The Crosswise Model for Surveys on Sensitive Topics: A General Framework for Item Selection and Statistical Analysis

Area of Science:

• Social Sciences
• Statistics
• Survey Methodology

Background:

• Direct survey questions on sensitive topics can lead to response bias.
• Indirect questioning techniques aim to improve data accuracy by protecting participant privacy.
• The Crosswise Model is an indirect technique pairing sensitive and non-sensitive items.

Purpose of the Study:

• To develop an integrated methodology for selecting appropriate baseline items in the Crosswise Model.
• To propose novel Bayesian estimation methods for analyzing data from the Crosswise Model.
• To enhance the efficiency and relax assumptions of existing Crosswise Model applications.

Main Methods:

• Developed a methodology for baseline item selection considering conceptual and statistical factors.
• Proposed four distinct statistical models within the methodology.
• Introduced novel Bayesian estimation techniques for implementing these models.
• Assessed model efficiency compared to common Crosswise Model applications.

Main Results:

• The new methodology provides a structured approach to baseline item selection.
• The proposed Bayesian methods offer improved efficiency and potentially relaxed statistical assumptions.
• Empirical application on LGBT attitudes demonstrates the Crosswise Model's utility.

Conclusions:

• The integrated methodology and Bayesian estimation enhance the Crosswise Model's applicability and reliability.
• This research facilitates more accurate data collection on sensitive issues.
• Available tools (app, code) support wider adoption of the improved Crosswise Model.