| Peer-Reviewed

Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the National Crime Victimization Survey

Received: 20 November 2018     Accepted: 13 December 2018     Published: 10 January 2019
Views:       Downloads:
Abstract

Panel surveys need to balance the benefits of repeated measurements (e.g., bounded interview, reduced cost, increased response rates) with the drawbacks that may eventually occur (e.g., respondent fatigue, mode effect). The optimal number of interview waves for a panel survey needs to maximize the advantages while minimizing the potential for bias due to incorporating sampling units for too many interview waves. In this paper, we develop cost models for two potential constraints: (1) keeping the number of interviews constant across designs, and (2) keeping the cost constant across designs. These models are applied to the National Crime Victimization Survey (NCVS). The NCVS currently uses a seven-wave or time-in-sample (TIS) design. In an effort to maintain or reduce costs and improve data quality, the Bureau of Justice Statistics commissioned a Panel Design Study to evaluate the effects of changing the NCVS from a 7-TIS design to a 5-TIS, 4-TIS, 3-TIS, or 1-TIS design. The study used a set of simulations to mimic different panel designs. The simulation assumptions were constructed using NCVS data from 1999 to 2011, and included assumptions about sample sizes, costs, response rates, household replacement, type of interview, demographics, and victimization propensities. Samples were simulated with different panel designs and summary victimization propensities, and standard errors were computed for key estimates. Simulations considered both keeping the cost constant and keeping the number of interviews constant across the different panel design options. In this paper, we show the impact of changing the number of panel TISs on property and violent victimization rates in terms of point estimates, variability, sample sizes, and costs, by several population characteristics. Simulation results found that a 4-TIS design is optimal for the NCVS.

Published in Science Journal of Applied Mathematics and Statistics (Volume 6, Issue 5)
DOI 10.11648/j.sjams.20180605.11
Page(s) 135-147
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2019. Published by Science Publishing Group

