Date: Wed, 9 Dec 1998 10:27:40 -0600
Reply-To: Jack Hamilton <jack_hamilton@FIRSTHEALTH.COM>
Sender: "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From: Jack Hamilton <jack_hamilton@FIRSTHEALTH.COM>
Subject: Re[2]: SAS certification program -- no rumor, just fact
Content-Type: text/plain; charset=US-ASCII
<pdorfma@FL6612MAILEX4.UCS.ATT.COM > wrote:
>13. I would benefit from becoming certified in SAS software.
>
>Since when any kind of certificate has become, even marginally, an
>equivalent of knowledge and actual ability to program?
A certificate, if issued by an organization that actually knows what
it's doing, would probably indicate some level of knowledge.
For example, the is a coding contest at SUGI every year. This year's
SUGI had good questions, so I decided to enter it. I was one of 6
people who got all the answers right. I didn't get a certificate, but I
would claim that "winning" that contest indicates real world knowledge
of SAS. Of course, it wasn't a multiple-choice questionnaire.
(Hearing other contestant's answers was interesting. Some of them were
better than mine, or showed a completely different approach to the
problem.)
>14. If SAS Institute offered a professional certification, I would pursue
>it.
>
>Will not move a finger until it becomes clear that I fail to be employed as
>a SAS professional without it.
If it were free, and tests were held at SUGI or some other convenient
location, I would probably go for it, but as a matter of curiosity, not
because I think it would help my career.
>15. The formal SAS Institute training I have acquired has improved my job
>performance.
>
>Not to mention its being ludicrously overpriced,
I don't think it's that bad. It's not as cheap as the local community college,
but most SAS training is well organized. The beginning classes seem to be the
least useful (based on reviewing the curricula). I took the SQL class many
years ago, and thought it was pretty good.
>it is not geared towards a
>professional programmer. Not very long ago, SI opposed the mere premise of
>professional SAS programming to the extent of having refused to publish the
>Book, "Professional SAS Programming Secrets". And, by the way, they have
>never changed their mind.
Are you sure that that's the motive? Perhaps they just didn't think
that there are any secrets to it.
>21. Multiple choice tests are an accurate and fair measure of an
>individual's knowledge.
>
>I wonder, who was the inventor of this nonsense?
Some multiple choice tests are probably fair and accurate measures, but
there are so many ways to do things in SAS, and so many exceptions to
most of the "rules", that such a test would be very difficult to devise.
I would guess that the people who would do poorly on the first round of
SAS tests would be (a) the people who really don't know anything, and
(b) the people who really know SAS well, and don't like any of the
oversimplified answers.
>Looks like this kind of
>"knowledge evaluation" has really gotten to many hearts. Recently, I was
>trying to interview a person for a mainframe position by proposing to write
>a SAS program, that is, exactly what he had claimed he was good at and
>wanted to be hired to do. Here are all the books, PC online help, ISPF
>session open, a couple of hours (I had asked several colleagues to write it,
>and it never took over 15 minutes). The guy refused and told me he would
>complain about "unfair hiring practice". Out of curiosity, I asked him what
>he had expected. The answer was "multiple choice". Really...
We interviewed someone for a position earlier this year. I asked
something like "what does the LAG function do?", and his response was
"Is that how you do things around here? You ask people questions you
already know the answer to?"
In a job interview, yes. He didn't get the job - partly because of that
answer, and partly because the answers he did deign to give us weren't
entirely correct.
Our interview questions are almost entirely "essay" questions, and
getting the right answer isn't as important as showing how the
interviewee approaches problems. We also ask to look at sample code.
