1 |
On Wednesday 09 August 2006 07:53, Eilverijus Kondratas wrote: |
2 |
> ) I don't think that this calculation can be used for future |
3 |
> planings: " what system will be better". Statisticaly we should |
4 |
> apply "z" or atleast "t" statistics instead of simple average. |
5 |
> |
6 |
> Generaly speaking, calculations given in an article are the simplest |
7 |
> ones tought in primary school. I did not find anything from |
8 |
> advanced statistics according to which the rating could be applied. |
9 |
|
10 |
So, perhaps we should request the data-set from the authors and apply some |
11 |
more (re)gressive statistics to the problem. |
12 |
|
13 |
Regards, |
14 |
|
15 |
- Brian |
16 |
-- |
17 |
gentoo-security@g.o mailing list |