Keywords

Panel Waves, Optimal Design, Cost Model, National Crime Victimization Survey

References
[1] Truman J, Langton L, Planty M. Criminal Victimization, 2012 (NCJ 243389). Washington, DC: Department of Justice, Bureau of Justice Statistics; 2013. Available from: http://www.bjs.gov/content/pub/pdf/cv12.pdf.
[2] U.S. Census Bureau. National Crime Victimization Survey: Technical Documentation (NCJ 247252). Washington, DC: U.S. Census Bureau; 2014.
[3] U.S. Census Bureau. Current Population Survey: Design and Methogology (Technical Paper 66). Washington, DC: Department of Labor, Bureau of Labor Statistics; 2006.
[4] U.S. Census Bureau. Consumer Expenditure Survey Anthology, 2011 (Report 1030). Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics; 2011.
[5] Cantwell PJ. Rotating Panel Design. In: Cantwell P, editor. The Encyclopedia of Survey Research Methods. Thousand Oaks, CA: Sage Publications; 2008. p. 772-4.
[6] Kasprzyk D, Duncan G, Kalton G, Singh MP, editors. Panel Surveys. New York: John Wiley & Sons; 1989.
[7] Kalton G, Citro CF. Panel Surveys: The Fourth Dimension. European Journal of Social Science Research. 1995; 8(1):25-39.
[8] Biderman AD, Cantor D. A Longitudinal Analysis of Bounding, Respondent Conditioning and Mobility as Sources of Panel Bias in the National Crime Survey. Joint Statistical Meeings, Survey Research Methods Section; Alexandria, VA: American Statistical Association; 1984. p. 708-13.
[9] Gaskell GD, Wright DB, O'Muircheartaigh CA. Telescoping of Landmark Events: Implications for Survey Research. Public Opinion Quarterly. 2000; 64:77-89.
[10] Anderson SJ. Longitudinal Study Designs. In: Liamputtong P, editor. Handbook of Research Methods in Health Social Science. Singapore: Springer; 2018.
[11] Jenkins SP, Smeeding TM. In praise of panel surveys, a Sonder-Panel, and a Sonder-Panel-Papa. In: Erlinghagen M, Hank K, Kreyenfeld M, editors. Innovation and Wissentransfer in der Empirischen Sozial- und Verhaltensforschung. Campus Verlag: Frankfurt am Main; 2018. p. 11-38.
[12] Lynn P, Lugtig PJ. Total Survey Error in Longitudinal Surveys. In: Biemer P, de Leeuw E, Eskman S, Edward B, Kreuter F, Lyberg L, et al., editors. Total Survey Error in Practice. Hoboken, NJ: John Wiley & Sons; 2017. p. 279-99.
[13] Addington LA. Disentangling the Effects of Bounding and Mobility on Reports of Criminal Victimization. Journal of Quantitative Criminology. 2005; 21(3):321-43.
[14] Berzofsky ME, Moore A, Couzens GL, Heller D, Burfeind C, Krebs C. A Total Survey Error Review of the National Crime Victimization Survey Sample Design. forthcoming.
[15] Couzens GL, Berzofsky ME, Krebs C, editors. Analyzing Potential Mode Effects in the National Crime Victimization Survey. Proceedings of the Joint Statistical Meetings, Survey Research Methods Section; 2014; Alexandria, VA: American Statistical Association. Available from: https://www.amstat.org/membersonly/proceedings/2014/data/assets/pdf/312744_89876.pdf.
[16] Bailar BA. The Effects of Rotation Group Bias on Estimates from Panel Surveys. Journal of the American Statistical Association. 1975; 70:23-30.
[17] Halpern-Manners A, Warren JR, Torche F. Panel conditioning in the general social survey. Sociological Methods & Research. 2017; 46(1):103-24.
[18] Hart TC, Rennison MC, Gibson C. Revisiting Respondent 'Fatigue Bias' in the National Crime Victimization Survey. Journal of Quantitative Criminology. 2005; 21(3):345-63.
[19] Cantor D. A Review and Summary of Studies on Panel Conditioning. In: Menard S, editor. Handbook of Longitudinal Research: Design, Measurement, and Analysis. Amsterdam: Academic Press; 2007. p. 124-37.
[20] Hardison-Walters J, Moore A, Berzofsky ME, Langton L. Household Burglary 1994-2011 (NCJ 241754). Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics; 2013. Available from: http://www.bjs.gov/content/pub/pdf/hb9411.pdf.
[21] Planty M, Langton L, Kreb C, Berzofsky ME, Smiley-McDonald H. Female Victims of Sexual Violence, 1994-2010 (NCJ 240655). Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics; 2013. Available from: http://www.bjs.gov/content/pub/pdf/hb9411.pdf.
[22] Berzofsky ME, Biemer P. Classification Error in Crime Victimization Surveys: A Markov Latent Class Analysis. In: Biemer P, de Leeuw E, Eckman S, Edward B, Kreuter F, Lyberg L, et al., editors. Total Survey Error in Practice. Hoboken, NJ: John Wiley & Sons; 2017. p. 387-412.
Cite This Article
  • APA Style

    Marcus Evan Berzofsky, Ivan Carrillo-Garcia. (2019). Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the National Crime Victimization Survey. Science Journal of Applied Mathematics and Statistics, 6(5), 135-147. https://doi.org/10.11648/j.sjams.20180605.11

    Copy | Download

    ACS Style

    Marcus Evan Berzofsky; Ivan Carrillo-Garcia. Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the National Crime Victimization Survey. Sci. J. Appl. Math. Stat. 2019, 6(5), 135-147. doi: 10.11648/j.sjams.20180605.11

    Copy | Download

    AMA Style

    Marcus Evan Berzofsky, Ivan Carrillo-Garcia. Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the National Crime Victimization Survey. Sci J Appl Math Stat. 2019;6(5):135-147. doi: 10.11648/j.sjams.20180605.11

