[ad_1]
Considerable research has been devoted to studying gender bias in the physical workplace, particularly in the fields of management and economics.
However, gender bias and its manifestations receive less attention within online platforms. There are some exceptions, including operational outcomes such as employment, performance, and compensation. However, how work process design can reduce or increase gender bias is largely ignored.
A new study from the University of Notre Dame investigates how gender bias impacts human-centered operations on online platforms.
“Gender Discrepancies and Bias in Human-Centered Work: Evidence from a Randomized Field Experiment,” by lead author Yoon-seok Song, assistant professor of information technology, analytics, and operations at the University of Notre Dame Mendoza College of Business; It will be published in the Journal of Operations Management. Co-authors include Caitlin Wowak, Robert and Sarah Lumpkins Associate Professor of Business Analytics at the University of Notre Dame, Corey Angst, Jack and Joanne McGraw Family Associate Professor of IT, Analytics, and Operations, and Sung Kyun. Includes Angela Choi of Kate University. South Korea.
For seven months, the team worked with an Asian online weight management platform that provides programs, products, and consulting services for healthy living.
The researchers created an experimental design that paired clients and consultants in various gender combinations to examine how gender bias differed during interactions.
Customers can ask questions through the website’s chat feature, and a consultant will respond and recommend products within 24 hours. Clients then rate their experience.
“Surprisingly, we found that customers have a better experience and spend more on products and services when there is gender disparity,” Son said. “In other words, this happens when a female client is paired with a male consultant, or a male client is paired with a woman.”
The team also found that the impact of gender nonconformity varies depending on customer behavior, such as whether they leave a review, buy a recommended product, or simply browse.
Knowing the consultant’s gender led customers to leave more and higher ratings and click on recommended products to purchase.
Ratings are even higher if the consultant is of the opposite sex, and even higher from female clients with male consultants.
Female customers had higher click-through rates when gender was reported, regardless of gender, but men had higher click-through rates on products only when the consultant was female.
Purchase rates increased only for products recommended by female consultants, with men purchasing significantly more than women.
“Past research suggests that the presence of male consultants may reduce women’s performance. However, in reality, women may perform better than male consultants. We know that there is,” Son said.
The study notes that, “Designing (or redesigning) business processes to account for gender bias is important in all sectors, but services are increasingly human-centered and increasingly move to online platforms that allow for face-to-face interactions. This is especially important in the service industry, where there are many Interaction is almost eliminated. ”
The findings apply to any situation where face-to-face interaction is not required and provide new tools for managers of platform businesses such as TripAdvisor, Yelp, Uber and Lyft. In ride-hailing services, ride decisions are made before the customer and driver meet.
A better understanding of how a customer’s gender influences a company’s bias toward its employees can help companies improve not only their business performance but also their customer service.
Companies can look at a customer’s name and browsing history to infer gender and design more effective customer-employee matching algorithms and training programs.
“Our research can help companies make the most of matches and exploit mismatches to maximize overall efficiency,” Son said. “This is a win-win situation for both customers and businesses.”
contact: Yun-seok Song, 574-631-1666, yson@nd.edu
[ad_2]
Source link
