AI-Adapted EFL Reading Texts for Mixed-Proficiency Learners

A study by Hesham Aldamen examined the effect of AI-based text adaptation on reading comprehension for 48 university EFL students. The data, published on figshare in 2026, includes original and adapted versions of source texts used in a controlled experiment. The study employed a 2x2 mixed factorial design to test performance across different proficiency levels.

Use Cases

• Training or evaluating NLP models for automatic text simplification based on proficiency levels mentioned in the study.
• Studying the interaction between text complexity and learner proficiency for educational research.
• Developing benchmarks for AI tools designed to support differentiated reading instruction in language classrooms.

Strengths

• The experimental design is clearly described, including a 2x2 mixed factorial design and counterbalancing to reduce order effects.
• Results are statistically grounded, with specific p-values (p < .001) reported for key findings.
• The dataset is openly licensed under CC-BY-4.0, facilitating reuse and sharing.

Limitations

• The dataset is very small at 42.2 KB, indicating limited scope and sample size.
• Column-level documentation is absent; field semantics must be inferred after download.
• The participant pool of 48 university students may limit generalizability of findings.

Provenance

Source

figshare

Collection Method

Data was generated from a controlled study where 48 university EFL students completed comprehension tasks using original and AI-adapted texts.

Freshness

Last updated 2026-04-21 05:34:02; freshness should be verified.