Using Graphical Analyses to Improve all Aspects of the Survey of Consumer Finances

Introduction Data providers and analysts are now equipped with sophisticated tools m a highly technical computmg environment. Surveys are at the same time becoming more intricate resulting in data files which may be both quite complicated and potentially very large. As a result. traditional data processing tasks have become much more complex and often are very data dependent. In order to better understand the underlying data, it is always helpful and sometimes critical for the data analyst to “look at” the data using graphical techniques. The Survey of Consumer Finances (SCF) is a complex survey with data on all aspects of a household’s financial characteristics. Additionally. the sample design IS not simple. The use of graphics has become a key Ingredient in producing high-quality data for the Survey of Consumer Finances. Different types of graphical plots have been used at all stages of processing. These stages include sample selection. imputation, editing. weighting. analyses. and disclosure review. The ob1ectiv.e of this paper is to detail how the graphics are used. hou their use has improvfed the surve! processing. and how the use of graphics will be expanded for future surveys. Including this ,ntroduction. this paper contatns four sections. In the next section. we detail the Survey of Consumer Finances data, the sample design and other processing uruque to the survey. The third section discusses the use of graphics in the areas ltsted above. The final section includes some closing comments to encourage the use of graphics. Examples are mcluded to show how the implementation works in the survey.