Tag Archives: trust

Can we trust their data?

Dear Readers,

Below is a slightly updated version of comments made by Michelle Fine at the Montclair Board of Education meeting on Monday, November 17, 2014:

A Closer Look at the Strategic Progress Report 

A year ago, on October 8, 2013, the Montclair Times reported that the achievement gap had widened in Montclair schools. David Deutsch referred to the gaps between economically disadvantaged African Americans and non-disadvantaged Whites, as “massive.”  Shelley Lombard said, “If anybody can address this and start to do something about it, it’s Montclair. That’s why some board members are very enthusiastic about Penny MacCormack.”

Then in October 2014, Gail Clarke presented the Achievement Gap analysis to the public. I draw your attention to one chart: High Expectations for All: Race and Ethnicity. Ms. Clarke told the public that the Black-White gap had narrowed on 9 of 15 measures.

Some might see this as cause to celebrate, but there are serious problems with the presentation of the data: problems of omission and commission. After consulting with a team of three statisticians, I summarize the errors in data presentation and interpretation below, and for this evening will identify at least one serious policy issue that Board must address. My comments derive from a simple comparison of the chart presented to the public and the raw data in the Appendix and on the MPS website.

What’s missing? In the full report for 2014, economically disadvantaged Black children are never compared to non-disadvantaged Black children nor to non-disadvantaged White children. This statistical omission is ironic, given that the race/class gap was such a cause of such concern in 2013 when the Board was suggesting that Superintendent Alvarez had underestimated the gap.

What’s wrong with the data? This year, when you are assessing the MacCormack administration, not only did you fail to provide critical comparisons of low- income African Americans to non-disadvantaged Whites, but the data you did present to the public consistently overstated the percent of Black students who scored as proficient when compared to data in your own Appendix. The discrepancy is as high as 13% on the 3rd grade ELA, 11% in 4th grade, 8% in 5th, 7% ELA 6, 14% in ELA 7, 13% in ELA 8 and 4% in HSPA for ELA 11. The same patterns can be found in math.

Stated plainly, the key comparisons from last year are not available to the public and the data presented to the public systematically and inaccurately skew in the direction of claiming substantial achievement gains among African American students.

I have no way of knowing if this is human error, cherry- picking or strategic misrepresentation of only the non-economically disadvantaged African Americans. I have contacted both Gail Clarke and George Glass with the hope that they can help me understand the conflicting information.

The public deserves accurate reporting and fiscal responsibility. At this point “correcting” the chart is necessary, but no longer sufficient. The larger issue at stake is one that has haunted this administration for years: questions of accuracy, honesty and lack of public trust.

As I understand, the Board voted to grant Dr. MacCormack and perhaps one of her Chiefs a “merit bonus” based, in part, on these presumably narrowed gaps.

My question to the Board is simple: Will these merit bonuses be held in escrow until we have numbers we can trust?

Michelle Fine