Causal Inference in Conjoint Analysis: Understanding Multidimensional Choices via Stated Preference Experiments
The DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Abstract
Survey experiments are a core tool for causal inference. Yet, the design of classical survey experiments prevents them from identifying which components of a multidimensional treatment are influential. Here, we show how conjoint analysis , an experimental design yet to be widely applied in political science, enables researchers to estimate the causal effects of multiple treatment components and assess several causal hypotheses simultaneously. In conjoint analysis, respondents score a set of alternatives, where each has randomly varied attributes. Here, we undertake a formal identification analysis to integrate conjoint analysis with the potential outcomes framework for causal inference. We propose a new causal estimand and show that it can be nonparametrically identified and easily estimated from conjoint data using a fully randomized design. The analysis enables us to propose diagnostic checks for the identification assumptions. We then demonstrate the value of these techniques through empirical applications to voter decision making and attitudes toward immigrants.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Conditional logit analysis of qualitative choice behavior
McFadden, D. L. (1972) · eScholarship (California Digital Library) · Cited by 14,203
Estimating causal effects of treatments in randomized and nonrandomized studies
Rubin, D. B. (1974) · Journal of Educational Psychology · doi:10.1037/h0037350 · Cited by 9,436
Cragg, J. G., & Zarembka, P. (1975) · The Canadian Journal of Economics · doi:10.2307/134075 · Cited by 8,342
Statistics and Causal Inference
Holland, P. W. (1986) · Journal of the American Statistical Association · doi:10.1080/01621459.1986.10478354 · Cited by 5,293
Eulau, H. (1960) · American Political Science Review · doi:10.1017/s000305540012235x · Cited by 4,722
Evaluating Online Labor Markets for Experimental Research: Amazon.com's Mechanical Turk
Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012) · Political Analysis · doi:10.1093/pan/mpr057 · Cited by 4,008
Constructive Consumer Choice Processes
Bettman, J. R., Luce, M. F., & Payne, J. W. (1998) · Journal of Consumer Research · doi:10.1086/209535 · Cited by 2,831
Simultaneous conjoint measurement: A new type of fundamental measurement
Luce, R. D., & Tukey, J. W. (1964) · Journal of Mathematical Psychology · doi:10.1016/0022-2496(64)90015-x · Cited by 2,007
Contingent Valuation: Is Some Number Better than No Number?
Diamond, P. A., & Hausman, J. A. (1994) · Journal of Economic Perspectives · doi:10.1257/jep.8.4.45 · Cited by 1,894
Causal Mechanisms in the Social Sciences
Hedström, P., & Ylikoski, P. (2010) · Annual Review of Sociology · doi:10.1146/annurev.soc.012809.102632 · Cited by 1,630
Research works printed but not linked: 7
- Green, 2002, Partisan hearts and minds
- Campbell, 1960, The American voter
- These journals are the American Journal of Political Science, the American Political Science Review, and the Journal of Politics.
- Hauser, 2007, MIT Sloan Courseware
- Brader, 2008, Is it immigration or the immigrants? The emotional influence of groups on public opinion and political action, American Journal of Political Science, 52
- 10.1287/inte.31.4.56.9676 · doi:10.1287/inte.31.4.56.9676 · Paper page
- Bechtel, 2013, Public support for global climate cooperation
Cited by
Items 601–700 of 1,000
What do aid recipients want? Public attitudes toward foreign aid in developing countries
Kim, S. E., Park, J. H., Rhee, I., & Yang, J. (2024) · World Development · doi:10.1016/j.worlddev.2024.106815 · Cited by 17
Magistro, B., Borwein, S., Alvarez, R. M., Bonikowski, B., & Loewen, P. J. (2025) · American Journal of Political Science · doi:10.1111/ajps.12959 · Cited by 26
What Do People Want From a Welfare System? Conjoint Survey Evidence From UK Adults
Nettle, D., Chrisp, J., Johnson, E., & Johnson, M. (2025) · Poverty & Public Policy · doi:10.1002/pop4.70018 · Cited by 18
“In recent years, conjoint analysis (Hainmueller et al. 2014) has become increasingly popular as a method for eliciting respondents' valuations of competing policy goods (Bremer and Bür-gisser 2025).”
InfluentialExplaining deference: why and when do policymakers think FDI needs tax incentives?
