4 ms·
I had a very interesting work-study job in my first year in college, teaching dyslexic students who were struggling with basic algebra - essentially, walking th
by photochemsyn 2y ago
I had a very interesting work-study job in my first year in college, teaching dyslexic students who were struggling with basic algebra - essentially, walking through their homework assignments with them. Success meant figuring out some way of presenting the problem that worked for them, typically with some kind of physical example. I remember one student in particular could simply not grasp the notion that feeding x into a function f(x) would generate a point y on a two-dimensional graph, not even for a simple straight line, let alone an exponential or sine curve. Eventually a machine analogy worked: here's a vegetable slicer, you throw a number into it, it spits out another number - reproducibly. To describe this machine, we write an algebraic equation.
However, this would have been absolutely impossible in a large class - these students needed individually tailored one-on-one attention for at least an hour each to get the ideas across to them. Some parents can afford to hire such private tutors, some can't.
Also, people seem to have natural affinities for different areas of mathematics, just like people naturally take to different sports. I have no idea of whether genetics plays a role, although clearly in physical sports body types matter - the long-distance runner is unlikely to also be a champion weightlifter, etc. I've also noticed that some people who claim to be 'bad at math' often have good mathematical skills that they don't even think of as math, e.g. combinatorics as applied to playing games like cards or Scrabble.
In the end, mathematics is a hard subject and students who want to master any aspect of it will need at least three hours of study on their own (or assisted) for each hour spent in class listening to an instructor. It's not for everyone, and that's okay.