Data Cart

Your data extract

0 variables
0 samples
View Cart
EXPWTP
Experimental person weight

Description

EXPWTP indicates how many persons in the U.S. population are represented by a given person in an IPUMS sample. This is an experimental person-level weight that was developed to correct for the impact of the COVID-19 pandemic on the 2020 American Community Survey data collection.

These weights incorporate additional administrative data, including income data from the Internal Revenue Service (IRS), benefit data from the Social Security Administration (SSA), demographic data from the 2010 Census, industry data for the Census Bureau’s Business Register, and third-party home value data. These data were then incorporated into a two-stage reweighting procedure called Entropy Balancing. Stage one reweights respondent housing units to control for selection into response. Stage two, an estimation of individual weights to match external population totals, applies constraints to preserve the distribution of housing unit characteristics, the notion of spousal equivalence, external population targets by age, race, sex, and Hispanic origin, and to balance weights within and across months. For more information on the creation of entropy-balanced weights, see section 5 of the working paper, Addressing nonresponse bias in the American Community Survey during the pandemic using administrative data.

In June of 2022, the U.S. Census Bureau released experimental weights for the 2019 1-year ACS PUMS to enable users to apply both production weights and experimental weights to analyses that utilize the 2019 1-year ACS PUMS data. For this reason, IPUMS USA made these experimental weights (EXPWTH and EXPWTP) available as separate weights from the standard ‘production’ weights (HHWT and PERWT). 

Users should note that for the 2020 ACS PUMS samples, the experimental weights were released in place of the standard production weights; therefore, the experimental weights are available for 2020 ACS samples in two locations, under standard weight variable names (HHWT and PERWT) and experimental weight variables names (EXPWTH and EXPWTP). 

Codes

EXPWTP is a 5-digit numeric variable which indicates how many persons in the U.S. population are represented by a given person in an IPUMS sample. EXPWTP specific variable codes for missing, edited, or unidentified observations, observations not applicable (N/A), observations not in universe (NIU), top and bottom value coding, etc. are provided below if applicable by Census year (and data sample if specified).

EXPWTP Specific Variable Codes
0 = N/A (Not in Universe)

Comparability

User Note: Users combining multiple samples from the same year must make an adjustment to EXPWTP in order to get accurate population estimates. When combining two 1% samples from a given year (for a total of a 2% sample), users need to multiply the EXPWTP values in both samples by 0.5, so that the weights will inflate the two samples together to represent that year's entire U.S. population. Similarly, when combining a given year's 1% and 5% samples together (for a total of a 6% sample), users must multiply EXPWTP values in the 1% sample by 0.167 (1/6) and EXPWTP values in the 5% sample by 0.833 (5/6).

Universe

  • 2019 ACS: Not group quarters.
  • 2020 - onward ACS/PRCS: All persons.

Availability

United States
  • 2021: --
  • 2020: All samples
  • 2019: ACS
  • 2018: --
  • 2017: --
  • 2016: --
  • 2015: --
  • 2014: --
  • 2013: --
  • 2012: --
  • 2011: --
  • 2010: --
  • 2009: --
  • 2008: --
  • 2007: --
  • 2006: --
  • 2005: --
  • 2004: --
  • 2003: --
  • 2002: --
  • 2001: --
  • 2000: --
  • 1990: --
  • 1980: --
  • 1970: --
  • 1960: --
  • 1950: --
  • 1940: --
  • 1930: --
  • 1920: --
  • 1910: --
  • 1900: --
  • 1880: --
  • 1870: --
  • 1860: --
  • 1850: --
Puerto Rico
  • 2021: --
  • 2020: All samples
  • 2019: --
  • 2018: --
  • 2017: --
  • 2016: --
  • 2015: --
  • 2014: --
  • 2013: --
  • 2012: --
  • 2011: --
  • 2010: --
  • 2009: --
  • 2008: --
  • 2007: --
  • 2006: --
  • 2005: --
  • 2000: --
  • 1990: --
  • 1980: --
  • 1970: --
  • 1930: --
  • 1920: --
  • 1910: --

Flags

This variable has no flags.

Editing Procedure

There is no editing procedure available for this variable.