Examining conservation compliance with randomized response technique analyses

Abstract

Understanding conservation non-compliance-violations of laws or social norms designed to protect natural resources from overexploitation-is a priority for conservation research and management. As direct questioning about stigmatized behaviors can be biased, researchers have adopted more complex indirect questioning techniques. The randomized response technique (RRT) is one of the most powerful indirect survey methods, yet analyses of these data require sophisticated statistical models. To date, there has been limited user-friendly software to analyze RRT data, particularly so for models that combine information from multiple RRT questions. We present an overview of three RRT models that cover single or multiple questions and provide an R package, zapstRR (ZoologicAl Package for RRT), that is accessible for experienced and new R users. Researchers can estimate the prevalence of non-compliance, the number of illicit activities performed by individuals, perform regression for univariate and multivariate RRT data, and correct prevalence estimates for evasive response bias. We show the strengths and limitations of different estimators for RRT data using two case studies. The first focuses on illegal bird hunting in Southwest China using a standard RRT question design and a novel implementation that was hypothesized to offer further anonymity to respondents. The second case study examined illicit bushmeat consumption in Madagascar after the onset of educational interventions (Randriamamonjy et al., 2015). The case studies illustrate how to apply functions from zapstRR and demonstrate how the models can work in tandem to uncover distinct patterns within RRT datasets. This article is protected by copyright. All rights reserved.