When to Protect? Using the Crosswise Model to Integrate Protected and Direct Responses in Surveys of Sensitive Behavior

Sensitive survey techniques (SSTs) are frequently used to study sensitive behaviors. However, existing strategies for employing SSTs lead to highly variable prevalence estimates and do not permit analysts to address the question of whether the use of an SST is actually necessary. The current article presents a survey questioning strategy and corresponding statistical framework that fills this gap. By jointly analyzing survey responses generated by an SST (the crosswise model) along with direct responses about the sensitive behavior, the article's framework addresses the question of whether the use of an SST is required to study a given sensitive behavior, provides an efficient estimate of the prevalence of the sensitive behavior, and, in its extended form, efficiently estimates how individual characteristics relate to the likelihood of engaging in the behavior. The utility of the approach is demonstrated through an examination of gender differences in proclivities towards corruption in Costa Rica.

[1]  G. Tian,et al.  Two new models for survey sampling with sensitive characteristic: design and analysis , 2008 .

[2]  Adam Glynn.  What Can We Learn with Statistical Truth Serum?Design and Analysis of the List Experiment , 2013 .

[3]  Aila M. Matanock,et al.  The Enemy at Home: Exploring the Social Roots of Criminal Organizations in Mexico , 2012 .

[4]  Jean-Paul Fox,et al.  A Mixed Effects Randomized Item Response Model , 2008 .

[5]  Peter G. M. van der Heijden,et al.  Meta-Analysis of Randomized Response Research , 2005 .

[6]  S. Edgell,et al.  An empirical test of the unrelated question randomized response technique , 1992 .

[7]  Kosuke Imai,et al.  An Empirical Validation Study of Popular Survey Methodologies for Sensitive Questions , 2016 .

[8]  James H. Kuklinski,et al.  Affirmative Action and the Politics of Realignment , 1998, British Journal of Political Science.

[9]  Jean-Paul Fox,et al.  Using Item Response Theory to Obtain Individual Information From Randomized Response Data: An Application Using Cheating Data , 2008 .

[10]  Paul E. Tracy,et al.  The Validity of Randomized Response for Sensitive Measurements , 1981 .

[11]  Stephen F. Knack,et al.  Gender and Corruption , 1999 .

[12]  Ulf Böckenholt,et al.  Do randomized-response designs eliminate response biases? An empirical study of non-compliance behavior , 2009 .

[13]  Axel Franzen,et al.  Anonymity in the Dictator Game Revisited , 2012 .

[14]  O. Azfar,et al.  Identifying Reticent Respondents: Assessing the Quality of Survey Data on Corruption and Values , 2005, Economic Development and Cultural Change.

[15]  Forrest W. Crawford,et al.  Combining List Experiment and Direct Question Estimates of Sensitive Behavior Prevalence. , 2013, Journal of survey statistics and methodology.

[16]  Stefan Stremersch,et al.  Analysis of sensitive questions across cultures: an application of multigroup item randomized response theory to sexual attitudes and behavior. , 2012, Journal of personality and social psychology.

[17]  Jochen Musch,et al.  Defection in the dark? A randomized-response investigation of cooperativeness in social dilemma games , 2011 .

[18]  A. Goetz.  Political Cleaners: Women as the New Anti‐Corruption Force? , 2007 .

[19]  Daniel W. Gingerich.  Understanding Off-the-Books Politics: Conducting Inference on the Determinants of Sensitive Behavior with Randomized Response Surveys , 2010, Political Analysis.

[20]  Aart C. Kraay,et al.  Misunderestimating Corruption , 2013, Review of Economics and Statistics.

[21]  A. Janus,et al.  The Influence of Social Desirability Pressures on Expressed Immigration Attitudes , 2010 .

[22]  S L Warner,et al.  Randomized response: a survey technique for eliminating evasive answer bias. , 1965, Journal of the American Statistical Association.

