July 12, 2001

10:24 AM

George W. Bush's Handwriting.handwriting analysis reveals: If you want something from George W. Bush, tell him how much you like what he's done so far. Tell him quickly. And don't try to push him around. You hear that Cheney?
posted by brucec (35 comments total)

And if you have real player you can Hear Gore's analysis. Gore is happy and looks at the bright side of life .

But his would his small ego had served him well as Commander in Chief?
And what about him keeping secrets?
posted by brucec at 10:32 AM on July 12, 2001

Is there also ReverseSpeech analysis of Ralph Nader's astrological chart, or something equally scientific?
posted by lileks at 10:36 AM on July 12, 2001

lileks. handwriting analysis is scientific. After the number of crimes /forgeries solved by graphology, with the companies that use it for employee slection, equating it to astrology is old-fashioned thinking.
posted by brucec at 10:53 AM on July 12, 2001

Handwriting analysis is scientific for forensic identification purposes. Personality analysis through graphology is utter nonsense .
posted by Skot at 10:59 AM on July 12, 2001

lileks. handwriting analysis is scientific. After the number of crimes /forgeries solved by graphology, with the companies that use it for employee slection, equating it to astrology is old-fashioned thinking.

Oh, please. Determining whether a signature is authentic or a forgery is scientific. Determining someone's personality from handwriting is reading tea leaves.
posted by anapestic at 10:59 AM on July 12, 2001

Determining someone's
personality from handwriting is reading tea leaves.

No, tea leaves were not created by the person. Writing was.
posted by brucec at 11:05 AM on July 12, 2001

I know other things created by people...
posted by mecawilson at 11:08 AM on July 12, 2001

....and many of those things may reveal something about the person. But handwriting is so common so as it could be examined and patterns established.
posted by brucec at 11:23 AM on July 12, 2001

<wide eyed disbelief> My wife changes her handwriting daily (sometimes in the middle of a page). Does this mean that she has dissociative disorder? </wide eyed disbelief>
posted by iceberg273 at 11:25 AM on July 12, 2001

No, mecawilson, I think that's created by bovines of the male persuasion.
posted by nprigoda at 11:26 AM on July 12, 2001

Handwriting University? Pshaw. Just go to Shorthand Summer School.

As for graphology, I'm curious. The Skeptic's Dictionary suggests it's similar to Myers-Briggs (about which they also list skeptical opinions). Is there any research comparing the results of a graphological exam to accepted personality tests? I think there's modest utility, if it's not overinterpreted, in knowing your personality type, but I'm not so sure about "unconscious" personality.
posted by dhartung at 11:34 AM on July 12, 2001

<wide eyed disbelief> What if you hate to write stuff out by hand and prefer to type everything instead? Does that mean you have no personality? </wide eyed disbelief>
posted by iceberg273 at 11:35 AM on July 12, 2001

*sigh*

From a position paper on graphology by the B.C. Civil Liberties Association:

It is ironic that some of the personnel managers for municipalities and corporations in B.C., who are most concerned about objectivity and fairness, have fallen prey to a personnel selection tool which, in fact, is most likely to result in unfair personnel decisions...Graphologists routinely fail fair and impartial tests of their exaggerated claims to discern traits that are opaque to orthodox assessment techniques.
posted by Avogadro at 11:38 AM on July 12, 2001

My wife changes her handwriting daily (sometimes in the middle of a page).
Does this mean that she has dissociative disorder?

Its a good point, and anyone who does this needs to look at a few writing samples. Chances are, the handwriting still has some common patterns.


What if you hate to write stuff out by hand and prefer to type everything
instead? Does that mean you have no personality?

Obviously not. Somewhere, you've written somehting. And by god, I'll find it and analyze you! Just kidding, but you do bing up a good point. People are writing less, and now its almost necessary to take a handwriting sample, because in some cases they are not available.

have fallen prey to a personnel selection tool
which, in fact, is most likely to result in unfair personnel decisions...

I dont necessarily disagree with you or the BC civil liberties union, but is it the analysis technique itself or how people will make decisions based on that analysis that's the danger? I would be scared of the latter. But I'm not afraid of the technique.
posted by brucec at 11:46 AM on July 12, 2001

I'd like to bing up a point as well.

I read GW's analysis pretending it was written about me. Lo and behold! It fit! Amazing!
posted by jragon at 11:59 AM on July 12, 2001

dhartung: Myers-Briggs may a little iffy and useless except on a highly general level and in re to self-awareness, but it's not nearly in the same league as graphology. (The skeptic's site mainly complains about MBTI because of its being used by business and industry, which is impossible to argue with. But business loves a management fad and takes it too far. Always has, always will.) Most handwriting variation is due to differences in fine motor coordination, not personality and tests have proved it a zillion times over.
posted by raysmj at 12:01 PM on July 12, 2001

*cries*
posted by baby jesus at 12:04 PM on July 12, 2001

Now see what you've done?
posted by iceberg273 at 12:07 PM on July 12, 2001

BabyJesus believes that we have no faith and have turned our back on Him, believing Him dead? Is there lost scroll or something with graphology mentioned in it somewhere, one thrown in with the Gnostic texts or something? We're sorry. But hey.
posted by raysmj at 12:10 PM on July 12, 2001

The Skeptic's Dictionary suggests it's similar to Myers-Briggs (about
which they also list skeptical opinions). Is there any research comparing the results of a
graphological exam to accepted personality tests? I think there's modest utility, if it's not
overinterpreted, in knowing your personality type, but I'm not so sure about "unconscious"

That's my view of it. Its a tool. Employers would be foolish (and would get sued) if they use it alone to predict an employee. Cops would be foolish to try and use it to convict. You use it to open up areas that you could question further, or or to know what you need to ask references about.

