[Paper Review] Negative Surveys

Fernando Esponda|arXiv (Cornell University)|Aug 7, 2006

Survey Sampling and Estimation Techniques9 references34 citations

TL;DR

This paper introduces Negative Surveys, a privacy-preserving survey method that estimates population proportions of a sensitive polychotomous variable by having respondents eliminate one of t−1 randomly selected alternatives from a set of t answers. Unlike traditional randomized response techniques, it avoids direct self-reporting, reducing information leakage and enhancing privacy through a choice-based randomization mechanism with provable statistical reliability.

ABSTRACT

In this paper we propose a strategy for administering a survey that is mindful of sensitive data and individual privacy. The survey in question seeks to estimate the population proportions of a sensitive, polychotomous variable and does not depend on anonymity, cryptography, or in legal guarantees for its privacy preserving properties. Our technique, called Negative Surveys, presents interviewees with a question and t possible answers, and asks participants to eliminate one of t-1 alternatives at random. The method is closely related to randomized response techniques (RRTs) in that both rely on a random component to preserve privacy; however, while RRTs require respondents to choose among questions and give an answer, negative surveys ask them to choose between possible answers to a single question. This distinction has important consequences for the privacy, methodology, and reliability of our scheme. In the course of the paper we quantify the amount of information surrendered by an interviewee, elaborate on how to estimate the desired population proportions, and discuss the properties of our method at length. We also introduce a specific setup that requires a single coin as a randomizing device, and that limits the amount of information each respondent is exposed to by presenting to them only a subset of the question's alternatives.

Motivation & Objective

• To develop a privacy-preserving survey method that does not rely on anonymity, cryptography, or legal safeguards.
• To estimate population proportions of a sensitive, polychotomous variable while minimizing the information revealed by individual respondents.
• To design a method that is simpler and more intuitive than existing randomized response techniques by focusing on elimination rather than direct response.
• To quantify the information loss and privacy trade-offs inherent in the survey design.
• To provide a practical implementation requiring minimal tools, such as a single coin, for randomization.

Proposed method

• Respondents are shown a question with t possible answers and asked to eliminate one of t−1 alternatives selected at random.
• The method uses a randomization device—such as a single coin—to select which alternatives are presented, limiting exposure to only a subset of the answer options.
• The survey design ensures that no single respondent reveals their true preference directly, preserving privacy through selective visibility and elimination.
• Population proportions are estimated using statistical models that account for the random selection of alternatives and the elimination process.
• The technique is mathematically related to randomized response but differs fundamentally in that it avoids direct self-reporting of sensitive attributes.
• The method is designed to minimize information surrender per respondent while maintaining statistical efficiency in population estimation.

Experimental results

Research questions

• RQ1 How can a survey be designed to estimate sensitive population proportions without relying on anonymity or cryptography?
• RQ2 What is the information leakage per respondent in a survey where participants eliminate answers rather than report them?
• RQ3 How does the elimination-based mechanism compare to traditional randomized response in terms of privacy and statistical reliability?
• RQ4 What is the optimal way to randomize the presentation of answer choices to minimize respondent exposure?
• RQ5 Can a single coin serve as a sufficient randomizing device in this framework?

Key findings

• The Negative Survey method successfully estimates population proportions of a sensitive polychotomous variable without requiring anonymity, cryptography, or legal guarantees.
• The method reduces information surrender per respondent by limiting exposure to only a subset of answer choices, enhancing privacy.
• The technique achieves statistical reliability through a well-defined randomization mechanism that preserves estimation accuracy.
• A single coin can be used as a randomizing device, enabling practical deployment with minimal resources.
• The elimination mechanism results in lower information leakage compared to direct-response methods, improving privacy protection.
• The method is mathematically sound and provides a provable framework for estimating sensitive proportions under privacy constraints.