[EMAIL PROTECTED] (Dennis Roberts) wrote in message
news:<[EMAIL PROTECTED]>...
> At 08:23 AM 7/23/02 -0700, Gordon Kenyon wrote:
>
> > I think I make it clear later in my post. What I am describing is
> >the density function presented to look really scary (with the number
> >"e" and pi and all), then hustled away never to spoken of again, its
> >role fulfilled by "table C". As a stand-alone density function it
> >isn't much use- taking the integral converts it into the distribution
> >function. Not that any of this is even hinted at in introductory
> >social/behavioral science stats texts.
>
>
> are you suggesting that the instructor should somehow .... talk about or
> discuss, even at a low level ... "... taking the integral converts it into
> the distribution function ... " ?
>
> how many students ... in intro kinds of classes ... would even know what an
> "integral" is?
I guess what I advocate is at least a brief introduction to the
concept of an integral. Gonick and Smith do an excellent job of this
in "The Cartoon Guide to Statistics" by the way. I truly believe the
intro student will be better served in the long run by some inkling of
the fact that continuous variables are not defined intuitively.
I am not an instructor and am certainly not dealing with the
practical issues you discuss- the students who are taking the intro
course are there to fulfill a requirement. Furthermore, the integral
that began the discussion cannot be solved by the methods one
generally picks up in the first couple of years of undergrad calculus.
Taking up too much time confusing and causing panic among one's
sophomore charges trying to explain the details of normal
distributions is not so much the point as legitimately preparing them
to go further if they should ever choose to do so.
The point I make that holds water is that incomplete, inaccurate
information about complex topics will prevent some people from
continuing to pursue those topics. Not that I think anyone around
here is speaking in defense of incomplete, inaccurate information
about complex topics.
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
