around one-tenth of the mean standard deviation in each of the score criteria (Table B6.6). While most country-language groups observed significant gender differences in at least one of the idea score criteria, there was substantial variation across country-language groups (see Box 6.3). Gender differences in the mean idea sum scores of boys and girls were more consistent across country-language groups, with significant results observed in all but four country-language groups (Brazil, Canada [English provinces], Chile and Colombia) and an average performance difference of 0.3 score points (or around one-third of the mean standard deviation in performance) in favour of girls. As for response flexibility, girls also showed a small but significant score advantage compared to boys, on average. Most notably, when considering the gender differences observed across the different criteria-based scores collectively, boys did not score significantly higher than girls in any of the idea or response score criteria, on average, in any of the country-language groups in the study.

Box 6.3. Cross-cultural variation in gender differences across the different dimensions of creativity

Criteria-based scores for each idea within a response

Appropriateness

Girls scored significantly higher than boys in idea appropriateness in 10 of the 15 country-language groups. Although significant, these differences generally remain small (around 0.1 score points on average), with the largest differences observed in Slovenia, Saudi Arabia and Korea (difference of 0.2 score points). No significant differences in this score criteria were observed in Belgium (French community), Brazil, Chile, Colombia and France.

Originality

Fewer gender differences across country-language groups were observed in the mean idea originality of boys and girls compared to the other score criteria. While girls scored 0.1 score points more than boys (a small but significant difference) on average across country-language groups, a similarly significant relationship was observed in only six country-language groups: Belgium (French community), Canada (French provinces), Saudi Arabia, Italy, Portugal and Slovenia. In Slovenia, the mean score point difference between boys and girls was close to 0.2 points in favour of girls.

Value

Around half of the country-language groups showed small and significant differences in mean idea value scores between boys and girls. These differences were significant and around 0.1 score points in Belgium (French community), France, Greece, Korea, Saudi Arabia and Portugal, and largest in Slovenia and Germany (around 0.2 score points).

Idea sum scores

Much larger score differences were observed between boys and girls with respect to their mean idea sum scores (i.e. their aggregated appropriateness, originality and value scores). These differences were significant in 11 of the 15 country-language groups, and on average across country-language groups, girls scored around 0.3 score points higher than boys overall. In some cases, differences in the mean idea sum scores of boys and girls are as large as 0.5 score points in Germany and Korea, 0.6 score points in Saudi Arabia, or even 0.7 score points in Slovenia.

Criteria-based scores for each response

Flexibility

Around half of the country-language groups also showed small, significant differences in mean response flexibility scores between boys and girls. Girls in Belgium (French community), France, Greece, Korea, Saudi Arabia and Portugal scored around 0.1 score points more than boys in idea flexibility, on average, and in Slovenia and Germany, they scored on average 0.2 score points more than boys.

Note: Mean idea and response scores are computed within each country-language on the pooled responses across all tasks.

Source: Table B6.6.

Taken together, the analyses using data from the two scoring methods indicate the persistence of gender differences in creative thinking on the PISA test. It seems that no matter how creative performance is evaluated, girls consistently and significantly outperform boys on average across most country-language groups – even if differences in some of the scores are generally small in magnitude. Other factors might explain some of this apparent gap in performance, for example differences in test engagement between boys and girls (Box 6.4), but the PISA CT Rescoring study data confirms that the gender differences observed in the PISA 2022 results were not simply an artefact of the PISA 2022 scoring method.

Box 6.4. Gender differences in engagement with the PISA 2022 creative thinking test

In the context of the PISA 2022 creative thinking assessment, task engagement emerged as a critical factor that may contribute to explaining some of the large observed gender differences in performance. The PISA report showed that boys displayed higher levels of disengaged behaviours than girls across a variety of task engagement indicators — a trend that was consistent across most PISA participating countries and economies (OECD, 2024[1]). In particular, boys were far more likely to engage in “rapid responding” behaviours compared to girls, on average. Boys also showed higher levels of disengagement than girls with all types of task in the assessment, on average, although gender differences in task engagement were relatively smaller in the scientific problem-solving domain than in the other three task domains. These differences in task engagement would non-discriminately affect the scores of boys and girls in any scoring method applied to the PISA creative thinking data.

Task engagement is not necessarily construct-irrelevant in the context of creative thinking: confluence approaches of creativity underline that engagement is an important factor that enables individuals to produce creative work. However, whether these differences in task engagement in the PISA creative thinking test would translate to similar differences outside of a testing environment remains an open question. For example, one study found boys tend to exhibit more disengaged behaviours in low-stake test environments (Leng and von Davier, 2025[14]). The gender differences observed in the PISA 2022 results and in this report should therefore be interpreted within this context.

