Your Lying Eyes

Dedicated to uncovering the truth that stands naked before your lying eyes.

E-mail Me

Twitter: yourlyingeyes

13 March 2011

But What About the Kids?

Surprisingly, in the imbroglio over the plan to strip away collective bargaining rights from Wisconsin public employees (at least those without the guns and fire hoses), most of the overheated rhetoric has been about how terrible it would be for the workers themselves, rather than for the school kids. That might have something to do with schools having to be shut down so teachers could march in the capital.

But that hasn't stopped others from making such claims. There was that silly little chart showing how badly the five states without collective bargaining for teachers do in the all important "Combined SAT/ACT Rankings".* I can't even imagine how such a ranking could be compiled in any meaningful way - indeed, the inventor of the ranking admits as much* - it's a rather dodgy statistic.

Then of course Krugman got into the act, claiming that Texas has the worst graduation rate in the nation compared to Wisconsin's stellar rank. Again, graduation rate is a highly malleable statistic, even assuming the rankings Krugman claimed are even true. And of course he failed to even consider the impact of demographics in a comparison of two states where one (Texas) has lots of minorities while the other is overwhelmingly white. IowaHawk then took Krugman to task, comparing Wisconsing unfavorably to Texas on NAEP scores, disaggregated by race.

And surely, if there are negative effects on education from denying teachers collective-bargaining rights, then we would see it in the NAEP scores. The NAEP gives the same test across the country to 4th grade, 8th grade, and 12th grade students. If lack of collective bargaining leads to poor educational results, I can't imagine these wouldn't be reflected in 8th-grade reading performance. So I looked at those. Of course we have to separate out the effects of race, since that one variable has such a profound statistical impact. Here are the results for white 8th-graders on the 2009 NAEP Reading tests by state:

I've highlighted the top 4 - the wealthy Eastern seabord states - and the bottom state, West Virginia for context. The 5 states without collective bargaining are in light green. Wisconsin is in light blue. As you can see, 2 of the 5 score rather well - Texas is 12th and Virginia is safely in the top half. Wisconsin trails these two, just above one of the other 5, North Carolina. Georgia and South Carolina, the other two, are in the bottom half, but far from the bottom. It's pretty tough to argue - especially if you're from Wisconsin - that lack of collective bargaining has negative effects on education.

But what about for African-American children? Again, there's no evidence of a poor effect of lack of collective bargaining on black childrens' education, either.

Again, I highlighted the top 4 - though here it's not quite so predictable - look who's #1! But this time 3 of the 5 non-collective-bargaining states are in the top half, and again the other two are far from the bottom. But look who is almost at the bottom - Wisconsin! Only Arkansas has worse reading results for African-American students (I guess that head start they got in desegregation didn't take).

I fully support teachers having a union, by the way. As a parent, I have seen how a few politically savvy parents can get real chummy with the school administration and get their way. I wouldn't care to see the fate of teachers in their hands - so there's nothing like a teacher's union to fend off such meddling. However, I don't think a very good argument can be made that collective bargaining provides any kind of educational benefit.

Also, I find it unlikely that Wisconsin teachers are themselves to blame for any of this - there are many compounding factors. For example, Steve Sailer pointed out in a Marginal Revolution thread on this (that I can't find) that generous welfare benefits in Wisconsin could have drawn a different minority population - one more interested in handouts than working - than Texas, with its limited welfare benefits. Texas might have more high-tech industry, drawing smart parents (and their smart kids). Generally, I'm pretty sure that what we measure as "quality education" is little more than a measure of how smart the kids are to begin with.

* He has also compiled what he calls the State Enlightenment Rankings - it's rather amusing to anyone who has heard of Daniel Patrick Moynihan's quip about "proximity to the Canadian border."

The source for the data is of course the NAEP - specifically the Data Explorer. It's a simple on-line application to select the data you want - try it out some time. One question you might have is "what about other tests and grades?" Well, I don't thing the 12th grade tests are worth much due to the distorting effects of dropouts. I took a quick look at the other tests and didn't find any surprises - you're welcome to look of course, but I doubt you'll find anything materially different from the above. The specific data behind these charts are shown here.

1 Comments:

Anonymous CJ said...

Gotta say something about those charts -- the visual presentation of quantitative data, if you want to get highfalutin about it.

In the first chart, New Jersey's red bar looks almost twice as tall as West Virginia's, implying a large difference between them. However, if you actually look at the numbers on the vertical axis, you see that NJ's average score is 281 and WV's is 256 -- a difference of only 10 percent.

The trick is, of course, that the bottom of the visible scale isn't zero, it's 240. You might also note that the second scale depicting black scores has the bottom of the graph set at 220. Incidentally, blacks in Hawaii, which tops the second chart, score an average of 256 -- exactly the same number as the West Virginia average which is the lowest datapoint on the first chart.

Strange presentation.

March 22, 2011 4:00 AM  

Post a Comment

<< Home