Letter: Somerville's Statistics Guru Takes Aim at 'Least Attractive School District' Designation
Somerville's SomerStats director fires back at an online analysis that determined Somerville was one the country's "least attractive school districts."
Daniel Hadley, director of SomerStat, the city's statistics department, took umbradge at an online study, posted by the real estate website Trulia, that determined Somerville was one of the "least attractive school districts" in the country.
In fact, Hadley writes in a letter to the editor, Somerville school enrollment numbers are rising. Furthermore, he called the study "an example of sloppy data analysis."
Here's his letter:
Our District is Desirable
When Trulia, the real estate website, recently published its list of “least attractive school districts,” many of us in Somerville were left questioning how we had earned such a dubious distinction. I appreciate the caveat Chris gave in his recent post covering the list: “’least attractive’ doesn't necessarily mean ‘worse.’ Rather, it's just an observation of population trends from the Census.” I would go a step further, however, to say that Trulia’s list doesn’t mean anything at all. It’s just an example of sloppy data analysis.
In a nutshell, Trulia examined census data and ranked each city by its ratio of schoolchildren to toddlers. Places that had a lot more toddlers than schoolchildren were labeled unattractive school districts, and the top ten were called “America’s least attractive.” It doesn’t take a data scientist to see why this label is misleading. For places on the list like Cambridge and Somerville, the ratio is seriously skewed by the number of young people they attract.
Somerville is an outlier, matched only by Cambridge, in terms of the number of students and young professionals that live within its borders. 25-34 year olds make up 31% of the population, the highest in the state. This means that Somerville is also an outlier in its ratio of toddlers to schoolchildren: 1.5/1
Like Trulia’s analysts, I also found a correlation between this ratio and standardized test scores in Massachusetts (R-squared = .4). Does this mean that young professionals are leaving the city for towns with better MCAS scores … or does it mean that they are leaving for larger homes and dragging down the scores as they go? Probably a bit of both.
As any longtime resident knows, families come and go for a variety of reasons. Cheaper homes, a new job, and even something as seemingly small as lead paint, can drive parents of young children to look elsewhere.
That said, one of the interesting things I found in the data is that Somerville families appear to be staying longer than you would expect from its demographics. Given its test scores, median housing value, and student population, I calculated that our toddler to schoolchildren ratio should actually be more like 1.7. The difference is small, but it suggests that many families grow attached to Somerville and stay despite the challenges.
My finding is confirmed by the school enrollment numbers, which are projected to rise significantly over the coming years. We think that the trend of “family flight,” if you want to call it that, is reversing in Somerville. We recently made a large investment in our schools, and we continue to work with groups like Progress Together to ensure their success. Ultimately, I still think the question to ask is not “why are so many families leaving,” but, “why are so many young people coming?” It’s not just our proximity to Harvard, MIT and Tufts. Somerville is really cool.
Daniel Hadley, Director of SomerStat