6.2. Are gender differences in creativity more pronounced in certain domains?

While gender differences in creative thinking skill generally appear consistent across the diverse national-linguistic contexts in the PISA CT Rescoring study, is this also the case when examining the performance of boys and girls across the different types of task included in the study? Prior research and reviews on gender differences in creativity measures have found mixed results (Baer and Kaufman, 2008[2]; Alabbasi et al., 2025[15]; Taylor et al., 2024[16]) with many studies using relatively general or abstract divergent thinking tasks and few studies focused on exploring gender differences using domain-specific creativity measures. Some such studies have found no significant differences in the performance of men and women in figural (visual) and creative writing tasks (Taylor and Barbot, 2021[17]), or small performance advantages for girls in storytelling and figural tasks (Ivcevic et al., 2022[18]).

The data from the PISA CT Rescoring study show that, on average across country-language groups, girls score significantly higher than boys in their standardised holistic creativity scores in the written expression, visual expression and social problem-solving tasks included in the study. In these three task domains, girls scored around 0.3 score points higher than boys (around one-fifth of a standard deviation), on average. However, the score gap is markedly smaller for tasks in the scientific problem-solving domain (difference of 0.1 score points on average) (Table B6.7). Overall, these domain-level gender differences in holistic creativity scores are generally consistent with the strengths and weakness of boys and girls reported in the PISA 2022 results (OECD, 2024[1]).

A more nuanced picture emerges when examining the mean scores of girls and boys by domain context within each country-language group. Figure 6.3 shows that girls tend to score significantly better than boys in the written expression and social problem-solving tasks in over half of the country-language groups participating in the study, respectively, compared to only four country-language groups in the single visual expression task included in the study. Interestingly, in the four country-language groups where girls do score significantly better than boys in the visual task (Canada [French provinces], Greece, Korea and Slovenia), the performance difference is large (between a third and a half of the mean standard deviation in performance) – and typically more so than most differences observed in the written or social problem-solving tasks across country-language groups. While the PISA 2022 results found a more consistent difference in the performance of boys and girls in the visual expression items on the PISA test, the fact that the PISA CT Rescoring project included only one visual item may explain this discrepancy.

For tasks in the scientific problem-solving domain, we observe a different general pattern in gender differences in student scores. In most country-language groups, and on average, there is no significant difference in the scores of boys and girls in these tasks: only in Saudi Arabia and Canada (English provinces) do girls still score significantly higher than boys in their standardised holistic creativity scores in scientific problem-solving. The opposite is true in Chile, where boys score significantly better than girls in scientific problem-solving – the only instance in our study where boys significantly outperform girls in a subset of tasks.

One reason for the different results in the scientific problem-solving tasks may be gendered differences in engagement with these items, relative to the other tasks in the PISA test. As described in Box 6.4, gender differences in task engagement persisted across all task types in the PISA creative thinking test, although they were relatively smaller in the scientific problem-solving tasks. It is not that boys are more engaged with scientific problem-solving tasks specifically – in fact, both boys and girls exhibited higher levels of relatively rapid responding and non-responding behaviours to the scientific problem-solving tasks compared to the other task types – but rather, girls show higher levels of relative disengagement than boys with the scientific problem-solving tasks compared to all other tasks (OECD, 2024[1]).

The relative disengagement by girls with creative scientific problem-solving tasks may stem from gender biases in the perception and attribution of creativity across different domains. Research has found that individuals typically consider some creative domains to be more suitable for women, such as crafts and literature, while other domains are more suitable for men, including science and engineering (Taylor and Barbot, 2023[19]). The same study also found that stereotypically “masculine-agentic” traits (e.g. adventurous, ambition, competitive, courageous, daring, decisive, self-reliant and willing to take risks) were rated as significantly more important for creativity in the science domain than stereotypically “feminine-communal” traits (e.g. cooperative, helpful to others, nurturing, sensitive, supportive, sympathetic, understanding of others, and warm in relations with others); in contrast, there were no significant differences in the perceived importance of the two groups of traits in the literature or crafts domains, despite gender biases in attributions of creativity in favour of women in these two domains. In other words, regardless of the gender of the individual, not only does creativity in scientific contexts tend to be viewed as a male domain, but stereotypically male traits are considered particularly beneficial for creativity in scientific contexts.

