Social, political, and economic issues are typically complex and influenced by many factors. In social science research, it is difficult to disentangle the importance of different factors and to test corresponding theories. This includes survey research when standard survey items are used to measure determinants of normative beliefs, attitudes, preferences, behavioral intentions, etc. Often all factors appear to be important, and responses are prone to social desirability. In multifactorial survey experiments (MFSEs) participants are asked to evaluate descriptions of a situation (factorial survey/vignette experiment), to choose between two or more mutually exclusive alternatives (discrete choice experiment), or to choose between two alternatives and to rate each of them (conjoint survey experiment). Because these situations and alternatives are experimentally varied on multiple factors/attributes and participants have to make trade-offs between attributes, MFSEs offer a way to disentangle the importance of attributes and reduce social desirability bias. MFSEs can also be used to test theories and mechanisms. They have been applied to a wide range of issues in both the Global North and the Global South, including policy preferences for immigration, acceptance of environmental policies, perceived fairness of the gender pay gap, preferences for health care, and normative beliefs about marriage and child education. This course will introduce different types of MFSEs (factorial survey, discrete choice, conjoint), addressing similarities, differences, and methodological challenges. Participants will become familiar with the theoretical underpinnings and empirical applications of these experiments, and will learn how to design experiments, collect and analyze MFSE data. The course will combine lectures with hands-on sessions using R (for some examples, Stata syntax can be provided). Participants will have the opportunity to discuss their own research projects/ideas.

 

Organizational Structure of the Course

Each day there will be three to four hours of input from the lecturers, combined with three to two hours of practical sessions. The inputs will introduce the main topics such as theoretical foundations, experimental design, questionnaire design and implementation, data analysis and interpretation. All topics will be illustrated with examples, and participants will be encouraged to ask questions and contribute to the discussion. The hands-on sessions will allow participants to apply these topics in groups or, if preferred, individually using R (for some topics, Stata syntax can be provided). During these sessions the lecturers will be available for questions and discussion. Participants will also have the opportunity to discuss their individual projects and/or research problems with the lecturers.

You will find the course useful if:

By the end of the course you will:

Software and Hardware Requirements

As a participant, you should bring your own laptop for use in the course. R, RStudio and the packages apollo and spdesign should be installed prior to the course. The latest version of R is available for free at https://cran.r-project.org/, and RStudio is available at https://www.rstudio.com/. While the course will mainly rely on R/RStudio for the practical exercises and we recommend that you use R/RStudio, you may also use Stata, if you have a strong preference for it. Make sure that Stata is installed on your laptop prior to the course. If needed, GESIS can provide short-term licences for Stata for the duration of the course.