    Copy | Download

  • @article{10.11648/j.sjams.20180605.11,
      author = {Marcus Evan Berzofsky and Ivan Carrillo-Garcia},
      title = {Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the National Crime Victimization Survey},
      journal = {Science Journal of Applied Mathematics and Statistics},
      volume = {6},
      number = {5},
      pages = {135-147},
      doi = {10.11648/j.sjams.20180605.11},
      url = {https://doi.org/10.11648/j.sjams.20180605.11},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.sjams.20180605.11},
      abstract = {Panel surveys need to balance the benefits of repeated measurements (e.g., bounded interview, reduced cost, increased response rates) with the drawbacks that may eventually occur (e.g., respondent fatigue, mode effect). The optimal number of interview waves for a panel survey needs to maximize the advantages while minimizing the potential for bias due to incorporating sampling units for too many interview waves. In this paper, we develop cost models for two potential constraints: (1) keeping the number of interviews constant across designs, and (2) keeping the cost constant across designs. These models are applied to the National Crime Victimization Survey (NCVS). The NCVS currently uses a seven-wave or time-in-sample (TIS) design. In an effort to maintain or reduce costs and improve data quality, the Bureau of Justice Statistics commissioned a Panel Design Study to evaluate the effects of changing the NCVS from a 7-TIS design to a 5-TIS, 4-TIS, 3-TIS, or 1-TIS design. The study used a set of simulations to mimic different panel designs. The simulation assumptions were constructed using NCVS data from 1999 to 2011, and included assumptions about sample sizes, costs, response rates, household replacement, type of interview, demographics, and victimization propensities. Samples were simulated with different panel designs and summary victimization propensities, and standard errors were computed for key estimates. Simulations considered both keeping the cost constant and keeping the number of interviews constant across the different panel design options. In this paper, we show the impact of changing the number of panel TISs on property and violent victimization rates in terms of point estimates, variability, sample sizes, and costs, by several population characteristics. Simulation results found that a 4-TIS design is optimal for the NCVS.},
     year = {2019}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the National Crime Victimization Survey
    AU  - Marcus Evan Berzofsky
    AU  - Ivan Carrillo-Garcia
    Y1  - 2019/01/10
    PY  - 2019
    N1  - https://doi.org/10.11648/j.sjams.20180605.11
    DO  - 10.11648/j.sjams.20180605.11
    T2  - Science Journal of Applied Mathematics and Statistics
    JF  - Science Journal of Applied Mathematics and Statistics
    JO  - Science Journal of Applied Mathematics and Statistics
    SP  - 135
    EP  - 147
    PB  - Science Publishing Group
    SN  - 2376-9513
    UR  - https://doi.org/10.11648/j.sjams.20180605.11
    AB  - Panel surveys need to balance the benefits of repeated measurements (e.g., bounded interview, reduced cost, increased response rates) with the drawbacks that may eventually occur (e.g., respondent fatigue, mode effect). The optimal number of interview waves for a panel survey needs to maximize the advantages while minimizing the potential for bias due to incorporating sampling units for too many interview waves. In this paper, we develop cost models for two potential constraints: (1) keeping the number of interviews constant across designs, and (2) keeping the cost constant across designs. These models are applied to the National Crime Victimization Survey (NCVS). The NCVS currently uses a seven-wave or time-in-sample (TIS) design. In an effort to maintain or reduce costs and improve data quality, the Bureau of Justice Statistics commissioned a Panel Design Study to evaluate the effects of changing the NCVS from a 7-TIS design to a 5-TIS, 4-TIS, 3-TIS, or 1-TIS design. The study used a set of simulations to mimic different panel designs. The simulation assumptions were constructed using NCVS data from 1999 to 2011, and included assumptions about sample sizes, costs, response rates, household replacement, type of interview, demographics, and victimization propensities. Samples were simulated with different panel designs and summary victimization propensities, and standard errors were computed for key estimates. Simulations considered both keeping the cost constant and keeping the number of interviews constant across the different panel design options. In this paper, we show the impact of changing the number of panel TISs on property and violent victimization rates in terms of point estimates, variability, sample sizes, and costs, by several population characteristics. Simulation results found that a 4-TIS design is optimal for the NCVS.
    VL  - 6
    IS  - 5
    ER  - 

    Copy | Download

Author Information
  • RTI International, Division of Statistics and Data Science, Research Triangle Park, USA

  • Statistics Canada, Ottawa, Canada

  • Sections