Working Paper: CEPR ID: DP14294
Authors: Abigail Adams
Abstract: This paper analyses gender differences in working patterns and wages on Amazon Mechanical Turk. Using information on 2 million tasks, I find no gender difference in task selection nor experience on the platform. Nonetheless, women earn 20% less per hour on average. Half of this gap is explained by differences in the scheduling of work; women have more fragmented work patterns with consequences for their task completion speed. A follow up survey shows that the wage gap is concentrated amongst women with young children, who also report that domestic responsibilities affect their ability to plan and complete work online.
Keywords: gender difference; flexibility; multitasking; telework; gig economy
JEL Codes: J16; J24; J31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Gender (female) (J16) | Hourly Wage (J31) |
| Gender (female) (J16) | Task Completion Speed (C69) |
| Task Completion Speed (C69) | Hourly Wage (J31) |
| Gender (female) (J16) | Fragmented Work Patterns (J29) |
| Fragmented Work Patterns (J29) | Hourly Wage (J31) |
| Gender (female) (J16) | Increased Interruptions from Domestic Responsibilities (D13) |
| Increased Interruptions from Domestic Responsibilities (D13) | Task Completion Speed (C69) |