Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We present results from a five-year effort to design promising virtual coaching interventions to improve college student achievement. Across nearly 20,000 students at three campuses, we find some improvement on study time, but no effect on academic outcomes. We interpret the results with unique survey data and a model of student effort. Treated students learn that more effort is needed to attain good grades and develop stronger preferences for high grades, but these effects are too small to translate into academic benefits. More comprehensive, social and better-timed interventions are needed for helping students outside the classroom.