March 2, 2000

Dr. Dan Wertz, Superintendent
Okemos Public Schools
4406 North Okemos Rd.
Okemos, MI 48864

Dear Dr. Wertz:

I am writing to administrators and board members for the Okemos Public Schools to solicit views about the results of the Iowa Test of Basic Skills administered to Okemos sixth graders in October 1999.  I am the parent of a sixth grader at Kinawa Middle School.  As I looked through  my daughter’s scores on the Iowa Test, I noticed that although she had done very well in Reading and Language (percentiles in the 80s and 90s), her Math Computation percentile (50) was notably below all of her other results.  In fact, her only other percentile rank below 82 was Math Problems and Data Interpretation (74).  I wondered if my daughter had some special difficulties with Math Computation so I inquired at the school about how the entire sixth grade had performed.  Mrs. Hoevel was kind enough to provide me with the summary report (aggregated results) for the entire sixth grade.  While I was pleased to see that my daughter’s overall performance compared favorably with her classmates’ results, I was very dismayed to see that the average result for Kinawa sixth graders in Math Computation was only at the 50^th percentile while the rest of the class’s scores generally placed them in the 70s for national percentile rankings (except for a 69 in Vocabulary and a 65 in Capitalization).  The initial question I pose to you is: Why are the Kinawa students’ results for Math Computation notably lower than their results for all other subject categories tested?  It appears to me that there is a curriculum problem–presumably one that begins in upper elementary school, since the test was administered during the students’ second month of middle school–which is making the Kinawa students relatively weaker than their counterparts around the country.  One might argue that they are merely average rather than weak compared to their counterparts, but it makes little sense to accept mediocre results for Okemos students in one subject when they are well above the national average in everything else.

Dr. Tweedy, the principal of Chippewa, was kind enough to supply me with the summary results for sixth graders at Chippewa (see attached tables).  While the results generally fulfilled my expectation that the Okemos curriculum would produce similar results for students at both middle schools, I was shocked to see that Chippewa students do not suffer from the same deficit in Math Computation that afflicts the Kinawa students.  In terms of national percentile rank, the Chippewa average (68) was only slightly below the rest of their scores but it was an astounding 18 points above the Kinawa score.  If analyzed in terms of Grade Equivalent Scores (see table), this means (as I understand it) that the average Chippewa student is one full grade (0.9) ahead of the average Kinawa student in Math Computation.   I am utterly perplexed.  How can this be?  The numbers of students taking the tests at each school were sufficiently large (156 at Kinawa and 181 at Chippewa) that the scores should not be skewed by a few poor performers at Kinawa or a few outstanding performers at Chippewa.  This leads to my second question for you: Is there something different about the materials or approach to teaching Math Computation at Hiawatha, Edgewood, Wardcliffe, and Kinawa as compared to Central, Cornell, Bennett Woods, and Chippewa?

Because I was curious about how the Okemos results for the Iowa tests compared with results in other, comparable school districts, I phoned school districts around Michigan to inquire about their results.  It turns out that many school districts in Michigan do not use the Iowa tests.  Except for Ann Arbor (which never sent me their results), I could not find another system that used the Iowa tests among the dozen districts that I contacted.  Relying on the Internet, I sent e-mail inquiries to several dozen school systems around the country.  Ultimately, I was able to obtain Iowa test results for two comparable school districts: Ames, Iowa, a college town,^[1] and West Des Moines, Iowa, an affluent suburb.  (I also found graphs summarizing Iowa test results on the web page for Rochester, Minnesota, but the graphs did not provide all of the national percentile scores or grade equivalents for purposes of comparability).  The West Des Moines results were comparable to the Okemos results for most subjects–but well above Kinawa in Math Computation.  The Ames results for all math categories were notably higher than those for Okemos.

I received an e-mail communication from Tony VanderZyl, the Assessment and Mathematics Specialist for the Ames Community School District, that informed me that their students produced low Math Computation scores on the Iowa tests in 1994.  As a result, the district took action:

A few years back, much attention was given to [Math Computation] in our district.  Following the attention, we added several different district math assessments to determine if and where we might have a problem in student performance in mathematics.  For three years, we have used the New Standards Mathematics Reference exam.  That assessment measured three broad areas of mathematics: Skills, Concepts, and Problem Solving.  On the New Standards, we consistently score significantly higher on Skills (90% meet standard) than on Concepts and Problem Solving (50% meet standard).  We also developed local assessments of computation and of problem solving.  On the local computation assessment, most of our students get 90% correct by spring of the year.  On the local problem solving assessment, about 60% of our students meet standards.
    We have also followed cohorts into high school.  There we find that our students score extremely well in mathematics on SAT, ACT, ITED which do not subdivide the math scores.
    After 6-7 years of discussions and study, here is our conclusion. Yes, we do have a deficient area in our student performance in mathematics.  It is not computational skill but rather problem solving and reasoning.  Our school board has targeted problem solving as a goal and we have been devoting professional development and other efforts toward that goal and are starting to show some gains.

I reproduce this message concerning Ames just to note that at least one other school system seems to have taken very concrete steps to address the issue that I detect in the Kinawa results.

Mrs. Hoevel has kindly offered to meet with me about this issue.  However, since this appears to be a district curriculum issue that has its roots in the elementary schools, I wanted to gather comments from officials throughout the district to aid in my understanding.  Thus, as I noted, I am writing to you to solicit your views about this issue.  I would be most grateful if you would respond to the following questions:

1.  Why are the Kinawa sixth graders’ Math Computation scores so much lower than their other scores–and so much lower than the scores of their counterparts at Chippewa?

2.  Do these Math Computation results at Kinawa indicate that there is a problem with the math curriculum?  If so, what is going to be done about it?  I have already heard that improving math results is a goal at Kinawa.  Frankly, however, I would like to know more than that it is a goal.  I would like to know what concrete steps will be taken to address this issue.

3.  Does this apparent problem have its roots in the way math is taught in the upper elementary grades since these test results were produced in the students’ second month at middle school?

4.  If Ames, Iowa can show math results on the Iowa tests that place their students at high percentiles (86 to 99), why can’t Okemos do the same?

Thank you very much for your prompt attention to this matter.  Meanwhile, I am informing other parents about this issue since they may have erroneously believed – as I originally did – that a relatively low Math Computation score reflected a unique weakness for their individual children rather than a systemic problem evident throughout the sixth grade at Kinawa.  Thank you again.

Sincerely,