Several statistic tests suppose the NORMALITY and/or EQUALITY OF VARIANCES of
the studied characteristic on the POPULATION (s). In general, we work on the
random sample with limited size.
Questions:
1- Is it necessary to test the normality EVEN THOUGH the distribution is NORMAL
in the population (morphological characteristics: height, weight... for
example)?
In this condition, If the statistical test reject the normality (with a sample
size >30 for example), what we decide ???
The sample size may help us to take a decision ?
2- Exactly the same questions for 'Equality of variances'.
Best regards,
Hassane.
|=======================================================|
| Hassane ABIDI (PhD) |
| Unite d'Epidemiologie; Centre Hospitalier Lyon-Sud |
| Pavillon 1.M; 69495 Pierre Benite Cedex, France |
| Tel: (33) 04 78 86 56 87 ; Fax: (33) 04 78 86 33 31 |
|=======================================================|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|