End-of-Year Testing

(I started this article several years ago. Just getting around to editing and publishing these drafts.)

I love the idea of End-of-Year testing in a purely hypothetical sense. Students should be able to demonstrate what they've learned. The teacher doesn't grade it or write it. The students can't weasel out of it. A group of math people decide what "Algebra I" should entail and write some questions to measure it.

The test is never perfect, as NY teachers will hurry to emphasize, but it is out of the hands of the teacher and that is good. We should not be afraid to let our students measure themselves against a common standard and we should be open to change if the unexpected happens. SATs serve this purpose as well.

Differences between what you expect them to get and what they get are the prime indicator. If your grades are all As and your kids can't succeed in appropriate tests, then you need to review what you are doing. If the majority of kids can't succeed on an EOY exam, then you the teacher needs to make a determination: is it the individuals, the exam, the curriculum, or me? If you are passing kids and the next teacher isn't, someone might need to adjust.

In theory, EOY exams should be perfect for this - it's just too bad that they'll be useless for it.

not linear.

The State of Oklahoma requires these tests and they will *attempt* to write them to be an honest assessment of the skills that should have been acquired in any particular course. Just like Texas, just like New York Regents.

What will happen is that End of Year testing, like WilyECoyote, is going to run full-speed into the Cliffs of Reality. The "passing" score will be set to 60%, then too many kids fail so it's quietly lowered to 45%, or 35%, or lower. If enough people still can't pass the test, the cut-score will be lowered again. Or you wind up with the weird raw score conversion charts of the NY Regents (right).

So ... is it poor preparation and teaching or poor test-making?

The graph that's been misunderstood
by admin everywhere.

Or could it be that the test is trying to apply the "Higher Standards" that everyone is crowing about? You know the trope: "Raising the Bar improves performance."

Unfortunate reality #1: If you raise the standard, more people will fail to reach it.
Unfortunate reality #2: Calculus kids do better on their SATs than Algebra I students. Selection bias. Duh.


What should we do?

Avoiding all EoC testing is silly. Pretending that some "3-week portfolio question is demonstrably superior" is the canard put forth by all those people who have never watched or judged a science fair. Individual teacher-written final exams are suspect because of quality-control issues and because of grading irregularities. Department-wide final exams are probably the best unless your state has Mr Honner holding your Regents exams to account, in which case, go with the Regents.

Unless your school just voted to eliminate all finals in light of the transition to Proficiency-Based Grading, but that's another post for another day.

New state requirements make end of year test a make or break affair

By Jennifer Lindsey, Ardmore (OK) Daily Ardmoreite Education Reporter
Jun 20, 2009

Ardmore, OK — The importance of doing well on state tests has increased dramatically for high school students who will be sophomores in the fall.

The pressure has gone up as well.

Due to changes in state law, members of the class of 2012 will have a new requirement in order to graduate.

They will have to obtain a satisfactory or better score on the End-Of-Instruction exams in Algebra I and English II, as well as in two of the five other EOI exams: Algebra II, biology I, English III, geometry and United States history.

The EOI tests aren’t new. Only the linkage between good performance on the tests and being allowed to graduate has been added.

Prior to the class of 2012, all students were required to take the EOI exams when they finished with a particular course, but the exams did not affect a student’s ability to graduate.