CrowdSurfer: Seamlessly Integrating Crowd-Feedback Tasks into Everyday Internet Surfing
Abstract
Crowd feedback overcomes scalability issues of feedback collection on interactive website designs. However, collecting feedback on crowdsourcing platforms decouples the feedback provider from the context of use. This creates more effort for crowdworkers to immerse into such context in crowdsourcing tasks. In this paper, we present CrowdSurfer, a browser extension that seamlessly integrates design feedback collection in crowdworkers' everyday internet surfing. This enables the scalable collection of in situ feedback and, in parallel, allows crowdworkers to flexibly integrate their work into their daily activities. In a field study, we compare the CrowdSurfer against traditional feedback collection. Our qualitative and quantitative results reveal that, while in situ feedback with the CrowdSurfer is not necessarily better, crowdworkers appreciate the effortless, enjoyable, and innovative method to conduct feedback tasks. We contribute with our findings on in situ feedback collection and provide recommendations for the integration of crowdworking tasks in everyday internet surfing.
Study specs
- Institution
- Karlsruhe Insititute of Technology
- Discipline
- Human-Computer Interaction
- Year
- 2023
- Human Data Platform
- Prolific
- Source
- View Source DOI Google Scholar
Peer Review & Critical Discussion
Potential Selection Bias in 2023 Cohort
The participant pool shows a concerning overrepresentation of users from high-income demographics. Looking at Table 3, we can see that 78% of respondents had annual incomes above $75k, which significantly limits the generalizability of these findings to broader populations.
Non-naive Participants Issue
I've noticed a methodological concern regarding participant naivety. Given that Prolific users often complete multiple studies, there's a real risk that participants had prior exposure to similar experimental paradigms, which could confound the results.
RLHF Applicability to This Study Design
The implications for RLHF training pipelines are understated. If we accept the authors' conclusions about preference stability, this has direct consequences for how we should structure reward model training. The temporal decay effect described in Section 4.2 is particularly relevant.
Verify your expertise to join discussion
Create an account and verify your credentials to participate in peer discussions.