Evaluating and improving the quality of survey data from panel and crowd-sourced samples: A practical guide for psychological research.

Abstract

The use of crowd-sourced and panel survey data in addiction research has become widespread. However, the validity of data obtained from newer panels such as Qualtrics has not been extensively evaluated. Furthermore, few addiction researchers appear to employ previously recommended guidelines for maximizing the quality of data obtained from panel samples. The goals of the present study were as follows: (a) to evaluate the quality of survey data obtained from Qualtrics including an evaluation of the company's internal data screening process and (b) to provide a practical implementation guide for data screening practices that maximize the quality of data obtained via panel and crowd-sourced samples. To address the goals, two panel samples evaluating vaping and video gaming behaviors were recruited in Canada via Qualtrics and underwent Qualtrics's internal data screening process before being rigorously rescreened by the authors. The results demonstrate that while Qualtrics's paid internal data quality process flags and removes many low-quality participants, there is still a large portion of participants presented by Qualtrics as high-quality that are likely low-quality responses that need to be screened out. The presented methodology provides a rigorous data screening protocol, including step-by-step application, for crowd-sourced samples in addictive behavior research for maximizing data quality. Researchers should be cautious in the use of Qualtrics data for administration of addiction survey research and are encouraged to use additional data screening procedures to maximize data quality. Public Health SignificanceThe findings demonstrate that while Qualtrics's paid internal data quality process flags and remove many low-quality participants, there is still a large portion of participants that should be screened for quality issues by researchers manually. The presented study provides a practical guide for implementing a rigorous data screening procedure to maximize the quality of panel and crowd-sourced survey data.

Economic resilience and polystrengths: Identifying protective factors associated with economic self-sufficiency among survivors of intimate partner violence.

“…On average, the survey took approximately 25 min for participants to complete. Women were compensated approximately $20 to participate, but the exact rate was designated by Qualtrics (see Belliveau & Yakovenko, 2022, for additional information on Qualtrics panels).…”

Crowdsourcing methods in addiction science: Emerging research and best practices.

“…They also provide practical recommendations to address these issues to include the use of both overt and covert fidelity measures. Belliveau and Yakovenko (2022) complement these findings by providing a practical implementation guide with step-by-step instructions for screening for speeding, straight-lining (i.e., tendency to make the same response in a group of questions), inconsistent responding, nonsensical responding, and missing data. Open-source code for conducting these procedures is provided for those looking to adopt these methods in their own work.…”

Development and validation of a reasons for electronic cigarette use questionnaire.

“…Compared with other crowdsourcing platforms, Prolific has been found to have fewer careless errors and better data quality than MTurk (Jones et al, 2022; Peer et al, 2022). Several strategies for evaluating and confirming the quality of data obtained from crowdsource platforms (Belliveau & Yakovenko, 2022) were utilized in this study, including attention to speeding (completing a questionnaire overly quickly), straight lining (responding the same across all questions), inconsistent responding, and missing responses.…”