Danzman, S. B., & Slaski, A. (2021) · Review of International Political Economy · doi:10.1080/09692290.2021.1885475 · Cited by 16
“The cjoint package, created by Hainmueller et al. (2014), provides a convenient algorithm to analyze this data and present the AMCEs.”
BackgroundMethodWhy Costly Rivalry Disputes Persist: A Paired Conjoint Experiment in Japan and South Korea
Lim, S., & Tanaka, S. (2022) · International Studies Quarterly · doi:10.1093/isq/sqac063 · Cited by 15
Support for climate unilateralism
Mildenberger, M. (2019) · Nature Climate Change · doi:10.1038/s41558-019-0416-x · Cited by 13
Euroscepticism and government accountability in the European Union
Schneider, C. J. (2019) · The Review of International Organizations · doi:10.1007/s11558-019-09358-w · Cited by 13
“12 See Hainmueller et al. (2014) for this method.”
Does descriptive representation increase perceptions of legitimacy? Evidence from Australia
Snagovsky, F., Kang, W. C., Sheppard, J., & Biddle, N. (2020) · Australian Journal of Political Science · doi:10.1080/10361146.2020.1804834 · Cited by 13
The Subconscious Effect of Subtle Media Bias on Perceptions of Terrorism
Feick, L., Donnay, K., & McCabe, K. T. (2020) · American Politics Research · doi:10.1177/1532673x20972105 · Cited by 19
“Following Hainmueller et al. (2014), we estimate Average Marginal Component Effects (AMCEs), which express the effect of a distinct attribute, on average, marginalizing over the joint distribution of the other attributes in the study.”
MethodWhere is the EU–UK relationship heading? A conjoint survey experiment of Brexit trade-offs
Hix, S., van der Linden, C., Massie, J., Pickup, M. A., & Savoie, J. (2022) · European Union Politics · doi:10.1177/14651165221123155 · Cited by 14
“…developed in market research to understand consumers’ preferences about different aspects of a product but is increasingly used in political science to assess voters’ preferences when choices across issue dimensions are connected (inter alia Bansak et al., 2021b; Hainmueller et al., 2014).”
BackgroundPublic Perceptions of Whistleblowing
Heumann, M., Friedes, A., Redlawsk, D. P., Cassak, L., & Kesari, A. (2015) · Public Integrity · doi:10.1080/10999922.2015.1093397 · Cited by 13
Policy Design and Domestic Support for International BailoutsPreprint
Bechtel, M. M., Hainmueller, J., & Margalit, Y. M. (2012) · SSRN Electronic Journal · doi:10.2139/ssrn.2163594 · Cited by 13
Self-citationConfounding in Survey ExperimentsPreprint
Dafoe, A., Caughey, D., & Zhang, B. (2015) · OSF Preprints (OSF Preprints) · doi:10.17605/osf.io/s49qh · Cited by 19
Preferences, Selection, and the Structure of Teacher PayPreprint
Johnston, A. C. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4114370 · Cited by 16
To What Extent Does Asylum Policy Match Public Policy Preferences?
Corre, T. L., & Tilley, J. R. (2024) · International Migration Review · doi:10.1177/01979183241253502 · Cited by 12
“First, conjoint experiments “ capture decision-making processes in information-rich environments more effectively than do traditional survey experiments ” (Hainmueller, Hopkins, and Yamamoto 2014, 27).”
Aoki, N. (2025) · Asia Pacific Journal of Public Administration · doi:10.1080/23276665.2025.2478554 · Cited by 18
“…five digital-era services from its corresponding traditional service (baseline), represented by the “no” status of the digital-era services in Table 2, on the probability of choosing a public administration (Hainmueller et al., 2014), interpreted here as the favourability of that administration.”
Race, Crime, and the Public’s Sentencing Preferences
Doherty, D., Dowling, C. M., Miller, M. G., & Tuma, M. D. (2022) · Public Opinion Quarterly · doi:10.1093/poq/nfac022 · Cited by 16
Carnahan, D., Ulusoy, E., Barry, R. C., McGraw, J., Virtue, I., & Bergan, D. E. (2022) · Journal of Communication · doi:10.1093/joc/jqac023 · Cited by 16
Framing affordable housing: an experimental test of changing attitudes
Matheis, M., & Sorens, J. (2022) · Housing Studies · doi:10.1080/02673037.2022.2101630 · Cited by 17
“Hainmueller, Hopkins & Yamamoto (2014) show that under a spare set of assumptions, the multiple regression model parameters properly identify causal quantities.”
