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 15.
The analysis of covariance was designed for situations
where the comparison groups are "intact groups."
The analysis of covariance is robust to the assumption
of equal regression slopes so long as the sample sizes are equal.
Data on the covariate variable and data on the dependent
variable must come from using the same measuring instruments at two
different points in time.
The more covariates the better.
Studies involving ANCOVA are inherently better that
studies involving ANOVA.
If the pretest means of a study's comparison groups
are compared and found not to be significantly different, the analysis
of covariance would have no advantage over an analysis of variance in
comparing the groups' posttest means.
In an analysis of covariance, the df values are computed
in the same manner as they are in an analysis of variance.
Because an analysis of covariance has more power
than an analysis of variance (presuming that the ANCOVA's covariates
are "good"), there's no need to worry about Type II errors.
When multiple covariates are used within the same
ANCOVA, it's good to have high correlations among the covariates and
between each of them and the dependent variable.