Re: A Test to Select Competency

[John Gear <catalyst -at- PACIFIER -dot- COM>]

>Interesting argument, John, but I must disagree. As you said, when you went
>to the Dr., you asked certain questions. In the same way, you can review the
>writing samples, and ask me questions about the process. If I am a "poseur"
>as one writer put it, this will quickly surface.

>Conducting an interview, based on the writing samples, with pointed questions
>should elicit a certain set of responses the employer is looking for. Based
>on this, the resume, the samples, the person's references, and the employer's
>gut instinct; they should be able to hire someone.

And no doubt you're familiar with research that suggests that, regardless of
their stated intentions, people tend to hire people who are as much *like
themselves* as possible.  That's the strongest operative bias in the hiring
situation and one that is quite difficult to overcome.  Performance testing
is surely helpful in that regard.

And you must be familiar with the research that shows that most hiring
decisions are made in the first several minutes of an interview, with the
entire rest of the process being used as face-saving time and during which
the interviewer *seeks* the information necessary to support the decision
already made.

But perhaps your experience is in some world where hiring managers have good
interviewing skills.  In this one however, *most* hiring managers have no
more experience or training at conducting hiring interviews than they do at
speaking Urdu.  Their entire qualification for conducting interviews
consists of having been interviewed some indeterminate number of
times--sometimes very few--and usually by people who themselves lacked any
skills or thought-out method.  Armed with their limited experience and a set
of "rules" from HR (designed to limit exposure to lawsuits, not to help the
interviewing process) they go off, wading through a stack of resumes from
strangers, with each side struggling to use the interview to represent
itself in the best possible light and to minimize any unfavorable
impressions.  Some system.

>No test is going to be right for everyone, so by using a standardized to test
>you could be eliminating some very talented people, who just happen not to be
>suited to the test you use.

Oh, there aren't any people who don't make a good first impression on your
world either?  You don't have extreme introverts?  How about "No one hiring
method is going to be right for everyone, so by using only a narrow range of
methods you could be eliminating some very talented people, who just happen
not to be suited to the methods you use."

All I suggest is that people who oppose any form of testing are, in effect,
saying that the current hiring "system" (HO HO HO!) is so good that it
cannot possibly be improved by the use of some exploratory tests intended to
verify/validate certain attributes.  This seems like a very extreme and
difficult to defend position.

I *don't* suggest (and don't think anyone else has either) that some form of
testing be *the* entire screen.  I only disagree with those who insist that
*any* form of *any* test is insulting and despicable, even if conducted as
an adjunct to the methods you've outlined.

>I used to work for a consulting firm that used the research of David
>McClelland of Harvard as the basis of their work. They identified four
>clearly different learning styles. People learn and process in a variety of
>ways. One test cannot possibly account for this.

Well I'm sure if you called Dave he'd say it would be foolish to rely on
"one test."  My Uncle Louie of the Duluth College night school always said
to use a *lot* of different avenues, perhaps including testing, to give--and
to *share*--information during the hiring process.



John Gear (catalyst -at- pacifier -dot- com)
"Advertising is a valuable economic factor because it is the cheapest way of
selling goods, particularly if the goods are worthless." -- Sinclair Lewis