• Login
    View Item 
    •   Home
    • UA Graduate and Undergraduate Research
    • UA Theses and Dissertations
    • Dissertations
    • View Item
    •   Home
    • UA Graduate and Undergraduate Research
    • UA Theses and Dissertations
    • Dissertations
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Browse

    All of UA Campus RepositoryCommunitiesTitleAuthorsIssue DateSubmit DateSubjectsPublisherJournalThis CollectionTitleAuthorsIssue DateSubmit DateSubjectsPublisherJournal

    My Account

    LoginRegister

    About

    AboutUA Faculty PublicationsUA DissertationsUA Master's ThesesUA Honors ThesesUA PressUA YearbooksUA CatalogsUA Libraries

    Statistics

    Most Popular ItemsStatistics by CountryMost Popular Authors

    Three Essays on the Incentives and Design of Survey Techniques

    • CSV
    • RefMan
    • EndNote
    • BibTex
    • RefWorks
    Thumbnail
    Name:
    azu_etd_13896_sip1_m.pdf
    Size:
    635.5Kb
    Format:
    PDF
    Download
    Author
    Flannery, Timothy
    Issue Date
    2015
    Keywords
    Privacy Measures
    Randomized Response
    Response Techniques
    Stigmatized Populations
    Survey Techniques
    Economics
    Correlated Response
    Advisor
    Blume, Andreas
    
    Metadata
    Show full item record
    Publisher
    The University of Arizona.
    Rights
    Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
    Abstract
    My dissertation focuses on the design and incentives of survey techniques. As many institutions use surveys to allocate funding or determine policy, ensuring surveys provide accurate information is essential. Though incentives certainly play a role in whether survey participants report information truthfully, economists have largely overlooked the issue while statisticians tend to focus on estimators without directly modeling incentive constraints. One of the chapters models and analyzes the incentives of a commonly used survey technique, randomized response, while the other two chapters of my dissertation design two response techniques which improve upon others found in the literature by obtaining more precise estimates and/or incentivizing participants better. In Chapter One "A Game Theoretic Analysis of the Randomized Response Technique," I explicitly model the decision of participants to truthfully respond in the randomized response survey as a game. Randomized response techniques are used to determine the proportion of a population that belongs to a stigmatized group and introduce noise so the surveyor cannot perfectly infer whether a participant belongs to a stigmatized group, regardless of how a participant responds. The interviewer wants to reduce noise as much as possible while maintaining enough noise to ensure participants respond truthfully. Unlike prior literature, I find that the incentives of a participant depend on the number of participants; therefore, the amount of noise required under randomized response decreases when the number of participants increases as adding respondents relaxes truth-telling constraints. However, adding respondents only relaxes incentive constraints to a limit, so some noise remains even when there are a large number of participants. I improve upon the original randomized response technique in two ways in Chapter 2: "Eliciting Private Information using Correlation: A Modification of Randomized Response." In standard randomized response techniques, participants receive questions independently by using a randomization device such as a die. With my technique, participants receive perfectly correlated questions which reduces the variance of the surveyor's estimator while still protecting the privacy of the subjects. Unlike with the randomized response technique, adding correlation allows the surveyor to use a dominant strategy mechanism though it provides limited information. In addition to correlation, my technique provides the surveyor with private information on the distribution of questions asked. Because of the private information, participants become more uncertain of which question is more associated with the stigmatizing characteristic giving them a stronger incentive to respond truthfully. My final chapter, Chapter 3 "A Response Technique with Dominant Strategies in Forced Responses," improves upon a randomized response technique commonly used in practice. In the forced response technique, a fraction of survey participants are directly asked whether they belong to the stigmatizing group while the remaining participants either simply state "yes" or "no" according to a privately observed command. Unlike the original randomized response technique, the surveyor must worry whether participants obey the command in addition to answering truthfully. Psychologically, participants may feel more inclined to disobey than to lie. Therefore, I design a technique where obeying the command is a dominant strategy by providing the surveyor with private information. The paper then discusses a more general response technique with private information and suggests restrictions on the mechanisms to ensure the surveyor does not have an incentive to try to "trick" respondents into believing they have more privacy protection than they actually do. The chapter concludes with a discussion on privacy measures.
    Type
    text
    Electronic Dissertation
    Degree Name
    Ph.D.
    Degree Level
    doctoral
    Degree Program
    Graduate College
    Economics
    Degree Grantor
    University of Arizona
    Collections
    Dissertations

    entitlement

     
    The University of Arizona Libraries | 1510 E. University Blvd. | Tucson, AZ 85721-0055
    Tel 520-621-6442 | repository@u.library.arizona.edu
    DSpace software copyright © 2002-2017  DuraSpace
    Quick Guide | Contact Us | Send Feedback
    Open Repository is a service operated by 
    Atmire NV
     

    Export search results

    The export option will allow you to export the current search results of the entered query to a file. Different formats are available for download. To export the items, click on the button corresponding with the preferred download format.

    By default, clicking on the export buttons will result in a download of the allowed maximum amount of items.

    To select a subset of the search results, click "Selective Export" button and make a selection of the items you want to export. The amount of items that can be exported at once is similarly restricted as the full export.

    After making a selection, click one of the export format buttons. The amount of items that will be exported is indicated in the bubble next to export format.