[MUD-Dev] Weaknesses in the HCDS player type model (was: 3rd Axis for Bartle's 2 axis theory of MUD players)

[Hans-Henrik Staerfeldt]

On Tuesday 22 October 2002 15:37, Ola Fosheim Grøstad wrote:

> Yes, this is a possibility, but there are some other problems than
> the ones you specify. First of all, graded scales are
> unreliable. You and I may feel the same way, yet you might give a
> 4 where I give a 2.

I agree, this is usually a problem, but on the other hand, I would
assume that this would mean that if you might give a -4 i would give
a -2. (its a bith of a stretch, but probably it would be a close
approximation).

> Still, one could probably develop a questionaire which largely
> consists of picking alternatives. Second, how do you scale the
> different dimensions, or is that of no importance in principal
> component analysis? It sounds like interesting approach, though I
> probably would have made lots of scatterplots and tried to see
> where that would get me first.

In the method each question is actually treated as its entirely own
dimention. The method then tries to make the best (linear)
multidimentional projection of a lower dimentionality (say we want 3
dimentions) that gives the best 'spread' of the points in the
projection (a point would then be a single persons answers in all
the dimentions/questions).

> *grin* Well, who says "other similar studies" are valid?

Those who did them, usually :-D

--
Hans Henrik Stærfeldt   |    bombman at diku.dk    | work:  hhs at cbs.dtu.dk      |
Address:                |___  +45 40383492    __|__       +45 45252471     __|
DTU, Kemitorvet,        | Scientific programmer at Center for Biological     |
bygn 208, CBS.          |  Sequence Analysis, Technical University of Denmark|


_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev