On Wed, 5 Oct 2016 22:30:39 -0500, Martin Eastburn
wrote:

Naturally, it pays when the classes pass with grade and you get a raise.

If not you might not at all or take more work or a cut.

For the last 40+ years there has been a measured value put on each
teacher. Did they do this or that grade. Did they do this or that
another... and the grade determines who is a good teacher. (HaHa Not.)

Classroom teacher for 9 years , College for 11.

Martin H. Eastburn

Martin, your introduction of "curves" to this discussion had nothing
to do with teaching or classrooms. You were commenting on the
evaluations of pollsters by 538.

I'll say again: The predictive values and the party biases are not
something that's graded on a curve. They're things you can measure by
counting. And the grades correspond to the accuracy and repeatability
of polls run by each organization.

No curves. No guesswork. Just measurement of objective criteria.

--
Ed Huntress



On 10/4/2016 10:54 PM, Ed Huntress wrote:
On Tue, 4 Oct 2016 20:19:20 -0500, Martin Eastburn
wrote:

As we all know grades are often dependent on the Professor/teacher/....
curves them to. Looks bad to fail all of the deadheads so one scales up
the grades. Been there seen it. Even 50 years ago.

Martin

It has nothing to do with "curves." Look into the calculations for
prediction rate. Those can't be fudged. They either got it right or
they didn't. If you looked into it further, you'd see that the bias
rates are similarly matters of objective calculation.

There is no argument with those ratings, Martin. You either understand
them and the hard facts from which they're calculated, or you don't.
Gunner was just bull****ting. He has no idea what he was talking
about.