Note on the Standardization of ACS/PRCS Income Variables and Other Dollar Amount Variables
In the decennial censuses, the reference period for all income variables is the previous calendar year. In the 2000 census, for instance, respondents reported their income for the 1999 calendar year.
The American Community Survey/Puerto Rican Community Survey is administered throughout the year, and many of the income and dollar amount questions refer to the 12 months prior to the date of survey. Thus, respondents will have one of 12 different reference periods (January to December, February to January, March to February, etc.), depending on the date when they were surveyed.
The Census Bureau provides a separate variable called ADJUST, which adjusts dollar amounts to the amount that they would have been had they been earned entirely during the calendar year. Ideally, this adjustment factor would be unique to each month of data. Consider the example of the 2008 ACS, released in the fall of 2009 but gathered throughout 2008: people surveyed in January 2008 earned all of their stated income during 2007 (January 2007 to December 2007), while people surveyed in December earned most of their stated income during 2008 (December 2007 to November 2008). However, month-specific adjustment factors would make it easier for individuals to be identified, so the Census Bureau does not provide them.
Instead, the Census Bureau averages the twelve month-specific adjustment factors to create a single adjustment factor that is constant for all cases. Census Bureau documentation advises users to multiply all income and dollar amount variables by the ADJUST constant, which converts numbers into calendar year dollars.1
Users should note that a constant adjustment factor cannot adjust dollar amounts to calendar-year dollars. Dollars earned during the previous year (as for January respondents) need to be adjusted the most; dollars earned mostly during the current year (as for December respondents) are essentially in calendar-year dollars already and need to be adjusted only slightly. Thus applying the Census Bureau's constant adjustment factor will still provide underestimates of incomes for January respondents and overestimates of incomes for December respondents. Furthermore, in IPUMS testing across multiple years of data, the unadjusted and the adjusted dollar amount variables were essentially perfectly correlated, and regression coefficients were not altered substantially: the differences across years in the adjustment factors are too small to affect results.
For this reason, IPUMS does not suggest that the Census Bureau's adjustment factors be used. However, they can be obtained by selecting the ADJUST variable. For multi-year samples, the adjustment factor is constant within unique values of MULTYEAR. The following dollar-amount variables with a year-long reference period may be adjusted if users desire:2
The table below shows the value of the Census Bureau's adjustment factor (available in ADJUST) for each ACS/PRCS sample year.
1. Until November 2009, IPUMS provided its own adjustment factor to convert unstandardized dollar amounts into calendar year dollars, based on inaccurate Census Bureau documentation claiming that ADJUST actually standardized dollar amounts into an unspecified month in the calendar year. Users who extracted such variables from ACS or PRCS samples before November 2009 should re-submit their extracts to get the most accurate data.