Teacher Unions: Useful? Um, Yes!

Back and Forth and Back on Teacher Unions

by Doug Noon

Before I fill in a missing piece from a wide-ranging discussion about teacher unions, we should review:

Diane Ravitch:

If getting rid of the unions was the solution to the problem of low performance, then why…. do the southern states — where unions are weak or non-existent — continue to perform worse than states with strong unions? And how can we explain the strong union presence in Massachusetts, which is the nation’s highest performing state on NAEP?

Mike Petrilli:

I’ve concluded that no, Diane isn’t right…. [W]hen it comes to union influence on the ground, at the district level, it’s not at all clear that the “strong states” versus “weak states” distinction makes any sense…. As Jay Greene told me, the unions’ goal “is to ensure as little policy variation across states as they can on their core issues.”

Jay Greene:

Many factors influence student achievement, so isolating the effect of teacher unions would require a rigorous social science research design that could identify the influence of unionization independent of other factors.

Rather than point to a state or district, which proves nothing, I would point people to a rigorous study [pdf] by Caroline Hoxby in a leading economics journal. The abstract states: “I find that teachers’ unions increase school inputs but reduce productivity sufficiently to have a negative overall effect on student performance.

Leo Casey:

Greene is up to his old “cherry picking” tricks here, citing the one study which supports his position while ignoring the many which do not. There is a small body of scholarly literature on the subject, and Hoxby’s essay is clearly the minority view; there are more noteworthy studies showing a positive relationship between teacher unionism and educational achievement.

Casey mentions a study by Lala Steelman, Brian Powell and Robert Carini, “Do Teacher Unions Hinder Educational Performance? which examines correlations between the presence of teacher unions and high SAT/ACT scores. Additionally, Casey cites F. Howard Nelson and Michael Rosen, “Are Teacher Unions Hurting American Education? [pdf], which makes a similar argument.

Casey also points to a couple of literature reviews on the topic, one of which comes from the Education Policy Studies Laboratory at Arizona State University, “School Reform Proposals: The Research Evidence,” by Robert Carini [summary pdf], [full report pdf].

Jay Greene:

Caroline Hoxby’s study, upon which I base my claims, employs a vastly superior research design…. But even if Leo insisted upon relying on the literature reviews he cites rather than the higher quality research, he would have to accept some results that aren’t very flattering to teacher unions. Those lit reviews find that unionization raises the cost of education by about 8% to 15%. In addition, they find that unionization tends to hurt the academic achievement of high-achieving and low-achieving students while benefiting more typical students found in the middle of the ability distribution.

Green likes Hoxby’s methodology because he believes that it separates causes from effects. But Greene ignores a criticism of Hoxby mentioned in the Carini paper:

Hoxby found that unionized districts had higher dropout rates than non-unionized districts from 1970 to 1990. Of the five studies examined in this section, Hoxby’s may offer the strongest evidence, although like the others, it too can be challenged on methodological grounds. In particular, Hoxby reported that she analyzed 10,509 school districts, and asserted that her sample constituted 95% of all districts in the United States in 1990. Given that there were 15,552 school districts in 1990, Hoxby’s research only covered 68% of the districts, not the 95% that she reported. It is not clear why nearly one in three districts were lost. More important, the missing districts were likely fiscally dependent districts, the bulk of which are located in strongly unionized Northeastern states. This is a potentially critical omission that may completely change her findings, particularly given the small gap in dropout rates that she found.

And even more significantly:

Further, Stone has argued that Hoxby’s finding that unionism led to higher drop-out rates is not necessarily inconsistent with research documenting favorable union effects. The argument is that, with a focus on high school dropouts, Hoxby essentially limited the scope of her study to lower-achieving students. In any case, three other studies discussed previously have reported that unionism did not increase dropout rates.

Nowhere do I see delusional people harder at work than I do when I read the contorted ravings of education policy wonks discussing harebrained ideas for how to fix schools. Choose your swamp. Then wade around in it. This is the beauty of the internet. Green reports about the achievement gap, which is attributed to the standardization of instructional settings that comes from unionization. But he ignores the obvious fact that this is precisely what the standards movement is all about, and which teachers recognize as exacerbating the problem.

The anti-union pro-corporate education reformers don’t have any actual solutions. Instead, they resort to changing the subject by criticizing unions for tying the hands of administrators. They don’t acknowledge the fact that administrators haven’t the slightest clue about how to stimulate academic progress for disadvantaged students without resorting to heavy-handed “motivational” approaches devoid of any educational merit.

Teacher unions haven’t been vocal enough in opposing these so-called reforms. But we can look to the teachers in Los Angeles for an example of teacher solidarity and activism. This is important because corporations are getting more militant and punitive in their efforts to prevent workers from unionizing. The “reformers” know that if the unions don’t slow them down, nobody will.

h/t Borderland