Here's my take on this Frankensteined assemblage of disjointed data sets they are trying to use (badly) to study the drivers of the gender gap:
It’s interesting to consider how different career paths and total time worked can affect compensation - for both men AND women. The topic certainly warrants further analysis.
That said, McKinsey’s narrative about their approach glosses over serious data challenges. While it’s clear their analysts made significant efforts (and leaps of logic) to transform suboptimal inputs into a grand, authoritative finding—it’s not clear WHY.
Knowing that errors compound with each step of a multi-stage analysis, and without additional visibility into the quality and utility of their source data and assumptions, I'm calling BS on McKinsey’s "calculations". You can read more about the problems a few of us have identified in the comments at https://www.linkedin.com/posts/sarahjacksonweaver_mckinsey-a...
It’s interesting to consider how different career paths and total time worked can affect compensation - for both men AND women. The topic certainly warrants further analysis.
That said, McKinsey’s narrative about their approach glosses over serious data challenges. While it’s clear their analysts made significant efforts (and leaps of logic) to transform suboptimal inputs into a grand, authoritative finding—it’s not clear WHY.
Knowing that errors compound with each step of a multi-stage analysis, and without additional visibility into the quality and utility of their source data and assumptions, I'm calling BS on McKinsey’s "calculations". You can read more about the problems a few of us have identified in the comments at https://www.linkedin.com/posts/sarahjacksonweaver_mckinsey-a...