{"id":"W4386000296","doi":"10.31235/osf.io/ewamx","title":"Quality Control for Quality Computational Concepts: Wrangling with Theory and Data Wrangling as Theorizing","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute on Governance","funders":"","keywords":"Formality; Quality (philosophy); Craft; Process (computing); Computer science; Control (management); Data quality; Management science; Data science; Epistemology; Artificial intelligence; Political science; Engineering; Law; Operations management","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1781822,0.001474054,0.00279393,0.0139947,0.006855861,0.02870829,0.006642409,0.005462416,0.004193919],"category_scores_gemma":[0.4847933,0.002102864,0.003167886,0.0104593,0.1139142,0.06001906,0.02181849,0.01282249,0.0005148734],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.01115841,"about_ca_system_score_gemma":0.01023795,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005029445,"about_ca_topic_score_gemma":0.002490258,"domain_scores_codex":[0.8175778,0.1236124,0.01079242,0.01630227,0.02944084,0.00227439],"domain_scores_gemma":[0.3509796,0.4873749,0.02714036,0.1063321,0.0249637,0.003209384],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00004163974,0.00002698518,0.001420275,0.000178416,0.00004168193,0.00005185313,0.006709617,0.002553102,0.0002249545,0.9713086,0.0006074692,0.01683536],"study_design_scores_gemma":[0.00001624891,0.00001500055,0.0001918243,0.000164557,0.00001450924,0.00003265593,0.0009160046,0.005939247,0.0005094135,0.9884553,0.003721523,0.00002377041],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02149154,0.001766517,0.9432096,0.02429768,0.0002222165,0.0002065439,0.0001196988,0.0003590502,0.008327094],"genre_scores_gemma":[0.5131421,0.00105769,0.4809625,0.00170453,0.0003611049,0.0007384556,0.000225969,0.0004823298,0.001325408],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8218178,"threshold_uncertainty_score":0.9423289,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2420541306345986,"score_gpt":0.5429487259625833,"score_spread":0.3008945953279847,"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."}}