School Information System

What a bunch of lies! Especially this line, "Today, Rainwater said, no statistical achievement gap exists between the 25,000 white and minority students in Madison’s schools."

As a statistician, I am, of course, aware of the claim that "there are lies, damned lies, and statistics." However, for the life of me, I can't figure out what the Superintendent is basing his claims on . It can't be graduation rates 86.8%(white) vs 60.9% (African-American). It can't be 4th grade reading scores 93% (white advanced and proficient) vs 57% (African-American advanced and proficient) or reading scores at any grade level. In fact, DPI data show the gap widening over the last four years. Now to be fair, I should note that all of this data is from November 2005, so something drastic might have happened this fall that we don't know about yet, but none of the state statistics show any decrease in the achievement gap...and that's no lie.

There is no data on the WINSS site for any year or for any grade that agrees with the comment attributed to Rainwater that there is no reading gap. Attributed to Rainwater is the language "no statistical achievement gap", but the sample sizes are too large to allow a gap of 30%+ to be statistically insignificant.

Even combining all minorities as a group and comparing to whites, the gap remains above 25%, still too large to be non-signficant statistically.

I'm at a loss.

The link to the WINSS graph is less telling than it initially looks, because there are a number of ways to define similar, and the criteria in each case is pretty broad (for instance, based on district size, Madison, at 24,000+ students falls into the same category as districts with 10,000+ students)

In the case of the link on this page, the districts that WINSS pulled up by % of economically disadvantaged students have enrollments ranging from 13 to 17 students.

It is possible to pull up similar districts by many different criteria. What is found in most cases, is that the districts Madison is compared to differ significantly in at least one key area (enrollment, % economically disadvantaged, % with Limited english proficiency, % with disabilities). The districts that show up on a variety of sorts and which are most similar by multiple factors seem to be fairly similar to Madison in terms of scores (Sheboygan comes up on multiple sorts and has a P&A of 80% to Madison' 79%). I found no other districts that matched in most areas.

My point is not that Madison's current percentages are acceptable: clearly we have not gotten where we want to get yet. However, if we are going to use comparisons to judge effectiveness, we need to take into account the range of factors known to impact test scores or it is difficult to find anything useful in the comparison.

Teacher L:

With all due respect (and I mean that sincerely, as per our previous conversation about teacher salaries), this is exactly the kind of thing that NCLB (which, I feel compelled to say to the Bush critics here, was passed with the approval of both Teddy Kennedy and Democratic Cal. Rep. George Miller) seeks to end. That is, the relative comparability of student achievement based on varying socio-economic factors.

NCLB may be viewed by many in the education community as unreasonable, underfunded, unrealistic, and uncompromising. But it stems -- and go read the Congressional testimony during its initial authorization, as I have -- from a belief that all children ought to be able to attain at least nominal levels of achievement in basic subject areas, regardless of socio-economic factors. A pretty broad, bi-partisan group of lawmakers agreed, and -- as the current debate over NCLB's reauthorization is finding -- that central belief is still at the core of what lawmakers expect schools and students to achieve.

While we can always find problems in the districts with whom we compare ourselves, the starting point for this thread was the claim that Madison has eliminated the achievement gap. The data clearly show that the gap has not disappeared. The link below shows 4th grade reading data for continuing students (the eliminates problems with students just entering the district). In 2002, there is a difference of 25% between the numbers of white students scoring advanced and proficient and the number of African-American students scoring at those levels. By 2005 the difference between the two groups of students was 34.3%. Clearly, the gap has not been eliminated.

http://data.dpi.state.wi.us/data/graphshell.asp?GraphFile=GEDISAPRIORYEARS&CompareTo=PRIORYEARS&Grade=4&Group=Race/Ethnicity&SUBJECTID=1RE&EligibleOnly=NO&WOW=WSAS&ORGLEVEL=DI&FULLKEY=02326903ZZZZ&DN=Madison+Metropolitan&SN=Show+Schools

To play devils advocate,
is he refering to the stats of comparing low-income white to low-income minorities and high-income white to high income minorities. There is a difference in the comparison of income as well as race.

I was at the School Board Meeting when the professor laid praise on our district for the closing of the gap due to the Hope program. I was a little confused then too.

Phil M,
I completely agree with the spirit of your post--and I was not trying to imply that demographs change our basic responsibility to our students. What I do think, is that comparisons are useful only if we can use them to learn and improve our own practices. Comparisons of a district like Madison to districts with fewer than 20 fourth graders, are not particularly useful. Comparisons to districts with similar demographics could be very useful if those districts were outperforming Madison by the standardized measures. However, Madison is a hard district to match in Wisconsin--we have high percentages in a variety of categories that are currently linked with low achievement. Those categories overlap, but do not fit neatly on top of one another. The closest match (according to WINSS)seems to be Sheboygan, which does not have significantly different scores. Thus, it doesn't jump out at me as being particularly useful either. It is certainly not unfair to have high expectations for the district, but as someone who is working hard every day to improve achievement for students, it feels discouraging and unfair to hang MMSD on its failure to meet the same standards as a district that is so completely different in makeup.

I am very confused by the quote that was attributed to Superintendant Rainwater. All I can say is that I have heard no one from district administration ever say such a thing. The messages that come out to us in the schools are that we have a lot of work still to do to close that gap. Given that, it is hard for me to take that particular quote at face value.