Feasibility and Utility of Multi-Mode Data Collection in Random-Digit Dialing Cellphone Surveys For Young Adults: Evidence From a Survey Experiment (Preprint)

2021 ◽  
Author(s):  
Daniel Alexander Gundersen ◽  
Jonathan Wivagg ◽  
William J Young ◽  
Ting Yan ◽  
Cristine D Delnevo
Keyword(s):  
Young Adults ◽  
Data Collection ◽  
Cigarette Smoking ◽  
Mixed Mode ◽  
Smoking Prevalence ◽  
Response Rate ◽  
Single Mode ◽  
Web Survey ◽  
Random Digit Dialing ◽  
Random Digit

BACKGROUND Young adults’ early adoption of new cellphone technologies have created challenges to survey recruitment but offer opportunities to combine traditional random-digit dialing (RDD) sampling with web-mode data collection. The National Young Adult Health Survey (NYAHS) was designed to test the feasibility of this methodology. OBJECTIVE (1) Compare response rates across telephone- and web-modes, (2) assess sample representativeness, (3) examine the impact of sampling design and sample weighting on statistical precision, and (4) compare cigarette smoking prevalence to a gold standard national survey. METHODS The sampling frame was randomized to: (1) single-mode telephone interviews, (2) telephone to web sequential mixed mode, and (3) single-mode web survey, and 831 young adults (18-34 years) were recruited via RDD at baseline. A soft-launch was conducted prior to main-launch. We compared web-mode to pooled single-mode telephone and mixed-mode (telephone-mode) on the American Association for Public Opinion Research (AAPOR) response rate 3 (RR3) and for the screening (S-RR3) and extended surveys (E-RR3). Base-weighted demographic distributions were compared to the American Community Survey (ACS). Telephone- and web-mode samples were pooled and calibrated to the ACS to calculate design effects (DEFF) and compare cigarette smoking prevalence to the National Health Interview Survey (NHIS). Consistency of estimates was judged by overlap of 95% confidence intervals (CI). RESULTS RR3 was higher in telephone- than web-mode (24% and 30% vs. 6.1% and 12.5% for soft- and main-launch, respectively), which was reflected in S-RR3 and E-RR3. The telephone-mode over-represented men (57.1% vs. 50.9%) and those enrolled in college (40.3% vs. 23.8%) and under-represented those with a BA/BS+ (34.4% vs. 55%). The web-mode over-represented white, non-Latinos (70.7% vs. 54.4%) and those with some college (30.4% vs. 7.6%) and under-represented Latinos (13.6% vs. 20.7%) and those with a HS/GED (15.3% vs. 29.3%). The DEFF was 1.29 (subpopulation range: 0.96 to 1.90). The NYAHS cigarette smoking prevalence was consistent with NHIS overall (15%, CI 12.4% to 18% vs. 13.5%, CI 12.3% to 14.7%), with notable deviation among 18 to 24 year-olds (15.6%, CI 11.3% to 22.2% vs. 8.7%, CI 7.1% to 10.6%), and those with <BS/BA as highest level of education (24%, CI 19.3% to 29.4% vs. 17.1%, CI 15.6% to 18.7%). CONCLUSIONS RDD sampling for a web survey is not feasible for young adults due to its low response rate. However, combining this methodology with RDD telephone surveys may have great potential for including media and collecting auto-photographic data in population surveys

scholarly journals The Implications of Alternative Allocation Criteria in Adaptive Design for Panel Surveys

2017 ◽  
Vol 33 (3) ◽  
pp. 781-799 ◽  
Author(s):  
Olena Kaminska ◽  
Peter Lynn
Keyword(s):  
Data Collection ◽  
Sample Size ◽  
Mixed Mode ◽  
Response Rate ◽  
Single Mode ◽  
Quality Measure ◽  
Alternative Methods ◽  
Effective Sample Size ◽  
Random Allocation ◽  
Sample Sizes

