Streamlining the Edit Review Process: Using Programming to Reduce Time Costs (303113)

Keywords: comment, review, data quality, programming

For complex surveys, one of the most effective tools for reducing nonsampling survey error is interviewer comments. In the Survey of Consumer Finances (SCF), these comments have been particularly useful in explaining unusual respondent situations, enabling editors to alter case data after the interview has been completed to reflect the household’s true state. However, this method of error reduction is also extraordinarily time consuming, requiring months of careful analysis by multiple editors, all of whom require extensive training. Given these issues, any method for speeding up the training and editing processes, while still maintaining the data quality improvements they generate, is worth exploring. With this in mind, a system was designed to incorporate survey data, interviewer comments, and a series of data checks into a single, easy-to-use program interface. Perhaps most helpfully, the program also generates financial summary sheets—such as a household balance sheet and income statement—for the quick identification of anomalies. Using this system, data editors can, at a glance, understand the basic fundamentals of a case, identify potential problems, make corrections, and then check to see that these corrections did not create further issues. The program also serves to encapsulate knowledge about the survey that previously had to be memorized during the training process. This program, named the Editor Assistant (EA), was fully employed for the 2013 SCF and used to swiftly train three new editors and speed up the editing process. The five-month reduction in required editing time—compared with previous years—has been credited in large part to the EA.