User Tag List

First 12345 Last

Results 21 to 30 of 46

  1. #21
    girl with a pretty smile Honor's Avatar
    Join Date
    Jul 2012
    MBTI
    ESTJ
    Enneagram
    3w2 so
    Posts
    1,671

    Default

    Quote Originally Posted by bologna View Post
    I haven't delved into the specific stats or sources, but I have bothered to use the equations.

    Scores of:
    N 2
    E 70
    O 95
    A 93
    C 92

    from another test (please don't kill me)

    Yield:
    EI 15.49 I
    SN 23.38 N
    TF 24.66 F
    JP -29.28 J

    Works well enough in this case.

    I'll have to mull over how the legitimacy of using correlations to derive equations and so on, but nothing too troublesome comes to mind off the top of my head.

    It's not a flaw in the algorithm; it's all up in the stats.

    Suppose we have some typology systems--perhaps Grant's Temperament System (GTS) and a Tinkerbell Classification (TC). Suppose also that we've had a bunch of people who've taken both the GTS and TC.

    Data could show that those who have scored as Extraverts on the GTS have also tended to score as Extraverts on the TC scale. There's no stopping the data from showing also that Extraverts on the GTS (perhaps slightly) tend to score high on the TC's "Caffeine Intake" factor.

    (Hell, the data could also show that Extraverts on the GTS tend to score as "Friendlies" on the GTS as well. This would mean that the GTS has factors that aren't quite independent from one another, which is not a good thing.)
    I thought we were all operating under the assumption that scales like extraversion/introversion and sensing/intuiting are discrete, which I believe they are. If we're not all operating under that assumption - and apparently we're not, then it's a whole different story.
    RobertCalifornia: TL thinks im black
    RobertCalifornia: shes my homegurl
    Hive: arent you
    SpankyMcfly: wait... you arent?

    thoughtlost: I am not really religious. I just like getting free stuff from churches.

  2. #22
    Senior Member reckful's Avatar
    Join Date
    Jul 2013
    MBTI
    INTJ
    Enneagram
    5
    Posts
    559

    Default

    Quote Originally Posted by mingularity View Post
    The test I'm using here was designed to explain as much of the variance in the original Big 5 as possible, given the number of questions. It's also free for non-commercial use.

    Why don't you lend a hand and try to make it actually work?
    Um, yeah, and, as you know, those correlation statistics were for a 500-person study involving McCrae & Costa's NEO-PI test and, as you also know (I assume), there's really no reason to think that a study involving the BFI test you're using would end up matching those statistics very closely. And in any event, McCrae & Costa have never (to my knowledge) suggested that it made sense to "convert" Big Five types to MBTI types using anything like the method you're using — and I suspect nobody else respectable has either, but feel free to surprise me with a citation or two.

    As for "lending a hand" and "mak[ing] it actually work," I already offered you my constructive solution, which was that you should "convert" Big Five to MBTI by means of the simple technique of matching Big Five Extraversion to MBTI E/I, Big Five Openness to MBTI S/N, Big Five Agreeableness to MBTI T/F and Big Five Conscientiousness to MBTI J/P.

    Are those going to be perfect matches? No, sir, but that's not the point. The point is that, as @Honor has also been trying to explain to you, you don't improve the conversion by trying to take the other dimensions into account in the way you're proposing. That's really not an appropriate use of those correlation statistics, and I can't help noting that you still haven't addressed my objection that your technique involves using those stats to give a same-directional boost to an affected dimension if you're converting in the opposite direction — which seems to me to mean (unless I'm missing something) that your conversion method must be fatally flawed.

  3. #23
    girl with a pretty smile Honor's Avatar
    Join Date
    Jul 2012
    MBTI
    ESTJ
    Enneagram
    3w2 so
    Posts
    1,671

    Default

    Quote Originally Posted by reckful View Post
    Um, yeah, and, as you know, those correlation statistics were for a 500-person study involving McCrae & Costa's NEO-PI test and, as you also know (I assume), there's really no reason to think that a study involving the BFI test you're using would end up matching those statistics very closely. And in any event, McCrae & Costa have never (to my knowledge) suggested that it made sense to "convert" Big Five types to MBTI types using anything like the method you're using — and I suspect nobody else respectable has either, but feel free to surprise me with a citation or two.

