{"id":"W2766861252","doi":"10.1093/reseval/rvx037","title":"Using contribution analysis to evaluate the impacts of research on policy: Getting to ‘good enough’","year":2017,"lang":"en","type":"article","venue":"Research Evaluation","topic":"Evaluation and Performance Assessment","field":"Decision Sciences","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"Impact; University of Waterloo","funders":"University of Waterloo; Canadian Cancer Society","keywords":"Management science; Regional science; Operations research; Public economics; Political science; Psychology; Sociology; Economics; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":["metaresearch"],"domain":"evaluation","study_design":"theoretical_or_conceptual","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":["metaresearch"],"domain":"evaluation","study_design":"observational","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","sts","scholarly_communication","insufficient_payload"],"consensus_categories":["metaresearch","insufficient_payload"],"category_scores_codex":[0.296196,0.0001604966,0.0003974755,0.004401063,0.003436812,0.001506077,0.002113811,0.0001109212,0.0009325754],"category_scores_gemma":[0.1453476,0.0001022252,0.0001839675,0.007794134,0.000296341,0.0006425466,0.0007466074,0.0005805732,0.0009983936],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001385186,"about_ca_system_score_gemma":0.002949105,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002147359,"about_ca_topic_score_gemma":0.001282602,"domain_scores_codex":[0.9600894,0.009900765,0.001104296,0.0008738126,0.02689984,0.00113188],"domain_scores_gemma":[0.9701758,0.005652245,0.0003634553,0.002838919,0.02055251,0.0004170698],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001559214,0.0005995834,0.08328746,0.00003205671,0.000878086,0.000004189877,0.01384685,0.3336892,0.1051595,0.06176222,0.007544671,0.391637],"study_design_scores_gemma":[0.0008857088,0.0007671983,0.4777434,0.0001004121,0.0001217213,9.040789e-7,0.002180474,0.4768889,0.01799669,0.02077687,0.002369218,0.0001684688],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9469531,0.00004612406,0.002495474,0.03895453,0.0001471709,0.002515417,0.00002596354,0.00001194049,0.008850327],"genre_scores_gemma":[0.9982846,0.0000172436,0.0004904523,0.0002223876,0.0003949599,0.0001792447,0.00001390282,0.00001353197,0.0003836332],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.394456,"threshold_uncertainty_score":0.9999807,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8233162217039798,"score_gpt":0.7626251681297992,"score_spread":0.06069105357418059,"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."}}