Actionable companion · Assessment & reasoning

Learning with agency: Validation of the Chinese version of the scientific imagination inventory in primary school students

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paper Search paper agent CC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Primary school, Children · stage unspecified. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Intensifying global competition in high technology and generative artificial intelligence (Gen AI) urgently calls for support to foster scientific imagination, which is essential for achieving breakthroughs in original innovation. As children enter primary school, their scientific imagination undergoes a rapid surge in development; however, few suitable tools are available to track and assess this growth. The Scientific Imagination Inventory (SII) demonstrated acceptable psychometric properties in samples of Korean students; however, its validity in other cultural contexts has remained unexplored. The present study was designed to evaluate the psychometric properties of the Chinese version of the SII, thereby addressing a critical gap in the availability of developmentally appropriate assessment instruments for primary school students in China. A total of 837 students in grades 2 to 6 were recruited from three private and public primary schools in Suzhou, China. Item analysis, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and criterion validity analysis were conducted. The results showed significant correlations between items and their respective dimension scores. EFA identified six factors that explained a cumulative variance of 55.99%. CFA supported a three-dimensional, six-factor structure (χ2/df = 1.553, RMSEA = 0.049, CFI = 0.906). Criterion validity was established through a significant correlation with the Vividness of Visual Imagery Questionnaire (VVIQ). Both internal consistency (α = 0.86) and split-half reliability (0.82) were satisfactory. The Chinese version of the SII showed adequate reliability and validity and is suitable for assessing the development of scientific imagination among primary school students in the sampled context. However, tests of measurement invariance indicated a lack of scalar invariance across gender and grade levels. Therefore, although the SII’s factor structure is generalizable to the Chinese context, our results underscore the critical importance of considering cultural and developmental response patterns when interpreting scores. Caution is advised against making direct mean comparisons across demographic groups.

What kind of evidence is this?

Supporting AI learning science · Survey / observational study

Stage: Primary school; Children · stage unspecified. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

Participants were 837 students recruited from three public and private elementary schools in Suzhou, China, using a combination of convenience and random sampling. This approach was chosen due to pragmatic constraints related to school access and administrative permissions, which are common in initial instrument validation studies. To partially mitigate sampling limitations and enhance the internal diversity of the sample, we deliberately selected schools of different types from various districts within Suzhou, aiming to capture a broad range of student backgrounds. A total of 850 questionnaires were distributed, yielding 837 valid responses after excluding invalid entries, corresponding to an effective response rate of 98.47%. The final sample comprised 837 participants, with 83 s graders (9.9%), 217 third graders (25.9%), 204 fourth graders (24.4%), 208 fifth graders (24.9%), and 125 sixth graders (14.9%). Of the total sample, 387 participants were female (46.2%) and 449 were male (53.6%). The entire sample was randomly divided into three subsamples.

Limitations and uncertainty · source passage

The present study has several limitations that should be considered when interpreting its findings and planning future research. First, the use of a convenience sample from a limited number of primary schools in Suzhou constrains both the diversity of the sample and the generalizability of the results. Although efforts were made to include schools from different districts and of different types, the findings primarily reflect an urban context in eastern China and may not extend to rural populations or other regional settings. Second, the psychometric evaluation revealed that certain items, although theoretically meaningful, contributed to suboptimal reliability indices. These items were retained to preserve content validity and theoretical integrity, but their impact on measurement precision warrants further investigation. The most notable methodological limitation pertains to measurement invariance: while configural and metric invariance were established across gender and grade groups, scalar invariance was not achieved. This indicates that, although the factor structure and factor loadings are comparable across groups, differences in item intercepts prevent statistically justified direct comparisons of the observed mean scores between these demographic categories. Consequently, the scale in its current form is not suitable for group-difference studies. Future research should recruit more representative and geographically diverse samples across China and investigate the causes of scalar non-invariance, such as cultural response patterns or developmental differences in item interpretation.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.

A 90-day district pilot

Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
Days 16–30 · Design. Pilot one teacher-led unit with short guided turns. The tutor asks for an explanation and offers a graduated hint before any worked solution. Use concrete examples and an equivalent task without AI. Name an accountable owner and set the baseline, success threshold and budget before implementation.
Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
Days 61–90 · Decide. Use an age-appropriate independent post-task and a delayed parallel task. Score explanation quality and transfer, not the number of agent messages or completion rate alone. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Duhong Peng; Junying Feng; Mingfeng Wang. Validation of the Chinese version of the scientific imagination inventory in primary school students. Frontiers in Psychology (2026-02-03). https://doi.org/10.3389/fpsyg.2026.1757878

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.