August 28, 2006

Irony Noted

The AFTies are at it again.

The AFTies take Edspresso to task for relying on a study on voucher effectiveness that supposedly shows:
A blip in year two, no impact after three years, "uncertain results"
and concludes:
that hardly makes a strong case for vouchers.
And yet ...

Hardly a post goes by on the AFTie blog that doesn't tout the effectiveness of gaining new union members reducing class size based largely on Project STAR which showed a blip in year one and no additional impact after three more years.

Hardly makes a strong case for reducing class size now does it.


SteveH said...

"...that hardly makes a strong case for vouchers."

No choice is better than choice? All private school parents are stupid or elitist? Poor parents cannot be trusted with choice? What is it?

The conclusion is that choice is not helpful and should not be allowed. I am always amazed at how eager they are to come to this conclusion. They want to make the decision and not leave it up to the parents. Parents are stupid. Parents, who take the effort to put their kids into a charter or private school, are not smart enough to decide for themselves. They need to be told what to do. Forget the fact that these parents could put their kids back into the public schools. They are apparently not smart enough to do that.

Rather than rate the performances of schools and let the parents decide, they want to decide for them. Of course, these people have no biases or agenda. Right.

rightwingprof said...

You're a better man than I. The AFT blog, along with the Chronicle, go way past my tolerance for moonbattery.

Mike in Texas said...

There is PLENT of research to indicate smaller class sizes are a meaningful reform that produces results.

For example, this study from the US Dept. of Education:

Here are some highlights:

Beside the STAR project you deride there is plenty of data from Wisconsin's SAGE program, which produced nearly the same results and is much more recent.

You CLAIM to be interested in research that shows what works, unless it disagrees with your pre-formed opinions.

Mike in Texas said...

Whoops, left out the highlights:

The STAR findings consistently showed a positive small-class effect. “At each grade level (K–3),
across all school locations (rural, urban, inner city, suburban), on every achievement measure
(criterion-referenced and norm-referenced tests), and for all subjects (reading, mathematics,
science, social science, language, study skills), the small-class students exceeded their peers in
regular and regular/aide classes. The results were both statistically and educationally significant”
1 (Boyd-Zaharias & Pate-Bain, 2000).

Small classes reduced the white-minority
achievement gap. While all students
significantly benefited from participation
in small classes, the greatest advantages
were found for minority, inner-city
students from low socio-economic
backgrounds (Word et al., 1990). The
benefit of small classes for minority
students (most of whom were African-
American in this study) was about twice
as large as that for white students (Finn,
1998). While students were in smaller
classes, the average test scores increased
by 7–10 percentile points for
African-American students and 3–4
percentile points for white students
(Krueger & Whitmore, 2001).

Smaller classes had the lowest percentage
of students retained in grade among
the three groups. For grades K–3, (S)
had an average of 4.9 students retained,
compared to 6.8 for (R) and 5.7
for (RA) (Word et al., 1990).

CTBS post-test scores showed that SAGE
students statistically outperformed their
comparison school counterparts in reading,
language arts, math, and total scores.

African-American SAGE students scored
lower on the pre-test in every sub-test,
except reading, than African-American
comparison students. Post-test results,
however, showed that African-American
SAGE students scored significantly
higher than African-American comparison
school students on every subtest and
had significantly higher total scores.

African-American students in both SAGE
and comparison schools scored significantly
lower on the pre-test than white
students, with a larger gap in the SAGE
schools. Post-test results, however,
revealed that African-American SAGE
students gained more than white SAGE
students in terms of CTBS total scale
scores, thereby reducing the achievement
gap. African-American students in
comparison schools achieved lesser
gains, and in these schools, the achievement
gap grew (Molnar et al., 1999).

Interesting that the Republicans, with all their talk of reducing the achievement gap, never seem to bring up class size reduction. In fact, little brother Bush has been doing all he can to circumvent the Florida constitution to increase class size.

KDeRosa said...

MiT I suggest you take a look at this CIS review of the class size research.

The current state of the research on class size reduction is methodologically flawed and shows small effect sizes. A 0.20 standard deviation increase in performance is too small to justify the expense. Your typical title I school performing at the 20th percentile would only be performing at the 28th percentile with project star-like effect size increases. Hardly a cause for celebration.

When you can show me an effect size about a standard deviation in a non-flawed study then we can talk.

KDeRosa said...

These are the commonly recognized methodological flaws in STAR:

The most important concerns are: 1) that not all students started the experiment at the same time, because kindergarten was not mandatory or universal in Tennessee; 2) sizable attrition occurred over the course of the experiment because of mobility and other factors, and this attrition was likely not random; 3) parents,teachers, and schools knew they were part of an experiment and, because of pressures from parents, part of the experiment was compromised by re-assignments of students;4) no achievement tests were given before kindergarten, making it difficult to analyze whether elements of the random-assignment process contributed to any subsequently
observed achievement differences; 5) approximately 6 percent of the students were transferred across treatment groups at the end of the first year of the experiment; and, 6) there was some drift from the target class sizes of 15 and 22 so that there is actually a distribution of realized class size outcomes over time in both treatment groups. Each of these issues has been raised by the initial researchers (e.g., Finn and Achilles, 1990) and by later interpreters of the results (e.g., Mosteller (1995) and Krueger (1997)), but the experimental data do not provide information that permits fully ascertaining the effects of such possible problems.

In fact, the data from this experiment have not been generally available to researchers who were not involved in the original design and analysis.

And lastly, the STAR findings have never been replicated.

Mike in Texas said...

The STAR findings have been replicated in Wisconsin's SAGE project.

California attempted an CSR program but it was so poorly done it was doomed to failure, and if I remember correctly class size was reduced from an average of 29.9 to 29, or basically not at all.

TThe problem with most class size reduction initatives is that they do not lower class sizes to the 15 or below mark, where the research indicates the real benefits appear.

Another problem with class size reduction is that it does not make any money for the politically connected buds of the Bush family, therefore it is not to be taken seriously.

The CIS review is not to be taken seriously either.The "reflections" merely point out what the researchers found and put a negative spin on them. Yes, a child needs to be in a reduced size class for several years to gain the full benefit. Yes, it works better in the younger grades, the younger the better. However, the effects ARE long lasting and measureable.

Mike in Texas said...

And let me direct you to this review of class size research.