Information Today, Inc. Corporate Site KMWorld CRM Media Streaming Media Faulkner Speech Technology Unisphere/DBTA
PRIVACY/COOKIES POLICY
Other ITI Websites
American Library Directory Boardwalk Empire Database Trends and Applications DestinationCRM Faulkner Information Services Fulltext Sources Online InfoToday Europe KMWorld Literary Market Place Plexus Publishing Smart Customer Service Speech Technology Streaming Media Streaming Media Europe Streaming Media Producer Unisphere Research



News & Events > NewsBreaks
Back Index Forward
Threads bluesky LinkedIn FaceBook Instagram RSS Feed
 



The Consumer Expenditure Survey Gets a Makeover
by
Posted On July 28, 2015
PAGE: 1 2


Options for question development and question order for the recall interview are being studied for their effects on both the process and responses. Do you ask general questions and circle back to gather more detail, or do you take a deep dive into purchases as people mention them? Every detail of the question and sequencing is being explored to assure that the data collected is the most accurate representation of household expenditures.

The diary survey covers expenditures on food eaten away from home; groceries purchased for the household; purchased clothing, shoes, jewelry, and accessories; and all other paid-for products, services, and expenses. The proposal suggests replacing the household paper diary with individual diaries maintained via electronic devices, i.e., smartphones (preferred) or desktops (via the internet). Results of tests conducted thus far indicate that “electronic diaries, specifically mobile diaries, do an effective job targeting groups generally underrepresented in CE data.”

Additional research efforts are underway to consult other data sources (e.g., housing data) to see whether they improve the quality of CE quarterly (CEQ) interview estimates. Also, when CEQ respondents are asked to recall expenses for the prior 3 months, providing them with a worksheet to use during those months may reduce errors in reporting.

The team is also considering alternate ways to impute missing asset/liability data. For example, if a respondent says he or she doesn’t know how much is left to pay on student loans or how much is in a retirement account, the government needs a way to make sure those accounts are checked. The goal is to collect this type of information starting in 2017. This would enable additional research into how savings (especially for retirement) and debt (credit cards and student loans) affect specific household purchases, particularly during economic downturns. How those questions are asked, i.e., in total or as ranges for the savings or debt, and which increments yield the most accurate data, may need to vary in terms of age or some other grouping model based on occupation or education.

At present, interview questions for the Telephone Point of Purchase Survey (TPOPS) are organized around expenditure categories similar to those used in the CE survey, but based on purchase frequency. The Gemini team is looking at the feasibility of integrating outlet (store name, store location, and price) questions into the CE diary survey and then assessing their impact on data quality and respondent burden. While further testing (for the mobile diary) needs to be carried out, it might be that if the CE diary were organized around transactions, the TPOPS survey could be eliminated.

The Gemini team continues to explore how to best assess respondents’ perceived level of burden and customize approaches (e.g., how often and when intervention is needed, such as reminders to enter data) designed to assure that accurate and detailed expenditures are reported. The team asked respondents’ feelings about the following:

  • Length of the interview
  • How interesting the survey is to the respondent
  • How easy/difficult it is to answer the questions
  • Appropriate number of interviews (e.g., too many or a reasonable number)
  • How burdensome was the survey to each respondent?
  • Sensitivity?
  • Would the respondent agree to take a similar survey in the future?
  • Would the respondent be willing to extend the survey for an additional 15 minutes?
  • Amount of effort that respondent put into answering the survey
  • Is the Census to be trusted to safeguard respondent information?

When asked which groups are particularly underrepresented in CE data (and how the BLS is addressing each), Erhard indicated that there is a higher-income nonresponse weighting factor; longer-term, Erhard sees the survey being distributed in other languages, beginning with Spanish.

About the Data

Data collected via the interview survey and diary survey is integrated to produce the published tables accessible on the CE survey website. It’s important to remember that spending is based on the consumer unit and not per capita and that income and expenditures are collected over different reference periods. Limited spatial data is public: Not all states are sampled, and most geographic data is suppressed.

Who Uses CE Microdata and How

Once data is collected, verified, and released, then it’s up to others to use the details to inform their work. Participants in this year’s workshop were treated to previews of research being done at several colleges and universities, such as:

  • “Driven to Poverty? An Analysis of Automobile Expenditures in Low-Income Households” by Stephen Brumbaugh, Ph.D. candidate at the University of California–Los Angeles’ Department of Urban Planning
  • “How Do Changes in Housing Wealth Affect Consumption Behavior?” by Taylor Smith, assistant professor of economics at Georgia Gwinnett College
  • “The 2011 Payroll Tax Cut and Household Spending: Evidence From a Quasi-Natural Experiment” by Naveen Singhal, Ph.D. candidate at the University of Illinois–Chicago
  • “Household Consumption Smoothing Between Monthly Housing Payments” by Li Zhang, Ph.D. candidate at the University of Virginia
  • “Income-Expenditure Elasticities of Less Healthy Consumption Goods” by Adam Hoffer, assistant professor at the University of Wisconsin–La Crosse

Attendees participated in four CE microdata practical training sessions. BLS staffers demonstrated how to use the PUMD files and documentation to calculate various statistics, including population means and aggregates, and to correct standard errors. In addition to covering the basics of data file structure, naming conventions, and record counts, there were several hands-on projects using annual healthcare expenditure data, including detailed health insurance costs.

More of These Types of Sessions Are Necessary

These days, we hear a good deal about hackathons, during which members of the public are invited to design innovative tools that would improve data collection, thus boosting federal agency transparency. Opportunities to learn how to make better use of the data available from these agencies are also helpful.

If there was a missed opportunity at this workshop, it was that no businesses using CE survey microdata were available to present on how they apply the data to industry sector analysis. Perhaps a cross-BLS or cross-agency workshop highlighting innovative uses of government data could give others ideas about how this data could give their firms an edge.


PAGE: 1 2


Barbie E. Keiser is an information resources management consultant located in the metropolitan Washington, D.C., area.

Email Barbie E. Keiser

Related Articles

3/14/2023Merged Companies Rebrand as Circana
4/5/2016New Report Offers Best Practices for Transparency Reporting
7/10/2014Interactive U.S. Map Shows Patron Demographics
9/20/2012APDU 2012 Conference Explores the Future of Federal Survey Data
8/27/2012Census Bureau Mobile App Now Available for iPhone and iPad
6/20/2011Easy and Affordable State-Based Research From RAND


Comments Add A Comment

              Back to top