The topic of this article is one-sided hypothesis testing for disparity, i.e., the mean of one group is larger than that of another when there is uncertainty as to which group a datum is drawn. For each datum, the uncertainty is captured with a given discrete probability distribution over the groups. Such situations arise, for example, in the use of Bayesian imputation methods to assess race and ethnicity disparities with certain insurance, health, and financial data. A widely used method to implement this assessment is the Bayesian Improved Surname Geocoding (BISG) method which assigns a discrete probability over six race/ethnicity groups to an individual given the individual’s surname and address location. Using a Bayesian framework and Markov Chain Monte Carlo sampling from the joint posterior distribution of the group means, the probability of a disparity hypothesis is estimated. Four methods are developed and compared with an illustrative data set. Three of these methods are implemented in an R-code and one method in WinBUGS. These methods are programed for any number of groups between two and six inclusive. All the codes are provided in the appendices.
References
[1]
McDonald, G.C. (2018) Statistical Testing When the Populations from Which Samples Are Drawn Are Uncertain. Health Services and Outcomes Research Methodology, 18, 155-174. https://doi.org/10.1007/s10742-018-0182-7
[2]
Elkadry, A. and McDonald, G.C. (2020) Hypothesis Testing When Data Sources Are Uncertain. Journal of Statistical Theory and Practice, 14, Article No. 66. https://doi.org/10.1007/s42519-020-00132-5
[3]
McDonald, G.C. and Oakley, R.H. (2023) Extending Computations for Disparity Testing When Data Sources Are Uncertain. Health Services and Outcomes Research Methodology, 23, 207-226. https://doi.org/10.1007/s10742-022-00286-8
[4]
McDonald, G.C. and Willard, J.F. (2023) Bootstrap Approach to Disparity Testing with Source Uncertainty in the Data. Health Services and Outcomes Research Methodology. https://doi.org/10.1007/s10742-023-00318-x
[5]
Navidi, W. (2024) Statistics for Engineers and Scientists. 6th Edition, McGraw Hill, New York.
[6]
Fiscella, K. and Fremont, A.M. (2006) Use of Geocoding and Surname Analysis to Estimate Race and Ethnicity. Health Services Research, 41, 1482-1500. https://doi.org/10.1111/j.1475-6773.2006.00551.x
[7]
Adjaye-Gbewonyo, D., Bednarczyk, R.A., Davis, R.L. and Omer, S.B. (2014) Using the Bayesian Improved Surname Geocoding Method (BISG) to Create a Working Classification of Race and Ethnicity in a Diverse Managed Care Population: A Validation Study. Health Services Research, 49, 268-283. https://doi.org/10.1111/1475-6773.12089
[8]
Brown, D.P., Knapp, C., Baker, K. and Kaufmann, M. (2016) Using Bayesian Imputation to Assess Racial and Ethnic Disparities in Pediatric Performance Measures. Health Services Research, 51, 1095-1108. https://doi.org/10.1111/1475-6773.12405
[9]
Consumer Financial Protection Bureau (2013, December 20) CFPB and DOJ Order Ally to Pay $80 Million to Consumers Harmed by Discriminatory Auto Loan Pricing. https://www.consumerfinance.gov/enforcement/actions/ally-financial-ally-bank/
[10]
Consumer Financial Protection Bureau (2014) Using Publicly Available Information to Proxy for Unidentified Race and Ethnicity. https://www.consumerfinance.gov/data-research/research-reports/using-publicly-available-information-to-proxy-for-unidentified-race-and-ethnicity/
[11]
Elliott, M.N., Morrison, P.A., Fremont, A., McCaffrey, D.F., Pantoja, P. and Lurie, N. (2009) Using the Census Bureau’s Surname List to Improve Estimates of Race/Ethnicity and Associated Disparities. Health Services and Outcomes Research Methodology, 9, 69-83. https://doi.org/10.1007/s10742-009-0047-1
[12]
McDonald, K.M. and Rojc, K.J. (2015) Automotive Finance Regulation: Warning Lights Flashing. The Business Lawyer, 70, 617-624.
[13]
McDonald, K.M. and Rojc, K.J. (2017) Accelerating Regulation of Automotive Finance. The Business Lawyer, 72, 559-566.
[14]
McDonald, K.M. and Rojc, K.J. (2022) Ladies and Gentlemen: Rev Your Regulatory Engines! The Business Lawyer, 77, 581-590.
[15]
Zhang, Y. (2018) Assessing Fair Lending Risks Using Race/Ethnicity Proxies. Management Science, 64, 178-197. https://doi.org/10.1287/mnsc.2016.2579
[16]
Zavez, K., Harel, O. and Aseltine, R.H. (2022) Imputing Race and Ethnicity in Healthcare Claims Databases. Health Services and Outcomes Research Methodology, 22, 493-507. https://doi.org/10.1007/s10742-022-00273-z
[17]
Voicu, I. (2018) Using First Name Information to Improve Race and Ethnicity Classification. Statistics and Public Policy, 5, 1-13. https://doi.org/10.1080/2330443X.2018.1427012
[18]
Albert, J. (2009) Bayesian Computation with R. 2nd Edition, Springer, New York. https://doi.org/10.1007/978-0-387-92298-0
[19]
Gill, J. (2015) Bayesian Methods: A Social and Behavioral Sciences Approach. 3rd Edition, CRC Press, Boca Raton.
[20]
Christensen, R., Johnson, W., Branscum, A. and Hanson, T.E. (2011) Bayesian Ideas and Data Analysis. CRC Press, Boca Raton. https://doi.org/10.1201/9781439894798
[21]
Albert, J. and Hu, J. (2020) Probability and Bayesian Modeling. CRC Press, Boca Raton. https://doi.org/10.1201/9781351030144
[22]
Lunn, D., Jackson, C., Best, N., Thomas, A. and Spiegelhalter, D. (2013) The BUGS Book: A Practical Introduction to Bayesian Analysis. CRC Press, Boca Raton. https://doi.org/10.1201/b13613
[23]
Rizzo, M.L. (2008) Statistical Computing with R. Chapman & Hall/CRC, Boca Raton.
[24]
Hahn, E.D. (2014) Bayesian Methods for Management and Business: Pragmatic Solutions for Real Problems. Wiley, Hoboken.
[25]
Woodworth, G.G. (2004): Biostatistics: A Bayesian Introduction. Wiley, Hoboken.
