On yet another challenge from Eduwonkette, this time on my post about the great job KIPP does with boys. I described how KIPP succeeds the old fashioned way, by ratcheting up the literacy instruction and refusing to let either boys or girls slip behind.

Eduwonkette suggests the data is skewed because more boys than girls drop out of KIPP. She raises a good question, and given the disastrous outcomes black boys are experiencing in inner-city schools (be SURE to watch this HBO documentary tonight on a Baltimore high school) I’d be shocked if there weren’t serious gender imbalances at KIPP schools.

According to KIPP, however, that’s not the case. Says KIPP spokesman Steve Mancini: There was no mass migration of female students at either the KIPP Newark or Bronx school school between 2004 and 2007. In fact, the percentage of female students at TEAM Academy held steady at 54 percent in October 2004 and December 2007, and also remained at 57 percent at KIPP Academy for both dates.

 A longer analysis by a KIPP practicioner follows below: 

Dear eduwonkette,
I am the founder of TEAM Academy and the Executive Director of TEAM Schools, the network of KIPP schools in Newark, NJ. It is rare that I have time to respond to articles or blog entries, but I feel that this is an important topic and would like to set the record straight on the subject of gender imbalances in attrition at TEAM Schools. To put it simply, boys do NOT leave TEAM Schools at a higher rate than girls.
I appreciate your acknowledgement that your conclusion - that TEAM Academy loses more boys than girls - is weakened by your lack of access to cohort data. I’m writing today to supplement the data you do have with some that you wouldn’t have been able to find on the state’s website, in the hopes of painting a fuller and more accurate picture of our attrition patterns.
As you mentioned, we have multiple schools in Newark under one charter. TEAM Academy is 6 years old and serves grades 5-8, while Rise Academy is only 2 years old and serves grades 5-6. For the sake of accuracy I will use numbers that disaggregate the TEAM students from those at Rise, and since Rise will not serve 8th grade for two more years, I’ll leave them out of this analysis. If you’re interested, overall attrition at Rise Academy has been under 4% for both years of their existence, and no gender patterns have emerged yet within this rather small sample.
The 5th and 8th grade enrollment data you used in your comparison were from last year’s 8th graders at TEAM and 5th graders at both Rise and TEAM. There are two problems with using these data that lead to a flawed and invalid conclusion:
1. These data compare students in different classes at different schools. You compared 5th graders at Rise and TEAM with 8th graders at TEAM only.
2. You compared data of two completely different groups of kids - last year’s fifth graders and last year’s 8th graders. As such, you do not account for fluctuations in the number of students from each gender who enter the school through our lottery each year. In other words, sometimes we have more boys in the lottery, and sometimes we have more girls.
Though you do acknowledge that the impurity of these data sets are problematic, you somehow nonetheless use these flawed data to come to an equally faulty conclusion: “These data don’t allow us to trace one cohort through school, but they do suggest that more girls are sticking with KIPP than boys.” I’m afraid that you underestimate how much of an apples-to-oranges comparison this actually represents.
To illustrate this point, consider the following data, from our current 5th and 8th grade classes at TEAM Academy (the ones you described were last year’s 5th and 8th grade classes):
Current 5th grade class: 48% boys, 52% girls
Current 8th grade class: 53% boys, 47% girls
Were we only to look at these numbers in performing the same analysis you performed, we would conclude that TEAM Academy actually does a BETTER job of holding onto its boys than its girls. Of course, this conclusion suffers from the same weakness from which your opposite conclusion suffers: neither describes the actual attrition of a single group of students over time.
To determine if more boys than girls left the school, one would need to track the same group of students who enter the school at the beginning of 5th grade and leave after 8th grade. We performed this analysis for TEAM Academy, and found that the facts lead to a dramatically different conclusion than the one you put forth.
Let’s take a look at the attrition data from a single cohort over time. For simplicity’s sake, I will refer to each cohort of students by the name we give them when they are in our schools, which is determined by the year most of them will go to college. The current class of 8th graders, then, is referred to as “2012″, the 7th graders are “2013″, and so on.
Tracking the class of 2012 from 5th grade to 8th grade, our data reveal that we started with 81 students, with a few more boys than girls (53% boys, 47% girls). Of those 81 students, 21 of them left over the four years, for a four-year attrition rate of about 26%, or an average annual attrition rate of about 6.5%. Of the 21 who left, 10 were girls and 11 were boys. Since the class of 2012 started with more boys than girls, this leaves us with attrition rates that are almost exactly equal between the two sexes.
Since the current class of 7th graders have been in our school for only three years, the data on their attrition are not as informative, but I’ll mention them anyway in case you’re interested. Our current 7th grade class, the class of 2013, has lost 12 students; 7 boys and 5 girls. These numbers are probably too small and too heavily impacted by random circumstance to be very meaningful (the majority of our students who leave do so because they are moving out of state), but we will certainly keep our eyes on 2013’s attrition patterns to try to make sure this trend doesn’t expand over the next few years.
So from the data we have, TEAM Academy does not in fact seem to be losing boys at a faster rate than we are losing girls. Though we’ll continue to monitor the relative retention of our boys and girls, we don’t currently see a gender imbalance.
Your article also mentions the overall gender imbalance of a couple of our classes. The ones you mention in particular are two of the most imbalanced classes we’ve ever had. Our overall numbers are as follows: 55% girls, 45% boys. Frankly, we don’t know why some classes have more boys and some have more girls, and we also don’t know why overall we have more girls than boys. As I’ve shown above, it’s not that we’re losing more boys than girls, but rather that more girls are signing up for our lottery. Since our lottery is blind, the only way we could equalize the numbers is through gender-targeted recruitment, which is fairly difficult but worth considering. We are currently trying to get to the root of our lottery pool’s imbalance in order to determine how to address it.
Again, thank you for drawing attention to this important topic. It is critical that all schools seek to serve all of their students equally well, and your vigilance on behalf of boys is appreciated. I appreciate this opportunity to set the record straight. If you would like to learn more about our schools, I would be happy to have you come out to Newark and visit at any time.
Thank you,
Ryan Hill

Executive Director

TEAM Schools, a network of KIPP schools

Tags: kipp

This entry was posted on Monday, June 23rd, 2008 at 10:31 am. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.