{"id":"W4416287994","doi":"10.2139/ssrn.5755162","title":"<p>Cheaper (and more effective) by the dozen: <span>Evidence from 12 randomized A/B tests optimizing tutoring for scale</span></p>","year":2025,"lang":"","type":"preprint","venue":"SSRN Electronic Journal","topic":"Global Educational Reforms and Inequalities","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Impact","funders":"","keywords":"Test (biology); Scale (ratio); Identification (biology); Measure (data warehouse); Educational testing; Randomized controlled trial; Test score","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":[],"consensus_categories":[],"category_scores_codex":[0.006295881,0.0007628069,0.001589488,0.001453898,0.0005886108,0.001520684,0.001507582,0.002390228,0.106988],"category_scores_gemma":[0.0369875,0.0003578652,0.001113019,0.001683718,0.001490271,0.003603018,0.00145095,0.002644489,0.006909617],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001776791,"about_ca_system_score_gemma":0.00394424,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0075605,"about_ca_topic_score_gemma":0.01045567,"domain_scores_codex":[0.9943838,0.003574599,0.0003268381,0.0006031538,0.0007337577,0.0003778187],"domain_scores_gemma":[0.976613,0.01353185,0.004098106,0.002033729,0.001008474,0.002714803],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"randomized_trial","study_design_scores_codex":[0.01148219,0.004694931,0.008398132,0.00344077,0.0004889612,0.00006301239,0.0002222294,0.00217794,0.001212974,0.01859828,0.1079058,0.8413147],"study_design_scores_gemma":[0.07831644,0.04335575,0.3538797,0.01102045,0.00446112,0.0005586055,0.002034896,0.01381574,0.01459842,0.1071765,0.3702408,0.0005417416],"study_design_candidate":"randomized_trial","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5457547,0.066534,0.03345214,0.1531742,0.007683539,0.004085348,0.01953274,0.003813302,0.1659701],"genre_scores_gemma":[0.914946,0.009623784,0.02914934,0.01295542,0.001255084,0.001985925,0.002764214,0.000564342,0.0267559],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.106988,"threshold_uncertainty_score":0.3579104,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01887907877773646,"score_gpt":0.3183404506011464,"score_spread":0.2994613718234099,"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."}}