• You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our free community, you will have access to additional post topics, communicate privately with other members (PM), view blogs, respond to polls, upload content, and access many other special features. Registration is fast, simple and absolutely free, so please join our community today! Just click here to register. You should turn your Ad Blocker off for this site or certain features may not work properly. If you have any problems with the registration process or your account login, please contact us by clicking here.

[MBTI General] Implicit-association test for MBTI Preferences (MBTI preference clarifier)

Seymour

Vaguely Precise
Joined
Sep 22, 2009
Messages
1,579
MBTI Type
INFP
Enneagram
5w4
Instinctual Variant
sx/so
Don't you hate it when someone totally "steals" the idea you've been mulling over by coming up with the same idea on their own, and then going one step farther by actually DOING something about it?

When, this just happened to me, but I can console myself that there's a new, useful way for determining MBTI preferences that you can take yourself (skip to TLDR at end if you don't care about the details).

The IAT

The Implicit-Association Test (IAT )is a instrument introduced in 1998 that attempts to measure a person's automatic (that is, unconscious) associations between concepts.

Typically, the IAT measures associations between two separate concepts by requiring the test taker to quickly categorize words or pictures.

Initially, the subject categorizes into a single concept. In the first two rounds, the subject categories items by one concept (gender, for example, with categories of "male" and "female"), and then by another concept (rationality, with categories of "logical" and "illogical").

Then, the subject must categorize the same items into paired categories (such as "female + logical" and "male + illogical"). Eventually the pairings are reversed (so one might have "male + logical" and "female + illogical"), and the subject re-categorizes words.

All responses are timed, and the timings are used to reveal an association. People tend to be faster at categorizing when the concept are associated, so the IAT can show (possibly unconscious) associations.

The IAT is sometimes used to get at racial bias (by combining categorization of black/white pictures with categorization of positive/negative words), and better predicts behaviors than self-report measures.


IAT meets MBTI


Recently, the Journal of Psychological Type published an article by Robert W. McPeek and Judith Breiner called
Implicit, Explicit, and Best-Fit Assessments of Psychological Type: Explorations in 'Shoes off' Measurement.

The article used two separate studies to compare preferences as measured by the MBTI, the IAT, and as determined by a facilitated "best fit" type process. Some of the interesting findings:

  • MBTI results showed a shift towards E, N, T, and J when compared to IAT results. (Research has shown that those preferences are more socially desirable.)
  • MBTI results shift toward T for men compared to IAT results. Surprisingly, women also tended to shift toward T on the MBTI, but less so than men (perhaps explained by the high value placed on Thinking in the workplace).
  • Best fit results show S->N, E->I when compared to MBTI.
  • Best fit facilitated by an MBTI professional tended to shift results toward the preferences of the facilitator for individuals with E, N, P (and females prefering F).
  • Facilitated fit results tended to shift men towards Thinking and women towards Feeling, when compared to the MBTI results.
  • Results tended to shift towards intuitive when going from IAT->MBTI->Faciliated Best Fit.
So, in general the IAT looks like a promising way to get at MBTI preferences that (perhaps) reduces the influence of cultural desirability. Farther research may show that the IAT gives a better "shoes off" typing result, and is more accurate that typically self-reporting instruments.

Furthermore, facilitated best-fit typing may be influenced by faciliator bias and may lead to less accurate (rather than more accurate) results in some cases. Consequently, perhaps a standardized, online best-fit process should be used before faciliator interaction.


IAT, MBTI and You

So, given the above, I used the terms from the research and free web-based IAT library to cobble together IAT Jungian Type Preference tests.

Each IAT test takes about five to ten minutes of uninterrupted attention. Since IAT tests are timed and one is supposed to go as fast as possible, they are somewhat tiring. I'd recommend using them mostly as an additional datapoint for clarifying middling preferences.

There is a test for each preference pair, and the tests are available here.

The original IAT library required a physical keyboard to work, so I modified it to give the (minimally functional) option of using on-screen buttons. The mobile version is still somewhat irritating, so if you have a physical keyboard you're better off using it (and likely to get more accurate results).


Summary/TLDR

New research indicates a way to determine preferences that may be less skewed by social desirability and perceived utility (and may get at unconscious preferences).

