Lessons Learned in Detecting and Correcting Response Heaping: Conceptual, Methodological, and Empirical Observations

Wildlife agencies have used hunter/angler surveys for decades to assist in estimating game populations and harvest. When hunters and anglers are asked to recall a frequency (e.g., days of participation) or a quantity (e.g., game bagged), some respondents round their answers to numbers ending in 0 or 5 (0–5 prototypes) or multiply days of participation by an integer to approximate harvest. Both types of survey responses result in response heaps (spikes in response frequency functions) and can result in biased estimates (e.g., means and totals). This article summarizes recent advances in detecting and correcting for response heaps in survey data. We address 11 lessons learned from previous research. These collective findings and observations are organized in three broad areas: (1) conceptual, (2) methodological, and (3) analytical lessons. We also discuss knowledge gaps where additional work is necessary. The identification of bias and development of corrective methods will hopefully enable wildlife managers to have greater confidence in survey estimates.