Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Various studies (e.g. Becker, 1962; Ariely et al., 2003) have noted anomalies concerning the relationship between observed demand and the preferences presumed to motivate it. We re-examine these findings using experimental choice data. After separating our subjects' choices into rational and irrational subsets based on consistency with the axioms of revealed preference, we estimate and compare demand coefficients. Mirroring Ariely et al.'s 'coherently arbitrary' choice, both rational and irrational demand estimates exhibit negative price and positive endowment coefficients. However, a comparison of the full set of demand coefficients indicates significant differences between the two, yielding an observable artefact of the preference hypothesis. Relaxing the goodness-of-fit of the revealed preference test (Afriat, 1987; Varian, 1994) does not alter our findings.