{"id":"W4411772179","doi":"10.5539/ijsp.v14n2p5","title":"Simpson’s Paradox: Aggregation Effects in Statistical and Machine Learning Models","year":2025,"lang":"en","type":"article","venue":"International Journal of Statistics and Probability","topic":"Advanced Statistical Methods and Models","field":"Mathematics","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Mathematical economics; Computer science; Statistical learning; Econometrics; Mathematics; Machine learning; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.04491188,0.001203501,0.002826717,0.002843195,0.003021926,0.006396556,0.00277877,0.004126145,0.004490454],"category_scores_gemma":[0.1871616,0.0008379143,0.002431985,0.004253835,0.008070198,0.00967902,0.007322612,0.008126831,0.0005835039],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002931744,"about_ca_system_score_gemma":0.002807959,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003488102,"about_ca_topic_score_gemma":0.00336298,"domain_scores_codex":[0.9750715,0.01523991,0.001201206,0.001934171,0.006012407,0.0005407648],"domain_scores_gemma":[0.7270069,0.242775,0.008397047,0.0140909,0.006006458,0.001723614],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00003901921,0.00003031079,0.002914699,0.0001726065,0.0001519789,0.0004296928,0.0004730494,0.01285865,0.0001726103,0.9546294,0.005811146,0.02231689],"study_design_scores_gemma":[0.00001563755,0.00002158469,0.0006348116,0.00008313475,0.00003188824,0.0002247669,0.00006752081,0.04013993,0.0001141921,0.9556678,0.002969373,0.0000293761],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03098494,0.01678907,0.8921625,0.028798,0.001871285,0.0001108969,0.0003699568,0.0003793583,0.02853391],"genre_scores_gemma":[0.7654631,0.01617242,0.1893935,0.0099815,0.006007139,0.0004513091,0.00041451,0.0004689838,0.01164763],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04491188,"threshold_uncertainty_score":0.2375196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06134224647241327,"score_gpt":0.3888597566167712,"score_spread":0.3275175101443579,"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."}}