I agree with the argument that that the Star Tribune Editorial Board made against the Higher Learning Commission’s decision to require teachers who teach college-level courses in high school to have master’s degrees or graduate credits in the subjects they teach (“Don’t make it harder to earn college credits,” Sept. 17). But some of the evidence that the Editorial Board used to make that argument is deeply flawed.

The editorial cited a Minnesota Department of Education report that found that the average test scores and graduation rates of students who take college-level courses are higher than the average test scores and graduation rates of students overall. This is a prime example of what is called “selection bias.” The vast majority of high school students who take college-level courses do so either because they think they have the ability to succeed in those courses or because someone else thinks they have that ability. As such, arguing that taking college-level courses is the only or even the primary cause of higher test scores and graduation rates is almost certainly wrong.

I don’t mean to suggest that we shouldn’t encourage high school students to take college-level classes. In fact, it is because I support that policy that I am taking this opportunity to highlight the danger of drawing conclusions based on data that suffer from obvious selection bias. In other instances, such flawed data is used to advance policies and practices that more rigorous experimental research does not support and that I very much oppose. So the next time you encounter someone citing data to advocate for a course of action in education, ask yourself if selection bias may be at work and if credit is being given or blame is being cast where it may not, in fact, be due.

Kent Pekel, St. Paul

The writer is president and CEO of the Search Institute, a nonprofit organization that conducts applied research in education and youth development.