How to detect heterogeneity in conjoint experiments
Conjoint experiments are fast becoming one of the dominant experimental methods within the social sciences. Despite recent efforts to model heterogeneity within this type of experiment, the relationship between the conjoint design and lower-level causal estimands is underdeveloped. In this paper, we...
Main Author: | |
---|---|
Format: | Journal article |
Language: | English |
Published: |
University of Chicago Press
2023
|
_version_ | 1797111762652233728 |
---|---|
author | Duch, R |
author_facet | Duch, R |
author_sort | Duch, R |
collection | OXFORD |
description | Conjoint experiments are fast becoming one of the dominant experimental methods within the social sciences. Despite recent efforts to model heterogeneity within
this type of experiment, the relationship between the conjoint design and lower-level
causal estimands is underdeveloped. In this paper, we clarify how conjoint heterogeneity can be construed as a set of nested, causal parameters that correspond to the levels
of the conjoint design. We then use this framework to propose a new estimation strategy, using machine learning, that better allows researchers to evaluate treatment effect
heterogeneity. We also provide novel tools for classifying and analysing heterogeneity
post-estimation using partitioning algorithms. Replicating two conjoint experiments,
we demonstrate our theoretical argument, and show how this method helps estimate
and detect substantive patterns of heterogeneity. To accompany this paper, we provide new a R package, cjbart, that allows researchers to model heterogeneity in their
experimental conjoint data. |
first_indexed | 2024-03-07T08:13:26Z |
format | Journal article |
id | oxford-uuid:f9c6bed0-fd2d-40d7-b378-f0d19d503087 |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-07T08:13:26Z |
publishDate | 2023 |
publisher | University of Chicago Press |
record_format | dspace |
spelling | oxford-uuid:f9c6bed0-fd2d-40d7-b378-f0d19d5030872023-12-07T10:18:43ZHow to detect heterogeneity in conjoint experimentsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:f9c6bed0-fd2d-40d7-b378-f0d19d503087EnglishSymplectic ElementsUniversity of Chicago Press2023Duch, RConjoint experiments are fast becoming one of the dominant experimental methods within the social sciences. Despite recent efforts to model heterogeneity within this type of experiment, the relationship between the conjoint design and lower-level causal estimands is underdeveloped. In this paper, we clarify how conjoint heterogeneity can be construed as a set of nested, causal parameters that correspond to the levels of the conjoint design. We then use this framework to propose a new estimation strategy, using machine learning, that better allows researchers to evaluate treatment effect heterogeneity. We also provide novel tools for classifying and analysing heterogeneity post-estimation using partitioning algorithms. Replicating two conjoint experiments, we demonstrate our theoretical argument, and show how this method helps estimate and detect substantive patterns of heterogeneity. To accompany this paper, we provide new a R package, cjbart, that allows researchers to model heterogeneity in their experimental conjoint data. |
spellingShingle | Duch, R How to detect heterogeneity in conjoint experiments |
title | How to detect heterogeneity in conjoint experiments |
title_full | How to detect heterogeneity in conjoint experiments |
title_fullStr | How to detect heterogeneity in conjoint experiments |
title_full_unstemmed | How to detect heterogeneity in conjoint experiments |
title_short | How to detect heterogeneity in conjoint experiments |
title_sort | how to detect heterogeneity in conjoint experiments |
work_keys_str_mv | AT duchr howtodetectheterogeneityinconjointexperiments |