Tracing Selection Effects in Three Non-Probability Samples

Abstract

Snowball sampling and targeted sampling are widely applied techniques to recruit samples from hidden populations, such as problematic drug users. The disadvantage is that they yield non-probability samples which cannot be generalised to the population. Despite thorough preparatory mapping procedures, selection effects continue to occur. This paper proposes an interpretation frame that allows estimating the direction of selection bias after data collection. Critical examination of the recruitment procedure and comparison with statistical and non-statistical external data sources are the core features of the interpretation frame. Applying the interpretation frame increases insight into the reliability of the results and allows to estimate where selection bias may have occurred.

A Comparison of Rural and Urban Nonmedical Prescription Opioid Users’ Lifetime and Recent Drug Use

“…Snowball sampling is a commonly used recruitment approach in studies involving hidden populations such as drug users (20). In this study, participants were initially recruited using flyers in local public places or by community key informants.…”

A local and regional Alcohol and Drugs Monitor in the Netherlands: a tale of three cities

“…The survey uses written questionnaires and (semi)structured and open interviews. Sampling methods used were snowball sampling, network sampling and targeted sampling [23]. In each city an ethnographic map was drawn based on the knowledge generated by the community fieldwork.…”

Yield and turnover of illicit indoor cannabis (Cannabis spp.) plantations in Belgium

“…Research on so-called hidden populations, particularly those involved in illegal drug markets, generally uses the snowball sampling technique [35][36][37][38][39][40][41][42][43]. Snowball sampling does not require a large population to obtain an acceptable level of validity for the research [37,39].…”