BackgroundMethodInfluentialWorking Hard or Hardly Working? Gender and Voter Evaluations of Legislator Productivity
Hargrave, L., & Smith, J. C. (2023) · Political Behavior · doi:10.1007/s11109-022-09853-8 · Cited by 12
“Following Hainmueller et al. (2014), we estimate the probability that a respondent chooses an MP via: where i indicates the respondent and j indicates the scenario.”
Measuring public preferences for government spending under constraints: a conjoint-analytic approach
Kölln, A., & Wlezien, C. B. (2023) · Journal of Elections Public Opinion and Parties · doi:10.1080/17457289.2023.2189255 · Cited by 16
“We estimate causal effects from the conjoint experiment using the cjoint package in R (see Hainmueller, Hopkins, and Yamamoto 2014; version 2.1.0) with clustered standard errors by respondent.”
BackgroundMethodInfluentialYamane, T., & Kaneko, S. (2023) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2023.137593 · Cited by 18
The effect of primaries on voter evaluations of candidate quality: Experimental evidence
Schwenk, J. (2023) · Party Politics · doi:10.1177/13540688231184627 · Cited by 12
“…marginal component effect (AMCE) is the marginal effect of one level of an attribute, averaged over the joint distribution of all other attributes (Hainmueller et al., 2014: 10), such that the estimates are the average change in probability that a pro fi le gets chosen based on an attribute level.”
Public preferences for phasing-out fossil fuels in the german building and transport sectors
Tröndle, T., Annaheim, J., Hoppe, J., Hanger-Kopp, S., & Patt, A. (2023) · Environmental Research Communications · doi:10.1088/2515-7620/acec39 · Cited by 13
“To analyse the causal effects of policy package attribute levels within our experiments, we use a standardmethod described byHainmueller (Hainmueller et al 2014) based on linear regression.”
MethodPedersen, M. J., Favero, N., & Park, J. (2023) · Public Management Review · doi:10.1080/14719037.2023.2245841 · Cited by 13
Yu, J. J., Shen, X., & Xie, Y. (2023) · China Population and Development Studies · doi:10.1007/s42379-023-00146-z · Cited by 15
“Therefore, drawing on the conjoint analysis method based on stated preference (Hainmueller et al., 2014), we explore the respondents’ assessments of the likelihood of childbearing under different scenarios, which are combinations of different conditions in multiple dimensions that are of interest…”
InfluentialTransparency and citizen support for public agencies: The case of foreign aid
Heinzel, M., Reinsberg, B., & Swedlund, H. J. (2024) · Governance · doi:10.1111/gove.12863 · Cited by 17
Perceptions of the social status hierarchy and its cultural and economic sources
Breyer, M. (2024) · European Journal of Political Research · doi:10.1111/1475-6765.12712 · Cited by 16
“While this is of course a simplified view on intersecting dimensions of social status, the conjoint approach is advantageous to disentangle the determinants of multidimensional concepts (Hainmueller et al., 2014).”
Unraveling a “Cancel Culture” Dynamic: When, Why, and Which Americans Sanction Offensive Speech
Dias, N. C., Druckman, J. N., & Levendusky, M. S. (2024) · The Journal of Politics · doi:10.1086/733004 · Cited by 25
Rudolph, L., Freitag, M., & Thurner, P. W. (2024) · Public Opinion Quarterly · doi:10.1093/poq/nfae038 · Cited by 17
Eady, G., & Rasmussen, A. (2024) · American Political Science Review · doi:10.1017/s0003055424001205 · Cited by 16
“To analyze the paired conjoint experiment, we calculate the Average Marginal Component Effects (AMCE) for each attribute of interest, and Average Marginal Component Interaction Effects (AMCIE) to test our conditional hypotheses (Hainmueller, Hopkins, and Yamamoto 2014).”
