David,

>It can't. I personally worked on the CBT versions and exams were serialized
>and sent to specific doctors. Results were encrypted and sent in to be scored.
>Results were accessed via serialization number.

So I assume "open-book" is OK for that test? Any time limits on the test to make that more difficult?

>I have never taken a M$ exam so let me plead ignorance if I make a dumb
>comment but how does M$ verify. I would hope at very least it is a State
>picture ID requirement but isn't that pretty easy to get around if you really want something.

Two forms of signed ID, with your picture on at least one of them. You then sign an acknowledgement form agreeing to the rules of the testing environment. The last two tests I took were in a room with 7 other computers, all visible through glass windows by the test administrator. My first two tests were in a room with 3 other computers, all visible through a small glass window in the door.

>Even if you can prove you are the individual that is supposed to take the
>test arn't there other ways around not really having the knowledge your
>score indicates?

Cellphones are not allowed, as well as no cheat sheets. I was tested on what I knew at the moment I took the test. How I got that knowledge was up to me. Some of it I learned by experience, some by studying and experimenting with unfamiliar concepts (following MS's official test outline of topics). Either way, I knew what I knew at the time of the test, but I did not know ahead of time what the specific questions would be.

Many of the questions on today's MS tests really do test your experience with logically analyzing situations and sifting through various trade-offs to arrive at the best answer. Very few of the questions on the four tests I have taken actually dealt very much with syntax, other than in the context of analyzing several slightly different versions of lengthy SQL SELECT statements or SQL Server stored procedures to determine which of them will produce the stated result requirement.

>In the end I believe a certain level of professional trust is assumed and any certification is just an attempt to differentiate one professional from another.

I agree in principle, but alas -- personal standards of integrity are not what they once were in my grandfather's day. Any certification test that is not done in a controlled environment is suspect today. That's why MS has changed their procedures and testing methods so much recently to restore credibility and balance out "exam-cramming". You gotta have some depth of experience to do well with most of the recent tests.