Free Lunch | Geographical Perspectives

Math TCAP Performance Highly Correlated with % Free-Reduced Lunch

In my initial post on the topic of TCAP scores I pointed out that Pueblo District 60 Schools (D60) have more students who qualify for Free Lunch than the total number of students in all of Pueblo County District 70 (D70). I also said D60 has a more difficult task not only due to the fact that they are a much larger school district overall but also because D60 students come from households with far lower incomes, on average.

Some readers may not be convinced that lower economic status translates into lower test scores. Let’s see if I can convince those folks that there is indeed a strong relationship between household income and academic performance.

Rather than focusing on Pueblo Schools I want to look at a bigger picture to better understand the relationship between test performance and economic status on a State-wide basis. So these data displayed in the charts below cover all over Colorado’s public schools. Many schools have too few test takers to be considered statistically significant so those small schools were eliminated from the data set. I also limited data selection to 3^rd grade, 7^th grade and 10^th grade. These grade levels are the middle grade (or close to it) for typical elementary, middle and high schools, so I think it’s an appropriate subset to explore. If I look at every grade level I worry the quantity of data displayed starts to become too much to digest. In addition, I’m only looking at Math scores and, for now, omitting Reading and Writing scores for two reasons. First, again, it’s difficult to look at too much data all at once and might lead to confusion. Second, in my opinion, Math scores tend to minimize cultural bias and, as such, provide a more accurate reflection of actual academic capability.

On each of the charts below the horizontal (x) axis represents the percentage of students who qualify for Free or Reduced Lunch programs. The vertical (y) axis represents the Math TCAP test score results. Each point on the chart represents one school’s Math TCAP scores and their corresponding % of students who are eligible for Free or Reduced Lunch. I’ve also included a trend line with the associated regression equation and R-Squared value (included in the upper right corner of each chart). Don’t worry if the equations make you nervous – these details are only there for reference.

As the charts clearly illustrate, or at least they clearly illustrate this to me, there is a very strong relationship between economic status and TCAP Math performance. As % Free-Reduced Lunch goes up, school TCAP Math performance goes down. Given the strength and consistency of this relationship, School District evaluation efforts and media coverage of local school performance results should focus on school performance within this context.

Furthermore, instead of blaming public school systems for failing our children perhaps we should consider blaming socioeconomic policy for producing too many students who are too hungry to succeed in school.

Read the Previous Post in this Series | Read the Next Post in this Series

Let’s Compare Apples to Apples When Evaluating Pueblo’s Public Schools

In recent editorials the Pueblo Chieftain has been singing the praises of Pueblo District 70 Schools (D70) and, in parallel both implicitly and explicitly, criticizing Pueblo District 60 (D60) for failure to achieve similar TCAP testing results. I needed a good set of data for my Statistics classroom so I decided to dig in to the TCAP data hoping to learn more about TCAP scores in general while exploring how well D70 was actually performing relative to D60.

The TCAP data are like most data sets produced by large government agencies, i.e., a big mess and far more difficult to interpret than necessary. But, I’ve seen a lot worse.

I downloaded the 2014 Math, Writing and Reading scores to see what I could learn. Boy, there’s a lot of data to sift through. Probably someone could do a doctoral dissertation analyzing the Colorado TCAP data. I don’t have time for that…at least not at the moment. So, for now, let’s look at the difference in economic circumstances between the two school districts.

The most readily available indicator of economic status by school and by district are the % Free and Reduced Lunch figures. Conveniently, the Colorado Dept of Education breaks down the testing results into 3 categories: (1) students qualifying for free lunch, (2) students qualifying for reduced lunch and (3) students who do not qualify for free or reduced lunch.

Here’s a table summarizing what I found:

Anything stand out to you?

What jumps off the page at me is that D60 educates more children receiving Free Lunch than the total number of students in D70. Somehow, I missed that in the Chieftain’s reporting.

Let’s say you’re a teacher and the principal asks you to choose between 2 classrooms, both with 30 students. The principal reminds you that the fate of your job, the school and the entire school district depends on you getting your students to succeed on a set of culturally biased standardized tests. As a professional educator you know that academic success is closely correlated with economic status. In classroom A you have 19 kids receiving free lunch and in classroom B you have only 9 kids receiving free lunch. Which classroom would you choose?

If we’re going to hand out trophies to D70 and castigate D60 let’s be sure to provide appropriate context. D60 has, by far, the more difficult task. To read the Chieftain you might mistakenly assume both districts are in the same boat.

I plan to dig further on this. Next post will likely explore the relationship between TCAP scores and % Free-Reduced Lunch eligibility. So stay tuned and, in the meantime, don’t be so quick to throw D60 under the big yellow bus.

Read the Next Post in this Series