• Graduate program
    • Why Tinbergen Institute?
    • Program Structure
    • Courses
    • Course Registration
    • Recent PhD Placements
    • Admissions
    • Facilities
  • Research
  • News
  • Events
    • Events Calendar
    • Tinbergen Institute Lectures
    • Annual Tinbergen Institute Conference
    • Events Archive
    • Summer School
      • Introduction in Genome-Wide Data Analysis
      • Inequalities in Health and Healthcare
      • Crash Course in Experimental Economics
      • Econometric Methods for Forecasting and Data Science
      • Behavioral Macro and Complexity
  • Times
Home | Events Archive | Item Response Privacy Protection Techniques: a Better Methodology to Assess Sensitive Attitudes and Behaviors
Seminar

Item Response Privacy Protection Techniques: a Better Methodology to Assess Sensitive Attitudes and Behaviors


  • Series
    Array
  • Speaker(s)
    Marco Gregori , Marco Gregori (Erasmus University Rotterdam)
  • Field
    Organizations and Markets
  • Location
    Erasmus University Rotterdam, Theil Building, Room C1-2
    Rotterdam
  • Date and time

    March 05, 2019
    13:00 - 14:00

Abstract:

This paper introduces Item Response Privacy Protection Techniques (IRPPT), a class of methods to obtain truthful responses to sensitive questions while protecting respondents' privacy. The methods proposed fully protect respondents' privacy, are easy to understand, and minimize the problem of noncompliance. The new methods also do not require a control group, which is cost effective, by leveraging information provided by participants elsewhere in the questionnaire. This improves on prior methods such as list experiments and randomized response techniques. The methodology is also suitable for polytomous variables, which is not possible with standard list experiments. We apply our method in a study about sensitive attitudes and workplace behaviors concerning the LGBT community. Direct questions result in substantial underestimation of sensitive attitudes. Further, only 9% of the sample found the method difficult to understand and only 6 % felt that their privacy was not protected. This opens up possibilities for wide-scale research on common but often hidden attitudes and behaviors of people.