AI-enhanced learning tools: impact of information quality characteristics on perceived ease of use analyzed with PLS-SEM and NCA

Purpose The purpose of this paper is to explore how the four dimensions of informational quality characteristics – accuracy, comprehensiveness, relevance, and timeliness – shape learners' perceptions of the ease of use of information in AI-enhanced learning environments. The study seeks to deepen understanding of how the quality of AI-generated instructional content influences students' ability to engage with, interpret, and apply information effectively. By extending the Technology Acceptance Model (TAM) to comprise informational attributes, the paper aims to clarify which informational quality characteristics make AI-supported learning more intuitive, cognitively efficient, and pedagogically meaningful. Design/methodology/approach A total of 197 responses were collected among undergraduate and graduate business students at a mid-sized Canadian university during the winter of 2024. The respondents were at least 18 years old. The researchers developed a survey instrument whose questions originated from prior research. SPSS and Partial Least Squares Structural Equation Modelling (PLS-SEM) with the Importance-Performance (IPMA) and Necessary Condition Analysis (NCA) were utilized as statistical techniques. Findings The results indicate that comprehensiveness is the only significant determinant of perceived ease of use of information in AI-enhanced learning environments. IPMA identifies comprehensiveness as a high-importance but underperforming factor, suggesting that improvements in this area would most enhance usability. NCA further reveals that both comprehensiveness and relevance are essential, must-have conditions for ease of use. In contrast, accuracy and timeliness exert limited influence on perceived ease of use. Originality/value This study extends the explanatory scope of TAM by incorporating informational quality characteristics as antecedents of perceived ease of use in AI-enhanced learning contexts. It is among the first to combine PLS-SEM, IPMA, and NCA to distinguish between determinant and must-have predictors of usability. The findings highlight that the comprehensiveness and relevance of AI-generated information – rather than its accuracy or timeliness – are the key factors shaping learners' perceptions of ease of use.