A study using machine learning to quantify gender discrimination found that the citation rate in the field of astronomy is not good for women. According to researchers at the Swiss Federal Institute of Technology in Zurich, due to gender discrimination, the citation rate of papers by first authors is about 10% lower than that of men.
Gender differences in citation patterns have previously been well documented in the scientific community. However, researchers have not previously tried to quantify how much the difference is due to gender discrimination. For example, men and women may publish different types of papers; women may work in different fields of science and have junior positions.
This latest article, which has not been peer reviewed but published on the preprinted server arXiv, attempts to explain and correct these factors. The authors declined to comment on the paper because they wanted to submit it to the journal Nature-Astronomy. However, other experts say the analysis looks very reliable.
“The novelty of this article is that it breaks the myth that gender differences in citations may be attributed to the specific content of the paper rather than gender,†said Cassidy Sugimoto, an information scientist at Indiana University in Birmingham.
The researchers analyzed 200,000 papers published in five journals from 1950 to 2015. First, they train machine learning algorithms to accurately calculate the citation rate of each paper in which the first author is male. This process takes advantage of as many gender-independent factors as possible, such as the journal, field, and year in which the paper was published, the location of the first author, and how many years the paper has been published.
Later, they asked the algorithm to analyze the first author's paper as a woman. Compared with the first author's paper for men, the actual citation rate of this part (since 1985) is about 6% lower. However, the algorithm predicts that these papers should have a further 4% citation rate.
Researchers say this is their best effort to measure gender discrimination, but their results should be treated with caution because the algorithm may need to take other factors into account.
Floor Standing Digital Signage
Floor Standing Digital Signage,Touch Digital Signage,Lcd Signage Indoor,Digital Signage Menu
Guangdong Elieken Electronic Technology Co.,Ltd. , https://www.elieken.com