This document provides information about performing a paired t-test, including the steps and an example. A paired t-test is used to compare two related samples, such as pre-test and post-test scores from the same group. The steps are: formulate hypotheses, determine if the test is one-tailed or two-tailed, calculate the degrees of freedom, select the t-test, compute the test statistic, state the decision rule based on critical value, and make a conclusion. An example tests if a new diet significantly reduces weight by comparing weights before and after for 7 women.
3. Statistical Tests
T TEST OF
INDEPENDENT
MEANS
*Compare scores of
2 different variables
but for two different
groups of cases
T TEST OF
DEPENDENT
MEANS
(PAIRED T-
TEST)
*Compare scores of
2 different variables
but for the same
group of cases
4. Statistical Tests
T TEST OF
INDEPENDENT
MEANS
Example:
Comparing Math Anxiety
scores of BS Math
Female and Male
Students
T TEST OF
DEPENDENT
MEANS
(PAIRED T-
TEST)
Example:
Comparing Science
Anxiety scores of BSEd
Math Students before
and after joining a
seminar about anxiety
5. T test of Dependent
Means
When to use this test?
In some research designs, it is necessary to
measure a certain variable of the same
people more than once
Commonly used to test improvements (or
decrements) of performance over a period of
time
Use to show differences in the two (2)
measures coming from the same group
6. T test of Dependent
Means
If we want to find out if the
difference between paired
observations is significant
or not, use this test.
7.
1
n
D
D
n
D
t
2
2
D = sum of the differences between
the paired observations
2
D
= sum of the squared difference
n = number of paired observation
T test of Dependent
Means
8. An investigator is interested in
assessing the effect of a film on the
attitude of students toward a certain
issue. A sample of five students is
administered an attitude scale before
and after viewing the film.
The before and after scores
are as follows:
Sample Problem 1
9. Sample
Before
Viewing
After
Viewing
Student 1 16 20
Student 2 7 27
Student 3 19 32
Student 4 11 33
Student 5 8 14
At 0.05 level of significance did the film
affect the attitude of the students?
10. T test of Dependent
Means
Step 1: Formulate the
Hypotheses
The film does not significantly affect the
attitude of the students.
The film significantly affects the attitude
of the students.
11. T test of Dependent
Means
Two-tailed (non-directional)
Step 3: Determine whether two-
tailed or one-tailed.
12. T test of Dependent
Means
Step 4: Compute the degrees
of freedom and determine the
critical value/tabular value
13.
14. T table comes
in different
format but the
process is still
the same.
15. T test of Dependent
Means
Step 5: Select the appropriate
statistical test and compute
for test statistic.
For this, the statistical test to be used is T test of
Dependent Means.
1
n
D
D
n
D
t
2
2
16. 65
D
Sample
Before
Viewing
After
Viewing
D D2
1 16 20 -4 16
2 7 27 -20 400
3 19 32 -13 169
4 11 33 -22 484
5 8 14 -6 36
Sample
Before
Viewing
After
Viewing
D D2
1 16 20
2 7 27
3 19 32
4 11 33
5 8 14
Sample
Before
Viewing
After
Viewing
1 16 20
2 7 27
3 19 32
4 11 33
5 8 14
1105
D2
T test of Dependent
Means
18. T test of Dependent
Means
Step 6: State the decision
rule.
19. T test of Dependent
Means
Step 6: State the decision
rule.
CRITICAL VALUE
20. T test of Dependent
Means
Step 6: State the decision
rule.
3.605
21. T test of Dependent
Means
Step 7: Make a conclusion.
22. It is claimed that a new diet will
reduce a person’s weight by 10
pounds on the average in a period of
2 weeks. The weights of 7 women
who followed this diet were recorded
before and after a 2-week period.
Sample Problem 1
24. T test of Dependent
Means
Step 1: Formulate the
Hypotheses
The new diet did not significantly reduce a
person’s weight.
The new diet significantly reduced a
person’s weight.
