|
Abstract:
|
We document our understanding of, and recommendations for, appropriate best practices in specifying the complex sampling design settings in statistical software that enables design-based analyses of survey data. We discuss features of complex sample survey data such as stratification, clustering, unequal probabilities of selection, and calibration, and outline their impact on estimation procedures. We demonstrate how statistical software treats them, and how the survey data providers can make data users' lives easier by clearly documenting accurate and efficient ways to make sure that their software properly accounts for the complex sampling design features. We provide rubrics that will aid complex sample survey data providers in aligning their level of documentation with best practices, and show how existing surveys and their documentation score based on these rubrics.
|
