Victor Champonnois

Séminaires internes
phd seminar

Victor Champonnois

AMSE
Testing anchoring on previous studies in meta-analyses: An application to the value of statistical life
Lieu

IBD Salle 16

Îlot Bernard du Bois - Salle 16

AMU - AMSE
5-9 boulevard Maurice Bourdet
13001 Marseille

Date(s)
Mardi 16 octobre 2018| 12:30 - 14:00
Contact(s)

Océane Piétri : oceane.pietri[at]univ-amu.fr
Morgan Raux : morgan.raux[at]univ-amu.fr
Laura Sénécal : laura.senecal[at]univ-amu.fr

Résumé

In this paper, I propose a formal test for a publication bias suggested by Viscusi and Masterman (2017). This bias arises from the selection of estimates based on their distance from the usual range in the literature. It implies that, in a given research area, there is a progressive concentration of estimates around this range as the literature grows. I test this hypothesis using OECD meta-data on the value of a statistical life (VSL) elicited via stated preferences. Using conditional variance regressions and quantile regressions, I am able to test for the effect of the number of prior studies on the dispersion of estimates, controlling for confounding factors. Consistently with the hypothesis, my results show a decrease in the dispersion of VSL estimates, which mostly happens for the lower part of the conditional distribution. These results imply that there is a systematic underestimation of the variance of mean estimates in meta-analyses of VSL. It might also affect meta-analyses in other research areas.