The National Intimate Partner and Sexual Violence Survey (NISVS)

The tables include a prevalence estimate, the estimated number of victims, and 95% confidence intervals. It is important to keep in mind that the prevalence estimates are based on a sample and not a census of the U.S. population. Estimates that are based on a sample always include some error. This uncertainty or error is estimated with a 95% confidence interval. The confidence interval provides a range of values that likely include the true prevalence estimate. The 95% confidence interval means that we can be 95% confident that the true prevalence is within the interval.

Readers are strongly cautioned against comparing estimates across states or by sex. Estimates that have overlapping confidence intervals might not be meaningfully different from each other and additional statistical analyses are needed to test for differences. Some confidence intervals are relatively large because the estimate is based on a small number of people who reported the particular type of victimization in the sample. Across all the tables, very few states have confidence intervals that do not overlap with those for the highest estimate in the table and even fewer have confidence intervals that do not overlap with the estimate for the entire U.S. population. Similarly, when data are available for men and women the confidence intervals tend to overlap and when they do not overlap the estimates are higher for women.

The confidence intervals will become smaller as additional years of data become available and can be aggregated across years. We also expect to be able to use the data from additional years to provide rolling 12 month prevalence estimates for states.