Seven Questions about Creativity and Creative Thinking

Introduction

Every three years, a sample of 15-year-old students around the world sits an assessment, known as PISA (Programme for International Student Assessment), that aims to measure how well their education system has prepared them for life after compulsory schooling. In 2022, for the first time, the PISA assessment included a test of creative thinking that was administered in 64 participating countries and economies. As part of PISA 2022, students also completed a background questionnaire about themselves, their homes and schools, and their experiences at school.

The PISA 2022 report on creative thinking was released in June 2024 and summarised the international results from the assessment (OECD, 2024[1]). The results provide a comparison of the extent to which 15-year-old students in each participating country and economy can engage in the process of creative thinking to generate diverse ideas or to generate and improve upon original ideas. Key findings from the report include that a significant gender gap exists in creative thinking (girls consistently outperformed boys); that proficiency in reading, mathematics and science supports performance in creative thinking but, beyond a baseline proficiency in these core literacies, all students can excel in creative thinking; and that performance differences associated with socio-economic status persist in creative thinking, despite an overall weaker association than the one observed between student background and performance in the core domains. Some countries and economies also performed better than expected in the creative thinking assessment, with the report identifying features of these high-performing systems.

While many national centres and governments try to ensure that educators and educational researchers get constructive feedback based on PISA results, most of the key messages published in the PISA reports don’t make it back to those working to support students in classrooms. What’s more, understanding how creative ideas develop and manifest has been the subject of several decades of research – but the PISA 2022 assessment represents the first large-scale, internationally comparable data collection of its kind. These data can shed new light on how successful 15-year-olds are at creative thinking around the world, the different ways that creative success can be measured, the variation in performance observed across student groups, and whether there are cross-cultural differences in how creativity is understood and evaluated.

Data and results in this report

This report analyses a sample of the data that was collected during the PISA 2022 Creative Thinking (CT) assessment. These data – students’ raw responses to a subset of the PISA 2022 CT items (see Insert) – were rescored as part of a research project (the “PISA CT Rescoring project”) conducted by the OECD and collaborating research teams in 16 country-language groups to further explore the rich information in students’ responses to the PISA test. The main goals of the PISA CT Rescoring project were twofold:

1. To investigate the feasibility of applying different scoring methods to the PISA 2022 data across diverse socio-cultural contexts (see Box A);

2. To analyse the rescored data sample and examine the extent to which key findings from the PISA 2022 report are replicated across different scoring methods.

The three scoring methods applied in the PISA CT Rescoring project conceptualise and prioritise creative value in different ways to each other and to the international scoring rules applied in the PISA 2022 CT assessment. In the PISA 2022 CT assessment, responses to creative thinking items were awarded full credit, partial credit (in some cases), or no credit. The scoring rules, described in the PISA 2022 CT assessment framework (OECD, 2023[2]), focused on evaluating students’ capacity to generate “sufficiently” original or different ideas (depending on the item type) across contexts – or in other words, distinguishing sufficiently creative from non-creative responses. This approach was adopted to maximise the scalability and reliability of the human coding effort in each country/economy and to facilitate cross-cultural comparability by focusing on criteria that could be evaluated most objectively across diverse contexts (Foster and Piacentini, 2025[3]; Foster and Piacentini, forthcoming[4]). For example, statistical infrequency was adopted as a proxy for the originality of student responses. However, the trade-off of the PISA 2022 CT scoring approach was that it could not identify the most creative responses within each participating country/economy, nor did it evaluate the overall “creative value” of students’ responses. In contrast, the alternative scoring methods applied in the PISA CT Rescoring project do provide different ways for the most creative responses in the sample to be identified. A first report using data from the PISA CT Rescoring project exploring students’ ideas on the test was released earlier this year (OECD, 2025[5]).

Box A. PISA CT Rescoring project: Data and scoring approaches used in the study

Data and item sample

The PISA CT Rescoring project data sample included 37,849 unique responses from 14,830 students that sat the PISA 2022 CT test, sampled from 14 participating countries and economies (16 country-language groups) (Annex A1). 7 items from the full 32 test-item pool were included in the study (see Insert). A maximum of 1000 students from each country-language group were randomly sampled, with at least 300 students sampled per item (where possible).1 Over 100 judges were involved in the rescoring effort, led by a team of experienced researchers in the field of creativity measurement in each of the participating country-language jurisdictions.

Scoring methods applied in the study

The three scoring methods applied to the PISA CT Rescoring project data sample were:

• The Holistic Judgement method: Valid task responses were given a single “creativity” score (score between 1-7 points), considering all relevant elements of their task response. The goal of this scoring method was to establish a relative ranking of responses within each country-language-task group.

• The Criteria-Based method: Each idea within a student task response was scored for its appropriateness (score between 0-2 points), originality and value (scores between 0-3 points, each). A cumulative idea sum score was also produced for each idea by aggregating the scores across these three criteria. In addition, responses to tasks requiring multiple ideas were scored for idea flexibility (one score per task response, between 0-2 points).

