New research by Curtin University has examined the misreporting of data surrounding illicit drug use in Australia, finding true participation rates are more likely to be about twice as high as the initial figures suggest.
Paper author Professor Mark Harris, from Curtin’s School of Economics and Finance, said researchers were often faced with extremely low recorded participation rates when trying to model, or explain, social ‘bads’ such as illicit drug use.
“Many people will have incentives to under-report their consumption including a fear of being caught, stigma and moral concerns, so it’s not sensible to look at raw numbers and assume they accurately represent what’s going on in the population,” Professor Harris said.
“In the case of illegal drug use in Australia, the reported participation rates of 12.2, 3.2 and 1.3 per cent for marijuana, amphetamines and cocaine respectively, mask true participation rates which are estimated to be almost double for marijuana (23 per cent), and more than double for amphetamines (8 per cent) and cocaine (5 per cent).
“The estimated chances that a user would misreport their participation is a staggering 65 per cent for cocaine, 31 per cent for marijuana and 17 per cent for amphetamines.”
Professor Harris said inaccurate estimates of the prevalence of certain behaviours could lead to significant implications if the data was used as a basis for policy making or funding allocation.
“The widespread use of data collected from individual and household level surveys by researchers and policy makers is clearly reliant on respondents supplying accurate and reliable information,” Professor Harris said.
“For this research, we assessed survey participants’ individual reporting behaviour gleaned from answers to unrelated questions and used this information to calculate the likelihood that their zero answer genuinely represented non-participation, deliberate misrepresentation (lying) or infrequent participation.
“It is apparent that the shortcomings of this type of data should therefore be well understood in order to make appropriate policy or health funding decisions.”
This research was conducted in collaboration with the University of Sheffield, Royal Melbourne Institute of Technology and the University of Exeter.
Funding for the research was provided by the Australian Research Council and the Bankwest Curtin Economics Centre.
The paper, Modelling Illegal Drug Participation, was published in the journal of The Royal Statistical Society, and can be viewed at http://onlinelibrary.wiley.com/doi/10.1111/rssa.12252/full.
