The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS: 1992-2003

Julia Soulakova, William W. Davis, Anne Hartman, James Gibson

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

This study identified whether survey administration mode (telephone or in-person) and respondent type (self or proxy) result in discrepant prevalence of current smoking in the adult U.S. population, while controlling for key sociodemographic characteristics and longitudinal changes of smoking prevalence over the 11-year period from 1992-2003. We used a multiple logistic regression analysis with replicate weights to model the current smoking status logit as a function of a number of covariates. The final model included individual- and familylevel sociodemographic characteristics, survey attributes, and multiple two-way interactions of survey mode and respondent type with other covariates. The respondent type is a significant predictor of current smoking prevalence and the magnitude of the difference depends on the age, sex, and education of the person whose smoking status is being reported. Furthermore, the survey mode has significant interactions with survey year, sex, and age. We conclude that using an overall unadjusted estimate of the current smoking prevalence may result in underestimating the current smoking rate when conducting proxy or telephone interviews especially for some sub-populations, such as young adults. We propose that estimates could be improved if more detailed information regarding the respondent type and survey administration mode characteristics were considered in addition to commonly used survey year and sociodemographic characteristics. This information is critical given that future surveillance is moving toward more complex designs. Thus, adjustment of estimates should be contemplated when comparing current smoking prevalence results within a given survey series with major changes in methodology over time and between different surveys using various modes and respondent types.

Original languageEnglish (US)
Pages (from-to)123-137
Number of pages15
JournalSurvey Research Methods
Volume3
Issue number3
StatePublished - Jan 1 2009

Fingerprint

smoking
human being
telephone interview
interaction
telephone
young adult
surveillance
regression analysis
logistics
methodology
education

Keywords

  • Multiple imputation
  • Multiple logistic regression
  • Race bridging
  • Replicate weights

ASJC Scopus subject areas

  • Education

Cite this

The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS : 1992-2003. / Soulakova, Julia; Davis, William W.; Hartman, Anne; Gibson, James.

In: Survey Research Methods, Vol. 3, No. 3, 01.01.2009, p. 123-137.

Research output: Contribution to journalArticle

Soulakova, J, Davis, WW, Hartman, A & Gibson, J 2009, 'The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS: 1992-2003', Survey Research Methods, vol. 3, no. 3, pp. 123-137.
Soulakova, Julia ; Davis, William W. ; Hartman, Anne ; Gibson, James. / The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS : 1992-2003. In: Survey Research Methods. 2009 ; Vol. 3, No. 3. pp. 123-137.
@article{7df6f181f19a4ba7b9bf4eb756d51569,
title = "The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS: 1992-2003",
abstract = "This study identified whether survey administration mode (telephone or in-person) and respondent type (self or proxy) result in discrepant prevalence of current smoking in the adult U.S. population, while controlling for key sociodemographic characteristics and longitudinal changes of smoking prevalence over the 11-year period from 1992-2003. We used a multiple logistic regression analysis with replicate weights to model the current smoking status logit as a function of a number of covariates. The final model included individual- and familylevel sociodemographic characteristics, survey attributes, and multiple two-way interactions of survey mode and respondent type with other covariates. The respondent type is a significant predictor of current smoking prevalence and the magnitude of the difference depends on the age, sex, and education of the person whose smoking status is being reported. Furthermore, the survey mode has significant interactions with survey year, sex, and age. We conclude that using an overall unadjusted estimate of the current smoking prevalence may result in underestimating the current smoking rate when conducting proxy or telephone interviews especially for some sub-populations, such as young adults. We propose that estimates could be improved if more detailed information regarding the respondent type and survey administration mode characteristics were considered in addition to commonly used survey year and sociodemographic characteristics. This information is critical given that future surveillance is moving toward more complex designs. Thus, adjustment of estimates should be contemplated when comparing current smoking prevalence results within a given survey series with major changes in methodology over time and between different surveys using various modes and respondent types.",
keywords = "Multiple imputation, Multiple logistic regression, Race bridging, Replicate weights",
author = "Julia Soulakova and Davis, {William W.} and Anne Hartman and James Gibson",
year = "2009",
month = "1",
day = "1",
language = "English (US)",
volume = "3",
pages = "123--137",
journal = "Survey Research Methods",
issn = "1864-3361",
publisher = "European Survey Research Association",
number = "3",

}

TY - JOUR

T1 - The impact of survey and response modes on current smoking prevalence estimates using TUS-CPS

T2 - 1992-2003

AU - Soulakova, Julia

AU - Davis, William W.

AU - Hartman, Anne

AU - Gibson, James

PY - 2009/1/1

Y1 - 2009/1/1

N2 - This study identified whether survey administration mode (telephone or in-person) and respondent type (self or proxy) result in discrepant prevalence of current smoking in the adult U.S. population, while controlling for key sociodemographic characteristics and longitudinal changes of smoking prevalence over the 11-year period from 1992-2003. We used a multiple logistic regression analysis with replicate weights to model the current smoking status logit as a function of a number of covariates. The final model included individual- and familylevel sociodemographic characteristics, survey attributes, and multiple two-way interactions of survey mode and respondent type with other covariates. The respondent type is a significant predictor of current smoking prevalence and the magnitude of the difference depends on the age, sex, and education of the person whose smoking status is being reported. Furthermore, the survey mode has significant interactions with survey year, sex, and age. We conclude that using an overall unadjusted estimate of the current smoking prevalence may result in underestimating the current smoking rate when conducting proxy or telephone interviews especially for some sub-populations, such as young adults. We propose that estimates could be improved if more detailed information regarding the respondent type and survey administration mode characteristics were considered in addition to commonly used survey year and sociodemographic characteristics. This information is critical given that future surveillance is moving toward more complex designs. Thus, adjustment of estimates should be contemplated when comparing current smoking prevalence results within a given survey series with major changes in methodology over time and between different surveys using various modes and respondent types.

AB - This study identified whether survey administration mode (telephone or in-person) and respondent type (self or proxy) result in discrepant prevalence of current smoking in the adult U.S. population, while controlling for key sociodemographic characteristics and longitudinal changes of smoking prevalence over the 11-year period from 1992-2003. We used a multiple logistic regression analysis with replicate weights to model the current smoking status logit as a function of a number of covariates. The final model included individual- and familylevel sociodemographic characteristics, survey attributes, and multiple two-way interactions of survey mode and respondent type with other covariates. The respondent type is a significant predictor of current smoking prevalence and the magnitude of the difference depends on the age, sex, and education of the person whose smoking status is being reported. Furthermore, the survey mode has significant interactions with survey year, sex, and age. We conclude that using an overall unadjusted estimate of the current smoking prevalence may result in underestimating the current smoking rate when conducting proxy or telephone interviews especially for some sub-populations, such as young adults. We propose that estimates could be improved if more detailed information regarding the respondent type and survey administration mode characteristics were considered in addition to commonly used survey year and sociodemographic characteristics. This information is critical given that future surveillance is moving toward more complex designs. Thus, adjustment of estimates should be contemplated when comparing current smoking prevalence results within a given survey series with major changes in methodology over time and between different surveys using various modes and respondent types.

KW - Multiple imputation

KW - Multiple logistic regression

KW - Race bridging

KW - Replicate weights

UR - http://www.scopus.com/inward/record.url?scp=80052968959&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052968959&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:80052968959

VL - 3

SP - 123

EP - 137

JO - Survey Research Methods

JF - Survey Research Methods

SN - 1864-3361

IS - 3

ER -