{"id":"W4366989641","doi":"10.1111/2041-210x.14110","title":"The data double standard","year":2023,"lang":"en","type":"article","venue":"Methods in Ecology and Evolution","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Vetting; Data science; Citizen science; Computer science; Data collection; Data quality; Benchmarking; Missing data; Data mining; Statistics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002066548,0.00004029791,0.00005747238,0.00001612534,0.0002412353,0.00001252333,0.000169832,0.00005467983,0.001881142],"category_scores_gemma":[0.0001480325,0.00002963415,0.000007416581,0.0002579579,0.0002110787,0.00008561506,0.0003933664,0.00007406314,0.00035692],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002128066,"about_ca_system_score_gemma":0.000005074367,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008045074,"about_ca_topic_score_gemma":0.003018333,"domain_scores_codex":[0.9993442,0.0001579301,0.00009157926,0.0001642798,0.00005649744,0.0001855308],"domain_scores_gemma":[0.9995472,0.0001903788,0.00002270309,0.0002139175,0.000002642745,0.000023152],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001184001,0.00002069835,0.9004933,0.000003916194,0.000005854572,0.00000389932,0.0001466371,0.0000589102,0.001052652,0.01189493,0.0551885,0.03101235],"study_design_scores_gemma":[0.0002355983,0.00001877824,0.8742592,6.297756e-7,0.000002590532,0.000002451106,0.0006031207,0.001257718,0.00005324824,0.004473418,0.1190559,0.00003739098],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.959235,0.0002142245,0.005974672,0.004196513,0.0008898047,0.0002251429,0.00005319036,0.00009470223,0.02911677],"genre_scores_gemma":[0.9885197,0.001278343,0.007139438,0.0002196439,0.0000437471,0.00006317096,0.0001427003,0.000009404605,0.002583901],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06386741,"threshold_uncertainty_score":0.9990312,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1012719692177708,"score_gpt":0.4046120094716364,"score_spread":0.3033400402538656,"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."}}