Comprehensibility, Coherence, and Engagement of AI-Generated Interactive Fiction for Educational Use: Questionnaire Item Pool

This technical note documents a candidate questionnaire item pool for measuring the perceived narrative quality of AI-generated interactive-fiction (IF) learning scenarios. The pool comprises nine five-point Likert items covering three constructs: narrative clarity, story-content coherence, and engagement. Seven items are adapted from established instruments (Narrative Engagement Scale, Transportation Scale Short Form, Intrinsic Motivation Inventory, MEEGA+); two coherence items were written from scratch to capture the intrinsic-integration principle. The note records the constructs, the item wordings in German (as administered) and English (working translation), per-item origin with original wordings, scoring rules fixed before data collection, and the design choices that scope this artefact to descriptive pilot use. The pool has not been psychometrically validated. Expert-panel content review, cognitive interviewing, and Beaton-protocol back-translation were not performed and are documented as prerequisites for any future scale development. The pool was first deployed in the companion empirical study on AI-generated interactive fiction in higher education (see related identifiers). This record contains only the methodological documentation and reports no empirical data of its own.