Seminar by Pamela Giustinelli

ore 12.30 Sala Seminari – I° piano, Palazzo Levi Cases, Via del Santo 33

05.12.2017

Tail and Center Rounding of Probabilistic Expectations in the Health and Retirement Study

Seminar by Pamela Giustinelli (Università Bocconi)

A large and growing number of surveys have been eliciting respondents’ expectations for future events on a 0-100 scale of percent chance. The evidence from different surveys and populations reveals that these data display substantial heaping at multiples of 10 and 5 percent, suggesting that respondents round their reports. The extent of rounding, however, is unobserved and its impact on inference unknown. In this paper, we study the nature of rounding in numerical reports of expectations by analyzing response patterns across numerous expectations questions and waves of the Health and Retirement Study (HRS). We discover a systematic tendency by about half of the respondents to provide more refined responses in the tails of the 0-100 scale than in its center. In contrast, only about five percent of the respondents provide more refined responses in the center than the tails. We also find that rounding practice varies somewhat across question domains, which range in the HRS from personal health to personal finances to macroeconomic events. We develop a two-stage framework to characterize person-specific rounding in each question domain and scale segment. Our framework incorporates the evidence from the first part of the analysis in the form of assumptions that partially identify respondents’ rounding. In particular, the first stage uses each respondent’s response pattern across questions and waves to bound the extent to which the respondent rounds responses in each question domain and scale segment. The second stage replaces each original point response with an interval, representing the range of possible values of the respondent’s true latent belief implied by the degree of rounding inferred in the first stage. Next we demonstrate how the interval data thus obtained can be employed as either an outcome variable or a covariate in prediction analyses of substantive interest. To assess the importance of rounding we compare empirical findings when rounding is ignored and when it is accounted for using our proposed approach.