    As for "lending a hand" and "mak[ing] it actually work," I already offered you my constructive solution, which was that you should "convert" Big Five to MBTI by means of the simple technique of matching Big Five Extraversion to MBTI E/I, Big Five Openness to MBTI S/N, Big Five Agreeableness to MBTI T/F and Big Five Conscientiousness to MBTI J/P.

    Are those going to be perfect matches? No, sir, but that's not the point. The point is that, as @Honor has also been trying to explain to you, you don't improve the conversion by trying to take the other dimensions into account in the way you're proposing. That's really not an appropriate use of those correlation statistics, and I can't help noting that you still haven't addressed my objection that your technique involves using those stats to give a same-directional boost to an affected dimension if you're converting in the opposite direction — which seems to me to mean (unless I'm missing something) that your conversion method must be fatally flawed.
    Thank you, reckful - what you wrote may be a clearer way of making the point. Thank God for INTJs.
    RobertCalifornia: TL thinks im black
    RobertCalifornia: shes my homegurl
    Hive: arent you
    SpankyMcfly: wait... you arent?

    thoughtlost: I am not really religious. I just like getting free stuff from churches.

  4. #24
    Senior Member
    Join Date
    Aug 2013
    MBTI
    ISTP
    Posts
    353

    Default

    Quote Originally Posted by reckful View Post
    here's really no reason to think that a study involving the BFI test you're using would end up matching those statistics very closely.
    The desire for a version of the Big 5 that explains as much variance of the full length Big 5 as possible in as few questions as possible is the entire reason the short version exists.

    Quote Originally Posted by reckful View Post
    McCrae & Costa have never (to my knowledge) suggested that it made sense to "convert"
    I suggested that it makes sense. Furthermore the very reason for the existence of their paper was to take the first step towards deriving a rotation matrix for converting between the two. And, I think they probably did.

    Quote Originally Posted by reckful View Post
    you don't improve the conversion by trying to take the other dimensions into account in the way you're proposing.
    Quote Originally Posted by McCrae & Costa
    These findings show that individuals categonzed as Introverts by the MBTI score slightly higher in NEO-PI Neuroticism and much lower in NEO-PI Extraversion, Intuitive types score higher in Openness to Expenence, Feeling types score higher in Neuroticism, Extraversion, and Agreeableness, and lower in Conscientiousness, and Perceiving types score higher in NEO-PI Extraversion and Openness, and lower in Conscientiousness
    You were saying?

    As I explained, we want to use all the available information to explain as much variance in the MBTI scores as possible. The algorithm for doing so is still up in the air. Please do download the paper and try to work out how to do it, as it will save me time later on.

  5. #25
    garbage
    Guest

    Default

    Quote Originally Posted by Honor View Post
    I thought we were all operating under the assumption that scales like extraversion/introversion and sensing/intuiting are discrete, which I believe they are. If we're not all operating under that assumption - and apparently we're not, then it's a whole different story.
    Under that assumption, you're pretty much right. Though, papers and such tend to view the factors as continuous scales ("scores") rather than dichotomies and calculate their correlations with those scores in mind.

    (Also, it appears that what I'd said had already been said anyway. Teaches me to not read threads.)

  6. #26
    Senior Member reckful's Avatar
    Join Date
    Jul 2013
    MBTI
    INTJ
    Enneagram
    5
    Posts
    559

    Default

    Quote Originally Posted by mingularity View Post
    You were saying?

    As I explained, we want to use all the available information to explain as much variance in the MBTI scores as possible. The algorithm for doing so is still up in the air. Please do download the paper and try to work out how to do it, as it will save me time later on.
    Do you honestly think that quote addresses my point? Yes, as I've already acknowledged, there are (relatively minor) correlations between the MBTI dimensions and what you might call the non-corresponding (for the most part) Big Five dimensions — just as there are (relatively minor) internal correlations among the MBTI dimensions (especially S/N and J/P) and (relatively minor) internal correlations among the Big Five factors and facets.

    None of which justifies applying those secondary correlations as part of a Big-Five-to-MBTI conversion in the way you're doing. You're not "adding available information" in a way that's consistent with the nature of that information, and I've never heard of McCrae & Costa or any other respectable source using those kinds of secondary correlations in that way.

    And you still haven't addressed the same-directional-boost-in-both-conversion-directions issue, and I assume that's because you realize something's fatally screwed up about your method but you're not willing to admit it.

