Harry Avis wrote:
"Your student made some very important points. Personality testing may have
an advantage over personal bias, but only from a nomothetic point of view.
Lie dectectors are not allowed in many states because of the risk of an of
both false positive and false negative interpretations. Presumably the
courts have decided that the risk to society and the potential harm to the
victim is too great. The use of personality tests to select employees (which
I believe is the focus of the student's paper) has a great potential for
harm to the invidivual, although much less harm to the employer. Although
personality tests may be able to predict with accuracy those who are
absolutely unqualified for a particular job, this lack of fit should be
obvious from any number of other, objective, sources. The problem, as I see
it, is the ceiling effect: the validity of a test in terms of its ability to
discriminate between two or more qualified applicants is questionable to
say the least. For example, does any psychology department give personality
tests to its prospective faculty candidates? If we don't trust them, how can
we advise others to do so."
I agree that the courts should not use a nomothetic perspective. Each person
needs to be tried in a criminal trial only on the evidence presented in that
trial. However, for other judgments, I think that judgments based on
nomothetic evidence are preferable. I don't believe I need to call witnesses
under oath to a jury trial every time a hiring decision is made.
The student did not specify personality tests. This was not a paper; it was
a response to a threaded discussion on the future of testing and what I
included in the message was the whole thing. I agreed with her point about
lie detectors in criminal cases because they don't measure up to the "beyond
a reasonable doubt" standard in judging individual cases. In fact, as many
as one-third of innocent people will be considered guilty by a lie detector.
However, I also don't look to the legal system as an authority on the kind
of evidence that is reliable. Much more credence is given to eyewitness
testimony in legal proceedings than research indicates is warranted. I would
also say that I would have to see the validity evidence before using a
personality test in hiring and that would include evidence that that
personality trait was required for the job and that there was a strong
correlation between that trait and success on the job. I would also look at
whether the test was only appropriate for screening purposes or whether it
could make higher level distinctions.
As to whether your personal judgment should replace information gathered in
a nomothetic framework, I would want to see the validity coefficient of your
personal judgment. If the test adds nothing, in a multivariate analysis, to
the prediction of outcomes, by all means, do not use the test. My main
objection to the student's statement was that tests are inappropriate to use
in such judgments if they are less than 100% accurate. By that standard, no
technique is satisfactory. Why is it that tests are singled out as needing
to be perfect and, when they are not perfect, they are replaced by something
that has less predictive validity than the test?
Rick
Dr. Richard L. Froman
Psychology Department
John Brown University
Siloam Springs, AR 72761
e-mail: [EMAIL PROTECTED]
http://www.jbu.edu/sbs/psych/froman.htm
