I've long been fascinated by what does and doesn't work in classrooms. I'm a graduate student in mathematics and I end up teaching elementary calculus courses t

I've long been fascinated by what does and doesn't work in classrooms. I'm a graduate student in mathematics and I end up teaching elementary calculus courses to undergrads. This also means that I tend to think about what does and doesn't work in college classrooms. Two things stuck out to me in this article. Firstly, Hattie says "Holding students back a grade really does hold students back, with an effect of -0.16." What does this actually mean? Does this mean that at graduation, these students perform 1/6 of a standard deviation below their peers? I'd be more interested if we were also presented with the alternative: how do students who didn't meet whatever minimal standards for grade advancement but were advanced anyway (and thus could have been held back a year, but were not) perform? As an early college math instructor, I'm very familiar with students who are incapable of performing mathematical operations from classes which they've previously passed. The more I think about it, the more I wonder about what it means to perform at some number of standard deviations above or below others. Is this based on standardized test performance? The second thing I wonder is related. As an alternative to the common policy of "Achievement Standards", he suggests "The alternative: a focus on growth and progress for each student, no matter where he or she starts." How might one implement this? For instance, suppose Albert and Beth are entering basic 6th grade math. Beth performs really well and for whatever reason Albert is very distracted and learns almost nothing. The next year rolls around, and Albert and Beth are entering basic 7th grade math. The teacher cannot teach them the exact same, as Albert doesn't understand prerequisite ideas. Is the teacher to instruct normally and then devote additional time to Albert, trying to incrementally increase his understanding? Does Hattie actually advocate Montessori-style schools with self-paced work? Perhaps schools should have a lowest-tier math class which is entirely based on individual students instead of central instruction? Or in this case, should Albert have been kept back a year? In many other ways, I like and/or understand Hattie's approach and ideas.