TY - JOUR
T1 - Competitor ability, sorting and overconfidence
T2 - An experiment
AU - Li, Tianyi
AU - Noussair, Charles N.
N1 - Publisher Copyright:
© 2025 Elsevier Inc.
PY - 2025/8
Y1 - 2025/8
N2 - While overconfidence in one's ability relative to others is common, the feeling that one is less qualified than one's peers is widespread in elite groups. In this paper, we show that both effects simultaneously exist for the same individuals and we propose the notion of a sorting bias to capture the overall pattern. We conduct an experiment in which individuals first take a mathematics test. They are then sorted into levels based on their performance, and matched with a competitor who scored at a similar level. The matched pairs then take a second mathematics test. Before the sorting into levels, they are asked to predict the probability that they perform better than the person that they are paired with, in a strategy method protocol. If they properly condition on the rule that sorts participants into pairs, they would predict a probability of .5 of being the better performer in their pair. We find that participants act as if they condition on the way their opponent has been sorted but do not sufficiently account for their own sorting. Individuals are less optimistic about outperforming a similarly selected peer, the higher performing the group to which they are assigned. This effect co-exists with a general pattern of overplacement, measured here as a belief that one has a greater than 50% chance of outperforming a peer with similar qualifications.
AB - While overconfidence in one's ability relative to others is common, the feeling that one is less qualified than one's peers is widespread in elite groups. In this paper, we show that both effects simultaneously exist for the same individuals and we propose the notion of a sorting bias to capture the overall pattern. We conduct an experiment in which individuals first take a mathematics test. They are then sorted into levels based on their performance, and matched with a competitor who scored at a similar level. The matched pairs then take a second mathematics test. Before the sorting into levels, they are asked to predict the probability that they perform better than the person that they are paired with, in a strategy method protocol. If they properly condition on the rule that sorts participants into pairs, they would predict a probability of .5 of being the better performer in their pair. We find that participants act as if they condition on the way their opponent has been sorted but do not sufficiently account for their own sorting. Individuals are less optimistic about outperforming a similarly selected peer, the higher performing the group to which they are assigned. This effect co-exists with a general pattern of overplacement, measured here as a belief that one has a greater than 50% chance of outperforming a peer with similar qualifications.
KW - Competitor ability
KW - Overconfidence
KW - Sorting bias
UR - https://www.scopus.com/pages/publications/105004244888
UR - https://www.scopus.com/inward/citedby.url?scp=105004244888&partnerID=8YFLogxK
U2 - 10.1016/j.socec.2025.102373
DO - 10.1016/j.socec.2025.102373
M3 - Article
AN - SCOPUS:105004244888
SN - 2214-8043
VL - 117
JO - Journal of Behavioral and Experimental Economics
JF - Journal of Behavioral and Experimental Economics
M1 - 102373
ER -