The Uptick in Income Segregation: Real Trend or Random Sampling Variation?

Area of Science:

• Sociology
• Urban Studies
• Demography

Background:

• Recent studies indicate a reversal in income segregation trends in U.S. metropolitan regions, shifting from decline to increase.
• This trend aligns with growing concerns about U.S. income inequality since the 1970s.
• However, evidence may be biased due to smaller American Community Survey (ACS) sample sizes compared to Census 2000.

Purpose of the Study:

• To investigate the impact of varying sampling rates on income segregation estimates.
• To propose and test bias-correction methods for segregation measures.
• To provide more accurate insights into U.S. income segregation trends.

Main Methods:

• Utilized 100% microdata from the 1940 census to simulate sampling rate impacts.
• Developed and tested bias-correction approaches for segregation measures.
• Applied correction methods to U.S. Census 2000 and American Community Survey (ACS) data (2007-2011).

Main Results:

• Reduced sample sizes in the ACS exaggerate the evidence of increasing income segregation.
• This exaggeration is particularly pronounced for specific subgroups, such as African Americans.
• Tested bias-correction methods showed potential for yielding more conclusive and unbiased results.

Conclusions:

• The apparent rise in income segregation may be partly an artifact of sampling biases in the ACS.
• Bias-correction techniques are crucial for accurate analysis of segregation trends.
• Further application of these methods to internal Census Bureau data is recommended for definitive findings.