Class War in the Voting Booth: Bias Against High‐Income Congressional Candidates
Griffin, J. D., Newman, B., & Buhr, P. (2019) · Legislative Studies Quarterly · doi:10.1111/lsq.12253 · Cited by 15
Detecting Voter Understanding of Ideological Labels Using a Conjoint Experiment
Miwa, H., Arami, R., & Taniguchi, M. (2021) · Political Behavior · doi:10.1007/s11109-021-09719-5 · Cited by 18
Coethnicity Beyond Clientelism: Insights from an Experimental Study of Political Behavior in Lebanon
Cammett, M. C., Kruszewska, D., Parreira, C., & Atallah, S. (2021) · Politics and Religion · doi:10.1017/s1755048321000201 · Cited by 17
“Hainmueller, Hopkins, and Yamamoto (2014) show that the AMCE is nonparametrically identified given the conditionally independent randomization of the attributes and can be estimated by regressing the binary outcome variable (candidate selected) on sets of indicator variables measuring the levels of…”
BackgroundSheares, A. M. (2022) · Sociology of Race and Ethnicity · doi:10.1177/23326492221125116 · Cited by 27
“For more information about the utility of conjoint experiments, please refer Hainmueller et al. (2014) and Schachter (2016).”
BackgroundMethodInfluentialBaron, D., Lauderdale, B. E., & Sheehy-Skeffington, J. A. (2023) · Political Psychology · doi:10.1111/pops.12891 · Cited by 21
“More recently, the use of conjoint experiments, involving the randomized presentation of candidates or party profiles that vary along multiple attributes, have yielded insights on which candidate characteristics cause respondents to select one candidate over another (Hainmueller et al., 2014).”
Is There an Idealized Target of Sexual Harassment in the MeToo Era?
Hart, C. G. (2023) · Social Problems · doi:10.1093/socpro/spad016 · Cited by 16
Turnbull-Dugarte, S. J., & Devine, D. (2023) · Electoral Studies · doi:10.1016/j.electstud.2023.102679 · Cited by 12
Fukuda, K., Isdwiyani, R., Kawata, K., & Yoshida, Y. (2018) · Waste Management & Research The Journal for a Sustainable Circular Economy · doi:10.1177/0734242x18793939 · Cited by 12
Do citizens provide political rewards to firms engaging in voluntary environmental action?
Kolcava, D., Scholderer, J., & Bernauer, T. (2020) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2020.123564 · Cited by 13
Vanhuysse, P., Jankowski, M., & Tepe, M. (2021) · Policy Sciences · doi:10.1007/s11077-021-09435-1 · Cited by 12
From the lab to the poll: The use of survey experiments in political research
Martini, S., & Olmastroni, F. (2021) · Italian Political Science Review/Rivista Italiana di Scienza Politica · doi:10.1017/ipo.2021.20 · Cited by 13
Cooperman, A. D., McLarty, A. R., & Seim, B. (2022) · Global Environmental Change · doi:10.1016/j.gloenvcha.2022.102512 · Cited by 12
Bughin, J., Cincera, M., Kiepfer, E., Reykowska, D., Philippi, F., Żyszkiewicz, M., Ohme, R. K., & Frank, D. (2022) · The European Journal of Health Economics · doi:10.1007/s10198-022-01450-0 · Cited by 11
“Furthermore, as the evidence shows that attribute weights match actual choices made by respondents [25, 27], conjoint studies are successfully applied in a variety of applications in health economics [26], or [16].”
BackgroundCrafting international apologies that work: A conjoint analysis approach
Kohama, S., Himichi, T., Inamasu, K., Mifune, N., Ohtsubo, Y., & Tago, A. (2022) · Conflict Management and Peace Science · doi:10.1177/07388942221094761 · Cited by 11
“The average marginal component-specific effects were calculated using the estimator and the Cjoint package developed by Hainmueller et al. (2014).”
MethodDesigning democracy: How does institutional structure affect civic participation?
Abbott, J., & Touchton, M. M. (2022) · Governance · doi:10.1111/gove.12710 · Cited by 15
Zeitlin, J., van der Duin, D., Kuhn, T., Weimer, M., & Jensen, M. D. (2021) · Regulation & Governance · doi:10.1111/rego.12483 · Cited by 27
“The results of the OLS regression are shown here, as estimating conjoint experiment results with OLS regression is the standard practice (Hainmueller et al., 2014) and facilitates interpretation of the coef fi cients.”
InfluentialRudolph, L., Freitag, M., & Thurner, P. W. (2023) · European Journal of Political Research · doi:10.1111/1475-6765.12617 · Cited by 17
Automation versus openness: support for policies to address job threats
Kuo, A. G., Manzano, D., & Gallego, A. (2023) · Journal of Public Policy · doi:10.1017/s0143814x23000260 · Cited by 15
Tutić, A., Grehl, S., & Liebe, U. (2023) · European Sociological Review · doi:10.1093/esr/jcad067 · Cited by 14
Hansen, P. (2023) · Public Management Review · doi:10.1080/14719037.2023.2270557 · Cited by 14
“As our overall research design, we opted for a conjoint survey experiment, since this design type is well suited for disentangling the individual causal effect of multiple client characteristics (Hainmueller, Hopkins, and Yamamoto 2014).”
