26 ms·
Here's the correlations between MBTI dimensions and the "big five" personality traits that are commonly used in social-sciences research: https://en.wikipedia.
by nostrademons 10y ago
Here's the correlations between MBTI dimensions and the "big five" personality traits that are commonly used in social-sciences research:
https://en.wikipedia.org/wiki/Myers%E2%80%93Briggs_Type_Indicator#Big_Five https://en.wikipedia.org/wiki/Myers%E2%80%93Briggs_Type_Indi...
When I look at those, I think "Not everything, but not nothing either". A correlation of 0.7+ is pretty high for anything in the social sciences - by comparison, the correlation between intelligence and genetics is about 0.4. It's certainly a lot higher than astrology, which has a correlation of roughly 0.0 with anything other than birthdate.
A lot of the criticism of MBTI is that it breaks people down into 16 arbitrary categories, and there's no way a single category can reflect everything about a person. This is true; there is no way a single category will reflect everything about a person. But we break people down into arbitrary categories all the time - race, social class, profession, job title, political party, ethnicity, place of residence, generation, etc. Categorization by MBTI brings all of the pitfalls of these - you're stereotyping, and you're going to miss important details about the person themself - as well as all the benefits, in the form of a quick mental shorthand to draw inferences about their future behavior on little data.
You can think of it in "Thinking fast and slow" terms: MBTIs and other social categories are in the "thinking fast" category, where you make quick but error-prone judgments on limited information, while actually getting to know someone is the "thinking slow" side, where you get a much more accurate picture of who they are at the cost of significantly higher effort. Indeed, the MBTI has a dimension for this (iNtuitive vs. Sensing), and I've found that online MBTI tests tend to overrepresent the number of Ns in the population compared to offline, professionally-given tests, because Ns tend to be more attracted to the MBTI as a way of thinking about the world.
- AstralStorm 10y agoCorrelation of 0.4 means generally "it could be" in proper statistics. As in "it is also probable for a normal variate that you're seeing a spurious one". You probably wanted effect size instead and a good statistical test.