Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The inferences drawn from the most widely used regression models of subjective welfare are subject to a “frame-of-reference bias,” stemming from non-ignorable heterogeneity in subjective scales, such as what it means to be “rich” or “poor.” To test for this bias, respondents in Tajikistan were asked to rank the economic status of theoretical vignette households, as well as their own. Respondents are found to hold diverse scales, but there is very little bias in either the economic gradient of subjective welfare or most other coefficients of interest. These results provide a foundation for standard survey methods and regression specifications for subjective welfare data.