The Skeptic's Dictionary does not give an accurate description (what do you expect, it is a skeptic's dictionary) It does not only show 'unconcious behaviors' --- many of the things it reveals are things we already know (if you are happy you know it, and it effects handwriting.) And they are things you would already know about a friend. But if you commit a crime, or if you are a famous person we can't ask, or a person who is no longer alive, then we need to know those personality traits and the subject is not necessarily going to give up them.

There are few tests on graphology. But there are few tests on any personality tests. One of the problems is: You cannot scientifically describe a personality. Yet they are very much a part of life, something we need to know about people we do business with or that may have commited a crime. Another problem is there are a lot of politics behind the use of graphology, and there are some low-grade practioners who are exploiting the fact that there is no legal definition of graphology.
posted by brucec at 12:17 PM on July 12, 2001

Actually, dhartung, a partial correction: No tests have ever come close to conclusively proving a link between personality and handwriting, and even if there are personality influences, they would have to be miniscule, given the clear role motor coordination.

Several studies from 1973 to 1986 (Schmidt and Hunter, Psychological Bulletin, 1998) showed that graphologists and untrained nongraphologists alike infers some limited information about personalities and job performance based on handwriting samples. But the untrained nongraphologists had as much luck in predicting performance as the "trained" graphologists. Also, when those whose writing was being accessed copied materials from a book -- that is, wrote by hand material from another source -- neither the graphologists or the non-g's were able to infer anything at all from the samples.
posted by raysmj at 12:26 PM on July 12, 2001

From the myhandwriting.com FAQ:

(Handwriting analysis) is a scientific method of identifying, evaluating, and understanding a person's personality via the strokes and patterns revealed by his handwriting.

But you tell us that there are few tests on graphology, where testing is a necessary condition in the scientific method. How then did one determine that certain strokes reveal certain personal characteristics, and why would graphology be more revealing than psychological testing?
posted by Avogadro at 12:27 PM on July 12, 2001

dhartung: Assessed, rather. Sorry. I was copying something from a book, sorta like in the tests mentioned.
posted by raysmj at 12:30 PM on July 12, 2001

"...where testing necessary condition in the scientific method"

If you can define 'happy,' 'lonely' or 'aggressive' in a scientific way, I'll start the test. In the meantime, they are still things I'd like to know about you before I hire you, or if you are the suspect for a crime.


"why would graphology be more revealing than psychological testing?"

Writing is more available. I just think its hard to do a psych test of every potential employee, or of a dead guy... but they probably wrote something. Writing is pretty common, although of course it is getting less so in our computer society.
posted by brucec at 12:44 PM on July 12, 2001

If you can define 'happy,' 'lonely' or 'aggressive' in a scientific way, I'll start the test. In the meantime, they are still things I'd like to know about you before I hire you, or if you are the suspect for a crime.

So on the one hand, you're dismissing the entire hypothesis as by definition unscientific, and then on the other hand using this as justification for finding out "things I'd like to know."

Hmmm.

Here's some cake. You can eat it, too.
posted by Skot at 12:51 PM on July 12, 2001

No one should make any judgements about personality at all because they are 'unscientific.'

Unfortunately, we live in the real world...these judgements are made. Deny yourself a tool to assist in making those judgements if you wish...If I was going up to George Bush to ask him for something, I'd like to use this.

As I said earlier. Use it as a tool.
posted by brucec at 1:00 PM on July 12, 2001

Is there no useful knowledge outside of what is scientifically proven? I think there is useful tools outside of things that can be put into neat scientific boxes. That does not mean its automatically astrology or tea leaves.

I you have patterns that you have culled from past examples, and if the rules are held consistently. Take this person's celebrity handwriting site I pick celebrities because they are personalities we know. We -- generaly speaking -- agree (although we can't scientifically agree) of the personalities of these celebrities. As long as he holds the rules consistently (that is doesn't change the rules for each celebrity, and these rules are generally the same as what other graphologists use ) it is useful knowledge that we can gain from.

If the same rules applying to several celebrities whose personality traits we generally agree on would provide some proof of graphology, but because you cannot define a personality, it is probably not science. Again, its still a tool I'd like.
posted by brucec at 1:15 PM on July 12, 2001

you cannot define a personality

You can operationally define a personality. You'll just get a lot of argument, that's all. Which was the basis of the horrible Theories of Personality class that I sat through as an undergraduate.