• The Conceptual Knowledge Tree method: Each idea within a student task response was mapped onto a branch and node of a conceptual knowledge tree that defines all possible solution path ideas in a problem space (Hatchuel and Weil, 2009[6]; Agogué et al., 2014[7]). This method was applied to two problem-solving tasks only (Food Waste and The Exhibit tasks – see Insert).

More information about the three scoring methods and the coding process that judges in each participating country-language group followed can be found in Annex A2. Information about the inter-rater reliability tests and results for each of the three methods can also be found in Annex A2.

Content of this report

This report analyses the results of the PISA CT Rescoring project and organises the findings into answers to 7 key questions about creativity and creative thinking (listed below). The report explores how these insights from the data might help educators and educational researchers to support the development of students’ creative thinking.

The questions can be grouped into three main topic sections:

1. Understanding creativity and creative thinking processes;

2. Unpacking cross-cultural differences in creativity and creative thinking; and

3. Variation in creative thinking performance across student groups.

Section 1. Understanding creativity and creative thinking processes

Question 1. What makes “creative ideas” creative?

Question 2. Divergent and convergent thinking: which tasks led to more creative ideas on the PISA test?

Question 3. Is creative thinking domain-specific, a more general skill – or both?

Section 2. Unpacking cross-cultural differences in creativity and creative thinking

Question 4. Are there cross-cultural differences in creative thinking?

Question 5. Thinking outside or inside the box? Patterns in idea flexibility and fixations in problem-solving tasks

Section 3. Variation in creative thinking performance across student groups

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

Question 7. Does socio-economic disadvantaged hinder students’ creativity?

PISA 2022 tasks included in the PISA CT Rescoring Study

PISA 2022 student model: ideation processes and domain contexts

The PISA creative thinking construct was broken down into three ideation processes for measurement purposes in the PISA 2022 test (Figure A). These three ideation processes encompass both divergent-exploratory and convergent-integrative thinking processes and were classified as follows: (1) ‘generate diverse ideas’; (2) ‘generate creative ideas’; and (3) ‘evaluate and improve ideas’. Every task on the PISA 2022 creative thinking test corresponded to one of these ideation processes.

Figure A. PISA 2022 creative thinking competency model

Ideation processes and domain contexts targeted in the PISA 2022 CT test

Source: (OECD, 2023[2])

The PISA 2022 CT test also situated tasks in four different domain contexts. Two contexts – written expression and visual expression – focused on students’ capacity to express their ideas creatively, whereas the other two contexts – social problem solving and scientific problem solving – focused on students’ capacity to apply creative thinking for a more functional purpose.

Every task in the test required students to submit a constructed response answer, either in the form of a written response (for the written expression and problem-solving tasks) or a graphic design (for the visual expression tasks).

Tasks included in the PISA CT Rescoring project

A balanced subset of 7 tasks from the wider 32 item pool were chosen for the PISA CT Rescoring study. These 7 tasks were chosen considering the following factors:

• Representation of the four domain contexts;

• Representation of divergent-exploratory and convergent-integrative thinking processes;

• Task position in the PISA 2022 test forms (to maximise the number of students in the sample that sat multiple items included in the study);

• Tasks already released into the public domain following the administration of PISA 2022.

Table A describes the balance of tasks included in the PISA CT Rescoring study by ideation process and domain context. Note that while no items classified as ‘evaluate and improve ideas’ items were included in the PISA CT Rescoring project, these tasks were similar in format and task demands to the ‘generate creative ideas’ items.

Table A. Balance of tasks included in the PISA CT Rescoring project

By ideation process and domain context

Tasks	Written expression	Visual expression	Social problem-solving	Scientific problem-solving	Total
Generate diverse ideas	1	0	1	2	4
Generate creative ideas	1	1	1	0	3
Total	2	1	2	2	7

Note: ‘Generate diverse ideas’ tasks required students to submit two or three different ideas. ‘Generate creative ideas’ tasks required students to submit only one original idea.

Each of the 7 tasks included in the PISA CT Rescoring study are shown below. For an in-depth exploration of genuine student responses to these 7 tasks, including some of the most creative responses in the project sample, see OECD (2025[5]).

Task 1 – Science Fair Poster

• Ideation process: Generate creative ideas

• Domain context: Visual expression

In the first task of the Science Fair Poster unit, students were asked to create an original poster for a school science fair that focused on the theme of “Life in Deep Space” (Figure B). The digital tool provided to students allowed them to drag and drop a selection of existing stamps and shapes in the workspace, as well as manipulate the colours of shapes and lines. Students were also allowed to provide a brief description of their posters to help judges interpret the meaning behind their designs.

Figure B. Science Fair Poster task

Note: This task is the first of two tasks in the Science Fair Poster unit.

Source: https://www.oecd.org/en/about/programmes/pisa/pisa-2022-creative-thinking-test-questions.html

Task 2 – Illustration Titles

• Ideation process: Generate diverse ideas

• Domain context: Written expression

For the second task in the Illustration Titles unit, students were asked to suggest creative title ideas for an abstract illustration showing a large book in a fantasy countryside landscape (Figure C). Students were instructed to think of three title ideas that were as different as possible to each other.

