Key points are not available for this paper at this time.
Abstract: High dropout rates at universities, often caused by false expectations and a lack of motivation, pose a serious problem in higher education. Online self-assessments (OSAs) assess expectations regarding a field of study (major) and provide feedback on the reality of the major, thus pointing out expectation-reality discrepancies as well as helping prospective students choose a major. Based on cognitive dissonance theory, pointing out expectation-reality discrepancies should be related to changes in motivation for the major (expectancies for success, subjective values, intention to choose a major) and this relationship should be strengthened by feedback. Past research has shown that OSAs can correct expectations and that expectation-reality discrepancies are related to motivation but has not investigated the role of feedback for this process. Therefore, we extend past research by examining whether the positive relationships between expectation-reality discrepancies and changes in motivation for a major are stronger for prospective students who receive feedback on their expectation-reality discrepancies than for prospective students who do not receive feedback after the assessment. We conducted a field experiment in which 234 prospective students were randomly assigned to one of two groups (EG1 = OSA including feedback; EG2 = OSA without feedback). As hypothesized larger expectation-reality discrepancies were associated with larger changes in motivation for a major (expectancies for success, subjective values, intention to choose a major). Beyond that, we found a moderation effect of the feedback condition showing that the positive relationships between expectation-reality discrepancies and expectancies for success were stronger when prospective students received feedback (vs. no feedback). As feedback only showed effects beyond expectation-reality discrepancies in one of the considered outcomes, both the development of assessment and feedback should be targeted to optimize the effectiveness of OSAs.
Merkle et al. (Mon,) studied this question.