Yes, I think that the main problem with the MBTI is not that it couldn't be scored in a way that would be more reliable but that the use of the typology, the traditional way it is scored, is not reliable.
The way I explain it to students is that, if these characteristics (such as introversion-extraversion) are normally distributed, about 2/3 of people will have a score on introversion/extraversion that is within one standard deviation of the middle of the distribution with only a few scoring out on the extremes. Using a typology in such a circumstance is going to result in many misclassifications because the great majority of individuals will have a score that is quite close to the other side of the curve. A drawing of the normal curve with the SEM interval makes this pretty obvious. So even a reasonably small standard error of measurement is likely to make the typological categorization unreliable even if the scale itself is quite reliable. Rick Dr. Rick Froman, Chair Division of Humanities and Social Sciences Professor of Psychology Box 3519 John Brown University 2000 W. University Siloam Springs, AR 72761 [email protected]<mailto:[email protected]> (479) 524-7295 http://bit.ly/DrFroman From: Stuart McKelvie [mailto:[email protected]] Sent: Tuesday, March 26, 2013 12:09 PM To: Teaching in the Psychological Sciences (TIPS) Subject: RE: [tips] Nothing Personal: The questionable Myers-Briggs test | Science | guardian.co.uk Dear Tipsters, I have followed this thread on the MBTI and wish to make one or two comments. 1. The original "article" was an opinion piece in The Guardian newspaper. This may or may not represent research on the test. Notably, one criticism at the end states that a yes-no test cannot capture the complexity of personality. Yet the Eysenck Personality Inventory (Questionnaire) with its two (or three) scales has been widely regarded as a very useful instrument. 2. A number of studies have examined psychometric properties of the MBTI. Generally, reliability seems acceptable (Capraro & Capraro, 2002; Gardner & Martinko, 1996) and validity, as reflected in relationships with other personality instruments, is reasonable (Furnham, Dissou, Sloan & Chamorro-Premuzic, 2007). 3. However, the typology does not receive much support. In particular, scores are not bimodal (Bess & Harvey, 2002). 4. When it comes to using the test, research does not seem favourable towards its value in choosing majors (Pulver & Kelly, 2008), improving leadership skills (Brown & Reilly, 2008) or developing team processes (Kuipersm Higgs, Tolkacjheva, & Witte, 2009). 5. However, there is evidence of a relationship between test scores and sales performance (Lewis, 1995). 6. On the other hand, using a person's scores may be helpful in improving self-understanding, communication and decision-making (Kuipers, Higgs, Tolkacheva, & Witte, 2009; Pulver & Kelly, 2008). 7. Some researchers recommend that it merits further study (Gardner & Matinko, 1996). 8. I think that the main problem with the test is that it may be used in ways that go beyond what research supports. To be fair to the Guardian piece, that sentiment is also expressed. And Tipsters who opine that money talks may not be far from the truth. --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=24536 or send a blank email to leave-24536-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
