A Letter to the
MathLand Folks

May 9, 1997

William Jarrett, President
Creative Publications
1300 Villa Street
Mountain View, CA 94041

Dear Bill,

Thank you for sending the Grade 2 and Grade 4 MathLand materials for review and for the offer to come to Mountain View to meet you and view the MathLand operation first-hand. I have other questions that would need to be answered first, all originating from the MathLand promotion document labeled "For Immediate Release, April 10, 1997", and entitled "Review of Math Test Scores Indicates Improvement in California Elementary Schools Using New Curriculum." It claims that MathLand is "currently the most widely used math curriculum in California" so congratulations on your marketing success.

This eight page document apparently received wide school and press circulation and it summarizes results of scores "drawn from standardized tests of 30,000 students in eight California public schools districts". The results of this study "revealed that students are learning what they should be learning in basic math skills as they progress through each grade level" and the report concludes that, "this refutes earlier charges by critics claiming that the program was destined to fail. It shows we're on the right track and on our way to reversing the long trend of declining math scores in this state." After looking at the actual second and fourth grade materials, I must admit that I would have come to a different conclusion; it would have been the same conclusion, in fact, as those unnamed critics. Since I am not a "data-proof ideologue", however, I hereby request the following:

1. Open access to the data base from which this data was sampled; at the very least, the names of the eight school districts involved.

2. The phone numbers or addresses needed to contact David Westmoreland and the other, unnamed, independent auditor who analyzed the results.

3. The phone number or address, or even just the school, needed to contact Catherine Au, "a Sacramento teacher for fifteen years", who attests to "dramatic improvements in my students" since she does not appear to be a teacher in the Sacramento Unified School District.

4. The source of the quote that MathLand, "has also been endorsed by the National Science Foundation" since it is my hope that the NSF is not in the business of making such endorsements, especially with no supporting data.

My suspicion - almost conviction, really - is that I will come to different conclusions about what is indicated in the data. This suspicion comes from looking at the MathLand materials directly, from seeing the documented and serious decline in the Santa Barbara Unified School District, and from the last page of this promotion document itself. That is a letter from Ruth Bunyan, principal of Roscoe Elementary School, documenting dramatic increase in CTBS math scores from 1995 to 1996 and crediting "MathLand materials and instructional strategies" for part of their "significant difference during this past year". My thanks to Cindy Lopez for asking Ms Bunyan to contact me. She called and we had a nice chat on April 30. She has retired from LA Unified since her November 4, 1996, letter but I have also talked with both the current principal, Mary Kurzeka, and the MathLand coordinator mentioned by both principals, Ellen Morgan.

Not mentioned in Ms Bunyan's letter was the longer term CTBS history at Roscoe. Following the same class from first through fourth grades, we have 49, 44, 22, 44. That is, the first year of MathLand (22) was a genuine disaster for these students but they somehow managed to recover and their recovery represents ALL of their progress. Even this data is misleading in that it is quite unrepresentative of the overall school performance. There were 27 fourth graders in the CTBS results at Roscoe in 1996, only 8 more the minimum that LA Unified requires to even report out the data. The letter makes quite a point of the fact that the school is "93% Hispanic with full bilingual classrooms in every grade." Given that situation at Roscoe, it seems strange that Ms Bunyan did not include the Aprenda results where 76 students were tested, nearly three times as many as were tested with the CTBS. Here are those results, also reported on a nationally normed percentile basis, and they are far closer to those "earlier charges that the program was destined to fail" than "reversing the long trend of declining math skills in this state."

                   Grade 2     Grade 3     Grade 4
                    93-94       94-95       95-96 

District Median       27          27          24  
Roscoe Elem.          32          23          15

A common explanation given by apologists for the decline in student performance data sometimes seen in implementations of reform in education is the lack of teacher familiarity with the philosophy, the methods, or the materials of the reform effort. "If only we had money to train the teachers", is often heard. According to Ms Bunyan, that does not explain the precipitous decline in the Roscoe Aprenda numbers. She told me that she, with Ellen Morgan, were on the front edge of the wave, so to speak. Roscoe Elementary is a LEARN "Phase 1" school within LA Unified and received extensive assistance from Xerox Corporation. MathLand training from the company itself was provided on three different occasions. They made their decision for MathLand because it was "the finest curriculum we could find" and not from ignorance of the methods nor the materials. In fact, they changed from Math Their Way, curricula already far down the reform movement road, so the teachers had had a great deal of experience in guiding children through the constructivist activities of MathLand before they ever started the program. Even so, according to Ms Morgan, a lot of time and a lot of money was spent in training teachers; "real mathematics training" was the way she put it.

There is another disturbing feature of the Roscoe situation that portends future problems unless immediate steps are taken to see that the anti-testing bias of some individuals in the public school environment are counteracted immediately. The comparable year-by-year Aprenda data for the next class, those who were fifth graders last year, is 23, 15, *, with the * indicating that the data for their fifth grade year was not collected; i.e., those students were not tested. When I asked Ms Morgan about the fifth grade data, she told me that "We only test at the 4th grade because that's the only place we have to." In regard to the CTBS data that was reported, "We were not disturbed that the scores went up but that's not what we are looking for. We put very little emphasis on CTBS scores." Even less, obviously, on Aprenda scores. It is appalling that these kinds of trends, both good and bad, that year-by-year standardized test data can indicate to parents and to policy makers, are being avoided within LA Unified, and perhaps statewide as well.

Sincerely,

Wayne Bishop
Mathematics & Comp. Sci.
Cal State Los Angeles