Public attitudes towards co-ethnic migrant integration: evidence from South Korea
Denney, S., & Green, C. K. (2023) · Journal of Ethnic and Migration Studies · doi:10.1080/1369183x.2023.2286207 · Cited by 12
“A method derived from the marketing profession, its full-scale inclusion in the quantitative social sciences is relatively new (Hainmueller, Hopkins, and Yamamoto 2014).”
MethodI’m from the Government, and I’m Here to Help: Public Perceptions of Coercive State Power
Blankshain, J. D., Cohn, L. P., & Lupton, D. L. (2024) · American Political Science Review · doi:10.1017/s0003055424000339 · Cited by 12
Ioku, T., Song, J., & Watamura, E. (2024) · Big Data & Society · doi:10.1177/20539517241290217 · Cited by 22
“We processed our Qualtrics conjoint experiment data using the “ q2c ” package in R (Song, 2022), which processes data fast and corrects column misalignment, issues present in the “ read.qualtrics ” function in the “ cjoint ” package (Hainmueller et al., 2014).”
Brückmann, G., Ruprecht, S., & Stadelmann‐Steffen, I. (2024) · Energy Research & Social Science · doi:10.1016/j.erss.2024.103819 · Cited by 15
“A conjoint experiment allows for the combined testing of various project attributes, which mirrors a more realistic setting than classic surveys, as respondents can make their decisions based on multiple criteria [107,108].”
InfluentialCan information, compensation and party cues increase mass support for green taxes?
Bürgisser, R., Stadelmann‐Steffen, I., & Armingeon, K. (2024) · Journal of European Public Policy · doi:10.1080/13501763.2024.2444332 · Cited by 15
“To implement the conjoint experiment, we follow the fully randomised approach proposed by Bechtel and Scheve (2013) and Hainmueller et al. (2014) except for a few logically implausible profile combinations.”
Pandemic Polity-Building: How Covid-19 Shaped the European Union
Truchlewski, Z., Oana, I., Moise, A. D., & Kriesi, H. (2025) · doi:10.1093/9780198951520.001.0001 · Cited by 14
Goplerud, M., Imai, K., & Pashley, N. E. (2025) · The Annals of Applied Statistics · doi:10.1214/24-aoas1994 · Cited by 22
Aruguete, N. (2021) · El Profesional de la Informacion · doi:10.3145/epi.2021.mar.18 · Cited by 30
Nelson, M. J., & Witko, C. (2021) · The Journal of Politics · doi:10.1086/715601 · Cited by 13
Weisshaar, K. (2021) · Socius Sociological Research for a Dynamic World · doi:10.1177/23780231211019861 · Cited by 16
“The conjoint experiment framework is not intended to simulate real application materials but instead to distill the decision-making process on the basis of easily interpretable treatments (Hainmueller et al. 2015; Hainmueller, Hopkins, and Yamamoto 2014).”
BackgroundMethodInfluentialHow political reception contexts shape location decisions of immigrants
Bennour, S., Manatschal, A., & Ruedin, D. (2022) · Journal of Ethnic and Migration Studies · doi:10.1080/1369183x.2022.2098468 · Cited by 13
“1 of R with the cjoint (Hainmueller et al., 2014) and cregg (Leeper, 2020) packages.”
BackgroundMethodInfluentialCOVID-19 health certification reduces outgroup bias: evidence from a conjoint experiment in Japan
Kubo, Y., & Okada, I. (2022) · Humanities and Social Sciences Communications · doi:10.1057/s41599-022-01324-z · Cited by 11
“…research (Cattin and Wittink, 1982; Green et al., 2001; Green and Rao, 1971; Green and Srinivasan, 1990), but later expanded as a causal inference method with randomized designs in political science (Clayton et al., 2021; Cuesta et al., 2021; Hainmueller et al., 2014; Leeper et al., 2020).”
