Typology Central - View Single Post

nightning · 05-15-2008, 12:05 AM

In science/psychology... we say there are two types of validity. Face/content validity and criterion validity.

Measurement Validity Types

Basically MBTI has one but not the other type of validity.

In terms of content validity... That the scale is measuring traits that actually exists in people... MBTI fails badly. It's like a self-fulfilling hypothesis. There has been no prove that these 16 types do exist... this remains true despite numerous experimental studies conducted by the scientific community. In this process, what they noticed was a different pattern appearing. That people's personalities can be broadly categorized into 5 different traits and that these traits are normally distributed across the population (which completely goes against the idea of having 16 distinct types). This system developed by McCrae et al is the five factor model: Big Five personality traits - Wikipedia, the free encyclopedia

What the test instrument for MBTI have is the other type, criterion validity... or more specifically test, re-test reliability. After so many version of MBTI... consistency between testing of the official test is up to some impressive 80+%. However this means nothing if what's being measured is incorrect.

It's like throwing darts... it doesn't matter whether you get all of them at the same spot... if they don't land on target, you get no points.

</ of rambling>

So why am I here? While I believe types are artificially imposed, the system does provide a structural framework as a way of looking at the vastly complex behavioral responses and interactions in people.

Theory's wrong... but it's still nonetheless useful.

05-15-2008, 12:05 AM	#30 (permalink)
nightning ish red no longer sad Join Date: Apr 2007 Type: INfJ Location: INTJ license revoked :( Posts: 3,346	In science/psychology... we say there are two types of validity. Face/content validity and criterion validity. Measurement Validity Types Basically MBTI has one but not the other type of validity. In terms of content validity... That the scale is measuring traits that actually exists in people... MBTI fails badly. It's like a self-fulfilling hypothesis. There has been no prove that these 16 types do exist... this remains true despite numerous experimental studies conducted by the scientific community. In this process, what they noticed was a different pattern appearing. That people's personalities can be broadly categorized into 5 different traits and that these traits are normally distributed across the population (which completely goes against the idea of having 16 distinct types). This system developed by McCrae et al is the five factor model: Big Five personality traits - Wikipedia, the free encyclopedia What the test instrument for MBTI have is the other type, criterion validity... or more specifically test, re-test reliability. After so many version of MBTI... consistency between testing of the official test is up to some impressive 80+%. However this means nothing if what's being measured is incorrect. It's like throwing darts... it doesn't matter whether you get all of them at the same spot... if they don't land on target, you get no points. </ of rambling> So why am I here? While I believe types are artificially imposed, the system does provide a structural framework as a way of looking at the vastly complex behavioral responses and interactions in people. Theory's wrong... but it's still nonetheless useful.