Paul Wilson reports "my belief that when you initially teach [service
statistics courses] to people you should first teach them away from the
computer, ... [because] if you sit down people at a computer from the
start they tend to see the object of the exercise as learning where to
click, and not learning statistics; whereas if you can get them to grasp
the basic [principles] of statistics the use of statistical software
comes easily."
My experience was always similar on the first point: teaching
statistical thinking and practical software manipulation in parallel
creates a mental conflict. I've also found many times that (a majority
of?) students forced to take "Stats 101" taught by an outsider see this
as an irrelevant distraction from their main interest. They do not
identify with the examples or the thinking. They see the point only when
they get their own data, so come sheepishly back for advice when doing
their projects six months later.
I'm less convinced on the last point, but can not identify why some
people take naturally to software use while others continue to struggle.
Part of it may be the need to accept that computers are normally
impersonal but logical and predictable, so do not make "sensible"
assumptions! As Paul mentions, coding and structuring data are
fundamental skills often overlooked.
Allan
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|