Chemistry Reference
In-Depth Information
Table 6.1
Can you identify the key components of this dataset?
0
0
0
6.5
5.0
5.5
6.0
6.5
7.0
6.5
0
1
1
5.5
5.0
7.0
6.5
6.5
7.5
6.0
0
0
2
4.5
5.5
6.0
6.0
5.0
7.0
7.0
1
1
0
5.5
5.0
5.5
5.5
7.0
7.0
5.0
1
0
1
6.0
6.0
6.5
6.0
6.5
6.0
7.5
1
1
2
5.5
5.5
6.0
7.5
6.0
6.0
8.0
1
0
0
6.0
6.0
6.5
6.0
4.0
4.5
8.0
1
1
1
6.0
7.0
7.0
7.0
6.0
7.0
6.0
1
0
2
6.0
6.5
5.0
6.5
6.0
4.5
7.0
0
1
0
6.0
7.5
6.0
7.5
7.5
7.0
6.5
0
0
1
6.0
6.0
5.0
6.0
6.0
7.0
7.5
0
1
2
6.0
7.0
7.0
5.0
7.5
7.0
6.0
Some of the details of experimental design and answers to these questions will
be revealed at the end of the chapter.
Common Statistical Techniques
There are many types of statistical tests. Choosing the right test to analyze our data
is critical. These selections should be done in consultation with the major professor,
advisory committee, and a statistician. If we use statistics extensively or are using
statistical techniques unique to our chosen topic, we should have a statistician on
our advisory committee. Generally, we choose the simplest test available that will
give a meaningful answer to the research questions posed. Three of the most com-
mon techniques include the following:
A
t-test
is conducted when comparing two treatments. It calls for a very simple
experiment, but if we want to know which of the two major colas or peanut but-
ters is best, incorporation of any other treatments will just add clutter.
ANOVA
(
Analysis of Variance
) is one of the most frequently used statistical tech-
niques. It is ideally suited for
multifactor
experiments and uses the variability in
the data to determine the
p-value
. It is particularly effective at drawing conclu-
sions about
interaction effects
, but these effects might obscure main effects of a
specific factor. For example, I designed several complex experiments on the
changes in quality of fresh vegetables from the field to the consumer. As we were
analyzing our data, one of my collaborators would always want to know the
main
effect
of temperature of handling and storage without the interfering
interaction
effects
of time and harvesting factors.
Our choice of experimental design is directly linked to the selection of our
statistical technique. Examples of sample design include
Latin-square
,
random-
ized block
, or
split-plot designs
. The power of our analysis and credibility of our
conclusions can be affected by our selection of the (in)appropriate design. There
