Lebanon Math Scores

I've been thinking about this and this on and off all weekend, as well as having conversations with others about it. I don't really have much to say, but I do wonder about this:

According to figures from Finch, the principal, 47 percent of Lebanon High School students failed Algebra I in the third trimester of the 2007-08 school year. Another 21 percent squeaked by with a D.

Finch said he did not check other classes or other trimesters for comparison.

First, the third-trimester thing. It seems to me that that's not going to be very representative (LT has another data point that supports this claim), since the third trimester is going to have the highest percentage of students who have already failed Algebra I at least once, which is going to create a skewed view of the scores, since the repeat rate for failing is pretty high. As well, I thought math classes at LHS were all two trimesters long - Algebra IA and IB, for example. I don't recall any IA classes being taught in the spring trimester, so what this probably really means is people who are failing IB.... right?

Oh, and yeah, those are high failure rates. That should go without saying.

The second thing is the assertion that Finch hadn't checked other trimesters or classes. Unless the LCSD is using incredibly outdated technology, it shouldn't have been hard to find more data. So either Finch did check, and told less than the whole truth, or didn't check, which is a mistake. Alternatively, I suppose he could have asked for more data but not had it by the time the story ran.... but why not just say that? I find it hard to believe Finch - or someone in the district - doesn't have that data on hand when the district declares an academic emergency.

In any case, as bad as the scores from the third trimester of last year are, what would seem to be more important (and more statistically significant) would be either the trend over three or more consecutive years of third-trimester Algebra I scores, or even the trend of scores by trimester over the course of the whole year. One data point has limited value.

Of course, there is this:

Although the number of students receiving passing AYP math scores improved slightly last year over the previous year - from 48.62 to 50.34 percent, the standard is 59 percent. In 2006-07, schools met standards if 49 percent of students passed the assessment tests. The percentage will continue to increase every two years until 2014 when 100 percent of students will have to pass.

This is another way in which NCLB is a pile of garbage: Standards change every year without any regard for what's actually happening on the ground. And it will never be the case that 100% of students pass a standardized test. That's a fantasy. (That we should be getting rid of asinine standardized testing altogether is a whole different blog post.) The other thing is the rate of increase: from 49% to 100% in seven years? Seriously? That's the stuff of legends.

One last thing: If I remember correctly, the majority of the math teachers at LHS who were teaching low-level math classes when I was there were very young and, as a group, very new. This was true with only one exception I can think of. I'm not suggesting that this alone can explain the scores, but I'd be surprised if it wasn't a factor.