The "ordered" hypotheses method was employed to avoid correction for the multiplicity of alpha error when multiple comparisons are performed [ 20].
Multiple comparisons are covered as are nested designs.
The adjusted p values of multiple comparisons are depicted in Table 4.
The results of Student's t test, corrected for multiple comparisons, are shown: *p < 0.05, **p < 0.001.
R functions for dealing with global tests as well as multiple comparisons are included.
First, it is outlined how the hypothesis of interest and nature of the factor of interest (ordinal or nominal) governs how many multiple comparisons are made and whether multiple comparisons are needed at all.
The significant differences (P < 0.05) for multiple comparisons are shown with different letters separately for each panel, whereas no letter or same letters indicate insignificant variation.
When multiple comparisons are made, the true alpha value for significance is lower than 0.05, which is the value assumed in the LSD test (Dielman [2005]).
The Bonferroni method, the method of Holm, and two 'False Discovery Rate'-controlling methods for adjusting P-values for multiple comparisons are compared.
Using the results of statistical simulations, the relative merits of the different methods that are available for making multiple comparisons are then discussed.
Moreover, because the data is analyzed simultaneously, multiple comparisons are avoided.
