The Urban Institute has automated the process of reading in and cleaning the Household Pulse Survey Public Use Microdata files. We have combined the available public use files across all weeks into one standardized CSV (
Public Use File All Weeks.csv) and generated a rolling two week summary file (
Two Week Rolling Avg Pulse Metrics.csv). that powers the Tracking COVID-19’s Effects by Race and Ethnicity feature. These datasets will be updated weekly as new public use files are released. Below are descriptions of all available files. To see the code for how we generated these data files, please visit our Github repo.
Public Use Files All Weeks.csv: A compiled CSV of public use files from all available weeks.. The file structure will be very similar to the Census Bureau’s published files but will have a few new columns appended. Each row represents a respondent from a given week. Please carefully note the
pweightcolumn which is the person weight assigned by the Census Bureau to each respondent. Any metric you create from the data has to sum across the
pweightcolumn to arrive at the individual level estimates. For additional help understanding what each of the columns in this file represent, please see the data dictionary files below.
PUF Census Data Dictionary.: The data dictionary as provided by the Census Bureau for the latest week in the data. This will be exactly the same as the Census Bureau's published data dictionary files. Note a couple of the Urban appended columns will not be in the data dictionary. For definitions of the appended columns, please see the PUF Urban Data Dictionary file
PUF Urban Data Dictionary: Urban has appended a few columns to the publicly available public use file. This file contains the descriptions of those appended columns. This file in conjunction with the PUF Census Data Dictionary should allow you to understand what each column means in the
Public Use File All Weeksfile.
Two Week Rolling Avg Pulse Metrics: This is a two week rolling average summary file of a few selected metrics from the Household Pulse Survey. Each row in this file represents a indicator/geography/week/race combination. For more information on what each of the columns in this file represent, see the
Two Week Rolling Avg Data Dictionary.
Two Week Rolling Avg Data Dictionary: The data dictionary for the above
Two Week Rolling Avg Pulse Metricsdataset.
If you have any questions, please reach out to firstname.lastname@example.org
Data and Resources
|Data Dictionary Files|
|Original Data Source|
|Citation Requirements|| |
Urban Institute. 2020. US Census Bureau Household Pulse Survey public use files. Accessible from https://datacatalog.urban.org/dataset/census-pulse-public-use-files. Data originally sourced from Census Household Pulse Survey, developed at the Urban Institute, and made available under the ODC-BY 1.0 Attribution License.