New Mexico and Utah have both used a version of the Relative Standard Error (RSE) statistic to provide query users with a sense of the statistical stability of the estimates in the query results. RSE (also known as coefficient of variation) is the standard error of the estimate divided by the estimate. The result may be thought of as the standard error expressed as a percentage of the estimate.

For instance, if you have an RSE of .50, then the standard error is 50% the magnitude of the estimate. At the 95% confidence interval lower limit would be close to zero (1.96 * s.e.) and the upper limit would be equal to nearly twice the estimate. Several national data sources (1) use RSE as a criterion for data suppression. in New Mexico, it is our policy to suppress data based on the threat of identification of individuals. We wanted to also provide users with an indication of the stability of the estimate.

If you go to either Utah's or New Mexico's query systems, you will see this reliability indicator in practice (look down the far right column of the table).

Here is a link to documentation for how New Mexico customized their View application to display the reliability warning flag.

(1) Klein RJ, Proctor SE, Boudreault MA, Turczyn KM. Healthy People 2010 criteria for data suppression. Statistical Notes, no 24. Hyattsville, Maryland: National Center for Health Statistics. June 2002.

Last modified 8 years ago Last modified on 09/04/10 20:49:45