EXPERIMENT 9: Implementing T-Test

AIM of the Experiment: The aim of the Experiment is to implement T-test in SPSS

Objectives:
1. To understand the concept of t-test in Statistics.
2. Train students how to apply t test and its Steps.
3. Student will understand why and on what kind of data t-test test will be applied in
Statistics.

Input/Apparatus Used:
Statistical Package for Social Science (SPSS) Software by IBM

Outcomes:
1. Students will learn the use of t test s in Statistics.
2. Student will able to perform the t test of Statistics.
3. Students learn how to Analysis and Interprets the Result with help of t test.

Introduction and Procedure

1. One Sample t test :

One sample t-test has only one sample test whether sample mean is equal to specified value
or not
The one sample t test determines whether the sample means is statistically different form a
known or hypothized population mean. It is a parametric test

Procedure
Step 1. Enter the data in spss worksheet
Step 2 .Click on Analyze then click on compare mean
Step 3 . Click on sample on one sample t test
You will get one sample t test dialog box
Step 4 . select the variable that you want to test and drag them to test variable box
Step 5 . Click on option in confidence interval procedure box use / input confidence interval
default

Step6 .Execute cases analysis by analysis

Step7. Click on continue
Step 8. Click ok and interpret you output
2. Independent Sample T- test

The Independent Sample t-test also called the two sample t-test. It is an inferential
statistical test that determines whether there is a statistically significant difference between
the means in two unrelated groups. The Independent Sample t-test is a parametric test.

Procedure
Step 1 : Enter the data in worksheet that you want to test.
Step 2 : Click on “Analyze” then click on compare mean.

Step 3 : Click on Independent Sample t-test.

You’ll get a dialogue box.
Step 4 : Select the variable that you want to test and drag them to test variable test.
Step 5 : In grouping variable box drag the variable you divided into groups.

Step 6 : Click “Define group” you will get a window having Group 1 and Group 2. Enter the
value that you have specified in value. For this example (1 for section A, 2 for section B)
Step 7 : Click on continue. If you want to change confidence interval percentage then click
on option. And change it.
Step 8 : Click on continue and then click on OK. Then interpret your calculated result.

NOTE : For making interpretation.

If the value of p (i.e. Sig. (2-tailed) > 0.05 ) , then Null hypothesis is accepted.
Otherwise, rejected.

.
3. Paired Sample T-test

The paired sample t-test compares two means that are from the same individual object or
related units. The two means typically represents two different times (i.e. before
experiment and after experiment) or two different but related conditions or units.
The purpose of the test is to determine whether there is statistical evidence that the mean
difference paired observation on a particular outcomes is significantly different from zero.

D̄−µ
t= S
√n
where,
D̄ = sample mean
µ = population mean
S = standard deviation
n = sample size
PROCEDURE
Step 1 : Enter the data in SPSS worksheet that you want to test.

Step 2 : Click on “Analyze”. Then click on compare mean.

Step 3 : Click on paired – sample t test.
You will get paired – sample t test.
NOTE : As you have specified before test and after test variable.
Step 4 : In paired variables ; drag 1 st variable (i.e. before test) in variable 1 and another
variable (i.e. after test) in variable 2.

NOTE : Pair column represents the numbers of paired sample t tests to run. You may choose
to run multiple paired sample t tests simultaneously by selecting multiple sets of matched
variables.
Step 5 : Click on Option and change confidence interval percentage if required.
Step 6 : Click on Continue and then click on OK. Interpret the results.

Discussion:

(i) T-test is used to compare the means of two groups. Discuss

Question: Viva Voce

(i) What are the various types of t-test?

(ii) What is the difference between paired sample t-test and independent sample t-
test?

Single sample t-test:

Whether the average mark of the class is 60??

One tail or two tail t-test?

Two tail: Whether the average marks of the class is 60 or not??
One tail: Whether the average mark of the class is more than 60??
One tail: Whether the average mark of the class is less than 60??

Null Hypothesis:
H0: Average marks of the class is 60.
 Alternative
H1: Average marks of the class in not 60

One-Sample Test
Test Value = 60
t df Sig. (2- Mean 95% Confidence Interval
tailed) Difference of the Difference
Lower Upper
Mark 1.074 9 .311 6.90000 -7.6337 21.4337
s
Significant value=??
Significant value = .311
.311 > 0.05 Therefore my hypothesis is accepted.
I Conclude that yes average marks of the class is 60

One-Sample Statistics
N Mean Std. Std. Error
Deviation Mean
Marks 10 66.9000 20.31666 6.42469

Paired sample??
Null hypothesis: There is no difference in the average marks in MST-1 and MST-2

Paired Samples Test

Paired Differences t df Sig. (2-

tailed)
Mean Std. Std. Error 95% Confidence
Deviation Mean Interval of the
Difference

Lower Upper

.0000 8.41955 2.66250 -6.02299 6.02299 .000 9 1.000

Pair MST1_Marks - 0
1 MST2_Marks

Significant value = 1.00

Since significant value i.e. 1.00 is more than 0.05 therefore my hypothesis is accepted
It means there is no significant difference in the performance of the students in MST-1 and
MST-2
 Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

MST1_Marks 66.9000 10 20.31666 6.42469

Pair 1
MST2_Marks 66.9000 10 17.87891 5.65381

Paired Samples Test

Paired Differences t df Sig. (2-

tailed)

Mean Std. Std. Error 95% Confidence Interval of

Deviation Mean the Difference

Lower Upper

MST1_Marks - 8.70000 11.88884 3.75958 .19524 17.20476 2.314 9 .046

Pair 1
MST2_Marks

Since significant value i.e. 0.046 < 0.05 so Our hypothesis is rejected

There is a significant difference in the average marks of the students in MST-1 and MST-2

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

MST1_Marks 66.9000 10 20.31666 6.42469

Pair 1
MST2_Marks 58.2000 10 17.38965 5.49909

In MST-2 the performance is poor as compare to MST-1

Average score of MST-1 66.900

Average score of MST-2 is 58.200