Update May 8, 2020: Initial results are scheduled for release May 14, 2020.
On April 22, OMB gave the U.S. Census Bureau emergency 180-day approval to conduct a weekly Small Business Pulse Survey (SBPS) of small nonfarm single-establishment firms (1-499 employees) on the impacts of the COVID-19 pandemic. There are 940,588 target businesses with a known email address. The SBPS will cover this universe over nine weeks -- approximately 105,000 firms will be contacted each week, beginning the week of April 26th. The estimated response rate is 15-20%. To the extent possible, SBPS data will be published by industry sector for the nation, state (including DC and Puerto Rico), and 50 largest Metropolitan Statistical Areas (MSAs). The Census Bureau will ask to repeat the survey cycle as needed and useful.
The 15 survey questions are available here: https://www.reginfo.gov/public/do/DownloadDocument?objectID=100501600
. Links to other key documents are at the end of this posting.
The Census Bureau has developed the SBPS as an experimental endeavor in cooperation with six other federal agencies. In response to Census Bureau outreach, content was recommended by the Small Business Administration (SBA), the Federal Reserve Board (FRB), the Minority Business Development Agency (MBDA), and the International Trade Administration (ITA). Proposed survey content also was reviewed by the Bureau of Labor Statistics (BLS) and the Bureau of Economic Analysis (BEA).
Testing this platform during the COVID-19 pandemic will allow the federal statistical system to demonstrate proof of concept with respect to the small business pulse survey platform by providing states with weekly data about the economic impacts on business operations. The ability to understand how small businesses are modifying their business models or shuttering and whether small businesses are seeking and receiving the financial assistance they need is important. 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.
This collection will be fielded with an expectation that there will likely be changes to the questions and weighting algorithms. The first week of data collection is designed to stress test the system and collect information about response to enable the development of the programs necessary to provide estimates that are representative at the national and state level. Between week one and week two of data collection, the programs to generate the weights will be run. Simultaneous with the first week of data collection, Census will cognitively test the identified new questions, with a focus on improving the comprehension and clarity, particularly in reference to aspects of the pandemic response. The goal is to have refined questions ready to field by week 2. It is also possible that the wording of some questions may be tweaked between weeks 2 and 3 and that new questions within the already approved question domains may be added.
Because this is an experimental launch of the Small Business Pulse Survey platform and approach, all results will be available on the Census’ Experimental Data Products site (https://www.census.gov/data/experimental-data-products.html#
). We clearly identify experimental data products and include methodology and supporting research with their release.
The only method of collecting information for this survey is electronically through the Census Bureau’s online reporting system, Centurion. The collection instrument is optimized for mobile response to further reduce respondent burden. We deem this the most efficient and least burdensome way to collect the information.
The universe will be subdivided into nine groups. Each group of ~100k businesses will receive an email with a survey link one week out of the nine week data collection period. This will minimize respondent burden, while also providing the data needed to produce detailed estimates by sector and state and for the top 50 MSAs. The collection instrument has been optimized for electronic response, including the option to respond by smartphone. This minimizes burden, such that we estimate 5 minutes or less to respond. All questions are checkbox responses.
We expect to complete 9 weeks of data collection. Additional weeks of collection could be possible, and we will work in coordination with OMB to make that determination.
Efforts will be made to account for potential biases in survey estimates due to the non-probabilistic sample by applying a weighting class adjustment. We are working to develop this adjustment in parallel with initiating data collection, and we expect that the adjustment methods may change as we learn through this collection process.
The Census Bureau envisions short, medium, and longer-term applications of the SBPS data. First, in the short-term data tabulations and visualizations will be disseminated within approximately 3 weeks of the reference period and weekly beyond that. Second, in the medium term, research data products linking census, survey and/or administrative records data to SBPS responses will be created to more fully understand and provide context for the SBPS data. Last, in the long term the SBPS linked data will be used to understand the relationship between these types of questions and important business outcomes including employment, productivity, growth, and survival in order to understand whether this platform can be used in real time to extract meaningful, policy relevant information. In the long term is when it will also be useful to link SBPS data to any program evaluation research.
In the short-term, data will be represented in a user-friendly graphical format and available at https://www.census.gov/businesspulsedata
as a featured experimental data product. All tabulations, including those for which visualizations are not created, will be downloadable so that users may create their own reports or visualizations. Users of the tabulations and visualizations may include policy makers within the above mentioned agencies as well academic experts and the media. The first publication of visualizations will not be available until week 3, as we will not close the processing cycle for week 1 until week 3. After week 3, we will establish a weekly publication cycle.
Medium-term research data products will allow us to provide more detailed information about the businesses during the COVID-19 pandemic without creating any additional reporting burden on small business owners. For example, linking the SBPS to the Annual Business Survey, which includes detailed information about the gender, race, and veterans status of business owners, will allow the Census Bureau to provide data on any differential impacts by these characteristics. Linking the SBPS to our data infrastructure will allow us to control for business size, age, access to credit before the pandemic, and myriad other characteristics of businesses and their owners in order to better isolate the impact of the pandemic on businesses.
In the longer term, the Census Bureau will use the linked SBPS to understand whether low burden, high frequency collections such as the SBPS are predictive of future state business outcomes. For example, if business applied for but did not receive federal assistance is this a significant predictor of that business’s future growth or survival? Because some of the business outcomes we are interested in are potentially longer term than the survey, we cannot fully evaluate the usefulness of the SBPS in the short or medium terms.
The Census proposal to OMB is at https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=202004-0607-003
Supporting Statements describing purpose, content, plans, and methods are at https://www.reginfo.gov/public/do/PRAViewDocument?ref_nbr=202004-0607-003
Attachments A-I are at https://www.reginfo.gov/public/do/PRAViewIC?ref_nbr=202004-0607-003&icID=241468
A. Screenshots of Content from Small Business Pulse Survey on Survey Instruments
B. Small Business Pulse Survey Data Visualization Mockup
C. Small Business Pulse Survey Landing Page
D. Crosswalk of Survey Content and Agency Requesting Item
E. Small Business Pulse Survey Content and Uses
F. Small Business Pulse Survey: Communicating Data Quality and Treatment as Experimental Data Product
G. Guiding Principles: Experimental Statistical Products
H. Small Business Pulse Survey Methodology
I. Small Business Pulse Survey Initial and Follow-up Emails