The recently implemented "star rating system" for public schools in Maryland is under attack for being too easy. Multiple factors determine a school's rating, but so far most of the state's schools (60 percent) are doing well — earning at least four stars or better on a five-star scale. Some have questioned why most schools rate this high and have pressured the state Board of Education to change the weightings of various factors in order to lower the average school rating. A better question not being asked is: What actionable insights about education in our schools do these ratings provide?
The star rating system is a furtherance of the pseudoscientific approach to education that has become so prevalent in the operation and management of public schools. Massive amounts of data on educational outcomes are collected, processed, analyzed, charted and reported, but no causal processes are identified nor predictive models constructed on which to base effective educational interventions. All this busy work is done in the name of "accountability," but the system is designed to ensure that no one will be held accountable because no fundamental changes in educational practices will take place. Instead numbers will be manipulated, weighting factors debated, and the sheer volume of data dumped on teachers and parents will bury any meaningful insights that might be obtained.
For many years now, Maryland public school websites have made available massive data-dumps on all manner of school metrics. For example, currently there are links to 175 school and program profiles on the website for the Baltimore City Public Schools (www.baltimorecityschools.org/schools). Each school profile has a link to its state report card, with its star rating, and links to the school's data profile, budget, parent survey, school performance plan, and effectiveness review. These documents are filled with colorful tables and charts showing data on teacher qualifications, enrollments, student demographics, suspensions, climate surveys, PARCC assessment results, learning goals, strategies, results indicators and so on. However, I cannot find any explanation of how all this data would fit into a valid predictive model for improving education outcomes.
Instead I find precise numbers with vague annotations. For example, the school performance plan for the four-star school Abbottson Elementary includes the goal to increase literacy proficiency, as measured by the PARCC assessment, from 10 percent of students in 2017-18 to 25 percent of students in 2018-19. There are three "evidence-based" strategies listed for achieving this goal — collaborative planning, strategic academic interventions, and ongoing analysis of Common Core Standards at grade level and across grade levels.
Multiply these kinds of suspect statistics, plus trendy jargon for each goal, by four goals per performance plan, by 175 plans and it's easy to see why public education is in peril. The star rating system adds more obfuscation and busy work, while failing to address root causes of educational failures. This approach epitomizes pseudoscience, in that it promotes a false narrative camouflaged with meaningless jargon, data and statistics to give an appearance of scientific validity. The falsehood is that measuring learning outcomes, articulating quantifiable goals and demanding "accountability" results in better educational outcomes.