{"id":"W2092746927","doi":"10.1080/13645579.2011.638221","title":"Exploring the validity of electronic newspaper databases","year":2011,"lang":"en","type":"article","venue":"International Journal of Social Research Methodology","topic":"Media Studies and Communication","field":"Social Sciences","cited_by":46,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Newspaper; Lexis; Content analysis; Database; Advertising; Political science; Computer science; Sociology; Business; Social science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":["metaresearch"],"domain":"methods","study_design":"observational","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"observational","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1591294,0.0004532546,0.001017661,0.02636445,0.003588109,0.01521926,0.00373247,0.002023935,0.003291944],"category_scores_gemma":[0.6317573,0.001187388,0.000980054,0.04269989,0.005405692,0.01414294,0.007946117,0.001411742,0.001045619],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005401093,"about_ca_system_score_gemma":0.004722565,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01567197,"about_ca_topic_score_gemma":0.009442055,"domain_scores_codex":[0.7263252,0.1481695,0.03941712,0.01720596,0.06439385,0.004488549],"domain_scores_gemma":[0.1250893,0.6717216,0.08011793,0.05091302,0.0704407,0.001717485],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0008959356,0.0003247974,0.8766078,0.001632359,0.0005044523,0.0004511008,0.02267676,0.001694738,0.0006570486,0.01865095,0.004784246,0.0711199],"study_design_scores_gemma":[0.0001946512,0.0004993845,0.7950506,0.004352431,0.0006454004,0.001858614,0.05425376,0.02104753,0.007990563,0.02235613,0.09145483,0.0002960875],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9255723,0.003467369,0.01715752,0.003924261,0.000216937,0.0009485096,0.0126984,0.0001859323,0.03582874],"genre_scores_gemma":[0.9813772,0.0005325094,0.008418269,0.0005241281,0.0001497004,0.0005736046,0.007408974,0.0001290914,0.0008865487],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8408706,"threshold_uncertainty_score":0.8415665,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.9349094824766678,"score_gpt":0.6276944662901474,"score_spread":0.3072150161865205,"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."}}