Summary Report of the AAPOR Task Force on Non-probability Sampling

Baker, Reg; Brick, J. Michael; Bates, Nancy; Battaglia, Mike; Couper, Mick P.; Dever, Jill A.; Gile, Krista J.; Tourangeau, Roger

doi:10.1093/jssam/smt008

Cited by 540 publications

(469 citation statements)

References 80 publications

(112 reference statements)

Supporting

Mentioning

452

Contrasting

Unclassified

Order By: Relevance

“…An external online survey company recruited the respondents using non-probability sampling. Nonprobability sampling has become an acceptable alternative to probability sampling due to coverage and non-response challenges encountered with probability sampling (Baker et al 2013). Non-probability samples, such as the one used in this study, are often gathered online by identifying and recruiting panels of individuals willing to complete surveys (Baker et al 2013).…”

Section: Methodsmentioning

confidence: 99%

Marketing Power Berries: An Importance-Performance Analysis of Blueberry

Qu¹,

Lamm²,

Rumble³

2017

Journal of Applied Communications

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Marketing Power Berries: An Importance-Performance Analysis of Blueberry

Qu¹,

Lamm²,

Rumble³

2017

Journal of Applied Communications

View full text Add to dashboard Cite

“…Coupons also play an important role in incentivizing participation, which requires redeeming coupons, and recruitment efforts, which is reflected in the number of redeemed coupons, equating to the number of recruits. Coupons not only decrease the data collection costs but also eliminate concerns with privacy, which led some to consider RDS innovative (for example, Baker et al 2013) and to advocate RDS consistent with the "voluntariness" spirit of the research participants, while criticizing probability sampling as being intrusive (for example, Constantine 2010). On the other hand, others have raised concerns about bias in RDS due to the incentivized nature of recruitment and the potential for unwarranted influence or coercion in the recruitment process (for example, Phillips 2010; Simon and Mosavel 2010).…”

Section: Introductionmentioning

confidence: 99%

Total Survey Error and Respondent Driven Sampling: Focus on Nonresponse and Measurement Errors in the Recruitment Process and the Network Size Reports and Implications for Inferences

Lee

Suzer‐Gurtekin

Wagner

et al. 2017

Journal of Official Statistics

View full text Add to dashboard Cite

This study attempted to integrate key assumptions in Respondent-Driven Sampling (RDS) into the Total Survey Error (TSE) perspectives and examine TSE as a new framework for a systematic assessment of RDS errors. Using two publicly available data sets on HIV-at-risk persons, nonresponse error in the RDS recruitment process and measurement error in network size reports were examined. On nonresponse, the ascertained partial nonresponse rate was high, and a substantial proportion of recruitment chains died early. Moreover, nonresponse occurred systematically: recruiters with lower income and higher health risks generated more recruits; and peers of closer relationships were more likely to accept recruitment coupons. This suggests a lack of randomness in the recruitment process, also shown through sizable intra-chain correlation. Self-reported network sizes suggested measurement error, given their wide dispersion and unreasonable reports. This measurement error has further implications for the current RDS estimators, which use network sizes as an adjustment factor on the assumption of a positive relationship between network sizes and selection probabilities in recruitment. The adjustment resulted in nontrivial unequal weighting effects and changed estimates in directions that were difficult to explain and, at times, illogical. Moreover, recruiters' network size played no role in actual recruitment. TSE may serve as a tool for evaluating errors in RDS, which further informs study design decisions and inference approaches.

show abstract

“…In addition, the data were weighted using the 2010 U.S. Census for age, gender, and race/ethnicity to ensure the respondents were representative of the population of interest (Kalton & Flores-Cervantes, 2003). This is a common procedure when using non-probability sampling to ensure accuracy and alleviate the impacts of selection, exclusion, and bias (Baker et al, 2013). The results were analyzed descriptively using the Statistical Package for the Social Sciences (SPSS) version 23 and Excel.…”

Section: Methodsmentioning

confidence: 99%