• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Is graphology science or pseudoscience.

There is an immense amount of research on it. Some things like gender and some types of illness or psychiatric conditions can be guessed from handwriting at higher than chance levels. Graphologists cannot determine personality traits from the handwriting any better than non-graphologists can from the content of the same writing.
 
I saw an example of a graphologist correctly identifying a serial killer by their handwriting once.

But as Randi would say, if they used graphology they did it the hard way, given the graphic nature of the confession used as a handwriting sample.
 
I used to have students try to design a study to test graphology as part of a research methods class because there are so many issues to consider that it makes a good exercise in experimental design.

Graphologists claim that they don't use the content of the writing. But they also say that the content cannot be standardised by having everyone copy a passage, because analysis of copied writing doesn't work.

This means that you have to control for content in any test, but not by having everyone copy the same passage. Any naturally produced sample (even a shopping list) has some content clues which might allow guesses about personality.

You can test whether personality can be inferred equally well from typed transcripts of the materials the graphologists are given. Then you need to have appropriate people to do the content-only control (ie you need to make sure that these people are just as good at judging personality from it's content as graphologists are). Otherwise the graphologists could do better, but just because they are better at guessing personality from content rather than analysing it from handwriting. You can't measure this by seeing how well the graphologists do with typed transcripts, because obviouly they would have no motivation to try to do well in with these (since it would undermine their own claims).
 
Elaedith said:
I used to have students try to design a study to test graphology as part of a research methods class because there are so many issues to consider that it makes a good exercise in experimental design.

Graphologists claim that they don't use the content of the writing. But they also say that the content cannot be standardised by having everyone copy a passage, because analysis of copied writing doesn't work.

This means that you have to control for content in any test, but not by having everyone copy the same passage. Any naturally produced sample (even a shopping list) has some content clues which might allow guesses about personality.

You can test whether personality can be inferred equally well from typed transcripts of the materials the graphologists are given. Then you need to have appropriate people to do the content-only control (ie you need to make sure that these people are just as good at judging personality from it's content as graphologists are). Otherwise the graphologists could do better, but just because they are better at guessing personality from content rather than analysing it from handwriting. You can't measure this by seeing how well the graphologists do with typed transcripts, because obviouly they would have no motivation to try to do well in with these (since it would undermine their own claims).
Click to expand...

That's an interesting problem.

Would it work if people were orally told what to write? Not copied off another paper, but instructed to write out a specific passage?
 
quixotecoyote said:
That's an interesting problem.

Would it work if people were orally told what to write? Not copied off another paper, but instructed to write out a specific passage?
Click to expand...

Maybe a passage everyone knows by heart---the Pledge of Allegiance, or "Mary had a little lamb", or something.
 
I think the general requirement is often that the content of the writing cannot be dictated to the writer by any means.
 
Elaedith said:
I think the general requirement is often that the content of the writing cannot be dictated to the writer by any means.
Click to expand...

Does this include vague suggestions like "re-tell the story of the Three Little Pigs"?
 
drkitten said:
Does this include vague suggestions like "re-tell the story of the Three Little Pigs"?
Click to expand...

It would probably depend on the graphologist. Wouldn't this allow quite a bit of variation in content that could be associated with personality, intelligence, education etc?
 
I highly recommend this book, The Write Stuff, by Barry Beyerstein. Don't be put off by the $70 list price; it is available used on Amazon from $4.34. For that price, you should not resist (even if they add shipping. Grrr.).

Beyerstein points out a critical difference between seemingly similar disciplines: graphology and forensic handwriting analysis. Graphology, used to analyze or predict personalities, is akin to astrology, cold reading, divination and dowsing, i.e., superstitious nonsense. Forensic handwriting analysis is a valid, if not perfect, science of comparing samples of handwriting to determine if they were written by the same person or not.
 
ben m said:
Would it work if people were orally told what to write? Not copied off another paper, but instructed to write out a specific passage?
Click to expand...
Maybe a passage everyone knows by heart---the Pledge of Allegiance, or "Mary had a little lamb", or something.
Click to expand...

Problem with that is that there might still be certain 'content' introduced... I'm thinking of things like spelling errors... which might give a clue to identity.

If you told them to write the words to "Mary had a little lamb", and they spelled it "Mary had a littl lam" the graphologist may (correctly) assume the individual is younger and/or uneducated because they spelled 'lamb' wrong rather than because of the way individual characters were written (part of what graphology is based on).
 
edd said:
I saw an example of a graphologist correctly identifying a serial killer by their handwriting once.
Click to expand...
Reminds me of the Russian psychics who, when shown a picture of serial killer Ted Bundy, said he looked like a fine young man but failed to detect his "darker side."
 
I remember reading about it in my psychology class. Apparently, I have low impulse control :rolleyes: . Did find a good study:

Correct assignments were made for 65% of the scripts, significantly more than chance expectation, with apparent consistency over the various subgroups and between graphologists. A control group of assessors, using typed versions of the scripts, performed significantly worse, with 54% successes, consistent with chance expectation. A second control group of lay assessors, using the handwritten scripts, had intermediate results, with 59% successes. It is suggested that the graphologists, while showing some discriminating ability, failed to substantiate claims made on their behalf
Click to expand...

Linky.
 
Freshman year in high school I had a woo-woo sex/health education teacher who didn't take a shine to me or the cynical, dykey lesbian girl I sat next to for some reason. We both thought she was weird and creepy but this teacher was very popular among students and parents for being "real."

One day, she asked the entire class to turn in a sheet of handwritten paper so she could perform graphology and determine our personality traits. I knew even then that was she was doing was taking her preconceived notions about the student and finding examples in the handwriting. It seems like an underhanded way to say cutting, mean things about students she didn't like. I was a lefty who wrote graffiti and was trying to find my own style and she ripped into me. LOL.

She wrote this book which was a big hit in the 1970s:

http://search.barnesandnoble.com/Ask-Me-if-I-Care/Nancy-J-Rubin/e/9780898155976
 
Last edited:
There is a huge difference between forensic handwriting analysis and graphology. The forensic identification of authorship (or determination of forgery) proceeds from an empirical study of the characteristics of the handwriting in the document in question in comparison to a known sample.
 
I'm skilled in typology.

"If some1 typez lik diz" then I can guarantee you they have a lower IQ than someone who types using proper grammar and diction.
 
You must log in or register to reply here.

Back
Top Bottom