2021
DOI: 10.1111/apps.12302
|View full text |Cite
|
Sign up to set email alerts
|

Data Collection via Online Platforms: Challenges and Recommendations for Future Research

Abstract: Online platforms such as Amazon's Mechanical Turk (MTurk) are increasingly used by researchers to collect survey and experimental data. Yet, such platforms often represent a tumultuous terrain for both researchers and reviewers. Researchers have to navigate the complexities of obtaining representative samples from online participant cohorts, ensuring data quality, ethically incentivizing participant engagement, and maintaining transparency. Reviewers, on the other hand, have to navigate the complexities of eva… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
103
1
1

Year Published

2021
2021
2024
2024

Publication Types

Select...
10

Relationship

1
9

Authors

Journals

citations
Cited by 176 publications
(119 citation statements)
references
References 79 publications
0
103
1
1
Order By: Relevance
“…Additionally, to deal with subject inattentiveness, during data screening we conducted multivariate outlier analysis and also observed the survey completion time of the survey for each respondent based on the approximate time suggested by Survey Monkey, our survey platform. Such practices comply with the recommendations from previous research (Harms & DeSimone, 2015;Meade & Craig, 2012;Newman et al, 2020). In order to avoid issue of repeated participation, we closely monitored the MTurk Worker IDs and IP addresses before approving the HITs.…”
Section: Context and Datamentioning
confidence: 89%
“…Additionally, to deal with subject inattentiveness, during data screening we conducted multivariate outlier analysis and also observed the survey completion time of the survey for each respondent based on the approximate time suggested by Survey Monkey, our survey platform. Such practices comply with the recommendations from previous research (Harms & DeSimone, 2015;Meade & Craig, 2012;Newman et al, 2020). In order to avoid issue of repeated participation, we closely monitored the MTurk Worker IDs and IP addresses before approving the HITs.…”
Section: Context and Datamentioning
confidence: 89%
“…The comparison between real-life, experimental, and survey data for financial decision making, such as investment behavior, has not been sufficiently addressed in the previous literature. There is evidence that data collection method affects data quality (Duffy et al, 2005;Jäckle et al, 2010;de Leeuw, 2005;Newman et al, 2021). The literature on the effect of incentivization on data quality is not coherent (Davern et al, 2003;van den Brakel et al, 2006;Stecklov et al, 2018).…”
mentioning
confidence: 99%
“…Limitations should be considered when interpreting these findings. The use of a nonprobability online panel reduced the ability to obtain results that can be generalized to the larger U.S. population ( Newman et al, 2020 ). This is because participants drawn from such panels tend to be younger, more liberal, more highly educated, and less religious than the general U.S. population ( Chandler & Shapiro, 2016 ).…”
Section: Discussionmentioning
confidence: 99%