Math Test Results

by Art Woolf

The U.S. Department of Education has just released the math results of the 2009 National Assessment of Education Progress (NAEP), or, as the DOE dubs it, the "Nation's Report Card."  The good news is that Vermont was one of only four states that had an increase in the test scores in both 4th and 8th grades (the two grades tested).   The bad news is that we're basically average.

In fourth grade, Vermont students scored 248 compared to 239 for the nation.  But the students who inhabit Vermont's schools look very different than students in the average school in the U.S.  Nationally, 58% of students are white; in Vermont it's 96%. (That's for 2004.)  Vermont's white test score was, not surprisingly, 248.  The national white test score was identical to Vermont, 248. 

In eighth grade, the results are similar.  Vermont students scored 293, compared to 282 for the nation.  But the white national score was 292, so Vermont students scored only marginally better than their peers. 

Vermont spends more than almost every other state to educate our students, but we don't seem to get a lot for what we pay for.

Comments

[Timothy Diette]

Art,

I love the graphic, it is one of my favorites for ed assessments.

Given your prior posts and the posts of others on the Tiger (and more broadly in the economic research) not much of a surprise.

Do you know if the test scores are broken down by district or county?

It would be interesting to try to tease out differences in achievement and find schools/districts that appear to be particularly successful (or unsuccessful) given the per pupil spending and school/district demographics.

[James Rude]

Just think, if the NPR radio show, The Prairie Home Companion, emanated from Vermont, then the closing of the show could end with: "Where all the women are strong, all of the men are good looking, and all of the children are...well, just average.

[G. Cross]

This is the National Assessment of Education Progress which has been going on for many, many years. It is a sampling of schools, grades, students and more; thus there are no individual student, school or district results. It is the only student assessment that is nationwide and carried out in a true research model. It is intended to be a longitudinal study. The real value is in the study of results over a number of years. Suggesting that it is a tool to evaluate a state's investment in education is akin to suggesting that the only economists of note(value) are those who sit in endowed chairs.

[trice]

"...intended to be a longitudinal study. The real value is in the study of results over a number of years. Suggesting that it is a tool to evaluate a state's investment in education is akin to suggesting that the only economists of note(value) are those who sit in endowed chairs."

You've got to be kidding! What is the purpose of assessment if you can't evaluate investment? What other tools do we have?

Research also shows that the graduation criteria of today's colleges don't even meet the level of high schools a hundred years ago. Even by your own standard, longitudinally, student scores in public schools have maintained (if that) only by lowering standards over time. Certainly employers and companies importing labor or outsourcing jobs will attest to that, but then I guess that doesn't qualify as a "true research"...courtesy of the same party that brings us the junk science of the IPCC global warming computer models.

As usual, no liberal program is ever to be judged by outcomes... only intentions.

[Chris Campion]

Interestingly, Art's breakdown of the data is not what you hear on the local news outlets. Also unmentioned is just how awful the metrics are in general - being average in a pack of lousy is not the KPI we're shooting for, with all the money we invest in education.

I thought educators hated standardized tests, by the way - so now this study is to suddenly be lauded by Vermont educators? There's a phrase for this out in the real world - "cherry-picking the data".

[Peter Joes]

The comparison we need to see is by income.

We might be white but I would bet that income is a more appropriate comparison.

So, Art, how does it stack up based on income.

PJ

[G. Cross]

Over the years only two factors have had a constant positive correlation with school performance; 1. socio-economic status of the family (income), and 2. educational attainment of the mother. More recent research suggests that educational attainment of both parents makes a difference. Thus, Peter Joe is on the right track.

[Art Woolf]

It gets difficult to analyze the NAEP test results based on socio-economic status. The only information that is collected (since kids don't know their parents' income) is whether the student tested is eligible for free or reduced price lunch. That's a good proxy for poverty. In Vermont, only 34% of 4th graders were eligible and 29% of 8th grade students. Those are both far below the national average of 48% and 43% and Vermont ranks 4th or 5th lowest in the nation.

Test results, for example in 8th grade math, show that Vermont students do better than the U.S. average for students who are eligible for free and reduced lunch (266 for the U.S. and 277 for Vermont) and for students who are not eligible (293 vs 300). So part of the reason we do better is that we have fewer lower income students. And since black and Hispanic families have lower incomes than whites, I think that race is a good proxy for income.

[G. Cross]

Art is close, but what he is not saying is that some schools in Vermont have 90% to 100% of the student body eligible for the free and reduced lunch program, while others have fewer than 10%to 15%. This creates a huge difference in these respective schools. On top of this a few schools in Vermont have between 20% and 35% of their student body residing in households where English is not the home language. Many other schools have no such students. In the last 10 - 15 years Vermont schools have become more diverse than at any time in history. This diversity, while beneficial in the long run, brings great challenges to some schools. Those who want to judge Vermont schools need to consider the many differences in our schools. I would urge all readers to visit schools across the state before making judgements about ALL schools.

[Chris Campion]

George, you're making a good case for school district consolidation.

[G. Cross]

Reducing the number of school boards in Vermont, thus having fewer school districts would help to reduce the basic inequities which currently exist. I have said this all along. The problem is that the arguments always gets into consolidating schools; whereas, the debate should be focused on reducing the number of school boards. We now have about one school board member for every 75, or less, students enrolled. Once there are fewer school boards all the other "consolidations" will flow in a slow systematic fashion.