{"id":"W2154725923","doi":"10.1215/00031283-80-1-22","title":"THE NORTH AMERICAN REGIONAL VOCABULARY SURVEY: NEW VARIABLES AND METHODS IN THE STUDY OF NORTH AMERICAN ENGLISH","year":2005,"lang":"en","type":"article","venue":"American Speech","topic":"Linguistic Variation and Morphology","field":"Social Sciences","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Vocabulary; Variation (astronomy); Geography; Nova scotia; American English; Regional variation; Lexical item; Linguistics; History; Regional science; Genealogy; Archaeology; Political science; Law","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002886717,0.0001618449,0.0004193712,0.00009628693,0.0003964365,0.00006665258,0.0006345216,0.00001618909,0.00001375611],"category_scores_gemma":[0.00275685,0.0001097467,0.00004401048,0.002189133,0.002106381,0.00005898201,0.00008598853,0.0002625533,0.000003082374],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006511351,"about_ca_system_score_gemma":0.000362501,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.2963145,"about_ca_topic_score_gemma":0.626948,"domain_scores_codex":[0.9952044,0.003137002,0.0004118683,0.000349775,0.0004780868,0.0004188089],"domain_scores_gemma":[0.9957132,0.002993437,0.0005257747,0.0004122776,0.0002171816,0.0001381255],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00004189863,0.0001628381,0.717829,5.234061e-7,0.00003534178,0.000004311858,0.02516127,0.00001633481,8.49577e-7,0.0004571425,0.001251868,0.2550386],"study_design_scores_gemma":[0.0002312217,0.0003630458,0.8869318,0.000001057358,0.00002500879,0.00000139444,0.02937041,0.00002291341,0.000001115048,0.00003259913,0.08289179,0.0001275965],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9943487,0.00003754596,0.0004515465,0.002238227,0.0001458205,0.0004274791,0.00001014862,0.00003554783,0.00230498],"genre_scores_gemma":[0.9845266,0.0002894286,0.01283017,0.001537303,0.0005552966,0.00002071177,0.00001002462,0.00001481946,0.0002155974],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3306336,"threshold_uncertainty_score":0.7761047,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03442550522972881,"score_gpt":0.3645259995402809,"score_spread":0.3301004943105521,"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."}}