Wikipedia

Roderick J. A. Little

Roderick Joseph Alexander Little is an academic statistician, whose main research contributions lie in the statistical analysis of data with missing values and the analysis of complex sample survey data. Little is Richard D. Remington Distinguished University Professor of Biostatistics in the Department of Biostatistics at the University of Michigan, where he also holds academic appointments in the Department of Statistics and the Institute for Social Research.

Roderick J. Little
R. J. A. Little
NationalityBritish
EducationUniversity of Cambridge
Imperial College London
Scientific career
FieldsStatistics
InstitutionsUSEPA
USCB
George Washington University
University of California, Los Angeles
University of Michigan
ThesisMissing Values in Multivariate Statistical Analysis (1974)
Doctoral advisors
Doctoral students

Education

edit

Little was born near London, England, and attended secondary school at Glasgow Academy in Scotland. He received a BA in Mathematics from Gonville and Caius College, Cambridge University, and an M.Sc. in Statistics and Operational Research and Ph.D. in Statistics at Imperial College of Science and Technology, University of London. His doctoral dissertation was on the analysis of data with missing values,[1] and was supervised by Professors Martin Beale and Sir David R. Cox.

Career

edit

After a two-year post-doc in the Department of Statistics at the University of Chicago in 1974-76, Little worked at the World Fertility Survey[2] from 1976–80, under the leadership of Sir Maurice Kendall. In 1980-82 he joined a group formed by Donald Rubin at the U.S. Environmental Protection Agency in Washington DC, and in 1982-3 he was an ASA/Census/NSF Fellow at the U.S. Census Bureau and an Adjunct Associate Professor at George Washington University. In 1983-93 he was Associate Professor and later Professor in the Department of Biomathematics at UCLA. He was appointed Professor and Chair of the Biostatistics Department at the University of Michigan in 1993 and chaired the department for 11 years between 1993 and 2009, a period of intensive departmental growth.

Statistical analysis with missing data

edit

Little’s primary research interest is the analysis of data sets with missing values. Many statistical techniques are designed for complete, rectangular data sets, but in practice many data sets contain missing values, either by design or accident. In 1987, Little co-authored a book[3][4] with Donald Rubin that was one of the earliest systematic treatments of the topic; the 2nd edition was published in 2002 and the 3rd edition in 2019. As detailed in that book, initial statistical approaches to missing values were relatively ad-hoc, such as discarding incomplete cases or substituting means. The main focus of the book is on likelihood-based inferential techniques, such as maximum likelihood and Bayesian inference, based on statistical models for the data and missing-data mechanism. The 1st edition focused mainly on maximum likelihood via the expectation-maximization (EM) algorithm, but later editions emphasize Bayesian methods and the related technique of multiple imputation. Little and Rubin were awarded the prestigious Karl Pearson Prize in 2017 by the International Statistical Institute (ISI), the leading international statistics society, for a research contribution that has had “profound influence on statistical theory, methodology or applications.” The citation for the award was as follows: “The work of Roderick J. Little and Donald B. Rubin, laid out in their seminal 1978 Biometrika papers and 1987 book, updated in 2002, has been no less than defining and transforming. Earlier missing data work was ad hoc at best. Little and Rubin defined the field and provided the methodological and applied communities with a useful and usable taxonomy and a set of key results. Today, their terminology and methodology is used more than ever. Their work has been transforming for the deep impact it had and has on both statistical practice and theory. It is one of the rare topics that has continued for the past thirty years to be studied and developed in academia, government and industry. For example, it plays a key role in the current work on sensitivity analysis with incomplete data.”

