Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This study extends earlier work identifying sequence order biases in contest outcomes determined solely by popular voting. Results for different contest evaluation formats are empirically scrutinised, where both expert panel scoring and popular voting determine contest ranking. Forms of sequence order bias exist separately in the expert panel voting even though they are undertaken after each individual performance, as well as in the popular vote at the end of the contest. We suggest that the biases observed in the expert voting can be explained as a type of ‘grade inflation’.