Date of Award

Fall 12-14-2016

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Managerial Sciences

First Advisor

Dr. Nikos Dimotakis

Second Advisor

Dr. Lisa Schurer Lambert


One important aspect of performance feedback that was found to have significant effects on work outcomes is feedback valence. However, the literature is inconclusive about which valence of feedback, positive or negative, is better for outcomes. To help solve this conundrum, this dissertation argues that it is important to look not just at the valence of external feedback but also the pattern of such feedback valence over time when studying the effects of feedback on important work outcomes.

Consequently, it identifies two main dimensions of feedback valence pattern in line with previous literature: trend (so whether valence is on average flat, increasing or decreasing over time) and consistency (the extent of variability in feedback valence over time), and explores the effects of all three dimensions, feedback valence at one point in time as well as the trend and (in)consistency of feedback valence over time, on important work outcomes at one point in time. Those work outcomes include motivation, learning and intentions to improve skills, satisfaction with the task and with performance, goals and expectations, and so consequently work performance. Mediators of the relationships between feedback valence and its patterns with outcomes, and moderators of the relationships between these feedback variables and mediators are also identified and tested.

One experiment and one field-like team-based study were conducted with one sample of undergraduate students each to test study hypotheses. Results show that feedback valence and its patterns do have significant (but small) effects on outcomes, with the effects on some outcomes stronger and more likely to generalize across contexts and analyses than others. Also, mediation analyses show that affect, self-efficacy and attribution are important mediators between feedback variables and outcomes, and moderation analyses provide evidence supporting the moderating effect of task characteristics, personality characteristics and factors that can be manipulated at the individual level like effort on the relationships between feedback variables and cognitive and affective mediators/outcomes. At the end, important results are discussed and analyzed followed by a summary of the study, suggestions for future research and implications for business practice.