RT Article T1 Estimating Marihuana usage using randomized response - some paradoxical findings JF Australian & New Zealand journal of statistics VO 23 IS 2 SP 139 OP 148 A1 Brewer, Ken R. LA English YR 1981 UL https://krimdok.uni-tuebingen.de/Record/192104506X AB A randomized response technique was used in parallel with direct questioning on the (illegal) usage of marihuana in a population survey of Canberra. Estimates of usage from the randomized questions were much lower than those obtained directly. A number of possible explanations for these paradoxical findings are considered. DO 10.1111/j.1467-842X.1981.tb00771.x