[23]  B. Torgler,et al.  Gender and Public Attitudes Toward Corruption and Tax Evasion , 2010 .

[24]  Ben Jann,et al.  Asking Sensitive Questions Using the Crosswise Model An Experimental Survey Measuring Plagiarism , 2012 .

[25]  Patrick D. Bourke,et al.  Estimating Proportions from Randomized Response Data Using the EM Algorithm , 1988 .

[26]  Ivar Krumpal,et al.  Estimating the prevalence of xenophobia and anti-Semitism in Germany: A comparison of randomized response and direct questioning. , 2012, Social science research.

[27]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[28]  Robert P. Berrens,et al.  Examining the Role of Social Isolation on Stated Preferences , 2004 .

[29]  Daniel W. Gingerich.  Political Institutions and Party-Directed Corruption in South America: Stealing for the Team , 2013 .

[30]  J. Hox,et al.  A Comparison of Randomized Response, Computer-Assisted Self-Interview, and Face-to-Face Direct Questioning , 2000 .

[31]  Kosuke Imai,et al.  Design and Analysis of the Randomized Response Technique , 2015 .

[32]  Daniel Corstange,et al.  Sensitive Questions, Truthful Answers? Modeling the List Experiment with LISTIT , 2009, Political Analysis.

[33]  S. Edgell,et al.  Validity of Forced Responses in a Randomized Response Model , 1982 .

[34]  Edmund J. Malesky,et al.  Monopoly Money: Foreign Investment and Bribery in Vietnam, a Survey Experiment , 2013 .

[35]  J. Fox,et al.  Mixture randomized item‐response modeling: a smoking behavior validation study , 2013, Statistics in medicine.

[36]  J. Esarey,et al.  “Fairer Sex” or Purity Myth? Corruption, Gender, and Institutional Context , 2013, Politics & Gender.

[37]  Jean-Paul Fox,et al.  Reducing Social Desirability Bias through Item Randomized Response: An Application to Measure Underreported Desires , 2010 .

[38]  Peter G. M. van der Heijden,et al.  A validation of a computer‐assisted randomized response survey to estimate the prevalence of fraud in social security , 2006 .

[39]  Virginia Oliveros.  A Working Machine: Patronage Jobs and Political Services in Argentina , 2013 .

[40]  J. Fox,et al.  Randomized Item Response Theory Models , 2005 .

[41]  Charles W. Lamb,et al.  An Empirical Validation of the Randomized Response Technique , 1978 .

[42]  James H. Kuklinski,et al.  Racial prejudice and attitudes toward affirmative action , 1997 .

[43]  Kosuke Imai,et al.  Multivariate Regression Analysis for the Item Count Technique , 2011 .

[44]  R. Fisman,et al.  Are women really the "fairer" sex? Corruption and women in government , 2001 .

[45]  Ulf Böckenholt,et al.  Item Randomized-Response Models for Measuring Noncompliance: Risk-Return Perceptions, Social Influences, and Self-Protective Responses , 2007 .

[46]  K. Imai,et al.  Comparing and Combining List and Endorsement Experiments: Evidence from Afghanistan , 2014 .

[47]  Ezequiel Gonzalez-Ocantos,et al.  Vote Buying and Social Desirability Bias: Experimental Evidence from Nicaragua , 2012 .

[48]  Ming Tan,et al.  Sample Surveys With Sensitive Questions: A Nonrandomized Response Approach , 2009 .

[49]  Edward G. Carmines,et al.  Reaching Beyond Race , 1997, PS: Political Science & Politics.

[50]  C. Ellertson,et al.  The Measure of Induced Abortion Levels in Mexico Using Random Response Technique , 2006 .

[51]  H. Sung.  Fairer Sex or Fairer System? Gender and Corruption Revisited , 2003 .

[52]  Kosuke Imai,et al.  Statistical Analysis of List Experiments , 2012, Political Analysis.