Figure C. Illustration Titles task

Note: This task is the second of two tasks in the Illustration Titles unit.

Source: https://www.oecd.org/en/about/programmes/pisa/pisa-2022-creative-thinking-test-questions.html

Task 3 – 2983

• Ideation process: Generate creative ideas

• Domain context: Written expression

In the written expression task 2983, students were asked to think of story idea for a book whose cover displays only the 4-digit number 2983 (Figure D). Students were asked to briefly describe what the story could be about and to focus on making their story idea as original as possible.

Figure D. 2983 task

Note: This task was the only task in the 2983 unit.

Source: https://www.oecd.org/en/about/programmes/pisa/pisa-2022-creative-thinking-test-questions.html

Task 4 – Save the Bees

• Ideation process: Generate creative ideas

• Domain context: Social problem solving

In the Save the Bees unit, students are told that the “Save the Bees” club at their school are trying to increase awareness about the important role that bees play in local ecosystems. In the first task of the unit, students were asked to suggest three different ideas for how the club could raise awareness about the importance of bees in their school. In this second task (Figure E), students were asked to suggest one original solution for achieving the same goal. Students were given the choice of providing a completely new idea or choosing one of the ideas they came up with in the previous task.

Figure E. Save the Bees task

Note: This task is the second of three tasks in the Save the Bees unit. Students could write a new response or choose to submit one of their ideas from the previous task.

Source: https://www.oecd.org/en/about/programmes/pisa/pisa-2022-creative-thinking-test-questions.html

Task 5 – Food Waste

• Ideation process: Generate diverse ideas

• Domain context: Social problem solving

In the Food Waste unit, students were presented with the problem scenario that supermarkets across the world waste tons of unsold fresh food every day. Students were then asked to suggest three different ways that supermarkets could reduce the amount of fresh food that they throw away at the end of the day (Figure F).

Figure F. Food Waste task

Note: This task was the only task in the Food Waste unit.

Source: PISA 2022 Creative Thinking test.

Task 6 – Save the River

• Ideation process: Generate diverse ideas

• Domain context: Scientific problem solving

The unit Save the River consisted of two tasks, and in this first task, students were asked to provide two different testable ideas that might explain a declining frog population in one part of a river flowing out of a city (Figure G). Students were explicitly asked to think of causes other than pollution from local farms and factories to encourage them to think of alternative hypotheses. As such, any ideas related to pollution were considered to have low appropriateness for this task.

Unlike the other divergent thinking tasks included in the PISA CT Rescoring project that ask students to propose three different ideas, students are only asked to provide two different ideas in this task.

Figure G. Save the River task

Note: This task is the first of two tasks in the Save the River unit.

Source: https://www.oecd.org/en/about/programmes/pisa/pisa-2022-creative-thinking-test-questions.html

Task 7 – The Exhibit

• Ideation process: Generate diverse ideas

• Domain context: Scientific problem solving

In The Exhibit task, students were asked to propose design ideas for constructing a separating wall in an animal park exhibit that would allow squirrels to pass from one room to the other while preventing the rats from doing so (Figure H). The item specified that while the squirrels and the rats had the same body size, squirrels are heavier, faster and more agile than the rats.

Figure H. The Exhibit task

Note: This task was the only task in the ‘The Exhibit’ unit.

Source: PISA 2022 Creative Thinking test.

References

[7] Agogué, M. et al. (2014), “The impact of type of examples on originality: Explaining fixation and stimulation effects”, The Journal of Creative Behavior, Vol. 48/1, pp. 1-12.

[3] Foster, N. and M. Piacentini (2025), “Assessing creative thinking at scale and across culture”, in Katz-Buonincontro, J. and T. Kettler (eds.), Oxford Handbook of Creativity and Education, Oxford Handbooks, Oxford University Press.

[4] Foster, N. and M. Piacentini (forthcoming), “PISA 2022 Creative Thinking: What it is, what it is not, and what next?”, Journal of Creative Behavior, https://doi.org/10.1002/jocb.70068.

[6] Hatchuel, A. and B. Weil (2009), “CK design theory: An advanced formulation”, Research in Engineering Design, Vol. 19, pp. 181-192.

[5] OECD (2025), Creative minds in action: Students’ imagination and ideas in storytelling, design and problem-solving tasks on the PISA test, OECD Education Policy Perspectives, No. 120, OECD Publishing, https://doi.org/10.1787/5b718eb2-en.

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

[2] OECD (2023), PISA 2022 Creative Thinking Framework, PISA, OECD Publishing, https://doi.org/10.1787/dfe0bf9c-en.

Note

← 1.  In Belgium (French community) and Hong Kong (China, English speaking), minimum sample thresholds were not met due to the smaller sample size in that country-language group. In these cases, the maximum number of students was sampled. Given the very small sample size of the Hong Kong (China, English speaking) dataset, these data have been merged with the Hong Kong (China, Chinese speaking) dataset for most of the analyses conducted in this report. For more information on the PISA CT Rescoring project samples, see Annex A1.