An important challenge for survey research is the elicitation of truthful answers to sensitive questions (e.g., racial prejudice, drug use, sexual behavior). In this project, I develop new statistical methods for survey methodology that can be used to achieve this goal. The first survey methodology is the item count technique or list experiment where respondents are asked to provide the total number of items on a list to which they answer affirmatively rather than to answer each item separately. For the randomly selected control group, the list only includes non-sensitive items. For the treatment group, the list contains a sensitive question in addition to these non-sensitive items. I show how to conduct a multivariate analysis in this setting by developing new estimators and applying them to the measurement of racial prejudice in the United States.
The second survey methodology I study is the endorsement experiment which is used to measure the levels of support for certain political actors. Here, respondents are asked to express their opinion about a particular policy endorsed by a randomly selected political actor. These responses are then contrasted with those from a control group that receives no endorsement. I show how to analyze such survey experiments by developing a Bayesian hierarchical measurement model. The proposed model uses item response theory to estimate support levels. I apply this methodology to recent survey experiments in Afghanistan and Pakistan in order to measure spatial variation in citizens' attitudes towards combatants and Islamist militant groups, respectively.
The third survey methodology is the randomized response method. This survey methodology asks respondents to use a randomization device, such as a coin flip, whose outcome is unobserved by the enumerator. By introducing random noise, the method conceals individual responses and consequently protects respondent privacy. In this project, I review standard designs available to applied researchers, develop various multivariate regression techniques for substantive analyses, propose power analyses to help improve research designs, present new robust designs that are based on less stringent assumptions than those of the standard designs, and make all described methods available through open-source software. I illustrate some of these methods with an original survey about militant groups in Nigeria.
Papers that develop methods:
Bullock, Will, Kosuke Imai, and Jacob Shapiro. (2011). ``Statistical Analysis of Endorsement Experiments: Measuring Support for Militant Groups in Pakistan.'' Political Analysis, Vol. 19, No. 4 (Autumn), pp. 363-384. (lead article)
Imai, Kosuke. (2011). ``Multivariate Regression Analysis for the Item Count Technique.'' Journal of the American Statistical Association, Vol. 106, No. 494 (June), pp. 407-416. (featured article)
Blair, Graeme and Kosuke Imai. (2012). ``Statistical Analysis of List Experiments.'' Political Analysis, Vol. 20, No. 1 (Winter), pp. 47-77.
Blair, Graeme, Kosuke Imai, and Jason Lyall. (2014). ``Comparing and Combining List and Endorsement Experiments: Evidence from Afghanistan.'' American Journal of Political Science, Vol. 58, No. 4 (October), pp. 1043-1063.
Imai, Kosuke, Bethany Park, and Kenneth Greene. (2015). ``Using the Predicted Responses from List Experiments as Explanatory Variables in Regression Models.'' Political Analysis, Vol. 23, No. 2 (Spring), pp. 180-196. Translated in Portuguese and Reprinted in Revista Debates Vol. 9, No 1.
Blair, Graeme, Kosuke Imai, and Yang-Yang Zhou. (2015). ``Design and Analysis of the Randomized Response Technique.'' Journal of the American Statistical Association, Vol. 110, No. 511 (September), pp. 1304-1319.
Chou, Winston, Kosuke Imai, and Bryn Rosenfeld. ``Sensitive Survey Questions with Auxiliary Information.''
Paper that empirically validates the methods:
Rosenfeld, Bryn, Kosuke Imai, and Jacob Shapiro. (2016). ``An Empirical Validation Study of Popular Survey Methodologies for Sensitive Questions.'' American Journal of Political Science, American Journal of Political Science, Vol. 60, No. 3 (July), pp. 783-802.
Papers that describe applications:
Lyall, Jason, Graeme Blair, and Kosuke Imai. (2013). ``Explaining Support for Combatants during Wartime: A Survey Experiment in Afghanistan.'' American Political Science Review, Vol. 107, No. 4 (November), pp. 679-705. Winner of the Pi Sigma Alpha Award.
Lyall, Jason, Kosuke Imai, and Yuki Shiraito. (2015). ``Coethnic Bias and Wartime Informing.'' Journal of Politics, Vol. 77, No. 3 (July), p. 833-848.
Hirose, Kentaro, Kosuke Imai, and Jason Lyall. ``Civilian Attitudes and Insurgent Tactics in Civil War.''
Example JAGS code for endorsement experiments is available at this dataverse.
R package for endorsement experiments: Shiraito, Yuki, and Kosuke Imai. ``endorse: R Package for Analyzing Endorsement Experiments.''
R package for list experiments: Blair, Graeme, and Kosuke Imai. ``list: Multivariate Statistical Inference for the Item Count Technique.''
See the Some examples of list experiments
Presentation slides that are used to describe this project at the 2010 Summer Political Methodology Conference are available for download at this link.
© Kosuke Imai