An Open Letter to Bill Gates: Why Not Measure This?
Yesterday I posted an essay that pointed out that Bill Gates apparently uses a different set of outcomes in choosing a school for his own children than the measurable ones his foundation advocates for the children of the less fortunate. I shared the thoughts of mathematician Cathy O'Neil, who points out
...the person who defines the model defines success, and by obscuring this power behind a data collection process and incrementally improved model results, it seems somehow sanitized and objective when it's not.
Don't be fooled by the mathematical imprimatur: behind every model and every data set is a political process that chose that data and built that model and defined success for that model.
As I pointed out yesterday, there are many outcomes that we value that are difficult to measure, leading to the first order of bias here, towards outcomes that are easily quantified. But there is a second order of bias at work here; deliberate choices that are made to define the parameters of the system within which we measure.
Bill Gates wrote in his letter:
Given a goal, you decide on what key variable you need to change to achieve it-the same way a business picks objectives for inside the company like customer satisfaction-and develop a plan for change and a way of measuring the change. You use the measurement as feedback to make adjustments. I think a lot of efforts fail because they don't focus on the right measure or they don't invest enough in doing it accurately.
The Gates Foundation has decided that the variable we can most readily change is the effectiveness of the classroom teacher. Therefore all their powers of measurement have focused on this single variable. We have explored in the past some of the problems with this. In particular, the fact that less than 15% of the differences in student growth can be attributed to their teacher suggests that perhaps we ought to be looking in the realm of the out of school factors, which have been found to account for more than 60% of these differences. Today I want to explore some of the areas that have remained unexamined.
What are some other things affecting our students that we should be measuring?
Here are some other things even more directly related to our children's performance in school and success in life beyond that could be measured, and improved. The first two are suggestions offered in this report: Ten Ways to Promote Educational Achievement and Attainment Beyond the Classroom:
Measure the number of Unplanned Pregnancies:
The report tells us:
Research conducted in Europe has found that children and teens born as a result of unwanted pregnancies tend to receive poorer grades and ratings from teachers in early childhood and adolescence.
We could address this through effective sex education and contraception. Also,
Intensive early childhood programs that promote child well-being have also been effective at lowering unintended pregnancy and birth rates among program participants more than a decade later, when they have entered the teen years. Similarly, programs geared towards the health and well-being of low- income first-time mothers have been found to be effective at preventing subsequent unintended pregnanciesMeasure the availability of Prenatal Care:
We also learn that:
Research has found that maternal health in both the prenatal and early childhood stages has substantial effects on early childhood academic and cognitive outcomes. Mothers who smoke cigarettes, drink alcohol in excess, or use drugs during pregnancy are at risk of having children who display relatively poor cognitive abilities in math and English during the elementary school years. Mothers who have insufficient prenatal care may also be at risk of having children with poor early childhood cognitive and language skills.
Measure the level of funding available to each school district.
As this article reveals, per pupil funding can vary significantly, even within a single state or region. And often the neediest students are the ones that receive the least support.
Measure the level of access to college.
The Gates Foundation has made it clear that, in their view, the primary purpose of a K12 education is to prepare students for college. However, in the past 30 years, the cost of attending a four year school has increased at four times the rate of inflation.
College debt recently passed the trillion dollar mark, eclipsing the total amount Americans owe on credit cards. This has prompted some to even suggest there may be a sort of bubble forming regarding the value of a college education. This puts college out of reach for many students, especially those lacking family wealth to back them up. It will do impoverished students little good to prepare them for college if they do not have the resources to stay and graduate, as this New York Times story illustrates.
Measure the levels of childhood poverty, neighborhood violence, homelessness, and family members incarcerated.
All four of these factors have huge impacts on children's lives, directly affecting their future prospects.
Measure economic and racial segregation.
Research shows that underprivileged students do best when not highly concentrated. The greatest advances in student achievement occurred during the years when the US expanded desegregation programs. Unfortunately this is now being reversed -- and some advocates of Charter schools are even suggesting this is a good thing.
Ken Bernstein suggests we measure exposure to lead:
Test all low-income children EARLY for lead - Michael Martin of the Arizona School Board Association has been harping on lead for years, and recently Kevin Drum has written on it. While in theory public housing "remediated" lead-based paint, in many cases it was just painted over. We know children will ingest peeling paint, and if that paint contains lead, the effect on learning downstream are quite severe.
In fact, we could do this nation a great deal of good by offering free inspection of all residences for sources of lead, and low-cost or free methods of remediating it.
Lisa Guernsey suggests we measure access to pre-school and libraries:
Measure whether a community and school district have provided children with access to preschool. Measure how many children have the chance to attend a full-day (versus a half-day) of kindergarten. Measure how much parents are paying for afterschool care and if siblings are being held responsible for that afterschool care instead. Measure whether libraries are partnering with school districts and vice versa. Measure how many parents have easy access to online tools and affordable broadband to be able to follow what is happening in their school districts and enable their children to do their homework.
When I challenged the Gates Foundation to address some of these things directly in our dialogue last summer, here was their representative's response:
What we can't do, however, is address all of the problems that put or keep families in poverty. We just don't have the resources to do that. But we are part of a community of donors who are committed to eliminating the causes of poverty.
I will accept that the Gates Foundation may not be able to address all of these problems directly. But according to Bill Gates' annual letter, the sheer act of measuring things and setting goals around them has tremendous power. I believe his school reform project is failing in large part because of the error he identified. It has been measuring the wrong things. If the Gates Foundation is unwilling to tackle the scourge of poverty directly, could it at least begin to actively measure and set goals for some of the things identified here? Any one of the things listed would be far more likely to be successful than the Foundation's misguided efforts to increase the capacity of teachers to raise test scores.
What do you think? What should we be measuring in order to improve education?
Continue the dialogue with me on Twitter at @AnthonyCody