Urban Crime Victimization Survey (UCVS)

The UCVS’s sampling design was based on a geographical stratification (seven strata retained matching each administrative district) and the use of a multistage-stage sampling procedure to select respondents. At the first stage, a total of 55 Primary Sampling Units (PSUs) within strata was randomly drawn with equal probability and without handing-over.

In the absence of listing of households’ addresses or any voter registration roll, a systematic sampling approach was used to select 15 households in each sampled PSU at the second stage. Thus, enumerators were told to walk in a given direction from a central point in the PSU and attempt to interview households at a rate of one over five without exceeding the quota of 15 households to visit in each PSU. In total, 812 households were visited.