{"id":"W4389577817","doi":"10.2196/47091","title":"Assessing and Improving Data Integrity in Web-Based Surveys: Comparison of Fraud Detection Systems in a COVID-19 Study","year":2023,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Survey Methodology and Nonresponse","field":"Social Sciences","cited_by":49,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Heart, Lung, and Blood Institute; National Institutes of Health","keywords":"Representativeness heuristic; Computer science; Preprint; Context (archaeology); Transparency (behavior); Data quality; Web application; Data integrity; Data science; Computer security; World Wide Web; Business; Statistics; Geography; Service (business)","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":["metaresearch","research_integrity"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2957584,0.0007765227,0.001207093,0.005011535,0.001619569,0.005371835,0.00227374,0.001845129,0.0009415541],"category_scores_gemma":[0.4854433,0.0009786643,0.002219633,0.004346185,0.00262747,0.004751243,0.008120978,0.002052571,0.000332345],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00365394,"about_ca_system_score_gemma":0.005525907,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004018266,"about_ca_topic_score_gemma":0.004095735,"domain_scores_codex":[0.6422523,0.2851237,0.03082384,0.007536744,0.03032473,0.003938735],"domain_scores_gemma":[0.3836072,0.4007322,0.08023008,0.05780338,0.0714817,0.006145499],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.005010617,0.001923888,0.8151847,0.001218247,0.001401562,0.0001320295,0.01670204,0.001540325,0.0008695732,0.001411019,0.001879242,0.1527269],"study_design_scores_gemma":[0.0007810202,0.008887986,0.9402921,0.001365782,0.0008600756,0.0004130393,0.009742074,0.02375636,0.00381153,0.001708357,0.008117287,0.0002643817],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9766088,0.0006669057,0.01479251,0.0009708303,0.000113013,0.003576499,0.0004777897,0.0002015775,0.002592112],"genre_scores_gemma":[0.9739811,0.0003933976,0.02202392,0.0005484066,0.00006661702,0.002165257,0.0005184493,0.00004771821,0.0002550595],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9981549,"threshold_uncertainty_score":0.8684552,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7110174167559848,"score_gpt":0.6488890374946584,"score_spread":0.06212837926132642,"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."}}