Read a Post for Creating Mathematical Futures through an Equitable Teaching Approach: The Case of Railside School   Reply to this Post   Re: A Prime Example of Data Cooking?  Posted By: Fred Welfare on October 1, 2013   After reading both Boaler's statement concerning harassment and Milgram et al's concerning invalidity, it seems clear that Bishop, Clopton and Milgram are reaching for straws.
This is from p5 of Milgram: "Among these schools, Railside had the second highest percentage of highly qualified teachers at 97%, tied for highest level of average parental education, had, by a full 8%, the lowest number of free lunch students and had the fourth smallest number of ESL students. On the basis of these rankings, there is reason to be concerned that Railside students would not do well in college."
This is an obvious non sequiter, the premises do not fit the conclusion. There is no reason to be concerned; the conditions for success appear to be in place except for parental education. Why not give the benefit of the doubt that there are conditions for educating students to achieve beyond their parents level, namely, teacher qualifications, nonpoverty and a low % of 1st generation immigrants, which in many cases is a reason for greater achievement!
The real reason for the problem which is perhaps the basis for Milgram and Bishop's criticism is on p8 where Milgram et al discuss faculty discontent and instability. The article is ambivalent about the acceptance of the new integrated math curriculum stating at one point that many teachers prefer it, at another that students have to choose it and then are persuaded, at another that parents prefer the traditional curriculum (the big difference is that geometry and algebra are taught concurrently rather than sequentially), and again at another point that teacher who had quit? returned when one of the schools reverted to the traditional curriculum implying that teachers can quit and then recover their position at whim! Here is the bane of the matter. Teachers are resisting the reform, the innovation while the administration and the State are requiring the new curriculum. The obvious parallel to student resistance is selfevident.
If by 'Data Cooking,' Wurman or Milgram and Bishop mean the nonsense on p9 where essentially similar statistics are presented for 3 schools and 2 are stated as 'better' than the 3rd and the 3rd school is stated as 'significantly the worst,' then it is their judgment which should be questioned. There are no ""significant"" differences between the 3.
Bishop et al take issue with Bouler over the "disaggregation of the data which they claim is disaggregated on the state website. So, take a look. There is a distinction between 91011 graders test results, but it is does not differentiate which 9th grades took the Integrated course or the Traditional course, it also does not explain how students were selected for the tests  did they have to pass the course or did everyone assigned to the course take the test. If so, why did 11th graders take the test; were they fails from 9th and 10th grade? So, it is hard to understand Milgram's criticism if they cannot show what disaggregation should entail along with the possible ambiguities of the data.
Milgram takes issue with Bouler's statistic that 41% of the 12th graders at Railside take PreCalc or Calculus. (This is Bouler's charge that Bishop and Milgram have acted unethically.) So, they call up the Principal and the Lead math teacher for an interview. They determine that no students have passed the AP Math Exam in the last 5 years. Then, they peruse the course catalog and presume on the blurb in the catalog that the courses are lowlevel. They go on to conclude that barely 30% of the graduates of Railside will be admitted to the California State University system on the basis of their math scores which they claim is poor and for some odd reason they add in that it is just as poor as the other 2 comparison schools. There is a graph on p14 which clearly indicates that Railside is considerably better that the other 2 schools and outperforms the CA average for several years. Perhaps, Milgram et al thought no one would look at the graph and compare it to their evaluation of the graph. Evidently, it is Bishop, Clopton and Milgram who are cooking the data.
Fred Welfare

 