{"id":"W4387532544","doi":"10.1098/rsos.230568","title":"Reducing bias in secondary data analysis via an Explore and Confirm Analysis Workflow (ECAW): a proposal and survey of observational researchers","year":2023,"lang":"en","type":"article","venue":"Royal Society Open Science","topic":"Health, Environment, Cognitive Aging","field":"Environmental Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Medical Research Council; Canadian Institutes of Health Research; Arnold Ventures","keywords":"Observational study; Workflow; Data science; Computer science; Data mining; Database; Statistics; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02032005,0.0001542588,0.0003897345,0.0001955479,0.0004946862,0.0002514972,0.001685111,0.00006963304,0.0005097351],"category_scores_gemma":[0.0008008837,0.0001489017,0.0000515869,0.009405593,0.001943944,0.001571646,0.003767743,0.0002912612,0.00001324338],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002057184,"about_ca_system_score_gemma":0.000291239,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03787276,"about_ca_topic_score_gemma":0.01526555,"domain_scores_codex":[0.9959631,0.0004997354,0.0004239435,0.001504742,0.001044955,0.0005635058],"domain_scores_gemma":[0.9980609,0.0004998915,0.0001711411,0.0009013182,0.00003723628,0.0003295371],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00001276208,0.00004936188,0.969993,0.000008941334,0.000101641,0.000003015274,0.003094819,0.005677977,0.0006943247,0.000004818694,0.00009544757,0.02026391],"study_design_scores_gemma":[0.0001613557,0.00002743751,0.6708151,0.000007747863,0.00006160168,2.433521e-7,0.001108592,0.3275619,0.00005900554,0.00006305626,0.00001989052,0.0001140764],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9975576,0.00002667559,0.00094868,0.0004470074,0.00001401764,0.0004482241,0.0001285168,0.00001409472,0.0004152126],"genre_scores_gemma":[0.9902695,0.00008630803,0.00900311,0.0001700169,0.000006323531,0.00002099068,0.0002965833,0.000009855651,0.0001373259],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3218839,"threshold_uncertainty_score":0.9685341,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3171527591360382,"score_gpt":0.4148605485946184,"score_spread":0.09770778945858016,"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."}}