Test
Use
Distribution
Degrees of Freedom
Chi square test
Testing for a
relationship between
two categorical
variables
(nominal/ordinal)
Chi square
(rows – 1)x(columns – 1)
One sample z-test for
proportions
Testing to see if a
population proportion
is different from a
standard value
Standard
normal
distribution
N/A
Two sample z-test for
proportions
Testing for a difference
between proportions
from two populations
Standard
normal
distribution
N/A
One sample t-test for
mean
Testing to see if a
population mean is
different from a
standard value
t
n-1
Two sample t-test for
mean
Testing to see if a
population mean is
different between two
populations.
Measurements from
distinct populations
t
n
1
+n
2
-2
Paired t-test
Pre/post testing for a
difference in means –
same subjects in
pre/post data
t
n-1
ANOVA
Testing to see if there is
a difference between
more than two
populations
F
Numerator df = # of groups – 1
Denominator df = N - # of
groups
Pearson/spearman
correlations
Test to see if correlation
is significantly different
from 0
t
n-2