AbstractAdaptive survey designs can be used to allocate sample elements to alternative data collection protocols in order to achieve a desired balance between some quality measure and survey costs. We compare four alternative methods for allocating sample elements to one of two data collection protocols. The methods differ in terms of the quality measure that they aim to optimize: response rate, R-indicator, coefficient of variation of the participation propensities, or effective sample size. Costs are also compared for a range of sample sizes. The data collection protocols considered are CAPI single-mode and web-CAPI sequential mixed-mode. We use data from a large experiment with random allocation to one of these two protocols. For each allocation method we predict outcomes in terms of several quality measures and costs. Although allocating the whole sample to single-mode CAPI produces a higher response rate than allocating the whole sample to the mixed-mode protocol, we find that two of the targeted allocations achieve a better response rate than single-mode CAPI at a lower cost. We also find that all four of the targeted designs out-perform both single-protocol designs in terms of representativity and effective sample size. For all but the smallest sample sizes, the adaptive designs bring cost savings relative to CAPI-only, though these are fairly modest in magnitude.

Is the European Health Interview Survey online yet? Response and net sample composition of a web-based data collection

2019 ◽  
Vol 30 (3) ◽  
pp. 595-601 ◽  
Author(s):  
Elise Braekman ◽  
Rana Charafeddine ◽  
Stefaan Demarest ◽  
Sabine Drieskens ◽  
Jean Tafforeau ◽  
...  
Keyword(s):  
Data Collection ◽  
Mixed Mode ◽  
Response Rate ◽  
Web Based ◽  
Health Interview Survey ◽  
Health Interview ◽  
Sample Composition ◽  
Interview Survey ◽  
European Health ◽  
European Health Interview Survey

Abstract Background The European Health Interview Survey (EHIS) provides cross-national data on health status, health care and health determinants. So far, 10 of the 30 member states (MS) opted for web-based questionnaires within mixed-mode designs but none used it as the sole mode. In the context of future EHIS, the response rate and net sample composition of a web-only approach was tested. Methods A Belgian study with a target sample size of 1000 (age: 16–85) was organized using the EHIS wave 3 model questionnaire. The sample was selected according to a multistage, clustered sampling procedure with geographical stratification. Field substitution was applied; non-participating households were replaced by similar households regarding statistical sector, sex and age. There was one reminder letter and a €10 conditional incentive. Results Considering all substitutions, a 16% response rate was obtained after sending one reminder. Elderly, Brussels Capital inhabitants, people living without a partner and those with a non-Belgian nationality were less responding. By design, there were no differences between the initial and final net sample regarding substitution characteristics. Nevertheless, people living without a partner, non-Belgians and lower educated people remain underrepresented. Conclusion There was a low response rate, particularly for some population groups. The response rate was lower than those of MS using mixed-mode designs including web, especially these comprising interviewer-based approaches. Despite the long and complex questionnaire, there was a low break off rate. So far, web-only data collection is not an acceptable strategy for population-based health surveys but efforts to increase the response should be further explored.

scholarly journals Estimating Components of Mean Squared Error to Evaluate the Benefits of Mixing Data Collection Modes

2017 ◽  
Vol 33 (2) ◽  
pp. 303-334
Author(s):  
Caroline Roberts ◽  
Caroline Vandenplas
Keyword(s):  
Data Collection ◽  
Mixed Mode ◽  
Mean Squared Error ◽  
Mail Survey ◽  
Single Mode ◽  
Sampling Variance ◽  
Sampling Frame ◽  
Relative Contribution ◽  
Survey Error ◽  
Squared Error

AbstractMixed mode data collection designs are increasingly being adopted with the hope that they may reduce selection errors in single mode survey designs. Yet possible reductions in selection errors achieved by mixing modes may be offset by a potential increase in total survey error due to extra measurement error being introduced by the additional mode(s). Few studies have investigated this empirically, however. In the present study, we compute the Mean Squared Error (MSE) for a range of estimates using data from a mode comparison experiment. We compare two mixed mode designs (a sequential web plus mail survey, and a combined concurrent and sequential CATI plus mail survey) with a single mode mail survey. The availability of auxiliary data on the sampling frame allows us to estimate several components of MSE (sampling variance, non-coverage, nonresponse and measurement bias) for a number of sociodemographic and target variables. Overall, MSEs are lowest for the single mode survey, and highest for the CATI plus mail design, though this pattern is not consistent across all estimates. Mixing modes generally reduces total bias, but the relative contribution to total survey error from different sources varies by design and by variable type.