You can take a version of those tests here. There's one test per preference pair. Each test takes about 5-10 minutes, and since you want to go as fast as you can, may be a little tiring.

This approach may be particularly good for clarifying middling preferences, and for working around the tendency to "overthink" one's type.
 

á´…eparted

passages
Joined
Jan 25, 2014
Messages
8,265
I am gonna try this again when I get home, but I got inconclusive results. Likely because I can't make myself go any faster to the point of making mistakes (I made three). :shrug:
 

Jaguar

Active member
Joined
May 5, 2007
Messages
20,647
All that clicking just to get this result?


You do not associate Self with Practical any more or less than you associate Others with Practical.


Fuck me.
 

Seymour

Vaguely Precise
Joined
Sep 22, 2009
Messages
1,579
MBTI Type
INFP
Enneagram
5w4
Instinctual Variant
sx/so
I am gonna try this again when I get home, but I got inconclusive results. Likely because I can't make myself go any faster to the point of making mistakes (I made three). :shrug:

Making the occasional mistake is fine. And it may be that some of your preferences are actually in the middle-ish. Empirically, preferences are distributed in a bell curve. so middling preferences are the most common. This is also backed up by measurable behavior, where stronger preferences are more predictive of behavior than middlings ones.

I also included the calculated tvalue for people who wanted a little more precision that just the English phrasing. The value ranges (applied to the absolute value of the result) are:

  • 2.89+: much more than
  • 2.64 to 2.89: more than
  • 1.99 to 2.64: a little more than
  • 1.66 to 1.99: just slight more than

with 1.00 meaning perfectly balanced/neutral/no associations.
 

prplchknz

Well-known member
Joined
Jun 11, 2007
Messages
34,397
MBTI Type
yupp
You associate Others with Extraversion and Self with Introversion much more than you associate Self with Extraversion and Others with Introversion.
(tvalue = -4.5794031269896225)

what does it mean if it's negative?
 

á´…eparted

passages
Joined
Jan 25, 2014
Messages
8,265
Making the occasional mistake is fine. And it may be that some of your preferences are actually in the middle-ish. Empirically, preferences are distributed in a bell curve. so middling preferences are the most common. This is also backed up by measurable behavior, where stronger preferences are more predictive of behavior than middlings ones.

I also included the calculated tvalue for people who wanted a little more precision that just the English phrasing. The value ranges (applied to the absolute value of the result) are:

  • 2.89+: much more than
  • 2.64 to 2.89: more than
  • 1.99 to 2.64: a little more than
  • 1.66 to 1.99: just slight more than

with 1.00 meaning perfectly balanced/neutral/no associations.

I only tested I/E on my phone and the T value was 0.64

I am gonna do them all on my keyboard at home.
 

Seymour

Vaguely Precise
Joined
Sep 22, 2009
Messages
1,579
MBTI Type
INFP
Enneagram
5w4
Instinctual Variant
sx/so
You associate Others with Extraversion and Self with Introversion much more than you associate Self with Extraversion and Others with Introversion.
(tvalue = -4.5794031269896225)

what does it mean if it's negative?

For the EI test, a positive value means you associate self with extraversion and others with introversion. A negative value means you associate self with introversion and others with extraversion.

For each test (EI, SN, TF, JP), a positive value means associating self with the first preference, and a negative value means you associate self with the second preference.

So a -4.57 means you associate the concept of self very strongly with introverted qualities.
 

HongDou

navigating
Joined
Nov 23, 2012
Messages
5,191
MBTI Type
ENFP
Enneagram
6w7
Instinctual Variant
so/sx
You associate Others with Extraversion and Self with Introversion a little more than you associate Self with Extraversion and Others with Introversion.

You do not associate Self with Practical any more or less than you associate Others with Practical. (go figure)

You associate Others with Thinking and Self with Feeling much more than you associate Self with Thinking and Others with Feeling.

You associate Others with Organized and Self with Spontaneous much more than you associate Self with Organized and Others with Spontaneous.

So according to this I have a slight preference for I>E, no preference for S and N, and a strong preference for both F>T and P>J. Well at least this is one of the first tests to peg me as an introvert - Similar Minds pegged me as an INFP on a JCF test although I think it was because I showed more Si than Te since I got equal scores on Ne and Fi.

