Description
Presented ongoing work:A Method to Scale-Up Interpretative Qualitative Analysis, with an Application to Aspirations in Cox's Bazaar, Bangladesh
The qualitative analysis of open-ended interviews has found limited use in economics, partly because the required inductive, interpretative human reading of text and coding is time consuming. We present a method with which to conduct qualitative analysis at scale by extending a small set of human-codes to a larger sample using natural language processing. We apply it to analyze 2,200 open-ended interviews on parent's aspirations for children with Rohingya refugees and their Bangladeshi hosts and show that economists’ focus on aspirations for material goals must be distinguished from moral and religious values, and the navigational capacity to achieve these aspirations. For example, while hosts are more ambitious than refugees, they display lower navigational capacity. We show how to assess the robustness and reliability of this approach, compare it to alternative methods including Large Language Models, and find that extending a sample of human coded interviews is likely to be optimal for a budget constrained researcher.
| Period | 16 Nov 2023 |
|---|---|
| Held at | Quantitative Economics |