Aryal, T. R., Ichihashi, M., & Kaneko, S. (2022) · Journal of Economic Structures · doi:10.1186/s40008-022-00287-3 · Cited by 11
Voters’ perceptions and evaluations of dynastic politics in Japan
Miwa, H., Kasuya, Y., & Ono, Y. (2022) · Asian Journal of Comparative Politics · doi:10.1177/20578911221144101 · Cited by 12
The personality is political (especially for populists)
Ortega, A. L. (2023) · Journal of Elections Public Opinion and Parties · doi:10.1080/17457289.2023.2189728 · Cited by 11
“…when the value of the attribute (component) of interest is changed from one level to another, averaging over all possible values of the candidates ’ remaining attributes and all possible values of the attributes of the other candidate in the choice task (Hainmueller, Hopkins, and Yamamoto 2014).”
BackgroundMethodState Absence, Vengeance, and the Logic of Vigilantism in Guatemala
Dow, D., Levy, G., Romero, D., & Tellez, J. (2023) · Comparative Political Studies · doi:10.1177/00104140231169026 · Cited by 12
“Conjoint experiments allow for the estimation of the average causal effects of multiple treatment components at the same time (Hainmueller et al., 2014).”
BackgroundInfluentialThe Electoral Consequences of Issue Frames
Peterson, E., & Simonovits, G. (2018) · The Journal of Politics · doi:10.1086/698886 · Cited by 15
“After the framing experiment, respondents evaluated candidates for the House of Representatives in a conjoint experiment (Hainmueller et al. 2014).”
BackgroundMethodInfluentialImplementing novel, flexible, and powerful survey designs in R Shiny
Kaufman, A. R. (2020) · PLoS ONE · doi:10.1371/journal.pone.0232424 · Cited by 14
“Conjoint designs are preferable for such comparisons in part because they produce stable estimates of many difficult-to-isolate effects simultaneously and show “which components of a multidimensional treatment are influential” [38].”
BackgroundMcGrath, L. F., Bernauer, T., Song, J., & Uji, A. (2021) · Climatic Change · doi:10.1007/s10584-021-03137-6 · Cited by 13
van der Does, R., & Kantorowicz, J. (2022) · Political Science Research and Methods · doi:10.1017/psrm.2022.3 · Cited by 11
“Following the logic of a conjoint analysis (Hainmueller et al., 2014), respondents then six times performed a choice task.”
MethodWhen and Why People Prefer Higher Educated Politicians: Ingroup Bias, Deference, and Resistance
van Noord, J., Kuppens, T., Spruyt, B., & Spears, R. (2022) · Personality and Social Psychology Bulletin · doi:10.1177/01461672221077794 · Cited by 14
“Several studies found a general preference for higher educated candidates (Arnesen et al., 2019; Hainmueller et al., 2014; Wüest & Pontusson, 2017).”
BackgroundTeacher Preferences, Working Conditions, and Compensation StructurePreprint
Johnston, A. C. (2020) · SSRN Electronic Journal · doi:10.2139/ssrn.3532779 · Cited by 25
In This Together? Support for European Fiscal Integration in the COVID-19 CrisisPreprint
Bremer, B., Kuhn, T., Meijers, M. J., & Nicoli, F. (2021) · doi:10.31235/osf.io/82cyw · Cited by 11
Human rights violations and public support for sanctions
Arı, B., & Sonmez, B. (2024) · Journal of Peace Research · doi:10.1177/00223433231201450 · Cited by 13
“…conjoint design allows respondents to choose or rate two or more hypothetical choices that have multiple attributes, with the objective of estimating the influence of each characteristic on the respondent’s preference intensity (Green, Krieger & Wind, 2001; Hainmueller, Hopkins & Yamamoto, 2014).”
BackgroundMethodThe Impact of EU Politicisation on Voting Behaviour in Europe
Lobo, M. C. (2023) · Palgrave studies in European Union politics · doi:10.1007/978-3-031-29187-6 · Cited by 12
Offringa, S., & Groeneveld, S. M. (2023) · Review of Public Personnel Administration · doi:10.1177/0734371x231182970 · Cited by 15
“In addition, we estimate the Average Component Interaction Effect (ACIE), which measures the effect of certain attributes conditional on a chosen other attribute, in our case a manager’s gender (Hainmueller et al., 2014).”
InfluentialPolicy preferences in response to large forced migration inflows
Allen, W., Ruiz, I., & Vargas‐Silva, C. (2023) · World Development · doi:10.1016/j.worlddev.2023.106462 · Cited by 15
Panel, S., Paulis, E., Pilet, J., Rojon, S., & Vittori, D. (2023) · Political Behavior · doi:10.1007/s11109-023-09904-8 · Cited by 13
“AMCEs represent the average difference in the probability of being chosen as a minister when comparing two attribute values (e.g., a candidate with or without field expertise), where the average is computed over all possible combinations of other attribute values.”
