{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RRXED6UIQL2WVTPHHUBHYWSFJF","short_pith_number":"pith:RRXED6UI","schema_version":"1.0","canonical_sha256":"8c6e41fa8882f56acde73d027c5a45495f7cedeb34e10cf2e18d9a5f872133ca","source":{"kind":"arxiv","id":"2309.17147","version":2},"attestation_state":"computed","paper":{"title":"Using Large Language Models for Qualitative Analysis can Introduce Serious Bias","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","econ.GN","q-fin.EC"],"primary_cat":"cs.CL","authors_text":"Aditya Chhabra, Julian Ashwin, Vijayendra Rao","submitted_at":"2023-09-29T11:19:15Z","abstract_excerpt":"Large Language Models (LLMs) are quickly becoming ubiquitous, but the implications for social science research are not yet well understood. This paper asks whether LLMs can help us analyse large-N qualitative data from open-ended interviews, with an application to transcripts of interviews with Rohingya refugees in Cox's Bazaar, Bangladesh. We find that a great deal of caution is needed in using LLMs to annotate text as there is a risk of introducing biases that can lead to misleading inferences. We here mean bias in the technical sense, that the errors that LLMs make in annotating interview t"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2309.17147","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-29T11:19:15Z","cross_cats_sorted":["cs.AI","econ.GN","q-fin.EC"],"title_canon_sha256":"9eaf5fa453c57d77b9061c74a52cf81810ba5eedb577d6e33573ab241302bd55","abstract_canon_sha256":"a23eca5d0ef28a6942904791f031c564219e185f2c97be412c4eeddf5386bc95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:57:26.870306Z","signature_b64":"2nl9yYphhD97IZIwCsb5PZYAhY/xPOKkqV5aBZW6Gjc1DvTNurh6Xykc10QbZEgx/5gdPU4smTLwZpnzA8XqAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8c6e41fa8882f56acde73d027c5a45495f7cedeb34e10cf2e18d9a5f872133ca","last_reissued_at":"2026-07-05T06:57:26.869718Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:57:26.869718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Using Large Language Models for Qualitative Analysis can Introduce Serious Bias","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","econ.GN","q-fin.EC"],"primary_cat":"cs.CL","authors_text":"Aditya Chhabra, Julian Ashwin, Vijayendra Rao","submitted_at":"2023-09-29T11:19:15Z","abstract_excerpt":"Large Language Models (LLMs) are quickly becoming ubiquitous, but the implications for social science research are not yet well understood. This paper asks whether LLMs can help us analyse large-N qualitative data from open-ended interviews, with an application to transcripts of interviews with Rohingya refugees in Cox's Bazaar, Bangladesh. We find that a great deal of caution is needed in using LLMs to annotate text as there is a risk of introducing biases that can lead to misleading inferences. We here mean bias in the technical sense, that the errors that LLMs make in annotating interview t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.17147","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2309.17147/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2309.17147","created_at":"2026-07-05T06:57:26.869802+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.17147v2","created_at":"2026-07-05T06:57:26.869802+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.17147","created_at":"2026-07-05T06:57:26.869802+00:00"},{"alias_kind":"pith_short_12","alias_value":"RRXED6UIQL2W","created_at":"2026-07-05T06:57:26.869802+00:00"},{"alias_kind":"pith_short_16","alias_value":"RRXED6UIQL2WVTPH","created_at":"2026-07-05T06:57:26.869802+00:00"},{"alias_kind":"pith_short_8","alias_value":"RRXED6UI","created_at":"2026-07-05T06:57:26.869802+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06468","citing_title":"EgoPolice: A Benchmark for Egocentric Video Understanding in High-Stakes Police Body-Worn Camera Footage","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF","json":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF.json","graph_json":"https://pith.science/api/pith-number/RRXED6UIQL2WVTPHHUBHYWSFJF/graph.json","events_json":"https://pith.science/api/pith-number/RRXED6UIQL2WVTPHHUBHYWSFJF/events.json","paper":"https://pith.science/paper/RRXED6UI"},"agent_actions":{"view_html":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF","download_json":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF.json","view_paper":"https://pith.science/paper/RRXED6UI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.17147&json=true","fetch_graph":"https://pith.science/api/pith-number/RRXED6UIQL2WVTPHHUBHYWSFJF/graph.json","fetch_events":"https://pith.science/api/pith-number/RRXED6UIQL2WVTPHHUBHYWSFJF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF/action/storage_attestation","attest_author":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF/action/author_attestation","sign_citation":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF/action/citation_signature","submit_replication":"https://pith.science/pith/RRXED6UIQL2WVTPHHUBHYWSFJF/action/replication_record"}},"created_at":"2026-07-05T06:57:26.869802+00:00","updated_at":"2026-07-05T06:57:26.869802+00:00"}