Description

REPWT provides 80 separate household-level weights that allow users to generate empirically derived standard errors. Person-level replicate weights are available in REPWTP.

More information about replicate weights is available on the IPUMS-USA replicate weights FAQ page, in the 2005 ACS PUMS Accuracy Statement, and in this Census Bureau document written for the Current Population Survey.

Calculating the standard error of an estimate enables the construction of a confidence interval around the sample estimate of interest and may also be used in hypothesis testing. In theory, the standard error of an estimate measures the variation of a statistic across multiple samples of a given population. Researchers can use replicate weights to mirror this theoretical approach when only sample data is available, and the resulting standard errors have a higher degree of precision than standard asymptotic standard errors.

The 2005-onward ACS and PRCS samples contain eighty replicate weights at the household level (variables named REPWT1 through REPWT80) and eighty at the person level (variables named REPWTP1 through REPWTP80). The Census Bureau produced these weights by using what is known as the successive difference replication (SDR) method. This involves repeated implementations of the initial (full-sample) weighting algorithm, such that full information about the ACS and PRCS samples are available in the replicate weights. Nevertheless, users should use these replicate weights only for generating variance estimates, not for obtaining unique parameter estimates.

User Note: The successive difference replication approach (SDR) is different from other methods for creating replicate weights such as balanced repeated replication (BRR) and jackknife estimation, and standard statistical software packages have no built-in method to handle them. However, Stata's jackknife standard error program can be adapted to calculate replicate standard errors for CPS data; see the IPUMS-USA replicate weights FAQ page for details.

Additionally, it is possible for replicate weights to take negative values for certain cases; again, users should use these weights only for variance estimation purposes and not to obtain independent estimates.

When REPWT is selected for data extraction, 80 replicate weights, REPWT1-REPWT80 are included in the data extract. REPWT1-REPWT80 are 4-digit numeric variables used to empirically derive standard errors. Selecting replicate weights will dramatically increase the size and processing time of extracts; users should request them only if they plan to use them. REPWT 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).

REPWT Specific Variable Codes

Comparability

Replicate weights are only available for the 2005-onward samples of the ACS and the PRCS. For all other years, users should use SUBSAMP.

User Note: Users should note that in the 2020 ACS 1-year file the standard REPWT variables were replaced with experimental weights that were designed to account for the effects of the COVID-19 pandemic on the 2020 1-year ACS data quality. The experimental weights use the same name and are in place of the standard REPWT variables, so selecting the REPWT variables for the 2020 ACS 1-year file will select the household-level replicate experimental weights for this sample. For more information, please see ACS and COVID-19: Guidance for Using the PUMS with Experimental Weights.

Universe

Availability

• 2022: All samples
• 2021: All samples
• 2020: All samples
• 2019: All samples
• 2018: All samples
• 2017: All samples
• 2016: All samples
• 2015: All samples
• 2014: All samples
• 2013: All samples
• 2012: All samples
• 2011: All samples
• 2010: ACS; ACS 3yr; ACS 5yr
• 2009: All samples
• 2008: All samples
• 2007: All samples
• 2006: All samples
• 2005: All samples
• 2004: --
• 2003: --
• 2002: --
• 2001: --
• 2000: --
• 1990: --
• 1980: --
• 1970: --
• 1960: --
• 1950: --
• 1940: --
• 1930: --
• 1920: --
• 1910: --
• 1900: --
• 1880: --
• 1870: --
• 1860: --
• 1850: --

• 2022: All samples
• 2021: All samples
• 2020: All samples
• 2019: All samples
• 2018: All samples
• 2017: All samples
• 2016: All samples
• 2015: All samples
• 2014: All samples
• 2013: All samples
• 2012: All samples
• 2011: All samples
• 2010: PRCS; PRCS 3yr; PRCS 5yr
• 2009: All samples
• 2008: All samples
• 2007: All samples
• 2006: All samples
• 2005: All samples
• 2000: --
• 1990: --
• 1980: --
• 1970: --
• 1930: --
• 1920: --
• 1910: --

Flags

Editing Procedure

There is no editing procedure available for this variable.