Missing data research

edit

Little’s main methodological contributions to missing-data methods, in collaboration with his students and colleagues, include methods for missing data for mixtures of continuous and categorical data using the general location model,[5] pattern-mixture models[6] for data that are missing not at random, penalized spline of propensity models for missing data[7] and causal inference,[8] subsample ignorable likelihood methods[9] in regression, proxy pattern-mixture models[10] for survey nonresponse, models for longitudinal data,[11][12][13] partially missing at random models,[14] and review papers on missing data in regression,[15] hot-deck imputation,[16] and masking data for confidentiality protection.[17]

Bayesian analysis of survey data

edit

Another research area is the analysis of data collected by complex sampling designs involving stratification and clustering of units. Since working as a statistician for the World Fertility Survey, Little worked on the development of model-based methods for survey analysis that are robust to misspecification, reasonably efficient, and capable of implementation in applied settings. Contributions with students and colleagues in this area include articles on survey nonresponse,[18][19][20][21][22] Bayesian methods for survey inference,[23][24] poststratification,[25] assessing selection bias,[26] and survey weighting from a Bayesian perspective.[27][28]

Calibrated Bayesian inference

edit

Little advocates the calibrated Bayesian approach to statistical analysis,[29][30] as proposed by George Box and Donald Rubin, among others. The idea is to develop Bayesian models for analysis that yield Bayesian inferences with good frequentist properties, such as posterior credible intervals that have close to nominal coverage when viewed as confidence intervals in repeated sampling. In the survey sampling arena, this leads to models that incorporate features of the sample design in the Bayesian model. Little argues that this Bayesian framework yields a more unified approach to survey sample inference than the design-based approach, which relies on the randomization distribution underlying sample selection as the basis for inference. Little’s applied interests in statistics are broad, including mental health, demography, environmental statistics, biology, economics, medicine, public health and the social sciences, as well as biostatistics.

Activities in U.S. federal statistics

edit

Little is a strong advocate of the importance of independent government statistical agencies for democracy. He served two terms on the Committee on National Statistics of the National Academy of Sciences, and in 2010-12 was the inaugural Associate Director for Survey Research and Methodology and Chief Scientist at the U.S. Bureau of the Census, a position that has elevated scientific aspects of Census Bureau operations. He has participated in many National Academy of the Sciences panels, in particular chairing a studies on multiple sclerosis and other neurologic disorders in veterans of the Persian Gulf and Post 9/11 wars, and on the treatment of missing data in clinical trials. He has been active in advising the U.S. Food and Drug Administration and pharmaceutical companies on methods for handling missing data in clinical studies[31][32][33][34][35]

Activities for the American Statistical Association

edit

Little served two terms on the Board of Directors of the American Statistical Association (ASA), first as Editorial Representative and then as a Vice President. Editorially, he was Coordinating and Applications Editor of the Journal of the American Statistical Association in 1992-4, and later, as Chair of the Survey Research Methods section of the ASA, helped to start a new academic journal on survey statistics, the Journal of Survey Statistics and Methodology. He served as the Statistics Co-Editor in Chief for that journal in 2016-18. In 2016, Little received a Founder’s Award[36] from the ASA for his contributions to the statistics profession.

Honors

edit

Little is a Fellow of the American Statistical Association and the American Academy of Arts and Sciences, and a Member of the International Statistical Institute and the U.S. National Academy of Medicine. In 2005 he received the ASA Wilks’ memorial award for contributions to statistics. Plenary talks include the 2005 President’s Invited Address and the 2012 COPSS Fisher Lecture at the Joint Statistical Meetings, and the President’s Invited Address at the 2018 Eastern North American Region Meeting of the International Biometric Society. In 2020 he received the Marvin Zelen Leadership Award in Statistical Science from Harvard University.