  7. #27
    Senior Member
    Join Date
    Aug 2013
    MBTI
    ISTP
    Posts
    353

    Default

    @Honor

    Here are some MBTI (Form F - discontinued in 2012) intercorrelations. I got them from this paper:

    Stricker, L. & Ross, J. (1963). Intercorrelations and reliability of the Myers-Briggs Type Indicator Scales. Psychology Reports.

    * means p < .05, ** means p < .01

    HTML Code:
    High school students
    
           E-I  S-N     T-F     J-P
    E-I         -.02   -.01    .04
    S-N    .05          .00    .18**
    T-F    .03   .07           .13*  
    J-P    .03   .21**  .14**
    
    University students
    
           E-I  S-N     T-F    J-P
    E-I         .08    -.04    .14*
    S-N   -.03          .07    .23**
    T-F    .05 -.02            .09  
    J-P    .13  .31**         -.05

  8. #28
    girl with a pretty smile Honor's Avatar
    Join Date
    Jul 2012
    MBTI
    ESTJ
    Enneagram
    3w2 so
    Posts
    1,671

    Default

    Quote Originally Posted by mingularity View Post
    @Honor

    Here are some MBTI (Form F - discontinued in 2012) intercorrelations. I got them from this paper:

    Stricker, L. & Ross, J. (1963). Intercorrelations and reliability of the Myers-Briggs Type Indicator Scales. Psychology Reports.

    * means p < .05, ** means p < .01

    HTML Code:
    High school students
    
           E-I  S-N     T-F     J-P
    E-I         -.02   -.01    .04
    S-N    .05          .00    .18**
    T-F    .03   .07           .13*  
    J-P    .03   .21**  .14**
    
    University students
    
           E-I  S-N     T-F    J-P
    E-I         .08    -.04    .14*
    S-N   -.03          .07    .23**
    T-F    .05 -.02            .09  
    J-P    .13  .31**         -.05
    ...

    you are still completely missing the point. if the data show correlations, which they may, it's a demonstration of the inherent bias in the test questions (i.e. the method of obtaining data is invalid). it doesn't make any sense for there to be correlations between the dichotomies because they are discrete, as far as i'm concerned. i'm not up for arguing about this anymore. if you don't get it, then you don't get it.
    RobertCalifornia: TL thinks im black
    RobertCalifornia: shes my homegurl
    Hive: arent you
    SpankyMcfly: wait... you arent?

    thoughtlost: I am not really religious. I just like getting free stuff from churches.

  9. #29
    Senior Member
    Join Date
    Aug 2013
    MBTI
    ISTP
    Posts
    353

    Default

    Quote Originally Posted by reckful View Post
    And you still haven't addressed the same-directional-boost-in-both-conversion-directions issue
    The word you are looking for is underdetermined, and, as you may have noticed (HINT: <--), I am aware of it. When you say fatal, do you mean that you have a proof that it is impossible for me to find a sufficient set of constraints? I believe such proofs are known to be NP-hard.

  10. #30
    Senior Member
    Join Date
    Aug 2013
    MBTI
    ISTP
    Posts
    353

    Default

    Quote Originally Posted by Honor View Post
    it doesn't make any sense for there to be correlations between the dichotomies because they are discrete
    Even if the dichotomies are discrete, language is messy.

Similar Threads

  1. Big 5 is more accurate than MBTI, and is favored by researchers
    By The Wailing Specter in forum Other Personality Systems
    Replies: 33
    Last Post: 08-20-2016, 04:05 PM
  2. MBTI and children
    By Ivy in forum Myers-Briggs and Jungian Cognitive Functions
    Replies: 30
    Last Post: 05-10-2010, 01:30 PM
  3. MBTI and relativism
    By rivercrow in forum Philosophy and Spirituality
    Replies: 10
    Last Post: 07-22-2007, 10:54 PM
  4. Web Link: MBTI and Christianity?
    By Usehername in forum Myers-Briggs and Jungian Cognitive Functions
    Replies: 8
    Last Post: 06-28-2007, 03:25 PM
  5. "The Shield" MBTI and E-types...
    By The Ü™ in forum Myers-Briggs and Jungian Cognitive Functions
    Replies: 2
    Last Post: 05-27-2007, 06:32 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
Single Sign On provided by vBSSO