Accepted for/Published in: JMIR Formative Research
Date Submitted: Nov 17, 2021
Open Peer Review Period: Nov 15, 2021 - Jan 10, 2022
Date Accepted: Jul 17, 2022
(closed for review but you can still tweet)
Gender Biases in Online Physician Ratings: Machine Learning Approach to Understand Differences in Review Ratings and Sentiment of Sanctioned Physicians
ABSTRACT
Background:
Prior research has highlighted gender differences in online physician reviews, however, to date no research has linked online ratings with quality of care.
Objective:
To compare a consumer-generated measure of physician quality (online ratings) with a clinical quality outcome (sanctions for malpractice or improper behavior), to understand how patients’ perception and evaluation of doctors differ based on the physician’s gender and quality.
Methods:
We use data from a large online doctor reviews website and the Federation of State Medical Boards. We implement paragraph vector methods to identify words that are specific to and indicative of the separate groups of physicians. We then enrich these findings by utilizing the NRC word-emotion association lexicon to assign emotional scores to the various segments: gender, gender and sanction, and gender and rating.
Results:
We find significant differences in the sentiment and emotion of reviews for male and female physicians. We find that numerical ratings are lower and the sentiment in text reviews is more negative for women who will be sanctioned than for men who will be sanctioned; sanctioned male doctors are still associated with positive reviews.
Conclusions:
Conclusions:
Given the growing impact of online reviews on demand for physician services, understanding the different reviews faced by male and female physicians is important for consumers and for platform architects in order to revisit their platform design.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.