Figure 6.3. Girls significantly outscore boys in tasks in all domains, on average, and in most country-language groups

Mean difference in standardised holistic creativity scores between boys and girls, by task domain and by country-language group

Note: Significant values are shown in colours of a darker shade for boys/girls, respectively. Country-language groups are ranked in descending order of the mean difference in standardised holistic creativity score (highest to lowest). Mean standardised holistic creativity scores of boys and girls are computed within each country-language on the pooled responses across relevant tasks in each domain.

Source: Table B6.7

These more general gender biases related to creativity in science mirror previous PISA findings that have found that girls report lower science self-efficacy than boys (OECD, 2019[20]). Are these findings also replicated in students’ creative self-efficacy beliefs, specifically girls’ and boys’ attitudes towards engaging in creative tasks in different domains? Figure 6.4 shows the mean difference in girls’ and boys’ self-reported creative self-efficacy in select tasks, by country-language group. On average, girls report feeling more confident than boys at being creative, in general, with large differences (around 10 percentage points or more) observed in Italy, Saudi Arabia and France. Girls also tend to score higher in the overall index of creative self-efficacy, on average across country-language groups (Table B6.8). When it comes to telling creative stories, the gender differences in favour of girls are particularly pronounced, with over 5 percentage point more girls than boys agreeing with the statement on average, and over 10 percentage point more girls than boys agreeing in Saudi Arabia, Slovenia, Greece and Germany. Yet when asked about their confidence in thinking creatively in science and invention contexts, girls are notably less confident – and even less confident than boys, on average. For example, many more boys than girls feel confident that they can think of many good ideas for science experiments (difference of 10 percentage points on average), and more boys than girls also report feeling confident to invent new things (difference of around 6 percentage points on average). In around half of all country-language groups, gender differences in creative self-efficacy related to thinking of good science experiments are significant and relatively large in favour of boys, with the difference around or greater than 15 percentage points in Greece, Hong Kong (China), Belgium (French community) and France, and over 20 percentage points in Portugal and Korea. Despite overall creative performance and creative self-efficacy than boys, girls nonetheless seem far less likely than boys to feel confident about being creative in scientific or engineering-type contexts. These findings therefore align with previous research on gender biases in how individuals perceive creative work and creative individuals in different domains (Taylor and Barbot, 2023[19]).

Figure 6.4. Girls are more confident than boys at being creative, in general, but report far less confidence than boys in creative scientific or invention tasks

Mean difference in the percentage of boys and girls that reported feeling confident or very confident with select items from the creative self-efficacy index, by country-language group

Note: Significant differences are shown in colours of a darker shade for each statement. Country-language groups are ranked in descending order of the mean difference in percentage of boys and girls (girl - boy) that reported feeling confident or very confident with each statement (highest to lowest).

Source: Table B6.8.

6.3. Do the data support the greater male variability hypothesis in creative thinking?

One hypothesis often explored in educational psychology is the Greater Male Variability Hypothesis (GMVH), which suggests that boys exhibit greater variability than girls in general across a range of performance contexts. Do our data confirm this hypothesis in the context of creative thinking? Results from our study show that, on average across country-language groups, there was marginally more variation in the standardised holistic creativity scores amongst boys than girls, with the largest differences in score variability observed in Saudi Arabia and Korea (Table B6.2). However, in most country-language groups, differences in the variation between boys and girls tended not to be statistically significant.

The GMVH suggests that boys are more likely to be represented at both high and low score extremes. Figure 6.5 shows the share of girls and boys amongst students within the 90th percentile of standardised holistic creativity scores in each task within each country-language group. In most country-language groups and on average, the share of girls amongst students within the 90^th percentile, represented by the proportion of the bar in green, exceeds the share of boys amongst students within the 90^th percentile, represented by the proportion of the bar in blue. In nine country-language groups, these gender differences are statistically significant. In Brazil, Greece and France, the gap in the share of girls and boys amongst the top 10% of performers in tasks is between 20-22 percentage points, and in Slovenia this difference reaches 29 percentage points.

Figure 6.5. On average, girls are overrepresented amongst the top 10% of performers in creative thinking tasks

Percentage of boys and girls amongst students in the 90^th percentile of standardised holistic creativity scores per task (pooled responses across tasks), by country-language group

Note: Significant differences are shown in colours of a darker shade for boys/girls, and values are rounded to the nearest integer. Country-language groups are ranked in ascending order of the percentage of girls (lowest to highest). The standardised holistic creativity scores were jittered, i.e. a random small digit (1e-6) was added to each score, so that equal deciles in each country-language-task group could be produced. The country-language average is computed amongst all responses within the top decile per task (pooled across tasks).

