{"id":"W4396515249","doi":"10.36834/cmej.76671","title":"Use the right words: evaluating the effect of word choice and word count on quality of narrative feedback in ophthalmology competency-based medical education assessments","year":2024,"lang":"en","type":"article","venue":"Canadian Medical Education Journal","topic":"Innovations in Medical Education","field":"Medicine","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Word (group theory); Quality (philosophy); Narrative; Computer science; Multiple choice; Natural language processing; Linguistics; Mathematics; Statistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00881535,0.0002137404,0.0004457755,0.0006310615,0.0002589408,0.00007247348,0.0003826538,0.0003173994,0.007730415],"category_scores_gemma":[0.07470084,0.0001250924,0.00009224973,0.001269329,0.0007833702,0.0001499456,0.00002160409,0.001985384,0.000007402159],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008855829,"about_ca_system_score_gemma":0.1121815,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007324358,"about_ca_topic_score_gemma":0.005139617,"domain_scores_codex":[0.9947169,0.001497475,0.001183987,0.0003067008,0.001949124,0.0003458138],"domain_scores_gemma":[0.9937978,0.003115918,0.0004383283,0.0004562045,0.0007755956,0.001416177],"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":[0.000448112,0.001262418,0.1184483,0.001206155,0.0003350304,0.00005629263,0.01000701,0.00003738327,0.00009917308,0.01196177,0.07274365,0.7833947],"study_design_scores_gemma":[0.003929789,0.002782799,0.9162033,0.01501417,0.0003958536,0.002509694,0.01344064,0.01828748,0.0001387858,0.003099437,0.02364057,0.0005575136],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8774251,0.0006883376,0.00003622553,0.1172073,0.003229692,0.0006211877,0.000004434038,0.000007391108,0.0007802474],"genre_scores_gemma":[0.9864779,0.00004123399,0.0007687787,0.01145073,0.0007936293,0.000125013,0.00005950289,0.00002308687,0.0002601107],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7977549,"threshold_uncertainty_score":0.9992859,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03865247458304513,"score_gpt":0.4519687203342231,"score_spread":0.413316245751178,"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."}}