Update May 8, 2020: Initial results are scheduled for release the week of May 18, 2020.
Update May 1, 2020: In a letter to OMB (
https://www.reginfo.gov/public/do/DownloadDocument?objectID=100792100), the Census Bureau indicates:
The Census Bureau will extend the first week of production into the second week. This will enable us to address a limitation in the initial week with the online data collection platform. Specifically, the platform could not support the need to send survey invitations to more than one email per sampled household, despite the fact that in many cases there are multiple emails associated with a household. This technical issue has been resolved and, in keeping with our proposed methodology, we are now able to reach households with alternative email addresses that were otherwise unreachable when the initial email address on record was not successful.
Consequently, we will be extending data collection for the first week of sample for an additional week to improve our ability to make effective contact with survey participants. With this extension, we will be delaying by one week the planned first release of the data. The first release of data will include this first wave of sample (with the adjusted collection period of two weeks, from April 23-May 6, 2020) and the second wave of sample (from May 7 - May 13, 2020).
In addition, the Census Bureau has posted the Spanish language version of the Household Pulse Survey --
https://www.reginfo.gov/public/do/DownloadDocument?objectID=100792200
From AEAStat staff:
On April 20, the Census Bureau sought and received OMB permission to conduct a weekly Household Pulse Survey (HPS) on the impacts of the COVID pandemic. HPS data collection began on April 23. HPS webpage
https://www.census.gov/householdpulsedata
The submission to OMB contains substantial details on the HPS effort. See
https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202004-0607-008. Click on View Supporting Statement and Other Documents for purpose, plans, methods, and survey instrument. In particular, see Attachment E for the rationale and sponsoring agency for each question, as well as further information on statistical methods and limitations:
https://www.reginfo.gov/public/do/DownloadDocument?objectID=100381801
Total sample size over the 12 week period will be approximately 13.8 million housing units. A 5% response rate is expected. 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 Census Bureau has developed the Household Pulse survey platform as an experimental endeavor in cooperation with five other federal agencies -- Economic Research Service (ERS), Bureau of Labor Statistics (BLS), National Center for Health Statistics (NCHS), National Center for Education Statistics (NCES), and the Department of Housing (HUD). Question domains include employment status, consumer spending, food security, housing, education disruptions, and dimensions of physical and mental wellness.
Testing this platform during the COVID-19 epidemic will allow the federal statistical systems to demonstrate proof of concept with respect to the household pulse survey platform by providing states with weekly data about the health, social, and economic characteristics of the population. Changes in these measures over time and geography will provide insight into the scope of the impact of the Covid-19 epidemic in the US. The ability to understand how individuals are experiencing business curtailment and closures, stay-at-home orders, school closures, changes in the availability of consumer goods and consumer patterns, and other abrupt and significant changes to American life. This experimental survey is designed to supplement the ability of the federal statistical system to rapidly respond and provide salient information based on a high quality sample frame, data integration, and cooperative expertise.
The Census Bureau will conduct this information collection online using Qualtrics as the data collection platform. Qualtrics currently is 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-protected data. The Census Bureau will utilize longitudinal data collection methodology to limit the need to re-ask questionnaire content once baseline information has been collected for continuing cases.
Weekly survey estimates will be produced by weighting the results to the estimate of the occupied number of housing units from the American Community Survey at the geographic levels of the nation, state, and metropolitan area. We expect to refine our weighting processes over time to make better use of available information and to reduce nonresponse bias in the estimates.
This survey will be deployed in a manner outside of the federal statistical system’s established, benchmark surveys. It is designed to respond to an emergent data need and produce near real-time data. The production and dissemination of these data will take place under the auspices of the Census Bureau’s Experimental Statistical Product Series (see
https://www.census.gov/data/experimental-data-products.html). As part of this process, the Census Bureau will be transparent about limitations in data quality. Despite anticipated limitations, the Census Bureau expects this survey will produce more comprehensive and timely data of higher quality than national phone surveys and anecdotal information as the agency will remain guided by its Quality Standards. The Census Bureau’s goal is accurate and unbiased estimates, and it will be transparent about the quality of the estimates obtained. The agency will continue to make improvements to the survey over time as it learns more about response rates and those responding to the survey to refine and improve statistical methods and make higher quality estimates.