Source: Table B6.9

When examining the same amongst the lowest 10% of performers on tasks (i.e. 10^th percentile), boys are significantly and consistently overrepresented – with 59% of students amongst the lowest 10% of scorers across tasks being boys. Boys are particularly overrepresented in this group in Saudi Arabia (71% students), Slovenia (68% students), Korea (66% students), Hong Kong (China) (66% students) and Germany (65% students) (Table B6.9). The only country-language group where girls are statistically significantly overrepresented amongst the lowest performers in tasks is Brazil, where 54% are girls.

Figure 6.6. On average, boys are overrepresented amongst the lowest 10% of performers in creative thinking tasks

Percentage of boys and girls amongst students in the 10^th percentile of standardised holistic creativity scores per task (pooled responses across tasks), by country-language group

Note: Significant differences are shown in colours of a darker shade for boys/girls, and values are rounded to the nearest integer. Country-language groups are ranked in ascending order of the percentage of boys (lowest to highest). The standardised holistic creativity scores were jittered, i.e. a random small digit (1e-6) was added to each score, so that equal deciles in each country-language-task group could be produced. The country-language average is computed amongst all responses within the bottom decile per task (pooled across tasks).

Source: Table B6.9.

The findings in this section provide limited support for the greater male variability hypothesis in creative thinking – though it could be argued that rather than boys exhibiting a significantly greater variation in performance than girls per se, the distribution of boys and girls differs mostly in that there is a higher incidence of low performers among boys than there are high performers. Again, these findings may be explained as much by gender differences in engagement with the PISA test (Box 6.4) as they are by the GMVH.

Alternatively, the overrepresentation of boys at the lower end of the holistic creativity scores may reflect broader gender differences in engagement with creative tasks, which have been argued to stem from certain notions of masculinity that run counter to the tacit attribution and promotion of many creative domains as a feminine pursuit (Taylor and Barbot, 2023[19]; Scholes and Nagel, 2010[21]). The OECD’s project on fostering critical and creative thinking examined the role of emotions while learning, and offers practitioners several ideas for differentiated engagement strategies that might help to support boys’ interest in and motivation to engage in creative thinking (OECD, 2022[22]).

Key findings and implications

• Girls consistently score higher than boys in the PISA creative thinking tasks.

  • This pattern was observed consistently across different country-language groups, task types and scoring methods, suggesting it is a relative stable finding that is not driven by any single scoring method, task feature, or sample anomaly.

  • However, the gender effect size observed in the PISA CT Rescoring data tended to be similar to or smaller than the equivalent gender effect size observed in both the PISA 2022 creative thinking and reading results. It may be that the more granular holistic scoring method used in the Rescoring study mitigates gender differences to some extent, for example by rewarding more subtle differences in idea quality and/or by being less driven by differences in test engagement.

• The only domain context in which girls did not show a significant and moderate performance advantage over boys in creative thinking is in scientific problem solving – although girls still significantly outperformed boys in the scientific problem-solving tasks, on average across country-language groups.

  • These results may rather reflect a relative weakness for girls in creative thinking, rather than a particular strength for boys. Indeed, data from the PISA 2022 assessment revealed that girls showed higher levels of relative disengagement than boys with the scientific problem-solving tasks compared to all other tasks.

  • Differences between boys and girls in creative thinking performance and task engagement in the scientific problem-solving tasks may reflect gendered attitudes and beliefs about creativity in science (Taylor and Barbot, 2023[19]). Girls reported greater levels of creative self-efficacy in general, but notably not in tasks in scientific or invention contexts.

  • Initiatives focused on supporting girls’ participation and confidence in Science, Technology, Engineering and Mathematics (STEM) education may help to redress the relative weakness of girls in creative thinking in scientific contexts.

• Boys show slightly more variation in creative thinking performance than girls, but boys are more overrepresented amongst low performers than girls are overrepresented amongst high performers.

  • Efforts should focus on raising the performance of boys at the lower end of the creative thinking score distribution. Initiatives that help to encourage boys’ engagement with creative tasks may be a productive strategy for practitioners (OECD, 2022[22]).

References

[7] Abraham, A. et al. (2013), “Gender differences in creative thinking: behavioral and fMRI findings”, Brain Imaging and Behavior, Vol. 8/1, pp. 39-51, https://doi.org/10.1007/s11682-013-9241-4.

[15] Alabbasi, A. et al. (2025), “Gender differences in creative potential: A meta-analysis of mean differences and variability”, Psychology of Aesthetics, Creativity and the Arts, Vol. 19/1, pp. 87-100, https://doi.org/10.1037/aca0000506.