References

edit
  1. ^ Beale, E. M. L.; Little, R. J. A. (1975). "Missing Values in Multivariate Analysis". Journal of the Royal Statistical Society. Series B (Methodological). 37 (1): 129–145. doi:10.1111/j.2517-6161.1975.tb01037.x. JSTOR 2984998.
  2. ^ Little, R.J.A. (1988). "Some Statistical Analysis Issues at the World Fertility Survey". The American Statistician. 42 (1): 31–36. doi:10.2307/2685258. JSTOR 2685258. PMID 12315059.
  3. ^ Mislevy, R.J. (1991). "Book Reviews: Statistical Analysis With Missing Data". Journal of Educational Statistics. 16 (2): 150–155.
  4. ^ Little, R.J.A. & Rubin, D.B. (2019). Statistical Analysis with Missing Data (3 ed.). New York: John Wiley.{{cite book}}: CS1 maint: multiple names: authors list (link)
  5. ^ Little, R.J.A. & Schluchter, M.D. (1985). "Maximum likelihood estimation for mixed continuous and categorical data with missing values". Biometrika. 72 (3): 497–512. doi:10.1093/biomet/72.3.497.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  6. ^ Little, R.J.A. (1993). "Patternmixture models for multivariate incomplete data". Journal of the American Statistical Association. 88 (421): 125–134. doi:10.2307/2290705. JSTOR 2290705.
  7. ^ Zhang, G. & Little, R. J (2009). "Extensions of the penalized spline of propensity prediction method of imputation". Biometrics. 65 (3): 911–8. doi:10.1111/j.1541-0420.2008.01155.x. hdl:2027.42/57686. PMID 19053998. S2CID 2145590.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  8. ^ Zhou, T., Elliott, M.R. & Little, R.J (19 April 2019). "Penalized Spline of Propensity Methods for Treatment Comparisons (with discussion and rejoinder)". Journal of the American Statistical Association. 114 (525): 1–38. doi:10.1080/01621459.2018.1518234. S2CID 146066305.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  9. ^ Little, R. J. & Zhang, N (2011). "Subsample ignorable likelihood for regression analysis with missing data". Journal of the Royal Statistical Society, Series C (Applied Statistics). 60 (4): 591–605. doi:10.1111/j.1467-9876.2011.00763.x. hdl:2027.42/86948. S2CID 53684702.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  10. ^ Andridge, R.H. & Little, R.J. (2011). "Proxy pattern-mixture analysis for survey nonresponse". Journal of Official Statistics. 27 (2): 153–180.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  11. ^ Little, R.J.A. & Yau, L. (1996). "Intent-to-treat analysis in longitudinal studies with drop-outs". Biometrics. 52 (4): 1324–1333. doi:10.2307/2532847. JSTOR 2532847. PMID 8962456.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  12. ^ Little, R.J.A. (1995). "Modeling the drop-out mechanism in longitudinal studies". Journal of the American Statistical Association. 90: 1112–1121. doi:10.2307/2291350. JSTOR 2291350.
  13. ^ Lange, K., Little, R.J.A. & Taylor, J.M.G. (1989). "Robust statistical modeling using the T distribution". Journal of the American Statistical Association. 84 (881896): 881–896. doi:10.2307/2290063. JSTOR 2290063.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  14. ^ Little, R.J., Rubin, D.B. & Zanganeh, S.Z. (2016). "Conditions for ignoring the missing-data mechanism in likelihood inferences for parameter subsets". Journal of the American Statistical Association. 112 (517): 314–320. doi:10.1080/01621459.2015.1136826. S2CID 126196078.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  15. ^ Little, R.J.A. (1992). "Regression with missing X's: a review". Journal of the American Statistical Association. 87 (420): 1227–1237. doi:10.2307/2290664. JSTOR 2290664.
  16. ^ Andridge*, R.H. & Little, R. J. (2010). "A review of hot deck imputation for survey nonresponse". International Statistical Review. 78 (1): 40–64. doi:10.1111/j.1751-5823.2010.00103.x. PMC 3130338. PMID 21743766.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  17. ^ Little, R.J.A. (1993). "Statistical analysis of masked data". Journal of Official Statistics. 9: 407–426.
  18. ^ Little, R.J.A. & Vartivarian, S. (2005). "Does weighting for nonresponse increase the variance of survey means?". Survey Methodology. 31: 161–168.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  19. ^ Little, R.J. & Vartivarian, S. (2003). "On weighting the rates in nonresponse weights". Statistics in Medicine. 22 (9): 1589–99. doi:10.1002/sim.1513. hdl:2027.42/34860. PMID 12704617. S2CID 25347022.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  20. ^ Little, R.J.A. (1988). "Missing data adjustments in large surveys". Journal of Business and Economic Statistics. 6 (3): 287–296. doi:10.2307/1391878. JSTOR 1391878.
