{"id":"W4210739057","doi":"10.6000/1929-6029.2022.11.01","title":"Multiple Imputation of Missing Race and Ethnicity in CDC COVID-19 Case-Level Surveillance Data","year":2022,"lang":"en","type":"article","venue":"International Journal of Statistics in Medical Research","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institutes of Health","keywords":"Imputation (statistics); Coronavirus disease 2019 (COVID-19); Missing data; Ethnic group; Race (biology); Statistics; Computer science; Political science; Sociology; Medicine; Mathematics; Gender studies","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01272141,0.00008317542,0.0003557203,0.0006615298,0.00006564289,0.00002320703,0.0007152347,0.00005271841,0.0005737655],"category_scores_gemma":[0.07557517,0.00008018329,0.0000233326,0.0004092045,0.0003619436,0.000118957,0.0008679351,0.001282698,0.000001145943],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005883366,"about_ca_system_score_gemma":0.002302239,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001513535,"about_ca_topic_score_gemma":0.001986188,"domain_scores_codex":[0.9937603,0.001069322,0.0009030322,0.000252331,0.003775886,0.0002391358],"domain_scores_gemma":[0.991065,0.007062426,0.0002981001,0.000257977,0.0008680155,0.000448457],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.004171193,0.00123591,0.5715573,0.0004064165,0.0001989209,0.1241639,0.001728075,0.0009964939,0.0004643833,0.001001737,0.02026426,0.2738113],"study_design_scores_gemma":[0.01535605,0.0007415249,0.6174093,0.0005986054,0.00002064379,0.02062381,0.003034447,0.3171814,0.00004861986,0.009308111,0.01540384,0.0002736643],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8463566,0.001519472,0.1222028,0.01790229,0.0006520748,0.0004564613,0.01074757,0.00001018423,0.0001524797],"genre_scores_gemma":[0.9861031,0.0005417982,0.01247707,0.0003352001,0.000105022,0.000004368134,0.0003995486,0.00001247085,0.00002149568],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3161849,"threshold_uncertainty_score":0.9322116,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2161426987061239,"score_gpt":0.5172899677975115,"score_spread":0.3011472690913876,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}