Under-representation in nationally representative secondary data

Abstract

There has been a significant increase in the use of secondary data sets. Many such data sets purport to be nationally representative. Secondary data sets include research commissioned by the National Center for Education Statistics, the Centers for Disease Control, and other public entities. Research increasingly utilizes these secondary data in their studies as these data sets are easily accessed by researchers. While these data sets are rich with opportunity, researchers must be aware of certain perils not only in the statistical analysis, but also in the proper use of data sets. While these data sets employ several techniques such as stratification and oversampling in order to maximize the representativeness of the data, certain sub-populations still have a tendency to be underrepresented. This study examines the subpopulation of children with autism spectrum disorders (ASDs) and whether they are underrepresented in various secondary data bases. In addition, this study reviews some published studies about children with ASDs which have used secondary data sets in which children with ASDs may be underrepresented.

Keywords:

• secondary data sets
• nationally-representative
• autism
• ASDs