U.S. flag

An official website of the United States government

Skip Header


Testing a Series of Categorical Earned Income Questions

Written by:
Working Paper Number: rsm2025-02

Abstract

The Annual Social and Economic Supplement (ASEC) of the Current Population Survey (CPS) includes a question on the amount of earned income from a person’s main job. In order to derive the amount for a full year, the respondent must provide the amount, designate the periodicity (yearly, monthly, bi-weekly, or weekly), and, for any periodicity less than a year, report how many times they were paid that amount during the calendar year. For those who do not provide all necessary information, this experiment tested a short series of unfolding categorical income follow-up questions. Respondents were asked three categorical income questions in total, with each subsequent question further narrowing the response provided in the previous question. Due to the narrower range for the middle income category, by the third categorical question, the response options had unusually small ranges. This specific pathway of unfolding questions was called the RSU (Really Small Unfolding) series. An additional element of the experiment was the option to choose “I remember the exact amount” on any of the categorical questions. Selecting this would prompt the respondent to enter an amount and not show them the rest of the income unfolding sequence.
The experiment was conducted on the Household Pulse Survey (HPS) between July and September 2024 and analyzed unweighted. Results indicate that most people provided complete income information (91%) and were not shown unfolding categorical questions. Of those routed to the unfolding questions, the highest percent of people broke off at the first categorical question. In a debriefing question, a majority of these respondents reported that they did not want to provide income information, although a portion also reported not knowing the amount. Of those who were presented the first income unfolding question, 62% answered all three categorical questions, thereby, providing the missing income data. Response options with particularly narrow income increments (the RSU Series) and offering the option to provide a finite income amount (the “I remember” option) both resulted in more missing data and are not recommended.

Page Last Revised - March 7, 2025
Is this page helpful?
Thumbs Up Image Yes Thumbs Down Image No
NO THANKS
255 characters maximum 255 characters maximum reached
Thank you for your feedback.
Comments or suggestions?

Top

Back to Header