Visualising Gender Inequality at Workplaces in Tableau

I recently finished reading Lean In by Sheryl Sandberg recently. And I must say that I am deeply moved and touched to the core by the book

#LeanIn

A photo posted by Ashish Chauhan (@ashishyoungy) on

I checked MakeoverMonday on Tuesday, a day late and I was in awe for a minute to see the data that was there for this week's dataviz because of the topic of the data. It was about gender inequality at workplaces, a topic that I deeply care about, especially since I read Lean In. It was a small dataset published by McKinsey with LeanIn.org, a foundation being run by Sheryl.

This is the data visualization published on McKinsey's website

http://www.mckinsey.com/business-functions/organization/our-insights/women-in-the-workplace

http://www.mckinsey.com/business-functions/organization/our-insights/women-in-the-workplace

Andy Kriebel had some issues with this visualization, so he asked us to recreate this visualization in this week's MakeoverMonday. I started to play around with the data and this is the first table that I created

And I was like what! Mere 17% women represent at the top-level in jobs in America and only 1% change over the period of three years. That is ≈ 0.33% per year. Assuming an oversimplistic model of linear constant change, It would take more than 100 years for women to represent an equal share at this level. 

From here on, I knew what story to present in my this week's dataviz entry. 

An interactive version of this data visualisation here

An interactive version of this data visualisation here

Key Message of My DataViz

Even though we are moving in the right direction, we are not moving fast enough

We simply cannot wait till 22nd Century for women to represent equally at workplaces

This post is part of MakeoverMonday DataViz Series, updated every Monday. You can see all of them here: