{"id":"W4413141452","doi":"10.2196/69956","title":"Improving Data Integrity in Samples Obtained From Web-Based Recruitment: Protocol for the Development of a Novel System for Assessing Participant Authenticity in a Remote Longitudinal Cohort Study of Polysubstance Use","year":2025,"lang":"en","type":"article","venue":"JMIR Research Protocols","topic":"Survey Methodology and Nonresponse","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Drug Abuse; National Institutes of Health","keywords":"Protocol (science); Baseline (sea); Web application; Internet privacy; Medicine; Applied psychology; Computer science; Psychology; World Wide Web; Alternative medicine","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.1212129,0.0001639808,0.0005615079,0.0003745101,0.0005821472,0.000130434,0.001222642,0.0001716362,0.0000061659],"category_scores_gemma":[0.03824991,0.0001234568,0.00005638624,0.00120212,0.0004032099,0.0002474166,0.0003561817,0.0004931225,1.497566e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004605424,"about_ca_system_score_gemma":0.004447257,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01022723,"about_ca_topic_score_gemma":0.08543207,"domain_scores_codex":[0.9805599,0.01545481,0.00140724,0.0008322535,0.0009168877,0.0008289189],"domain_scores_gemma":[0.9570839,0.04096255,0.0003927459,0.0008810541,0.0006141308,0.00006564038],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.04616744,0.004599769,0.9137533,0.001887004,0.0001466624,0.000004018014,0.01521931,0.00002624075,0.007556392,0.000491206,0.0000402109,0.01010841],"study_design_scores_gemma":[0.01211548,0.0003851254,0.9383827,0.005381627,0.00001646461,5.941094e-8,0.01879821,0.01854004,0.0041999,0.0001770898,0.001825779,0.0001775597],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.285893,9.288403e-7,0.02617313,0.00008035584,0.00001008424,0.687741,0.00007781631,0.00001668207,0.000006994993],"genre_scores_gemma":[0.2341506,1.859837e-8,0.03408669,0.000002919144,0.00001484288,0.7317088,0.000006579979,0.000009240986,0.00002032037],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.082963,"threshold_uncertainty_score":0.9963638,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.9146022405428403,"score_gpt":0.6889550407892063,"score_spread":0.225647199753634,"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."}}