InfluentialA difficult test for hard propaganda: Evidence from a choice experiment in Venezuela
Lutscher, P., & Donnay, K. (2024) · Journal of Peace Research · doi:10.1177/00223433231220269 · Cited by 14
“Asking respondents for relative, pair-wise comparisons is a well-established technique for the robust measurement of their perceptions (Hainmueller, Hopkins and Yamamoto, 2014).”
How do citizens define and value the rule of law? A conjoint experiment in Germany and Poland
Gutmann, J., Kantorowicz, J., & Voigt, S. (2024) · Journal of European Public Policy · doi:10.1080/13501763.2024.2326189 · Cited by 20
“To examine the baseline effects of treatments in the conjoint experiments, we estimate average marginal component effects (AMCEs) with standard errors clustered at the respondent level (see Bansak et al ., 2023; Hainmueller et al ., 2014).”
Someone Like Me? Disability Identity and Representation Perceptions
Reher, S., & Evans, E. (2024) · Political Behavior · doi:10.1007/s11109-024-09969-z · Cited by 14
“To test Hypotheses 1 we estimate the Average Marginal Component Effect (AMCE) (Hainmueller et al., 2014) of candidate disability on citizens’ feeling of whether a candidate represents people like them.”
Countering Authoritarian Behavior in Democracies
HOBOLT, S. B., & Osnabrügge, M. (2024) · Political Behavior · doi:10.1007/s11109-024-09971-5 · Cited by 14
“We start our analysis of the conjoint experiment by estimating the AMCEs in line with our preregistration (Hainmueller et al., 2014).”
InfluentialZhang, Y., & Wang, H. (2024) · Public Administration · doi:10.1111/padm.13042 · Cited by 12
“Unlike traditional survey methods that typically focus on a limited number of variables, conjoint experiments allow researchers to simultaneously identify the causal effects of multiple bureau-cratic traits (Hainmueller, Hopkins, and Yamamoto 2014).”
InfluentialIntroducing the Visual Conjoint, with an Application to Candidate Evaluation on Social Media
Vecchiato, A., & Munger, K. (2024) · Journal of Experimental Political Science · doi:10.1017/xps.2024.15 · Cited by 13
“We implement both a standard conjoint and our new Twitter visual conjoint in a candidate preference experiment based on the now-classic conjoint in Hainmueller, Hopkins, and Yamamoto (2014).”
InfluentialIt’s the Economy: The Effect of Economic Policy Appeals on Latino Independents
Wakefield, D. (2025) · Political Behavior · doi:10.1007/s11109-025-10004-y · Cited by 13
“As with other conjoint studies, I calculate average marginal component effects (AM-CEs) to estimate the relationship between treatment assignment and candidate vote choice (Hainmueller, Hopkins, and Yamamoto 2014; Kirkland and Coppock 2018).”
Why can't we be friends? Untangling conjoined polarization in America
Norman, J. M., & Green, B. (2025) · Political Psychology · doi:10.1111/pops.13084 · Cited by 14
“Although the choices in a conjoint experiment are artificial, research has shown that conjoint experiments have high external validity and reliably predict real-world behavior (Hainmueller et al., 2014).”
Citizen preferences for online hate speech regulation
Munzert, S., Traunmueller, R., Barberá, P., Guess, A. M., & Yang, J. (2025) · PNAS Nexus · doi:10.1093/pnasnexus/pgaf032 · Cited by 26
Large-scale moral machine experiment on large language models
Zaim bin Ahmad, M. S., & Takemoto, K. (2025) · PLOS One · doi:10.1371/journal.pone.0322776 · Cited by 9
“For analyzing LLM responses to the Moral Machine scenarios, we employed the conjoint analysis framework as described in [49] and implemented in the original study [10] on November 12, 2024 the Moral Machine experiment and our previous study [33].”
Costa, M., & Wallace, I. (2021) · American Politics Research · doi:10.1177/1532673x211006386 · Cited by 13
“As in typical candidate-choice conjoint designs (Hainmueller et al., 2014), respondents were presented with two hypothetical candidates side-by-side (Candidate A and Candidate B) in a table with a few attributes listed for each candidate.”
