Answers are grounded in this paper and cite their sources.
Summary Fifteen hundred adults selected for a population survey from the City of Edinburgh Electoral Registers by probability sampling were allocated to randomised response interview or direct interview at random. These alternative interviewing strategies were adopted for two questions relating to attributes which might be considered stigmatising. There were no significant differences in reporting between the two methods either at aggregate or at subgroup level. The apparent failure of the randomised response method to increase estimates of the proportions with the attributes hypothesised indicates either that traditional methods are satisfactory for the collection of this sensitive information, or that randomised response was not an acceptable alternative.
No takes yet. Share an insight, caveat, or question.
Duffy et al. (1988) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: