StatCrunch Creating a Contingency Table from Summary Data Video Lecture

In this video, you will learn
to create contingency tables
from summary data
using StatCrunch.
The dataset I am using
is called Pairwise Counts
for Categorical Variables.
This consists of a contingency
table loaded directly
into the StatCrunch data table.
The table shows six
values of A, that
are paired with
four values of C,
and two values of D.
The four B values are
paired with two values
of C and two values of D.
To analyze the summary
data, under the Stat menu,
choose Tables,
Contingency, With Summary.
We need to select both
the C and the D columns.
To do so, hold down Command
on a Mac, or Control on a PC.
In Row Labels, choose the var1
column, and for Column Label,
type in var2.
I'll now Click Compute.
The resulting contingency
table shows the original table
as well as the row-wise
totals, the column-wise totals,
and the total number of pairs.
The output also shows the result
of a standard chi-squared test
for independence.
Since the cell counts
are low, in this case,
a warning message is displayed
for the test results.
I will discuss a better
alternative for this data later
in the video.
StatCrunch also allows
additional information
to be displayed in
the contingency table.
To view the different options,
under the Options menu,
choose Edit.
This will take us back
to the original window
where we built our
contingency table.
In this window, StatCrunch
has other options
for the user to customize
their contingency table.
Under the Display options, you
will see the different options
that can be added.
For this example, I will choose
row percent and click Compute.
Now, the resulting table shows
that 2/3, 66.6% of the A values
are paired with the C
values, and 1/3, 33.3%
are paired with the D values.
Table shows a 50/50 split
between the C and D pairings
for the B values.
As mentioned prior,
the chi-square test
is not appropriate for this data
due to the small cell counts.
StatCrunch offers
a number of tests
that can be computed from
the contingency table output.
To view these,
back under Options,
there is a box labeled
Hypothesis Tests.
Note, some of these
tests can only
be restricted to a
two-by-two table.
They are labeled accordingly.
For this example, I'm going to
de-select the chi-square test
for independence and
choose the Fisher's
exact test for independence.
Also, notice further
down in the window,
StatCrunch can also calculate
confidence intervals.
For this example, I'm not
going to select the confidence
interval option.
Now, Click Compute, and
the resulting output
shows the new results for
the Fisher's exact test.