When people read, hear, or prepare research summaries,
they sometimes have misconceptions about what is or isn't "sound
practice" regarding the collection, analysis, and interpretation
of data. Here are some of these common (and dangerous) misconceptions
associated with the content of Chapter 11.
The statistical focus of a one-way ANOVA is the variance
of each comparison group.
Since the null hypothesis for a one-way ANOVA always
stipulates that the population means are equal, researchers should feel
no obligation to refer to this null hypothesis in their research summaries.
If the Bonferroni adjustment technique is used because
separate one-way ANOVAs are being conducted on each of several dependent
variables, the same "reduced" alpha level should be used for
each and every one-way ANOVA.
Researchers should never waste time checking on the
underlying assumption of normality is tenable.
The only thing that influences the size of the calculated
F-value in a one-way ANOVA is the degree to which the group means differ.
If the sample data conform to the homogeneity of
variance assumption, there will not be much variability among the scores
within any of the comparison groups.
Of a one-way ANOVA is conducted with alpha set at
.05, there is a 5 percent chance that a Type I Error will be made.
The assumption of independence is not a very important
assumption compared to the others.