OMB Information Collection Request
Supporting Statement B
U.S. Department of Commerce
U.S. Census Bureau
Household Pulse Survey
During the Coronavirus Pandemic
B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS
As noted in Part A, the Household Pulse Survey was initially fielded as a proof of concept to test the Federal statistical system’s ability to produce data in near real-time in the face of a national emergency. The survey is conducted under the auspices of the U.S. Census Bureau’s Experimental Data Products Series (see https://www.census.gov/data/experimental-data-products.html).
Planning for the survey began March 23, 2020, and with OMB emergency clearance granted on April 19, 2020 for 3 months of data collection (OMB No. 0607-1013, Exp. 7/31/2020), the Census Bureau launched the Household Pulse Survey on April 23, 2020. Data were first released on May 20, 2020, with updated releases made subsequently on a weekly basis. Since then, the data have been widely used to guide response and recovery efforts on the part of Federal agencies and state governments. Given the continuing nature of the pandemic, the Census Bureau seeks approval to exercise the balance of the 180 days allowable for emergency clearance under 5 CFR Part 1320, Section 1320.13, Emergency Processing. For ease of reference herein, the Census Bureau refers to the initial deployment of the Household Pulse Survey through July 31, 2020 as “Phase 1;” plans described here to extend the data collection through the balance of time allowable under the emergency clearance are referred to as “Phase 2.”
Given the rapid response nature of this survey, the Census Bureau has continued to make adjustments in content and methods, and documented those through the submission of non-substantive change requests to OMB. In the ensuing months, however, there has been increasing demand from other Federal agencies to revise or add content to the Household Pulse questionnaire that would produce data in support of their own missions and response to the pandemic. Rather than changing the survey on a continuous basis, the Census Bureau has decided to consolidate the requests into one comprehensively revised questionnaire.
The Census Bureau thus will cease collection on the current version (i.e., Phase 1) of the Household Pulse Survey in late July, 2020, and take a brief two-week hiatus in data collection to prepare for Phase 2 collection. The Census Bureau will deploy the Phase 2 instrument the first week of August, 2020. This will allow us to make the necessary revisions to the questionnaire instrument, as well as adjustments to the sampling approach, protocols for contacting survey respondents, and data processing cycle.
This ICR sets forth our approach to Phase 2. Adjustments are based on lessons learned from the Census Bureau’s experience conducting the survey in Phase 1. As with Phase 1, we will provide all information about sampling, weighting, post-survey processing, and cognitive testing results on our website to maximize transparency for the public. This transparency is especially important given the predicted value to decision makers of having rapid bi-weekly state-by-state estimates of the health, social, and economic characteristics of their package.
Universe and Respondent Selection
The universe for the Household Pulse Survey is the approximately 140,000,000 housing units in the United States. For operational efficiency, only housing units linked to one or more email addresses or cell phone numbers will be eligible for the sample. One respondent from each housing unit is selected to respond for him or herself. The sample for all interview periods is approximately 1,165,000 housing units. We expect to successfully receive 105,000 responses each interview period. The total sample size over the 6 interview periods will be approximately 7 million housing units. This represents a 9 percent response rate. One change from the Phase 1 sample design is we will no longer have a longitudinal component to the sample. The degree to which the longitudinal sample was used, and the quality of the longitudinal data given low response led us to conclude that it was not worth the burden on households selected for interviewing across multiple weeks. In Phase 1, we expected to see 60 percent of the respondents from the first week in sample respond in their second week in sample, but saw only between 15-25 percent respond in the second week. This high attrition rate did not allow us to meet the survey objectives of high correlation between the week-to-week samples to lower the variance. Each interview periods sample will be independent from the prior samples.
