Unanticipated Performance Effects: Evidence of Expectancy Bias in an Adaptive Mobile Learning Intervention in Sub-Saharan Africa

This dataset was collected as part of a quasi-experimental study designed to evaluate an adaptive mobile learning system targeting tertiary students in Nigeria.In Survey 2, two parallel cohorts were formed: An experimental group, notified in advance that they would later interact with a prototype adaptive mobile learning system. A control group, Were not informed of the next phase of the survey. Both groups studied identical print-based materials on a selected topic, after which they completed a post-test survey assessing multiple dimensions: Concept recall Formula understanding Real-world application Material engagement Motivation Exercise completion Perceived effectiveness Likelihood of future adoption Unexpectedly, statistical analysis revealed that the experimental group performed significantly better on baseline measures, despite using the same materials as the control group.This difference was later attributed to expectancy bias: prior awareness of the upcoming adaptive prototype appeared to elevate engagement and perceived value, leading to measurable improvements before the intervention itself. The dataset captures these pre-intervention baseline differences, providing empirical evidence of Unanticipated Performance Effects that informed subsequent experimental design and interpretation.