[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
At 10:51 am +0200 23/6/99, VIOLLON Stephanie wrote:
>I have conducted an experiment where there was a dependent variable (S)
>and two independent variables (A and V). I have carried out an analysis
>of variance, which showed a significant main effect of both variables (A
>and V) on the dependent variable (S), and a significant interaction
>between these both variables (A x V).
>I'd like to assess whether the effects of the variables A and V were
>very strong (to assess the "strength" of these effects).
>I was adviced to use a statistical variable that I do not know at all.
>This is: "omega square" for estimating the degree of association between
>independent and dependent variables. I was told that most statistical
>packages use a simpler formula (for instance, in SPSS, there is "eta").
>The problem is that my stats package is Statistica and I can't find
>"omega square" neither "eta".
>Is there someone who has yet calculated this variable (omega square?
>eta?) and who knows how calculating it with Statistica?
>Thank you in advance.
I know SPSS but not Statistica - I hope this helps.
Reply if it doesn't or if you need further clarification.
Eta squared is the effect size of a treatment (or independent variable, IV)
in an ANOVA. The simplest way to work this out is the proportion of the
overall sum of squares which can be attributed to the treatment (IV) and
gives you the proportion (or percentage) of overall variance which is
attributable to that given treatment.
How to work it out:
So, you simply divide the Sum of Squares (SS) for an independent variable
(e.g. A) by the total SS to get a ratio (or percentage) this would tell you
how much overall variance in the scores (the DV of S) can be explained as
being due to the differences manipulated by the IV of A.
For your purposes, get the SS for A, the SS for V and the SS total from the
F-table output of the ANOVA (I presume Statistica gives you this as output
for an ANOVA). And then simply compare SSA/SStotal and SSV/SStotal to find
which one has a greater effect on the DV S.
Eta squared is often called the correlation ratio (in the context of the
ANOVA) in some statistical software packages this may be the case in
Statistica (although I doubt it as it is confusing to most people).
Eta squared is really a simplified version of the ratio of variances of
effects called Wilk's Lambda. It is simplified in the case of a univariate
ANOVA (i.e. when you have only one DV). It may be that Statistica uses
Wilk's Lambda for effects in univariate ANOVAs: it just so happens that for
a univariate ANOVA, Wilk's Lambda = 1- eta squared. In multivariate
ANOVA's (i.e. more than one DV) the 1 in this equation is another ratio of
variance. (If you need an explanation this other ratio of variance in
Wilk's Lambda I can give it to you).
Kevin L. Baker .
Senior Lecturer in Psychology mobile 0778 852 0323
Dept of Human Communication office 0116 257 7761
De Montfort University fax 0116 257 7008
Leicester LE7 9SU, U.K. http://www.dmu.ac.uk/~klb