Not quite ordinary, yet.

Friday, TEA released preliminary data on 3rd and 5th grade reading scores. The Lancaster ISD reported their own gains for 5th graders at LISD at about 20% over last year. The crowd in attendance jumped to their feet and cheered (Except for those on crutches, of course. Even Larry Lewis's miraculous powers only extend so far.)

Now, last year for 5th grade the calculated “gains” in Lancaster were actually DOWN from 04 by about 10% So we might have expected / predicted a pretty good bounce this by the purely statistical pattern of "regression to the mean". And this 5th grade "bounce" is different from the 3rd grade numbers, which seem to be too high to be ordinary for two years in a row.

But a 30% bounce is the 2nd highest gain in my sample over some 111 data pairs.

Crandall, the small district, had a 58% gain one year (the sample max) and a 26% gain another year (the next biggest gain in the data set). Crandall also, predicitably, had big losses offsetting these gains in other years, and represented the minimum FALLS (21% and 19%) among the data. Wilmer-Hutchins ISD also had near max gains of 14% and 15% gains in their “good” (fraudulent) years and crashed back to minimum (falling) gains in their off years. (a 65% fall in the 2004-05 tests.)

Part of the current gain for Lancaster 5th grade results from starting at a low baseline. On percent passing, we fell one percent last year, from 52% passing to 51% passing, at the same time the rest of the state was gaining from 75% to 81% passing. So our relative score fell from 69% (of the state value) to 62% of that value, for a minus ten percent result. Negative gains are reported as a "fall".

Even now, the district is gloating about a passing rate of roughly 66% … which is up 30% over 2005's 51% passing rate. But 66% is still well below the state passing rate of 80% passing. Comparing to the state, 66/80 is, for 2006, about 82%, up from 2005's 62%. This is a 20% gain. It's a great gain but it is not exorbitant; it shows better passing rates, but rates that still are about 20% below "ordinary" for Texas 5th Graders.

TEA reports a significant number of students in all grades who fall short of standards during the first administration of TAKS tests nevertheless prove their ability to pass on the second or third attempt. It's possible the Lancaster 5th graders will continue to gain against state averages as later results are accumulated.