Oh, thank you Kate, for your latest post, from which all this flows. It’s reassuring that the luminaries, the veterans of this biz, are facing some of the same demons I am. It’s also profoundly scary to know that they will not go away (the demons, not the luminaries). At any rate, musing on her post made me think about the various ways that we are all tacking against strong prevailing winds.
And isn’t that a noble calling: to buck the system, to stick it to The Man? Sure is, but it only counts if what we do actually works. And that’s why good assessment is such a sterilizing light; as the punk said, we gots to know.
So, to the nitty-gritty. I teach “regular” stats; my colleague teaches “honors” stats. He sent me some problems he was giving to his kids. Here’s one:
A market research company employs a large number of typists to enter data into a computer. The time taken for new typists to learn the computer system is known to have a normal distribution with a mean of 90 minutes and a standard deviation of 18 minutes. The proportion of new typists that take more than two hours to learn the computer system is
If you are appalled, let me say that it is not his fault: if you were an AP Stat teacher, you would recognize this sort of thing as preparation for the AP exam. The Normal distribution is Chapter 2 in the book, and the exam has questions like this.
But if you are not appalled, let me try to convince you at least to be uncomfortable: Who are they kidding? Do you really think that some company knows this mean and standard deviation? How do they assess when a typist “knows” the new computer system? Do they really have enough data to establish whether the distribution is Normal? And why, in this day and age, does this company use an army of typists for data entry? [see note at end of post]
Doesn’t that argument sound good? Yeah! All power to realistic contexts! Promote authenticity! Stick it to The Man!
But here’s what really went on in my head when I read that problem:
“OMG, I’m so far behind.”