[2] Baer, J. and J. Kaufman (2008), “Gender differences in creativity”, The Journal of Creative Behavior, Vol. 42/2, pp. 75-105, https://doi.org/10.1002/j.2162-6057.2008.tb01289.x.

[4] Baer, J. and J. Kaufman (2005), “Bridging generality and specificity: The amusement park theoretical (APT) model of creativity”, Roeper Review, Vol. 27/3, pp. 158-163, https://doi.org/10.1080/02783190509554310.

[6] Bem, S. (1981), “Gender schema theory: A cognitive account of sex typing.”, Psychological Review, Vol. 88/4, pp. 354-364, https://doi.org/10.1037/0033-295x.88.4.354.

[9] Byrnes, J., D. Miller and W. Schafer (1999), “Gender differences in risk taking: A meta-analysis”, Psychological Bulletin, Vol. 125/3, pp. 367-383, https://doi.org/10.1037/0033-2909.125.3.367.

[8] Feist, G. (1998), “A meta-analysis of personality in scientific and artistic creativity”, Personality and Social Psychology Review, Vol. 2/4, pp. 290-309, https://doi.org/10.1207/s15327957pspr0204_5.

[12] He, W. (2018), “A 4-year longitudinal study of the sex-creativity relationship in childhood, adolescence, and emerging adulthood: Findings of mean and variability analyses”, Frontiers in Psychology, Vol. 9, https://doi.org/10.3389/fpsyg.2018.02331.

[18] Ivcevic, Z. et al. (2022), “Gender and creative ability: Mean differences and”, Thinking Skills and Creativity, Vol. 46, pp. 1-12, https://doi.org/10.1016/j.tsc.2022.101186.

[3] Kazemian, N. et al. (2024), “Some effects of sex and culture on creativity, no effect of incubation”, Empirical Studies of the Arts, Vol. 42/2, pp. 536-559, https://doi.org/10.1177/02762374231217638.

[5] Kim, K. (2006), “Can we trust creativity tests? A review of the Torrance Tests of Creative Thinking (TTCT)”, Creativity Research Journal, Vol. 18/1, pp. 3-14, https://doi.org/10.1207/s15326934crj1801_2.

[11] Lau, S. and P. Cheung (2010), “Developmental trends of creativity: What twists of turn do boys and girls take at different grades?”, Creativity Research Journal, Vol. 22/3, pp. 329-336, https://doi.org/10.1080/10400419.2010.503543.

[14] Leng, D. and M. von Davier (2025), “Gender differences and types of test-taking behaviors in PIRLS 2021”, Applied Measurement in Education, Vol. 38/1, pp. 75-93, https://doi.org/10.1080/08957347.2025.2486032.

[1] OECD (2024), PISA 2022 Results (Volume III): Creative Minds, Creative Schools, PISA, OECD Publishing, Paris, https://doi.org/10.1787/765ee8c2-en.

[13] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/53f23881-en.

[22] OECD (2022), How do girls and boys feel when developing creativity and critical thinking? Lessons for gender equity, OECD Education Spotlights, OECD Publishing.

[20] OECD (2019), Why don’t more girls choose to pursue a career in science?, PISA in Focus, 93, OECD Publishing.

[21] Scholes, L. and M. Nagel (2010), “Engaging the creative arts to meet the needs of twenty-first-century boys”, International Journal of Inclusive Education, Vol. 16/10, pp. 969-984, https://doi.org/10.1080/13603116.2010.538863.

[10] Slobodskaya, H. (2021), “Personality development from early childhood through adolescence”, Personality and Individual Differences, Vol. 172, pp. 1-7, https://doi.org/10.1016/j.paid.2020.110596.

[19] Taylor, C. and B. Barbot (2023), “Gender biases in attributions of creativity across domains”, Creativity Research Journal, Vol. 37/3, pp. 395-405, https://doi.org/10.1080/10400419.2023.2285546.

[17] Taylor, C. and B. Barbot (2021), “Gender differences in creativity: Examining the greater male variability hypothesis in different domains and tasks”, Personality and Individual Differences, Vol. 174, pp. 1-9, https://doi.org/10.1016/j.paid.2021.110661.

[16] Taylor, C. et al. (2024), “Gender differences and variability in creative ability: A systematic review and meta-analysis of the greater male variability hypothesis in creativity”, Journal of Personality and Social Psychology, Vol. 126/6, pp. 1161-1179, https://doi.org/10.1037/pspp0000484.