Publication |
2010.
|
Summary/Abstract |
The Defense Manpower Data Center of the U.S. Department of Defense launched the 2006 Survey of Active Duty Spouses to assess attitudes of the spouses of U.S. active-duty military members. Severe problems existed with the sampling, weight adjustments, and estimation (including variance-estimation) procedures. Stratification of the sample without proper consideration of the survey objectives made it impossible to achieve reportable information for many desired population subgroups. Excessive stratification caused many of the sampling strata to have very small numbers of respondents, both expected and actual. Consequently, nonresponse bias was probably enormous across many of the strata. Absurd weight adjustments likely contributed toward rendering many survey estimates unreliable. To make the survey estimates seem more precise, sampling strata were collapsed together to form new "variance strata" for variance estimation. Caution is advised in using results from this and other Defense Manpower Data Center surveys.
|