2014 National Survey on Drug Use and Health: Mental Health Estimates Computed Directly from the Clinical Sample of the Mental Health Surveillance Study and Measures of their Standard Errors

This document focuses on how the prevalence estimates and their standard errors were derived from the 2008 to 2012 MHSS clinical sample. In particular, it describes how the prevalence estimates covering the 2008 to 2012 time period were computed using sampling weights that had undergone a number of calibration adjustments with an emphasis on the last adjustment: poststratification—the annual calibration of the clinical sample to the NSDUH control totals. It then discusses several alternative methods for measuring the standard errors of those estimates.