The sign test is used to compare two populations (A and B) by examining pairs of observations from each population. The number of times population A exceeds population B (X) is the test statistic. Under the null hypothesis, the two populations are identical and the probability of A exceeding B is 0.5. To perform the sign test: (1) assign "+" if A>B, "-" if A<B, discard if equal; (2) count remaining pairs (n) and times the less frequent sign occurs (r); (3) compare r to critical values - if r is below the critical value, reject the null hypothesis that the two populations are identical. The example tests if there is a

2. The sign test is used to determine if there is a
significant difference between the mean characteristics of two
populations (say A and B). Responses on each pair of A and B
are compared. The number of times A exceeded B is used as
the test statistic. It is denoted as the letter X. This is called the
sign test because X is the number of positive (or negative)
signs associated with the difference between the pairs in
population A and B.
In such cases, the implied null hypothesis is that the two
population distributions are identical or there is no significant
difference between the two population distributions A and B.
For any given pair, the probability that A exceeds B is p=0.5
when the null hypothesis is true.

3. The following are the steps in doing the sign test:
a. State the null hypothesis.
Ho : There is no significant difference between the two
populations being compared.
b. State the alternative hypothesis.
HA : There is a significant difference between the two populations
being compared.
c. Test the null hypothesis using the following procedure.
1. Examine each pair (A and B) of observations. If A>B, assign
a plus (+) sign, if A<B, assign a minus (-) sign, if A=B,
the pair must be discard.
2. Count the number of pairs remaining and denote it by n.
3. Count the number of times the less frequent sign occurs
and denote this by r.
d. To test the null hypothesis, compare r with the critical values tabulated
n the table for critical values of r for the sign test in the Appendix.
e. Make your decision. If r is less than or equal to the tabulated value for
the chosen significance level, reject the null hypothesis.

4. Example:
The response time, in seconds, for two
different stimuli (S1 and S2) were recorded for 10
subjects who participated in a psychological
experiment. The following data were obtained.
Subject
S1
S2
1
2
3
6.4 9.4 7.8
7.8 10.3 8.9
4
7.7
5.2
5
8.8
11.3
6
7
8
5.6 12.1 6.9
4.1 14.7 8.7
9
4.2
7.1
10
5.6
8.1
Use the sign test to test the null hypothesis that
state that there is no significant difference between
stimulus 1 and stimulus 2 in terms of mean
response to them by the subjects.

5. Solution:
Step 1. Ho: There is no significant difference between
stimulus 1 and stimulus 2 in terms of mean response to
them by the two samples or the two stimuli are identical.
Step 2. Ha: Stimulus 1 and stimulus 2 are not identical.
Step 3. Get the sign of the difference between stimulus
1 and stimulus 2, then find r or the number of times the
less frequent sign occurs.
Subject
D
1
2
3
6.4 9.4 7.8
7.8 10.3 8.9
-
4
5
6
7
8
7.7 8.8 5.6 12.1 6.9
5.2 11.3 4.1 14.7 8.7
+
+
-
9
4.2
7.1
-
10
5.6
8.1
-

6. Step 4. Compare r=2 to the critical value of r.
Step 5. Make the decision. Since r=2 is
greater than rcritical =1, Ho is not rejected.
There is no significant difference between the
two stimuli in terms of the reaction time of the
subjects on them.

7. Thank you for
listening! 