Procedures for Collecting Information
Housing units linked to one or more email addresses or cell phone numbers are eligible for the sample. These housing units are stratified by state and the top 15 metropolitan areas. The sample size was determined to produce for an estimate of 40 percent a 90 percent margin of error of 2.8 percent for most states and 3.3 percent for the 11 smallest states with eligible sample units (Alaska, Delaware, District of Columbia, Hawaii, Maine, Montana, North Dakota, Rhode Island, South Dakota, Vermont, and Wyoming). The sample was then allocated to states and metropolitan areas proportionally, using the number of housing units as the measure of size. Samples cases will be matched to the Census Bureau’s Contact Frame to assign email addresses and cell phone numbers. See Attachment A for information on the Contact Frame.
The Census Bureau will conduct this information collection online using Qualtrics as the data collection platform. Qualtrics is currently used at the Census Bureau for research and development surveys and provides the necessary agility to deploy the Household Pulse Survey quickly and securely. It operates in the Gov Cloud, is FedRAMP authorized at the moderate level, and has an Authority to Operate from the Census Bureau to collect personally identifiable and Title 13-protected data.
Attachment B sets forth details on the Census Bureau’s weighting plan, which will use a multi-step process to help reduce nonresponse bias in the estimates.
Methods to Maximize Response
The survey is designed to meet the goal of accurate and timely bi-weekly estimates. It will be conducted by an internet questionnaire, with links sent by email and SMS text message. Multiple email addresses and cell phone numbers will be used to increase response for nonrespondents. These modes are expected to yield response rates much lower than traditional surveys. The benefits are implementation efficiency, cost, and timeliness of responses.
Standard errors will be large but will be available for data users to understand the quality of the estimates. Nonresponse bias is likely to be an issue, but measures such as the demographic distribution of the survey respondents compared to benchmarks will be produced for data users to consider in their analyses. There are no other known data sources that can provide comprehensive, relevant, and timely information. At the very least, this survey will provide non-anecdotal evidence about the effects of the pandemic on the population even if the quality is lower than would be traditionally acceptable.
Testing of Procedures
Prior to Phase 1,the Center for Behavioral Science Methods completed a thorough expert review that consisted of nine independent experts reviewing the questionnaire independently, followed by a consensus meeting to discuss comments and come up with recommendations. The recommendations were then passed to the Demographic Directorate, who accepted most of the expert recommendations and responded with subject-matter justifications for those recommendations that were not accepted. In the end, the questionnaire comprised a consensus of the expert survey methodologist recommendations and the subject matter needs. A similar process was followed prior to Phase 2.
Given the rapid response nature of this effort and in keeping with the agile approach in which the Household Pulse Survey is being deployed, the Census Bureau in partnership with BLS and NCHS concurrently launched a web probing study with participants who had opted in to participate in a research study with the Census Bureau via email. These sampled participants were sent email invitations to the survey with embedded probes. Two rounds of web probing were conducted in Phase 1. An additional round of web probing will be conducted prior to the fielding of Phase 2. The Household Pulse Survey Team will use results from this probing study to refine content and cut questions to stay within a 20-minute survey for Phase 2.
Questions identified as priorities for testing in all rounds include those that were developed specifically to address the Covid-19 epidemic; items developed for potential future deployment to understand households’ decision-making in light of receiving economic stimulus payments; and questions deemed to be overly complex from a cognitive standpoint, likely as a consequence of asking agencies contributing content to keep the number of items to a minimum to reduce burden (e.g., questions on food security, questions on educational disruption).
The Census Bureau will release these data under the auspices of its Experimental Statistical Product Series. Information on the Series is available at https://www.census.gov/data/experimental-data-products.html#.
Contacts for Statistical Aspects and Data Collection
Statistical Design:
Anthony Tersine
Demographic Statistical Methods Division
Demographic Programs Directorate
Data Collection/Survey Design:
Jason Fields
Social Economic and Housing Statistics Division
Demographic Programs Directorate
Jennifer Hunter Childs
Center for Behavioral Science Methods
Associate Director Research and Methodology
Statistical Analysis:
David Waddington
Social Economic and Housing Statistics Division
Demographic Programs Directorate
Attachments
Census Bureau Contact Frame
Household Pulse Survey: Weighting Approach
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Dumas, Sheleen (Federal) |
| File Modified | 0000-00-00 |
| File Created | 2021-01-13 |