  21. ^ Little, R.J.A. (1982). "Models for nonresponse in sample surveys". Journal of the American Statistical Association. 77 (378): 237–250. doi:10.2307/2287227. JSTOR 2287227.
  22. ^ Little, R.J.A. (1988). "Missing data adjustments in large surveys". Journal of Business and Economic Statistics. 6 (3): 287–296. doi:10.2307/1391878. JSTOR 1391878.
  23. ^ Little, R.J. "Calibrated Bayes: An alternative inferential paradigm for official statistics (with discussion and rejoinder)". Journal of Official Statistics. 28 (3): 309–372.
  24. ^ Little, R.J.A. (2004). "To model or not to model? competing modes of inference for finite population sampling". Journal of the American Statistical Association. 99 (466): 546–556. doi:10.1198/016214504000000467. S2CID 49574932.
  25. ^ Little, R.J.A. (1993). "Poststratification: a modeler's perspective". Journal of the American Statistical Association. 88: ification: a modeler's perspective. Journal of the American Statistical Association 88. doi:10.2307/2290705. JSTOR 2290705.
  26. ^ Little, R.J., West, B.T., Boonstra, P.S. & Hu, J. (2019). "Measures of the Degree of Departure from Ignorable Sample Selection". Journal of Survey Statistics and Methodology. 8 (5): 932–964. doi:10.1093/jssam/smz023. PMC 7750890. PMID 33381610.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  27. ^ Zheng, H. & Little, R.J. (2003). "Penalized spline model-based estimation of the finite population total from probability-proportional-to-size samples". Journal of Official Statistics. 19 (2): 99–117.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  28. ^ Elliott, M. R. & Little, R.J.A. (2000). "Model-based alternatives to trimming survey weights". Journal of Official Statistics. 16 (3): 191–209.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  29. ^ Little, R.J.A. (2006). "Calibrated Bayes: A Bayes/frequentist Roadmap". The American Statistician. 60 (3): 213–223. doi:10.1198/000313006X117837. S2CID 53505632.
  30. ^ Little, R.J. "Calibrated Bayes: An alternative inferential paradigm for official statistics (with discussion and rejoinder)". Journal of Official Statistics. 28 (3): 309–372.
  31. ^ Little, R.J. & Rubin, D.B. (2000). "Causal effects in clinical and epidemiological studies via potential outcomes: concepts and analytical approaches". Annual Review of Public Health. 21: 121–145. doi:10.1146/annurev.publhealth.21.1.121. PMID 10884949.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  32. ^ Little, R.J., Wang, J., Sun, X., Tian, H., Suh, E-Y., Lee, M., Sarich, T., Oppenheimer, L., Plotnikov, A., Wittes, J., Cook-Bruns, N., Burton, P., Gibson, M., & Mohanty, S. (2016). "Oppenheimer, L., Plotnikov, A., Wittes, J., Cook-Bruns, N., Burton, P., Gibson, M., & Mohanty, S. (2016). The treatment of missing data in a large cardiovascular clinical outcomes study". Clinical Trials. 13 (3): 344–351. doi:10.1177/1740774515626411. PMID 26908543. S2CID 41268081.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  33. ^ Little, R.J. & Kang, S. (2015). "Intention-to-treat analysis with treatment discontinuation and missing data in clinical trials". Statistics in Medicine. 34 (16): 2381–2390. doi:10.1002/sim.6352. hdl:2027.42/112012. PMID 25363683. S2CID 8735358.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  34. ^ Little, R.J., Long, Q. & Lin, X. (2009). "A comparison of methods for estimating the causal effect of a treatment in randomized clinical trials subject to noncompliance". Biometrics. 65 (2): 640–9. doi:10.1111/j.1541-0420.2008.01066.x. hdl:2027.42/65200. PMID 18510650. S2CID 4843005.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  35. ^ Little, R.J., D’Agostino, R., Cohen, M.L., Dickersin, K., Emerson, S.S., Farrar, J.T., Frangakis, C., Hogan, J.W., Molenberghs, G., Murphy, S.A., Rotnitsky, A., Scharfstein, D., Neaton, J.D., Shih, W., Siegel, J.P., Stern, H. (2012). "Special Report: The prevention and treatment of missing data in clinical trials". New England Journal of Medicine. 367 (14): 1355–1360. doi:10.1056/NEJMsr1203730. PMC 3771340. PMID 23034025.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  36. ^ "Founders Award". www.amstat.org. American Statistical Association.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Roderick_J._A._Little&oldid=1217724455"