Misreporting and econometric modelling of zeros in survey data on social bads: An application to cannabis consumption

Abstract

When modelling “social bads,” such as illegal drug consumption, researchers are often faced with a dependent variable characterised by a large number of zero observations. Building on the recent literature on hurdle and double‐hurdle models, we propose a double‐inflated modelling framework, where the zero observations are allowed to come from the following: nonparticipants; participant misreporters (who have larger loss functions associated with a truthful response); and infrequent consumers. Due to our empirical application, the model is derived for the case of an ordered discrete‐dependent variable. However, it is similarly possible to augment other such zero‐inflated models (e.g., zero‐inflated count models, and double‐hurdle models for continuous variables). The model is then applied to a consumer choice problem of cannabis consumption. We estimate that 17% of the reported zeros in the cannabis survey are from individuals who misreport their participation, 11% from infrequent users, and only 72% from true nonparticipants.

No More Glass Bottles? Canned Wine and Italian Consumers

“…The first decision is a binary choice and is modelled using a probit model, while the second is an ordered choice and is modelled through an ordered probit model [49]. ZIOP model has been widely used in studies on tobacco consumption [49], cannabis consumption [52], truck drivers' violation frequency [53], sports participation [54], and health-related issues [55].…”

Marijuana and alcohol: Evidence using border analysis and retail sales data

“…First, apart from Anderson et al (2013), most of the literature has investigated the question using survey data. While surveys suffer from potential bias resulting from underreporting issues of self-reported substance abuse (e.g., Greene et al 2018), data on alcohol sales do not suffer from that typical bias. Moreover, existing literature using sales and not self-reported consumption data reports finding from traditional DID analysis that exploit cross-section variation (across states) in MMLs over time to identify the effect of the policy change.…”

Buy flexible, pay more: The role of temporary contracts on wage inequality

“…Finally, survey data may suffer biases related to the misreporting and recalling of respondents. As for misreporting, it may be associated with perceived social stigma or systematic under-reporting behaviours, and it is more common among highly educated individuals, the self-employed, and wealthier households (Cannari and D'Alessio, 1993;Hurst et al, 2013;Greene et al, 2017). As for recall bias, it depends on the fact that, as is well known, people forget past events and details so reported values tend to be less and less accurate the longer the recall period (Scott and Amenuvegbe, 1990;Stull et al, 2009).…”