With this test I'd be IxFP. Well that's the first time I've ever been pegged as a potential ISFP. :D Fun test! Thanks [MENTION=8074]Seymour[/MENTION].
 

Seymour

Vaguely Precise
Joined
Sep 22, 2009
Messages
1,579
MBTI Type
INFP
Enneagram
5w4
Instinctual Variant
sx/so
I only tested I/E on my phone and the T value was 0.64

I am gonna do them all on my keyboard at home.

I think the physical keyboard is faster and more accurate and more likely to give solid results. When testing on my cell phone I found the on-screen buttons slower (including a perceptible delay before the press registered) and somewhat irritating.

So according to this I have a slight preference for I>E, no preference for S and N, and a strong preference for both F>T and P>J. Well at least this is one of the first tests to peg me as an introvert - Similar Minds pegged me as an INFP on a JCF test although I think it was because I showed more Si than Te since I got equal scores on Ne and Fi.

With this test I'd be IxFP. Well that's the first time I've ever been pegged as a potential ISFP. :D Fun test! Thanks [MENTION=8074]Seymour[/MENTION].

Yay! Glad it was fun!


I do feel sorry for the folks who still get a middling result (after all that effort). No guarantee this approach will reveal a strong preference, but seems somewhat likely it may remove some biases that lead to mistyping.
 

á´…eparted

passages
Joined
Jan 25, 2014
Messages
8,265
So the computer/keyboard DEFINITELY is the way to do it

You associate Self with Extraversion and Others with Introversion much more than you associate Others with Extraversion and Self with Introversion.
E > I (tvalue = 7.90586309633312)

You associate Others with Practical and Self with Imaginative a little more than you associate Self with Practical and Others with Imaginative.
N > S (tvalue = -2.443052095368782)

You do not associate Self with Thinking any more or less than you associate Others with Thinking - (though the tvalue is closer to F, so there is a very slight lean to F).
F ≥ T (tvalue = -1.3354008927190406)

You associate Self with Organized and Others with Spontaneous much more than you associate Others with Organized and Self with Spontaneous.
J > P (tvalue = 6.218638002271763)


Welp, I gotta hand it to this test, it pegged me as ENFJ (though technically ENxJ), with F being the weakest, and both my E and J very very high. If we get enough valid data from this from the forum I, I might make this the go-to test to use considering how accurate it was for me. Not just with type, but how close the relative levels are for myself. It really does get rid of bias. Excellent find Seymour.

[MENTION=8936]highlander[/MENTION], this is a really good one to look at. It's even research backed.
 

Galena

Silver and Lead
Joined
Mar 12, 2013
Messages
3,786
Enneagram
4w5
Instinctual Variant
sx/so
I >> much more >>E
S = no more or less = N
F >> much more >> T
P = no more or less = J

Nothing new, but not a surprise.
 

miss fortune

not to be trusted
Joined
Oct 4, 2007
Messages
20,589
Enneagram
827
Instinctual Variant
sp/so
EI= -2.2840865102195353

PI= -2.568067016320553

TF= 6.504722145699493

OS= -6.322533944042736

apparently my brain has something against the words "plan" and "me" going into the same category... I SUCKED at that test :doh:

on another note... the sensing words that they chose were much more Si than Se and I don't particularly identify with Si as being something I'm all that good at

kept clicking the wrong thing...


and for design purposes, I did it with the onscreen buttons, but my laptop is touchscreen
 

pluviophile

Member
Joined
Jun 8, 2015
Messages
76
MBTI Type
INFP
Enneagram
NFI
Instinctual Variant
sx/so
I (tvalue = -4.776975498505994)
N (tvalue = -2.091332872989884)
F (tvalue = -2.7659055589939006)
P/J (tvalue = -1.0539501849192752) You do not associate Self with Organized any more or less than you associate Others with Organized.
 

Seymour

Vaguely Precise
Joined
Sep 22, 2009
Messages
1,579
MBTI Type
INFP
Enneagram
5w4
Instinctual Variant
sx/so
So the computer/keyboard DEFINITELY is the way to do it.

I agree, physical keyboard is definitely superior.

I just integrated in a work-around to avoid the 300ms delay on mobile devices (iOS and Android, in particular). So mobile results should now be significantly more accurate (although physical keyboard is still likely to be better).

