{"id":"W4400529059","doi":"10.1145/3626772.3661362","title":"Reflections on the Coding Ability of LLMs for Analyzing Market Research Surveys","year":2024,"lang":"en","type":"article","venue":"","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Coding (social sciences); Market research; Data science; Computer science; Business; Marketing; Social science; Sociology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.161537,0.001252954,0.0009735391,0.00468992,0.002151794,0.01274183,0.004102604,0.004024032,0.005571684],"category_scores_gemma":[0.4899532,0.001159433,0.001676519,0.003483714,0.01753501,0.02213381,0.006581122,0.01124682,0.002276228],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006172147,"about_ca_system_score_gemma":0.004751913,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01424414,"about_ca_topic_score_gemma":0.009379117,"domain_scores_codex":[0.8547095,0.1241674,0.003905563,0.005582637,0.01032716,0.001307793],"domain_scores_gemma":[0.265463,0.6651448,0.006275048,0.04289755,0.01800097,0.002218682],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003853739,0.0002070662,0.02671059,0.0007398333,0.0002980133,0.0001679112,0.01909666,0.01983332,0.003169654,0.6602178,0.03894658,0.2302272],"study_design_scores_gemma":[0.00006856336,0.0001190578,0.006649991,0.0009085847,0.0000548134,0.000271211,0.006148961,0.1976364,0.004725001,0.7356653,0.04753209,0.0002198046],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"commentary","genre_scores_codex":[0.06444389,0.003800938,0.713205,0.1972889,0.0008875272,0.0002734329,0.002069295,0.001494637,0.01653624],"genre_scores_gemma":[0.5201293,0.001630208,0.455438,0.01414096,0.001712655,0.0007487107,0.001339626,0.001345073,0.003515445],"genre_candidate":"commentary","genre_consensus":null,"teacher_disagreement_score":0.161537,"threshold_uncertainty_score":0.8542997,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.449297537108599,"score_gpt":0.5977982595760435,"score_spread":0.1485007224674444,"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."}}