A Comparison of ABS Mail and RDD Surveys for Measuring Consumer Attitudes

2014 ◽  
Vol 56 (6) ◽  
pp. 737-756 ◽  
Author(s):  
Mahmoud Elkasabi ◽  
Z. Tuba Suzer-Gurtekin ◽  
James M. Lepkowski ◽  
Uiyoung Kim ◽  
Richard Curtin ◽  
...  
Keyword(s):  
Data Collection ◽  
Experimental Studies ◽  
Mail Survey ◽  
Consumer Attitudes ◽  
University Of Michigan ◽  
Telephone Surveys ◽  
Random Digit Dialing ◽  
Random Digit ◽  
Landline Telephone ◽  
The University

The increasing cost and decreasing coverage of Random Digit Dialing (RDD) landline telephone surveys motivated The Surveys of Consumer Attitudes (SCA) at the University of Michigan to conduct monthly experimental mail survey studies using address-based sampling (ABS). The primary objectives of the experimental studies were to evaluate the feasibility of transitioning the data collection operations from telephone to mail and to investigate differential survey errors between the two modes. Overall mail survey response rates were comparable to the RDD landline survey. Coverage improved using ABS, with more than 20% of the mail responses from non-landline telephone households not covered by the RDD landline telephone surveys. Mail survey respondents from households without landline telephones were more likely to be younger, have lower income, be renters and live in one-person households. There were no apparent measurement or reporting differences between the telephone and mail self-administered modes of data collection. Furthermore, inclusion of non-landline telephone households did not result in any substantial demographic or economic attitude differences between the two approaches.

An alternative measure of response rate in random-digit-dialing surveys that screen for eligible subpopulations

2000 ◽  
Vol 26 (2) ◽  
pp. 99-109 ◽  
Author(s):  
Trena M. Ezzati-Rice ◽  
Martin R. Frankel ◽  
David C. Hoaglin ◽  
John D. Loft ◽  
Victor G. Coronado ◽  
...  
Keyword(s):  
Response Rate ◽  
Alternative Measure ◽  
Random Digit Dialing ◽  
Random Digit

scholarly journals Examining differences in cigarette smoking prevalence among young adults across national surveillance surveys

PLoS ONE ◽  
2019 ◽  
Vol 14 (12) ◽  
pp. e0225312
Author(s):  
Peter Messeri ◽  
Jennifer Cantrell ◽  
Paul Mowery ◽  
Morgane Bennett ◽  
Elizabeth Hair ◽  
...  
Keyword(s):  
Young Adults ◽  
Cigarette Smoking ◽  
Smoking Prevalence ◽  
National Surveillance

scholarly journals Vaccination Coverage of Adults Aged above 19 Years using Mixed-Mode Random Digit Dialing (RDD) Survey

2015 ◽  
Vol 19 (1) ◽  
pp. 58-70 ◽  
Author(s):  
이석구 ◽  
박은영 ◽  
김미영 ◽  
양현일 ◽  
전소연 ◽  
...  
Keyword(s):  
Vaccination Coverage ◽  
Mixed Mode ◽  
Random Digit Dialing ◽  
Random Digit

scholarly journals Random Digit Dialing versus Address-Based Sampling using Telephone Data Collection

2016 ◽  
Vol 9 (3) ◽  
pp. 1-9
Author(s):  
Bonnie E. Shook-Sa* ◽  
Lauren Klein Warren* ◽  
David Roe ◽  
Douglas Currivan
Keyword(s):  
Data Collection ◽  
Random Digit Dialing ◽  
Random Digit

scholarly journals Assessing the Feasibility and Sample Quality of a National Random-digit Dialing Cellular Phone Survey of Young Adults

2013 ◽  
Vol 179 (1) ◽  
pp. 39-47 ◽  
Author(s):  
D. A. Gundersen ◽  
R. S. ZuWallack ◽  
J. Dayton ◽  
S. E. Echeverria ◽  
C. D. Delnevo
Keyword(s):  
Young Adults ◽  
Cellular Phone ◽  
Sample Quality ◽  
Random Digit Dialing ◽  
Random Digit ◽  
Phone Survey
Sign in / Sign up

Export Citation Format

Share Document