Thanks again for the feedback and for other folks who have posted results here. I'm also happy to take bug reports in order to make improvements.
 

Raspberry_rain

New member
Joined
Apr 28, 2015
Messages
84
MBTI Type
ENFP
Enneagram
4w3
Instinctual Variant
sx/sp
No more or less on any of them??? I think I did something wrong :unsure:
 

Lunar Light

New member
Joined
Jun 26, 2015
Messages
114
MBTI Type
ENFP
Enneagram
6w7
Instinctual Variant
so/sx
I have no idea how much ADHD might affect this because of impulsivity, ahaha, but this is what I got. I was surprised with my intuitive/sensing preference. Not surprised at all with my perceiving/judging preference. xxFP, though I guess it technically leans ENFP :p

You do not associate Self with Extraversion any more or less than you associate Others with Extraversion.
(tvalue = 0.5521272396366753)

You do not associate Self with Practical any more or less than you associate Others with Practical.
(tvalue = -1.0422784516655288)

You associate Others with Thinking and Self with Feeling a little more than you associate Self with Thinking and Others with Feeling.
(tvalue = -2.291180661878695)

You associate Others with Organized and Self with Spontaneous much more than you associate Self with Organized and Others with Spontaneous.
(tvalue = -3.074523963992244)
 

Vasilisa

Symbolic Herald
Joined
Feb 2, 2010
Messages
3,946
Instinctual Variant
so/sx
implicitly

You do not associate Self with Extraversion any more or less than you associate Others with Extraversion.
(tvalue = -0.46469499422840876)

You do not associate Self with Practical any more or less than you associate Others with Practical.
(tvalue = 0.8919039222422438)

You associate Others with Thinking and Self with Feeling much more than you associate Self with Thinking and Others with Feeling.
(tvalue = -3.191125820168382)

You do not associate Self with Organized any more or less than you associate Others with Organized.
(tvalue = 1.5188581302374706)
 

reckful

New member
Joined
Jul 6, 2013
Messages
656
MBTI Type
INTJ
Enneagram
5
You associate Others with Extraversion and Self with Introversion much more than you associate Self with Extraversion and Others with Introversion.
(tvalue = -9.981)

You associate Others with Practical and Self with Imaginative a little more than you associate Self with Practical and Others with Imaginative.
(tvalue = -2.254)

You associate Self with Thinking and Others with Feeling much more than you associate Others with Thinking and Self with Feeling.
(tvalue = 3.447)

You do not associate Self with Organized any more or less than you associate Others with Organized.
(tvalue = 0.877)

In the middle on J/P? Only in my dreams. :alttongue:

Just like when I take a typing test, I found there were stretches when I'd get into a groove and respond very quickly for a while, and other times when I'd have a few stuttery pauses for no apparent reason, and I suspect the timing of those differences didn't have a tremendous amount to do with my type.

By the time I took the fourth test (J/P), I'd learned the trick that things went fastest if I imagined I was whatever type was on my side of the screen, so I pictured myself as a P among a bunch of Js (during the stretch when that was the relevant configuration), and that may be the reason that I came out especially middlish on that dimension.
 

Ghost

Megustalations
Joined
Jul 10, 2013
Messages
1,042
MBTI Type
INFP
Enneagram
5w4
Instinctual Variant
sp/so
You associate Others with Extraversion and Self with Introversion more than you associate Self with Extraversion and Others with Introversion. (tvalue = -2.8033798393539455)

You associate Others with Practical and Self with Imaginative much more than you associate Self with Practical and Others with Imaginative. (tvalue = -4.546506138321923)

You do not associate Self with Thinking any more or less than you associate Others with Thinking.
(tvalue = -0.568)

You do not associate Self with Organized any more or less than you associate Others with Organized.
(tvalue = -0.182)


So...INFP? I made a lot of mistakes on that last one.
 

Bush

cute lil war dog
Joined
Nov 18, 2008
Messages
5,182
Enneagram
3w4
Instinctual Variant
sp/so
MBTI results showed a shift towards E, N, T, and J when compared to IAT results. (Research has shown that those preferences are more socially desirable.)
Huh. This sure does put a damper on the whole 'woe-is-me and society doesn't value me and they value the S way and conformity instead, so let's have an N circlejerk about it' mentality.
 
Top