Actionable companion · Assessment and evidence of learning
From evidence to district action: Can artificial intelligence read our handwriting? A comparison of humans and artificial intelligence in the future of assessment
AI-assisted editorial implementation guide. The district actions are proposed applications, not findings or recommendations attributed to the original authors. No additional peer review is asserted.
Evidence and context
Confirm the study population, setting and task before applying its findings to your district.
Original authors’ abstract, reproduced under the linked source license:
The purpose of this study is to evaluate the legibility of fourth-grade elementary school students’ handwriting and to examine the differences and relationships between two evaluators: a human researcher and artificial intelligence (ChatGPT). To this end, the study was conducted using the descriptive survey model, one of the quantitative research approaches. The study sample consisted of 67 fourth-grade students attending a public elementary school. The students’ handwriting was evaluated using the Multidimensional Legibility Scale. Writing samples were collected from each student, including copying samples, dictated samples, and free writing samples. One class period was allocated for each writing task. The 201 handwriting samples obtained from the 67 students were evaluated by a researcher (the second author) and artificial intelligence (ChatGPT). According to the findings, no significant differences were detected between the artificial intelligence and the researcher in any writing type or subdimension, except for the slant subdimension of free writing. However, a high level of positive and significant correlation was identified between the two raters. Findings indicate that artificial intelligence can produce results comparable to those of expert raters when evaluating handwriting legibility. The fact that differences among raters are not statistically significant suggests that AI-based handwriting evaluations could serve as an objective and reliable alternative.
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. Redesign one low-stakes assessment: require a first attempt, an AI-assisted revision, a source-check log and a short oral explanation. Use the same rubric for the comparison task. 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. Blind-score reasoning, factual accuracy and unaided transfer on a fresh task. Record suspected misconduct separately from demonstrated evidence. 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
İsmail Yaşartürk; Halil İbrahim Öksüz; Ramazan Deniz; Mustafa Öztürk. Can artificial intelligence read our handwriting? A comparison of humans and artificial intelligence in the future of assessment. PLOS One (2026-09-21). https://doi.org/10.1371/journal.pone.0358694
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.