25. T test of Dependent
Means
One-tailed (directional)
Step 3: Determine whether two-
tailed or one-tailed.
26. T test of Dependent
Means
Step 4: Compute the degrees
of freedom and determine the
critical value/tabular value
27.
28. T test of Dependent
Means
Step 5: Select the appropriate
statistical test and compute
for test statistic.
For this, the statistical test to be used is T test of
Dependent Means.
1
n
D
D
n
D
t
2
2
29. 57
D
639
D2
Woman X Y D D2
1 129 130 -1 1
2 133 121 12 144
3 136 128 8 64
4 152 137 15 225
5 141 129 12 144
6 138 132 6 36
7 125 120 5 25
Woman X Y D D2
1 129 130
2 133 121
3 136 128
4 152 137
5 141 129
6 138 132
7 125 120
Woman X Y
1 129 130
2 133 121
3 136 128
4 152 137
5 141 129
6 138 132
7 125 120
T test of Dependent
Means
31. T test of Dependent
Means
Step 6: State the decision
rule.
32. T test of Dependent
Means
Step 6: State the decision
rule.
CRITICAL VALUE
33. T test of Dependent
Means
Step 6: State the decision
rule.
3.99
34. T test of Dependent
Means
Step 7: Make a conclusion.
35.
36. A training director of a large chain of
department stores in Makati wants to
know if a weeks training course will help
12 salesladies’ rapport with the
customers. One month prior to the
training and one month after training
there were rated in a 10 point scale. At
0.05 level of significance is there a
reason to believe that the training has
improved the rapport of salesladies with
the customers.
37. Sample Before the
Training
After the
Training
1 6.9 7.0
2 6.9 7.1
3 7.2 8.3
4 6.8 7.0
5 6.5 6.6
6 6.0 6.0
7 5.3 5.6
8 7.0 7.7
9 6.6 7.0
10 6.8 6.8
11 6.3 6.3
12 5.2 5.3
Editor's Notes
T TEST OF INDEPENDENT MEANS is also known as independent samples t test or t test of independent samples
-compares means of two independent groups (totally different groups not related to one another)
T TEST OF DEPENDENT MEANS is also known as paired samples t test, paired t test, dependent samples t test, t test of dependent samples, or repeated measures design
-compares means (scores) of two related samples (same group of samples)
NOTE: In rare cases, this test can be used also with two different groups of samples having the SAME attribute that you need to observe and relevant in your study)
T test
The main task here is to determine whether the film affects the attitude of the students. We can say that the film affects the attitude of students if there is a significant difference between the two scores (before and after viewing the film)—either the effect is positive (significant increase on the AFTER scores) or negative (significant decrease on the AFTER scores)
Thus, this problem can be solved using paired t test.
𝑯_𝟎 means null hypothesis. The scores of the two groups have no significant difference
𝑯_𝒂 means alternative hypothesis. The scores of the two groups have significant difference. (two tailed)
The level of significance is the probability of rejecting the null hypothesis. A 0.05 level of significance means that there is a 5% chance (risk) of concluding that difference between the two groups exist.
*NOTE: The level of significance is given in the problem. 𝜶=𝟎.𝟎𝟓
Two-tailed or nondirectional since the problem only asks if the film affects the attitude. Suppose the question is “Does the film positively affect the attitude of the student?” then it becomes one-tailed (one-directional).
𝑯_𝟎 means null hypothesis. The weight of the two groups have no significant difference.
𝑯_𝒂 means alternative hypothesis. The weight of the second group (after the diet) is significantly lower than the first group (before the new diet). (this is one-tailed)
The level of significance is the probability of rejecting the null hypothesis. A 0.05 level of significance means that there is a 5% chance (risk) of concluding that difference between the two groups exist.
*NOTE: The level of significance is given in the problem. 𝜶=𝟎.𝟎𝟓
One-tailed or directional since the problem only if the claim is true. The claim is “The new diet will reduce a person’s weight . Thus, the second group is expected to be lower than the first group. (One-tailed).