The sample size for each country VACS is calculated based on an estimated prevalence of childhood sexual violence from existing data in each country (ie, the Demographic and Health Survey or other surveys), relative standard error and margin of error. Childhood sexual violence is used as the basis of power estimations because it is typically the least prevalent type of violence. This yields sample sizes that are more likely to be powered to detect and report results for the least prevalent violence type. The effective sample size is bolstered for the cluster design of the survey and adjusted for non-response, including household and individual response rates. The sample size for the VACS is robust, with a range of 891–7912 females and 803–2717 males in each country (table 1 ).
Weighting is used to obtain representative parameter estimates from survey data. It accounts for the probability that each respondent came into the sample, the differential effects of non-response and imperfect sampling frames that affect the composition of the sample.5 Final sample weights are calculated by (1) determining base weights to account for all steps of random selection that led to the sample of population members, (2) adjusting for non-response and (3) adjusting the final set of adjusted weights to the distribution of the population.
Finally, response rates are calculated using formulas from the American Association of Public Opinion Research.6 These rates are computed for the entire sample (both household and individual). Estimated individual-level eligibility rates are calculated separately for females and males of responding households. The final response rate is the multiplication of household and individual-level response rates. The response rates for the VACS range from 72.0% to 97.9% among females and 65.7% to 98.1% among males (table 1 ).
Weighting is used to obtain representative parameter estimates from survey data. It accounts for the probability that each respondent came into the sample, the differential effects of non-response and imperfect sampling frames that affect the composition of the sample.5 Final sample weights are calculated by (1) determining base weights to account for all steps of random selection that led to the sample of population members, (2) adjusting for non-response and (3) adjusting the final set of adjusted weights to the distribution of the population.
Finally, response rates are calculated using formulas from the American Association of Public Opinion Research.6 These rates are computed for the entire sample (both household and individual). Estimated individual-level eligibility rates are calculated separately for females and males of responding households. The final response rate is the multiplication of household and individual-level response rates. The response rates for the VACS range from 72.0% to 97.9% among females and 65.7% to 98.1% among males (
