O.K.  Here's a challenge:  rewrite the problem so that it meets your
standards, retaining as much of the original setting as possible, and
permitting the naive student to justify to him/herself the use of a
normal approximation in calculating the desired confidence interval.

(I'll be teaching at this level again this fall, and while the textbook
in use does not contain examples quite like this one, I'm willing to use
all the help I can get!)

On 13 Jul 2003, User968758 wrote:

> There were so many problems with the wording of this question that
> -- if this actually is an "homework" exercise (forget the
> assumptions about normalily etc)

Don't see how one can "forget about" such assumptions.  How else can one
justify using the standard Gaussian distribution in calculating a
confidence interval?
 [Or do you disdain the idea of "exercise" (see my earlier post on this
point) utterly, and argue that a student ought not to be required to
address a numerical example until he has learned enough theory to
encompass all possible peculiarities and exceptions that might apply to
real problems in the real world?]

> -- it is enough to make one despair about the quality of instruction
> of statistists at the introductory level -- an issue that is not
> new.

No, indeed.  So:  what have you to contribute to the improvement of that
quality?

Cheers!    -- Don.
 -----------------------------------------------------------------------
 Donald F. Burrill                                         [EMAIL PROTECTED]
 56 Sebbins Pond Drive, Bedford, NH 03110                 (603) 626-0816

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to