What exactly is the CTSonline feature that does approximate significance tests?
Along with each table is provided a link to a window that allows you to do an approximate test of whether the difference between two estimates is statistically significant. To do the test, you click on the link and then enter the two estimates and their associated standard errors. Those values are used to calculate a Z-statistic and p-value for the test of whether the difference between the estimates is zero. Please note, however, that this test will not yield exact results because of the complex sample design of the CTS surveys, which incorporates stratification, clustering, and oversampling. If you would like exact tests of whether two estimates differ, you need to obtain the survey data files.
