Pay for performance, satisfaction and retention in longitudinal crowdsourced research.

Area of Science:

• Social Sciences
• Cognitive Sciences
• Human-Computer Interaction

Background:

• Crowdsourcing is vital for participant recruitment in social and cognitive sciences, supplying up to 50% of research subjects.
• Researchers often overlook the gig-economy worker status of these participants, impacting study design and ethical considerations.

Purpose of the Study:

• To investigate the influence of pay and pay raises on the satisfaction, performance, and retention of worker-participants in crowdsourced research.
• To examine the relationship between compensation and psychological contract elements like perceived choice and justice.

Main Methods:

• A longitudinal study involving 359 participants recruited from Amazon Mechanical Turk (MTurk).
• Experimental manipulation of initial pay and pay raises over time.
• Analysis of performance, satisfaction, perceived choice, justice perceptions, and attrition.

Main Results:

• Compensation levels (initial pay, pay raises, overall pay) did not significantly affect participant performance.
• Pay was a significant predictor of perceived choice, justice perceptions, and attrition rates.
• Worker-participants demonstrated vulnerability due to limited power in pay negotiation.

Conclusions:

• While lower pay may not directly decrease performance in crowdsourced research, ethical considerations regarding fair treatment are paramount.
• Researchers must acknowledge the ethical obligations towards gig-economy worker-participants to ensure fair practices.
• The findings underscore the need for ethical frameworks addressing compensation and worker rights in crowdsourced research.