The Persistent Gender Pay Gap: How Wage Inequality Affects Anonymous Workers on Amazon’s Mechanical Turk

By Cheskie Rosenzweig, MS, Aaron Moss, PhD, & Leib Litman, PhD

The American workplace has changed a lot since the 1960s, but at least one thing remains the same: men often earn more money than women. This gender wage gap has lingered for decades despite increased public attention and legislative focus. Today, women earn about 20% less than men (1,2).

Throughout the decades, researchers have conducted numerous studies investigating the causes of the gender wage gap. Among the causes these studies have identified are gender differences in education, segregation of men and women into different industries, decisions that men and women make surrounding time off for family care, and discrimination. Although most researchers agree that these factors contribute to the wage gap, there is strong disagreement about how much each factor matters and which is the most important.

Traditional work environments make it difficult or impossible for researchers to ascertain how much each of the factors described above contribute to the gender wage gap. However, some non-traditional work environments offer a unique opportunity to study the gender wage gap by making it possible for researchers to control various factors that are traditionally understood to cause the difference in earnings between men and women. In a recent study, the CloudResearch team collaborated with outside experts to investigate whether a gender wage gap exists in an anonymous crowdsourcing platform called Mechanical Turk (MTurk).

MTurk is an online platform in the gig economy that allows anonymous “workers” to complete small tasks posted by “requesters” in exchange for money. The nature of the MTurk environment removes many of the factors traditionally understood to cause the gender wage gap. For example, discrimination and the flexibility of workers’ schedules cannot explain a wage gap on MTurk because workers are anonymous and free to complete tasks whenever they want. In addition, on MTurk researchers can control for worker education and experience using data about each worker’s tenure on the platform. Finally, researchers can control for task heterogeneity by examining whether men and women cluster into tasks with different titles and descriptions.

To examine whether there is a gender wage gap on MTurk, our research team worked with outside experts to examine data from over 5 million microtasks completed on MTurk from January 2016 through June 2017. These tasks were completed by 12,312 female workers and 9,959 male workers.

The study found that men earned an estimated wage of $5.70 per hour, whereas women earned $5.10 — 60 cents less per hour. These earnings, which were calculated based on average completion times, don’t account for multitasking (when workers hold several tasks in their queue and work on them one at a time) yet still show a 10.5% wage gap. Further analyses indicated that there was even a wage gap present in task selection, with women accepting tasks that paid 5.8% less than men did. Even after controlling for numerous covariates (experience, age, income, education, family composition, race, number of children, task length, the speed of accepting a task, and thirteen types of subtasks), the wage gap was still 4.3%.

Can the Wage Gap on MTurk be Eliminated?

How does a wage gap emerge on an anonymous online platform? The research indicates that women may be willing to work for less money than men, perhaps because of societal stereotypes that undervalue the labor of women.

Although the gender wage gap is especially difficult to eliminate in traditional work environments, eliminating it on MTurk may be relatively easy. Researchers who post studies and requesters who post other tasks can place a gender quota onto each HIT. This relatively simple fix will ensure that men and women have equal opportunities to complete high-paying HITs. And, for research studies, gender quotas will ensure that samples are not biased by participant gender.

At CloudResearch, we have a variety of features that make it easy to set up a fair wage study, including the ability to sample participants using gender quotas. Visit our MTurk Toolkit or Prime Panels webpages to learn more.

References

1 United States Department of Labor (DOL), Bureau of Labor Statistics (BLS) (2016) Women’s earning 83 percent of men’s but vary by occupation. TED Econ Dly, available at https://www.bls.gov/opub/ted/2016/womens-earnings-83-percent-of-mens-but-vary-by- occupation.htm, accessed on 11/04/2019.

2 Davis A (2015) Women still earn less than men across the board (Economic Policy Institute, 2015), available at http://www.epi.org/publication/women-still-earn-less-than- men-across-the-board/, accessed on 11/04/2019.

3 Brown S, Roberts J, Taylor K (2011) The Gender Reservation Wage Gap: evidence from British panel data. Econ Lett 113(1): 88-91.