How do you keep your job postings from limiting your applicant pool?
Introduction
Writing job postings that attract a diverse talent pool is essential in today’s workforce. Depending on the field, the difference between the number of men and women in a field can be huge. The U.S. Bureau of Labor Statistics shows that some industries are male-dominated by as much as 99%, while other fields show the opposite trend, employing up to 99% of women.
Gender diversity in the workplace brings together unique perspectives, skills, and experiences, contributing to a more innovative and productive environment. Part of building a positive offline community is ensuring that the same inclusivity and understanding extend to the online world and vice versa. To achieve this, creating online job posts that appeal to both men and women is crucial, recognizing the differences in their job search priorities.
This article addresses how men and women look for distinct aspects in a job post and investigates how to create inclusive job descriptions that speak to both genders.
The Issue of Diverse Job Postings
There have been several studies to date that have shown that job postings not only commonly show gender bias but that this bias may be keeping men and women out of specific job fields. The problem seems to be that hiring managers are using words in their job ads that either don’t appeal to one gender or cause some level of concern in those who might otherwise apply.
How I Examined Job Posting Language
To investigate this further, I created a corpus of 376 job ads. Half of the postings were for construction management. This field is 91.5% men, according to the U.S. Bureau of Labor Statistics, and the other half of the ads were for registered nurse positions, which is 87.9% women, according to the same statistics.
With this corpus created, I could search through 200,000 words easily, a task that would have taken much longer to sort and categorize by hand. Using the corpus, I could compare the language of these two very non-diverse fields through their job postings.
Using a separate list of words that, in experiments, were shown to influence which type of applicants applied, I created a list of keywords that appeared in the male-dominated field postings but did not appear in the female-dominated field postings and vice versa. Many of these postings significantly lean toward one gender or the other.
With nearly the same token count in each corpus, I could create a simple count of the male or female-biased words in each set of postings.
Results of Corpus Search
See the figure below for a breakdown of those words in each set of postings. I summed the number of each kind of biased word, then calculated the percent difference between both corpora’s totals. The data illustrates how the postings in a male-dominated field contain 12.22% more male-biased words. At the same time, postings in a female-dominated field included 34.86% more female-biased words.
Conclusion
This article illustrates that a problem exists in the ways that job postings are created. Postings can unfairly influence which people will be willing to apply. This influence may be contributing to specific job fields becoming less gender diverse.
So, how do we use this research to increase diversity in these fields?
We can think about the language we use when we post a job ad online to fill a need in our companies. Are we fairly representing the position you are trying to fill? Or are we writing with someone specific in mind? Instead of directing our ads at a single person who fits our image of an employee, we can reconsider the words we use to target the many types of qualified people who will create a more egalitarian work environment.
–Jared Kennedy