BackgroundVoters Use Campaign Finance Transparency and Compliance Information
Wood, A. K. (2022) · Political Behavior · doi:10.1007/s11109-022-09776-4 · Cited by 20
“Conjoint analysis allows researchers to estimate how much a given candidate attribute (here, policy position, transparency, compliance, and valence attributes) increases or decreases the chance that a given profile is selected, a quantity called the Average Marginal Component Effect, or AMCE (Hainmueller et al., 2014;…”
BackgroundInfluentialCanales, K. L., Kropf, M. E., Leland, S. M., & Maestas, C. D. (2022) · Urban Affairs Review · doi:10.1177/10780874221103765 · Cited by 10
“This means it is possible to estimate an “average” marginal effect of each attribute on the decision to purchase (Hainmueller, Hopkins, and Yamamoto 2014, 10–11).”
BackgroundMethodInfluentialHankla, C., Banerjee, S., Bohlken, A. T., & Banerjee, A. (2022) · Comparative Political Studies · doi:10.1177/00104140221141838 · Cited by 12
“Gender, race, sexual orientation, age, and other identities are not simply additives, but they interact in complex ways, and these interactions have implications for politics (e.g., Hancock, 2007).”
BackgroundMapping out the motivational basis of active representation as intergroup behavior
Kang, I., & Jilke, S. (2022) · Public Administration · doi:10.1111/padm.12908 · Cited by 11
Causal effect estimation for multivariate continuous treatments
Chen, J., & Zhou, Y. (2023) · Biometrical Journal · doi:10.1002/bimj.202200122 · Cited by 12
“Some research is focused on multivariate categorical treatments, such as factorial designs and conjoint analysis that estimate the main or interaction effect of any combination level of treatments (Dasgupta et al., 2015; Hainmueller et al., 2014).”
BackgroundHumble, S., Sharma, A., Rangaraju, B., Dixon, P., & Pennington, M. (2023) · BMJ Open · doi:10.1136/bmjopen-2022-067680 · Cited by 16
Sorensen, A., & Chen, P. (2023) · Politics Groups and Identities · doi:10.1080/21565503.2023.2231894 · Cited by 11
Deservingness and the Politics of Student Debt Relief
SoRelle, M. E., & Laws, S. (2023) · Perspectives on Politics · doi:10.1017/s1537592723001457 · Cited by 11
“16 The AMCE captures the average change in the probability that a borrower ’ s pro fi le will be selected when a particular attribute level is listed relative to a baseline category (Hainmueller, Hopkins, and Yamamoto 2014).”
BackgroundA Systematic Review of Experimental Research on Public Service Motivation
Chung, K. H., Rhee, I., & Liu, C. (2023) · Public Performance & Management Review · doi:10.1080/15309576.2023.2269386 · Cited by 14
“Since different attributes are assessed simultaneously, it can better capture respondents’ real preferences compared to other commonly used survey experiments that focus solely on one dimension (Hainmueller et al., 2014).”
Rasmussen, A., & Otjes, S. (2024) · Journal of European Public Policy · doi:10.1080/13501763.2024.2306226 · Cited by 13
Metadata
- Authors
- Hainmueller, J., Hopkins, D. J., & Yamamoto, T.
- Affiliations
- Harvard University (US)
- Corresponding author
- Yamamoto, Teppei
- Venue
- Political Analysis
- Publisher
- Cambridge University Press (CUP)
- Published
- 2014
- Volume
- 22
- Issue
- 1
- Pages
- 1-30
- ISSN
- 1047-1987 (ISSN-L)
- 1476-4989
- OpenAlex
- W2745098867
- Cited by
- 1,990
- References
- 51
- Citations by year
- 2026: 273 · 2025: 357 · 2024: 275 · 2023: 225 · 2022: 217 · 2021: 163 · 2020: 168 · 2019: 110 · 2018: 90 · 2017: 47 · 2016: 31 · 2015: 16 · 2014: 16
- FWCI
- 18.88
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Political Analysis (publishedVersion)
- Funders
- University of Cambridge
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Economic and Environmental Valuation
- Social Sciences › Sociology and Political Science › Names, Identity, and Discrimination Research
- Keywords
- conjoint analysis, causal inference, stated preference experiment, causal estimands, nonparametric identification, potential outcomes framework, randomized experiments, vote choice, attitudes toward immigrants
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.