Pitfalls in the analysis of complex surveys using Stata

The purpose of this presentation is to show the common mistakes in the analysis of complex surveys. In Mexico, we have a significant number of complex surveys available, which cover (among other issues) household income and expenditures, the labor market, consumer confidence, public security perception, and family life. The heart of the matter is the following: if you ignore the sampling design of a complex survey (basically, the probability weights, the clustering, and the stratification), inevitably you will get an erroneous estimation of whatever you are dealing with. Stata is a fully survey-capable software that takes into account the sampling design. I explore Stata’s survey methods capabilities and, as far as I know, illustrate the best practices in the analysis of complex surveys for the following topics: descriptive statistics, variance estimation methods, hypothesis testing, and econometric models.