{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.06124321,0.001185196,0.00248496,0.0031501,0.001391927,0.002452042,0.003149709,0.001392304,0.002869177],"category_scores_gemma":[0.1540936,0.001273092,0.003165914,0.007169828,0.0007044628,0.001432355,0.002226352,0.003345574,0.00102508],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001607162,"about_ca_system_score_gemma":0.00396381,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02344902,"about_ca_topic_score_gemma":0.02181973,"domain_scores_codex":[0.947026,0.03456777,0.006566262,0.005355237,0.004648597,0.00183614],"domain_scores_gemma":[0.9261656,0.03742551,0.01231163,0.01404735,0.009439471,0.0006104172],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00109948,0.0004568563,0.693222,0.00186293,0.003489369,0.001297512,0.002329181,0.03629732,0.001101251,0.0149354,0.05687619,0.1870326],"study_design_scores_gemma":[0.0008052703,0.001272028,0.3969399,0.00506696,0.004556836,0.002194769,0.003343602,0.3863215,0.00896538,0.07430086,0.1157559,0.0004769824],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2698846,0.006741065,0.6418858,0.005851067,0.001888174,0.002151761,0.06102037,0.001560897,0.009016327],"genre_scores_gemma":[0.7103536,0.002306244,0.2343169,0.003230772,0.0004368002,0.003290981,0.04249426,0.0003404483,0.00323007],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.06124321,"threshold_uncertainty_score":0.3238889,"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."}}