(the handwriting analysis rubric is an operational definition of personality. It should be internally self consistent. The article from Psych Bull that raysmj mentioned above indicates that it is not (it is confounded with content, among other things). Scientifically speaking.)
posted by iceberg273 at 1:26 PM on July 12, 2001

One can certainly develop graphology using scientific method: have a large, representative sample of people rate their own personalities ("on a scale of 1-10: how extroverted are you?") and also have them rated by others. Then see which handwriting features correlate to which personality traits, both as self-perceived and other-perceived.

If you tell me that graphology was developed using just such a method, then I'll grant it may be valid. However, it was not, as far as I have ever heard, developed this way.
posted by kindall at 1:27 PM on July 12, 2001

My grandmother taught my mother and her sisters how to write, my mother taught me and my siblings. (The fine tradition of homeschooling.) My youngest sister and I have tested our handwriting against our mother's and it is similar to the point that it is often difficult to tell, if we write our own names and then one another's, which are the real signatures and which are the forgeries.

In addition, my mother's handwriting (and mine as well, I suppose) is only vaguely discernable from her sister's. Both are reasonable facsimiles, I'm told, of my grandmother's handwriting before she was stricken with arthritis.

A scientific handwriting expert had difficulty discerning between my writing and my sister's after a lengthy inspection. Similarly, I don't doubt that a graphological "personality" analysis of our writing would likely suggest that similar traits are common amongst the lot of us. That analysis would be wrong. The exceptions, IMO, disprove the hypothesis.
posted by Dreama at 1:51 PM on July 12, 2001

One can certainly develop graphology using scientific method: have a large, representative sample of people rate their own personalities ("on a scale of 1-10: how extroverted are you?") and also have them rated by others. Then see which handwriting features correlate to which personality traits, both as self-perceived and other-perceived.

Let's say this were done (and as Kindall points out, there's no indication it has been done), and it turned out that there was a correlation between handwriting features and personality traits. Let's say that a slant in a particular direction correlates with aggression. How strong is that correlation likely to be? 60%? 70%? At that level of association, what is the predictive value of graphology when applied to an individual? Zippo. So you can't really "use it as a tool."
posted by anapestic at 1:56 PM on July 12, 2001

I really don't think you should be hiring dead guys, no matter how nice their handwriting is.
posted by CrazyUncleJoe at 2:00 PM on July 12, 2001

kindall's test is interesting but still has flaws. I think Benjamin is an extrovert 10, while Marie thinks he's an 8. We now have a 20% difference. Over a study of many people that could enlarge further. You could do averages, I suppose, or just allow people to say Extrovert or non-Extrovert, and I believe the graphlogists would come out well.

The more likely was that Dr. Smith knows that Extroverts do A and B on a normal basis and Benjamin does A and B, and then I see Benjamin's writing and compare it to Tom, Brad, Bill, Bob, Luke, and hundres of other people. That's the foundation of graphology detailed in this history There have been studies and research, but as I understand it, not conclusive scientific tests.

anapestic, no a 60% or 70 level of correlation would probably not be useful. However, you forget that there are multiple facets of handwriting. If I can apply three facets that show extroverts - big, forward slant, using lots of the page and all are at 70% -- its very useful.

Graphologists make these interpretations based on the multitude of factors. Just like a jury would make his decision on a multitude of patterns that indidcate guilt without knowing scientifically. Or a doctor makes his diagnosis without , except in the rarest of cases, knowing scientifically that he's correct. (there is a move for doctors to make more statistical decisions, but so far, it still a lot of interpretation.)
posted by brucec at 3:10 PM on July 12, 2001

I've been trying to find an article from an unbiased source (such as a scientific journal) that supports the use of graphology, but have been unsuccessful. brucec, if know of such a source, please link it. I guess that I can't really take much credence in what a website on graphology says to support it.

I'll keep looking for an independent link.
posted by Avogadro at 4:07 PM on July 12, 2001

I have found the same thing, that most of the web sites that come up are those selling services. What does this tell me? There is a thriving market and many have satisfied clients, like this one. So it does take some searching to find indepnendent links. But they exist, and I've listed some. Now, they don't all rave about graphology, but I'd say the consesus supports my position. Use it as a tool

real science journals, require a subsciption and so won't come up on the web search.. If anyone has a subscription to them or to an agregat like OVID, that do a search for me. But that's a little beyond the call of duty for MeFi.


Here's a crime website - not a graphology site. Does a pretty good job of presenting it as a crime, and he's not selling services, nor is he a raving fan. why files


< a href="http://www.daily.umn.edu/daily/1998/11/30/news/job/"> minnestoa daily


scripps howard news article.


business monthly article



Las Vegas Sun

There is a Washington post article that will cost me $1.50 to read, I aint paying for it. Point is, there's plenty of independent stuff out there, although I agree there are an awful lot of people selling services.
posted by brucec at 9:09 PM on July 12, 2001