Diversity Analytics
Equal in Numbers, Unequal in Power
You’ve probably heard a lot about diversity and inclusion, and that’s because these concepts have become more and more important in creating a workplace where everyone can thrive. This includes things like race, ethnicity, gender, age, sexual orientation, religion, and disability, as well as differences in education, work experience, and skills.
But it’s not just about recognizing that people are different — it’s about valuing those differences and creating an environment where everyone feels respected and included. That’s where inclusion comes in. Inclusion means making sure that everyone feels like they belong and that they have an equal opportunity to succeed.
Why are diversity and inclusion so important? Well, they make the workplace a better and more productive place. When people feel like they’re part of a team and that their contributions are valued, they’re more likely to be engaged and motivated. And when you have a diverse team, you get a wider range of ideas and perspectives, which can lead to more creative and innovative solutions.
What is the role of HR Analytics in D&I? Mainly, they can work with senior leadership to develop metrics and goals for D&I initiatives, and track progress over time to ensure that the organization is making meaningful progress toward its objectives. By partnering with other stakeholders across the organization, HR analysts can play a critical role in creating a workplace culture that values diversity and promotes inclusivity at all levels.( Check out this article where I explained how D&I affects organisational creativity: https://www.linkedin.com/pulse/creativity-series-1-özgür-özcan/ )
In this blog, I’ll focus on gender diversity by exploring diversity data and track possible glass-ceiling effect or discrimination/bias.
Analysis
We’re looking into a management consulting firm that operates in a dozen countries, and our initial findings were quite positive. With a total of 1491 employees, split almost perfectly down the middle between male and female workers, it seemed like this company was doing a great job of promoting gender diversity and inclusivity.
As we dug deeper into the data, we started to see a different picture emerge. When we broke down the gender count by job tier, we found that the higher up the ladder you went, the proportion of male employees tend to increase. At the senior level, there were far more men than women, and the same trend continued all the way up to the top.
However, we cannot simply rely on these observations to draw conclusions about the existence of a gender-based glass ceiling or discrimination/bias. The observed pattern of data could have occurred purely by random variation.
Get Özgür Özcan’s stories in your inbox
Join Medium for free to get updates from this writer.
This is where the chi-square test comes in — it allows us to determine whether the observed pattern of results is statistically significant or simply due to chance. In other words, the chi-square test helps us to assess whether there is a real relationship between gender and job tier or whether the observed pattern could be explained by random variation.
By comparing the observed frequencies to the frequencies we would expect to see under the null hypothesis of independence, the chi-square test helps us to calculate a p-value. This p-value tells us the probability of observing the current data or more extreme data if the null hypothesis were true. If the p-value is very low (usually less than 0.05), we reject the null hypothesis and conclude that there is a statistically significant relationship between gender and job tier.
Pearson Chi-Square results reveals that the relationship between gender and job tier is significant. In other words, this relationship is meaningful in 1 in 1000 level of certainty and cannot be explained by random variation alone. This finding could offer valuable insights into the nature of the company’s workforce and potential disparities or inequalities that may exist.
What is the next step?
While statistical significance is a good starting point to understand the relationship between gender and job tier within a company, it’s important to keep in mind that it doesn’t reveal the direction or strength of the relationship. For this reason, it’s recommended to conduct additional analyses, such as effect size calculation, post-hoc testing, and loregression analysis. These methods can help provide a more complete understanding of the relationship and whether gender is still a significant predictor of job tier after controlling for other variables.
In addition to quantitative analysis, it’s also useful to consider qualitative research to gain insights into the experiences and perspectives of employees within the company. For instance, conducting interviews or focus groups can help identify the factors that contribute to gender-based disparities and potential solutions. By combining quantitative and qualitative analyses, a more nuanced and actionable understanding of the situation can be gained.
