Teacher Narratives on AI Integration in Turkish Education

283 Turkish educators provided narratives on artificial intelligence in education, totaling 88,296 tokens. Mustafa Işıkgöz created this corpus-based study dataset, which was last updated in April 2026. The data includes anonymized participant demographics and a thematic lexicon for sentiment and discourse analysis.

Use Cases

• Analyze sentiment dimension scores from the thematic lexicon against participant demographics like age and professional seniority.
• Train models to predict thematic clusters, such as Pedagogical Actors or Contrast Markers, from the lemmatized Turkish narrative text.
• Conduct keyword-in-context (KWIC) analysis on the 88,296-token corpus to identify discourse patterns related to AI usage frequency.
• Examine correlations between gender or subject area and the use of specific sentiment-laden lemmas from the provided glossary.

Strengths

• Corpus contains 283 valid teacher narratives, providing a substantial qualitative sample.
• Dataset includes 88,296 tokens of text data for linguistic analysis.
• Provides a detailed thematic lexicon mapping Turkish lemmas to English translations and thematic clusters.

Limitations

• All narratives are in Turkish, limiting direct analysis for researchers without Turkish language skills or translation tools.
• Sample is geographically limited to educators from a single country (Turkey), reducing generalizability.
• Data collection method and specific time range of the study are not detailed, affecting temporal context.

Provenance

Source

Mustafa Işıkgöz via figshare.

Collection Method

Collected via a survey form from educators; corpus processed using AntConc and Zemberek-NLP with manual lemmatization.

Time Range

null

Freshness

null

Geography

Turkey (participants are Turkish educators).