716 / 762
716
Chapter 16: One-Way Analysis of Variance
16-3 Terminology Associated with Analysis of Variance
(ANOVA)
In this section, you will be introduced to some terminology frequently used
in analysis of variance.
In
Example 16-1
, if we assume that these are four samples from four
potential populations of freshmen, sophomores, juniors, and seniors from the
campus, then the statistical methods which will be used to compare the four
population means is called
Analysis of Variance
. These methods are often
referred to by the acronym:
ANOVA
. The techniques of analysis of
variance will enable us to determine whether any observed differences
among the sample means are due to sampling errors or whether there are
indeed significant differences between the population means.
Definition: ANOVA
ANOVA is a group of statistical methods which will help us to determine
the “existence” of any significant differences among several population
means.
Suppose a researcher would like to determine the effectiveness of three
different drugs of equal dosage on lower back pain. The structure which
will be used to collect the data and test the statement or the hypothesis of the
problem is called the
experimental design
.
Definition: Experimental Design
The term experimental design in ANOVA is simply a structure which will
be used to collect the data and to test the statement of the problem (usually
hypotheses) to be solved.
Generally, experimental designs are reliable and efficient procedures within
a structure for planning experiments so that data can be collected and
analyzed to yield valid and objective conclusions about the statement of the
problem.