{"id":"W4200456451","doi":"10.31234/osf.io/j76eg","title":"Challenges and Solutions for Data Analysis in an Adult Lifespan Study of over 100,000 Online Cognitive Test Completions","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Health, Environment, Cognitive Aging","field":"Environmental Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Baycrest Hospital","funders":"","keywords":"Computer science; Outlier; Visualization; Data mining; Smoothing; Data science; Data visualization; Information retrieval; Machine learning; Artificial intelligence","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":[0.227476,0.001158573,0.001484561,0.007692432,0.004861979,0.008811534,0.005078336,0.003894724,0.004047397],"category_scores_gemma":[0.5194777,0.001703112,0.002513738,0.01288773,0.006228569,0.008542529,0.00938154,0.006925744,0.002280386],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004223446,"about_ca_system_score_gemma":0.009843552,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006565704,"about_ca_topic_score_gemma":0.01574683,"domain_scores_codex":[0.7505862,0.1923671,0.02141359,0.009600398,0.02451529,0.001517305],"domain_scores_gemma":[0.3493708,0.4957547,0.03191956,0.06538917,0.05266236,0.004903459],"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":[0.0006941162,0.0009709144,0.1586924,0.008909274,0.001468827,0.00237637,0.09960237,0.009003937,0.005961569,0.06695871,0.1806194,0.4647421],"study_design_scores_gemma":[0.000386347,0.0007475479,0.1542174,0.01157097,0.0004664223,0.003510458,0.09006467,0.02729848,0.005429458,0.3205137,0.3849938,0.0008007319],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09812774,0.005191314,0.6818709,0.1805643,0.00377119,0.006826708,0.0119514,0.002136221,0.009560294],"genre_scores_gemma":[0.1359987,0.00173875,0.8231767,0.01516541,0.00113133,0.0161952,0.004292384,0.0008883163,0.001413192],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.227476,"threshold_uncertainty_score":0.9526595,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1780504610705116,"score_gpt":0.377899496763216,"score_spread":0.1998490356927044,"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."}}