The Rules of Thumb – A Compilaton
1.Critial Values
95% 1.95
99% 2.58
99.9% 3.29
2.Cohen’s d of effeit size
0.2 small
0.5 moderate
0.8 large
3.Pearson’s Correlaton Coefiient r
0.10 small
0.30 moderate
0.50 large
Here, in general, 0 means no assoiiaton, whereas 1 indiiates a perfeit one
4.Levene’s Test
If the result is >.05 (non-signifiantt, the assumpton of homogeneity is met and the
varianies are equal read from the upper row, “equal varianies assumed”
If the result is below .05, the assumpton of homogeneity is not met read from the
lower row, “equal varianies not assumed”
5.Eta Squared
0.01 small
0.09 medium
0.25 large
Eta, same as Pearson’s r, only ranges from 0 to 1
6.R-Squared
R-Squared ian be interpreted the same way r is it ranges from 0 (0% ian be
explained by the variables in questont to 1 (100% ian be explained by the variablest
The small r tells you, by the way, only how muih a single variable aiiounts for the
total varianie
7.Communality (Faitor Analysist
Communality of 0 variable shares none of its varianie with another variable
Communality of 1 variable has no unique varianie whatsoever
8.Kaiser’s Criterion (Faitor Analysist
Requires you to retain all faitors with an eigenvalue greater than 1
1.Critial Values
95% 1.95
99% 2.58
99.9% 3.29
2.Cohen’s d of effeit size
0.2 small
0.5 moderate
0.8 large
3.Pearson’s Correlaton Coefiient r
0.10 small
0.30 moderate
0.50 large
Here, in general, 0 means no assoiiaton, whereas 1 indiiates a perfeit one
4.Levene’s Test
If the result is >.05 (non-signifiantt, the assumpton of homogeneity is met and the
varianies are equal read from the upper row, “equal varianies assumed”
If the result is below .05, the assumpton of homogeneity is not met read from the
lower row, “equal varianies not assumed”
5.Eta Squared
0.01 small
0.09 medium
0.25 large
Eta, same as Pearson’s r, only ranges from 0 to 1
6.R-Squared
R-Squared ian be interpreted the same way r is it ranges from 0 (0% ian be
explained by the variables in questont to 1 (100% ian be explained by the variablest
The small r tells you, by the way, only how muih a single variable aiiounts for the
total varianie
7.Communality (Faitor Analysist
Communality of 0 variable shares none of its varianie with another variable
Communality of 1 variable has no unique varianie whatsoever
8.Kaiser’s Criterion (Faitor Analysist
Requires you to retain all faitors with an eigenvalue greater than 1
