Yes, you "should" have understood this.  Along with all the rest of us.....
:)  And we often don't.

Seriously, this is a subtlety that often gets past many of us.

You need to ask what exactly does your datum (individual item) tell you?  A
categorical data point simply says the item is one of a category.  An
ordinal category means that the category is less than some, more than others
in the total group (dataset, if you will).  when you recode to a 'dummy'
variable, you are forcing the ordinal data into an interval measurement.
Now each 'point' is spaced apart from each other one by a set, and equal,
amount.

Is a 'fancy' apple exactly one notch higher than an 'A' apple, and that one
notch higher than a 'good' apple, and all of them above a 'cider' apple:?
Yes, they are spaced apart in that order, but are they equally spaced?
Well, I can change the ordinal categories into dummy variables, 1, 2, 3 ,and
4.  Will that make them equally spaced?  Well, no, it just makes them look
like it.  The added information is not 'real.'

So, using dummy variables is possible, certainly.  Just don't let them fool
you into thinking your data tells you more than it did at the start, OK?

Cheers,
Jay

Robert Lundqvist wrote:

> I found in one of the textbooks we use that calculating correlation
> coefficients is not meaningful when you have categorical data. However,
> using dummy variables should be possible, shouldn't it? Either when you
> have one ordinary numerica variable and one dummy, or even when you have
> two dummy variables. If not, could someone please put me in the right
> direction so I can stop be so hesitating in class...Comments are welcome,
> even if it turns out that I should have understood this.
>
> --robert
> .
> .
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> .                  http://jse.stat.ncsu.edu/                    .
> =================================================================

--
Jay Warner
Principal Scientist
Warner Consulting, Inc.
4444 North Green Bay Road
Racine, WI 53404-1216
USA

Ph: (262) 634-9100
FAX: (262) 681-1133
email: [EMAIL PROTECTED]
web: http://www.a2q.com

The A2Q Method (tm) -- What do you want to improve today?




.
.
=================================================================
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