Rounding in Recreation Demand Models: A Latent Class Count Model

A commonly observed feature of visitation data, elicited via a survey instrument, is a greaterpropensity for individuals to report trip numbers that are multiples of 5's, relative to otherpossible integers (such as 3 or 6). One explanation of this phenomenon is that some surveyrespondents have di±culty recalling the exact number of trips taken and instead choose toround their responses. This paper examines the impact that rounding can have on the estimateddemand for recreation and the bias that it may induce on subsequent welfare estimates. Wepropose the use of a latent class structure in which respondents are assumed to be members ofeither a nonrounding or a rounding class. A series of generated data experiments are providedto illustrate the range of possible impacts that ignoring rounding can have on the estimatedparameters of the model and on the welfare implications from site closure. The results suggestthat biases can be substantial, particularly when then unconditional mean number of trips isin the range from two to four. An illustrative application is provided using visitation data toSaylorville Lake in central Iowa.[...]