NJ Charter Closings: Where's the Logic?

I am just starting to look at the New Jersey charter schools that either had their charters renewed or will be forced to closed in this latest approval round. But already, I fail to see any logic in the NJDOE's decisions:

Oceanside opened in 1999, and Middleton has poured her heart and soul into the school. She acknowledged that test scores were not where they should be, but said she wished the state had given more consideration to the children she teaches; they come from some of the poorest sections of the city.

In its rejection letter to the school, the Department of Education said Oceanside’s state test scores were below that of the Atlantic City public school district. The data show Oceanside trailing the district by 2 percentage points in language arts and 1 percentage point in math.

Middleton prepared 2012 test data reports showing that while the school did not outperform the district overall, her students did score higher than their peers in the individual public schools in neighborhoods where her students live.

The data show Oceanside scoring better than five of the eight district schools in language arts and better than three schools — the Dr. Martin Luther King Jr. School Complex, Uptown Complex and New York Avenue — in math. The three public schools that consistently perform better than Oceanside are Chelsea Heights, Richmond Avenue and Sovereign Avenue, but Oceanside gets only 2 percent of its students from those neighborhoods, Middleton said.

“This is really not a valid comparison and has a negative impact on how Oceanside scores in the new comparative analysis,” she said. [emphasis mine]

Is this true? Do the students at Oceanside Charter School outperform their peers?

Let me, once again, steal a couple of graphs from Bruce Baker. The Rutgers professor has run regression analyses on charters based on their performance in statewide tests and the characteristics of their students (how many receive free lunch, gender, race, etc.). This allows him to make a prediction about how well the school should do on tests, and compare it to how well they actually do. (The numbers are from 2010-11.)

Here are the results in 8th Grade Math, with my annotations:

Oceanside is way over-performing in this test, given its high enrollment of Free Lunch eligible students (our best way of measuring a school's population of students in poverty). Notice how poorly Liberty Academy, another charter slated for closure, is doing in comparison.

Let's also consider some schools that had their charters approved. Here's Baker's analysis for charters in Newark on the same test, again with my annotations:

Marion Thomas and Maria Varisco-Rogers were approved, but their expansions were denied; Discovery was also approved. Again, where is the logic? These schools' relative performance is all over the place, even though the percentage of students in poverty they serve is relatively even.

Of course, we're looking at only one test here; maybe there is more variation in others. Still, you can understand the frustration of charter administrators and parents when they ask the NJDOE: "Where's the logic? Why do they get to stay open while we have to close?"

Unless there's more at work here than test scores...

Thursday, May 24, 2012 

Two boys, 9 and 11, are facing charges of unlawful possession of a weapon after bringing a handgun to Oceanside Charter School in Atlantic City on Thursday. 

They were released to the custody of their parents Thursday evening. Police are still investigating whether any adults will face charges as well. 

Atlantic City Police say one single shot was fired inside the school bathroom earlier in the day.

Maybe high test scores aren't enough to overcome some blots on a charter's record. More on the closures in a bit.