Lifting ‘ceiling’ helped Christel House, other schools

It was widely reported last week that Tony Bennett boosted the grade for Christel House Academy by finding a way to disregard scores on high-school-level algebra and English assessments. But that only got the school’s grade from a C to a B. How did it get to an A?

Here’s the answer, thanks to Cynthia Roach, director of assessment for Indianapolis Public Schools: Indiana Department of Education staff also removed a “ceiling” that had been used in calculating grades.

This is a pretty big deal. The change improved final grades not only for Christel House but for more than 140 others schools. Some school officials may have been aware of the new approach, but I can’t find evidence that DOE officials discussed it as a policy matter with the State Board of Education or shared it with the public.

Indiana’s grading system gives schools 4 points for an A, 3 points for a B, 2 for a C and so on. Elementary-middle schools get a base grade for the percentage of students who pass ISTEP exams in math and English/language arts. Additionally, they get up to 2 bonus points if a high percentage of certain students show “high growth.” Sub-grades for math and English/language arts are averaged to produce the school’s overall grade.

The state initially put a ceiling of 4 points (an A) on the math or English sub-grade for any school; in other words, a school couldn’t get extra credit for high scores and high growth in the same subject. State board members said this would keep schools from getting an A if they didn’t excel in both math and English. You can see an explanation and the rationale for the ceiling in items No. 11 and 29 from an old FAQ document for the state’s grading metrics. But those items were deleted from the current version of the FAQ.

The ceiling was still in place last summer, according to information provided to school officials at the time. And it was still there when Jon Gubera, the DOE’s chief accountability officer, emailed Bennett with the bad news that Christel House had earned a C. The school’s elementary-middle students earned 3.5 points for their math passing rate and got 1 point for growth, a total math sub-grade of 4.5 But Gubera capped the math score at 4.

Once the ceiling was lifted, however, Christel House had just enough points to meet Bennett’s expectation that the school get an A. Its students got 3 points for their English/language arts passing rate; add 4.5 points for math, divide by 2 and you get 3.75 points, enough for a low A.

Roach re-calculated grades for all the state’s elementary-middle and combined schools – both with and without the ceiling — and she found it made a big difference at the top. Primarily, it produced more than 140 additional As for elementary, middle and combined schools. It had no effect on schools that got a D or an F. (See summary)

One wonders what the rationale was for making such a significant change shortly before the 2012 grades were released last October. Is it conceivable that DOE staff set out to get Christel House an A, and others schools received collateral benefits?

Did Bennett explain the revision to the State Board of Education or other stakeholders? (It wasn’t addressed when the board discussed and approved the grades last Oct. 31). Did it need federal approval, since the school grading system was tied to Indiana’s No Child Left Behind Act waiver? Was it transparent enough for Secretary of Education Arne Duncan? Does it explain why the release of school grades was delayed several times?

The task force that Indiana legislative leaders created to review what happened with the A-to-F grading system should have plenty of questions to pursue.

About those 13 schools

The other change that raised Christel House’s grade was to base it exclusively on scores for students in grades 3-8, disregarding test scores for its 10th-grade students.

Bennett claimed the approach was intended not to help Christel House but to make the grading system fair for it and 12 other formerly elementary-middle schools that had added high-school-age students but weren’t yet able to get credit, in their high-school grades, for graduation rate or advanced courses.

Christel DeHaan, president and CEO of Christel House, echoed that claim in a statement last week. Some reporters understood Bennett to say that 13 schools had their grades changed or benefited as a result.

But that’s not true.

The 13 schools with the same grade alignment as Christel House are listed in an attachment to one of the emails between Bennett and his staff that the Associated Press unearthed. Only Christel House got a better grade as a result of dropping high-school students from the calculation.

And one school appears to have gotten a worse grade thanks to the change.

The other schools listed in the email are: Richmond Academy, Hammond Academy, Indianapolis Lighthouse Charter School, Hoosier Academy Virtual Charter School, Gary Lighthouse Charter School, West Gary Lighthouse, LEAD College Prep School, Anderson Preparatory Academy, Indianapolis Math & Science Academy, Indiana Connections Academy, International School of Columbus and Eman School.

By my calculations, confirmed by Roach, Hammond Academy would have had a D if high-school grades counted. Instead it got an F. Grades for all the other schools stayed the same, regardless of whether high-school scores were part of the mix.

Also, for what it’s worth, Christel House was the only one of the 13 schools that got a higher grade as a result of lifting the ceiling on math or English/language arts scores.

About these ads

12 thoughts on “Lifting ‘ceiling’ helped Christel House, other schools

  1. Some very interesting “findings” here.
    What does the actual A-F rule say about capping?
    Were there any positive changes for IPS schools that Cynthia Roach mentioned?
    Of the “140 schools” that benefitted were they mostly traditional public schools?
    For part two, you say “that’s not true,” yet you have no factual basis – you are still speculating.
    If you don’t run the actual models with the correct data, then you are speculating.

    • Good questions. I can’t tell that the ceiling was included in the formal rule approved by the state board. In other words, I don’t think the change violated state law. It seems to me it was a pretty major change (but quiet) in how DOE interpreted and executed the ruled.

      The 140 schools would include a mix of TPS and charters – I can’t think of a reason this approach would favor one type of school over the other. Probably some IPS, but they would be underrepresented because the change mainly affected schools that ended up with As.

      As for the 13 schools, I did run the calculations using test data for high-school grades that’s publicly available (mainly DOE COMPASS). It’s true that the data for accountability purposes are a bit different (e.g., only students enrolled for 162 days are counted; kids who passed Algebra I in previous years count). But from what I can figure, using the ACTUAL data would be much more likely to strengthen my conclusions than to undermine them. At the very least, five of the schools still ended up with Fs, so the suggestion that dropping the high-school scores “helped” 13 schools is clearly off base.

  2. What is his angle? Has he read Hammond’s appeal, or is he just going on what they say? The “reporting” on this is nothing less than amateur. Besides 162 day rule are there other reasons a student’s scores may not be used in accountability?

    Do you have evidence that the cap only “affected A schools?” What if it moved a D school to C? Does that make you a liar?

    Additionally, per your statement the Compass is a “bit” different than the accountability shows just how little you know. Take a look for example at the pass rate of George Washington school in IPS under ECAs and then under accountability and note the difference. Your comment that using the “ACTUAL” data strengthens your argument shows you know ABSOLUTELY nothing about statistics. Explain to me how that “strengthens your argument” using math. I won’t wait because you cannot do it. No one but you said it “helped 13 schools.”

    This really is amateur hour. It is painful to read analyses from people without analytical skills.

    Please stop citing Ann Hyslop as some kind of expert – she clearly has no idea about how the models work. Remember cowards throw grenades because they cannot stand to get into the real mix.

  3. Pingback: Detail still missing from Indiana grade-change story | School Matters

  4. Pingback: Records request leads to wait | School Matters

  5. Pingback: Public records quest meets dead end | School Matters

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s