Perhaps under Obama we'll stop measuring student ability through standardized testing period. Okay, that'll never happen, but every year I get students coming in from high school who say they've always tested well in English and are confused, then, that they're getting C's and D's on their essays. I started in Illinois and now I'm in Wyoming - the students are equally bad at writing.

Here's a small and maybe meaningless anecdote - I subbed for a year in a school district in southern Illinois, and one day taught an AP English course. These were advanced placement juniors and seniors reading To Kill a Mockingbird. This should not be a difficult book for eighth graders - which isn't to say it's not an excellent book - yet these highly placed students were reading it in their second semester.

There was an interesting point brought up in the comments of the Eduwonkette post--should we be more concerned about closing the gap, or about seeing all scores rise at the same rate? It's a fair enough point to argue that without closing the gap, there can be no equality in access to powerful/well paying positions, but surely there is a point at which it would be preferable to have all students reach a certain skill level than to avoid having some groups of students outperform others? If for no reason than academic skill level is only going to be one component in making a candidate an attractive hire--in a genuinely meritocratic system, I would think adequate (adequate meaning adequate for performing a high level job, not basic) skills plus other attractive work qualities would at least balance out against higher than necessary skills but a lack of other desirable traits.

Its also worth pointing out that the difficulty of tests changes from year to year which often results in the appearance of improvement without actual improvement.

But sure, lets keep taking fantastical claims of improvement at face value. Everyone benefits when test scores go up!

In NYC I know that some teachers "cook" their tests, and gives students the answers during the state, city, and predictive states. How does this prove tests are effective educationally and statistics show the achievement gap disappearing?

OK - after 6 years teaching 1st grade in an "inner city" (I use the term loosely because of Los Angeles' strange geography) school, this is what I found.

First, year by year, we got better at teaching little 6 year olds to fill in bubbles correctly and to fill in only one bubble per questions.

The tests had many of the same questions each year, so we got better at making sure the kids knew particular facts.

Our test giving skills improved (even when we didn't do things like emphasize particular words in the questions, or read the questions again).

So surprise surprise, the kids' scores went up.

Also, the school encouraged us to have the parents of particularly slow students sign waivers exempting their kids from the tests. (We had to have the waivers because otherwise, the absent kids' scores would be counted as 0.)

One year, the school counselor had us all meet in the auditorium where we were encouraged to "review" the tests a week before they were given. Very illegal. But I'm sure our school was not the only one.

Then there were stories about teachers who erased and corrected their students' answers.

The school administration turned a blind eye, and indeed, seemed to encourage "anything that will get our test scores up."

Some years, the school would be given extra money from the state for "improving" its scores -- we teachers never found out where that money went.

Even without the money, the administrators were judged by their schools' scores, so the incentive was to get them up by hook or by crook.

Meantime, we had to jettison our most stimulating lessons and methods, because they took time from test prep activities which went on all year, and were monitored by surprise classroom visits.

Yes, I am bitter. We were doing better teaching before "accountability" hit.

There has to be a better way.

signed, ex-teacher