A
few months back, I got an email with the header “Our Test Identifies an Asshole
Before They Are Hired.” As I dug into
it, I realized that the author (who I won’t name, as that would make me an
asshole) was claiming that he and others in his company could reliably
identify and screen out future workplace assholes by analyzing handwriting
samples. He even sent me an example of a
report that –- based on only a handwriting sample – concluded that a job
candidate was a “bad apple,” who had problems including “self-centered,” “aggression,”
“moody”, “ and a “poor team player.”
I
try to be open-minded and love strange ideas when they are treated as
hypotheses, hunches, and interesting experiments. BUT I also believe strongly in evidence-based
management, and that managers have a responsibility to act on the best evidence
available. Unfortunately, although graphology
is used routinely in some countries to help select new employees (notably in France),
existing research suggests that it is not a valid method for selecting new
employees.
Perhaps the most
damning study was published by Frank Schmidt and the late John Hunter in the Psychological Bulletin in 1998. These two – very skilled and very careful researchers–
analyzed the pattern of relationships observed in peer reviewed journals during
the prior 85 years to identify which employee selection methods were best and
worst. General mental ability (IQ and related tests) was the best predictor and work sample tests (e.g., seeing if people can
actually do the job) were the best of the 19 examined. Two predictors stood as the
worst, graphology and age. Here is the
rank order of the 19 predictors they examined – the key thing to know here is
that neither graphology nor age provided any
valid information at all about whether future employees would do better or
worse on the job.
1. GMA tests
2. Work sample tests
3. integrity tests
4. Conscientiousness tests
5. Employment interviews
(structured)
6. Employment interviews
(unstructured)
7. Job knowledge tests
8. Job tryout procedure
9. Peer ratings
10. T & E behavioral
consistency method
11. Reference checks
12. Job experience (years)
13. Biographical data measures
14. Assessment centers
15. T & E point
method
16. Years of education
17. Interests
18. Graphology
19. Age
Perhaps some rigorous research demonstrating the predictive power of graphology will eventually be
published, or even has been published. I searched for new research that contradicted Schmidt & Hunter, but I didn’t
find any; perhaps there is some that I don’t know about. I could only find
research that examined why the process of looking at handwriting samples
and hearing expert opinions might fuel the illusion of validity, even if graphology has
no validity at all.
Practicing evidence-based management is about
acting on the best knowledge that you have right now, but being open to new
evidence and information. Based on the
best research that I know of, my advice is don’t
use graphology to screen out workplace assholes, or for any other employee selection
decisions. It appears to be useless.
I
am sure that this will make some graphologists unhappy, but then again, astrologers
have huge faith in their methods as well.
Leave a Reply