Data and Decision Research Paper :

T – Test Data Set – Identify the Variable with the Data Set

Submitted by

K.U.PARINATHA

MBA

Section – G

Department of Management

PES University
 ABSTRACT :

The subject of this Research Paper is T-Test Hypothesis Testing.

It is one of the most widely used hypothesis testing techniques of
statistics. There are various classifications of the T-test hypothesis,
this includes the T-test for single mean- It helps to compare a sample
with a known population. The other most prominently used T-test
hypothesis testing techniques are – T-test for Independent mean and
The Paired T-test for sample mean.
In this research paper, we will be discussing in detail about the
Independent T-Test Hypothesis Testing Method of a known
population.
 INTRODUCTION:

The T-test Hypothesis is a statistical tool that is used to ascertain the

means of one or two datasets by using hypothesis testing and to find
out if there is a significant difference between the means of one or
two datasets . The student's T-test or T-test hypothesis , was originally
introduced by William Sealy Gosset in 1908. Later on, this tool
transformed data analysis and was widely utilized for testing
hypothesis in a variety of domains, such as business psychology,
economics, biology, and more.

 The T-test was developed because of the requirement to analyze

the quality of small samples for industrial quality control. Gosset
created it so that it may be used in cases with a restricted sample
size where the population standard deviation is unknown.

 The T-test functions on the basis of certain presumptions, which

are as follows :

• The normality of the data within each dataset should be

approximately normally distributed.

• The sample size of the T-test should be 30 or less.

 • The observations are chosen at random and independent of
one another.

• The two parameters used in a T-test hypothesis are the

sample means and standard deviations (or standard error).

• The variance in the different populations should be almost

equal.

 There are different Types of T-test Hypothesis, this includes the

following:
1). One – Sample T-test :
Using this , we facilitate a comparison between a sample
mean and its poplution.

Formula :-
t = ( x̄ - μ) / (s / √n – 1 )

Where,
1). T is the T-Statistics
2). x̄ is the sample mean
3). μ is the population mean
4). s is the standard deviation of the sample
5). n is the sample size
6). n – 1 is the degree of freedom
  T-test for Difference of Means :-

Using this , we facilitate a comparison between two

independent sample mean drawn from two normal population.
Here , population variance being equal but unknown.
x− y
t=
Formula :-
(
√ S 2 11 + 12
n n )
Where ,
1). x,y are the two mean values
2). n is the sample size
3). n1 +n 2−2 is the degree of freedom
2 2
n1 s 1 +n2 s 2
And , 2
s=
n1 +n2 −2

 Paired T-test for difference of means :-

Using this , we facilitate a comparison between the means of the

paired population. It helps to figure out if the two sets of the
population are significantly different from each other.

d
Formula :- t = s / √ n

( ( ))
2
Σd 2 1 2 ∑d
d=
n and s=
n−1
∑ d −¿
n
Where ,
1). d = Average of difference of the two values
2). S = Standard Deviation
3). D = x – y is the degree of freedom
 DATA ANALYSIS AND INTERPRETATION

This section presents the statistical analysis of the data used in this research.
The data was collected by conducting a survey among the MBA students in
Section G at PES University.

This data consists of the GPA’s of 26 students who have participated in the
survey and the number of hours spent on social media platforms. As this data
will help us understand the impact of social media on students' academic
performance. I have chosen this subject for my research.

A comparison between the students is done to see whether there is a significant

difference between the GPA's for the grades and the social media usage of the
students. The independent sample test is used to see this significance. The data
is given in Table 1.

Table 1. The GPA's and the social media usage of the students, are as
follows:
 Table 2. The Descriptive analysis of the GPA’s of the students
according to the social media usage is given below:
t-Test: Two-Sample Assuming Unequal Variances
Calculation High Social Media Users ( 2.5hr Low social media users ( 1hr -
- 4.5hr) 2hr)
Mean 8.020769231 8.567769231
Variance 0.660124359 0.660124359
Observations 13 13
Hypothesized Mean 0
Difference
GPA of GPA of
Df 24
High Social Media Usage ( 2.5hr - Low Social media usage ( 1hr -
t Stat -1.701532781
4.5hr) 2hr)
P(T<=t) one-tail 0.050882385
8 9.31
t Critical one-tail 1.71088208
7 7
P(T<=t) two-tail 0.101764769
8 9.5
t Critical two-tail 2.063898562
8.49 8.5
7.1 8.5
9.51 9.5
9.1 9.101
8.5 8.5
7.37 8.27
8 7.2
8.7 8
7 9.5
7.5 8.5
 Figure 1:- The percentage of students according to GPA, are as
follows :

Figure 2:- The percentage of the usage of various social media

platforms by the students, are as follows :-

Figure 3 :The Descriptive analysis of the GPA’s of the students

according to the social media usage is given below
 High Social Media Users ( 2.5hr - 4.5hr)
10
9
8
7
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9 10 11 12 13

Low social media users ( 1hr - 2hr)

10
9
8
7
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9 10 11 12 13

Interference :-

Table 3: Two sample T-test between the Grades of the social media
usage of the students :-

GPA :- N Mean Variance t-value p-value

High 13 8.020769231 0.660124359 - 1.701532781 0.1017647
usage of 69
Social
media
Less 13 8.567769231 0.660124359
usage of
social
media
Two sample T-test between the Grades of the social media usage of
the students :-

1) The null hypothesis is that there is no significant difference

between the two grades of the two groups of students. i.e.,
H 0=¿u =u ¿
1 2

2) The alternate hypothesis is that H 1 : u1 ≠ u2

x− y
t=
3) Formula :-
(
√ S 2 11 + 12
n n )
2 2
n1 s 1 +n2 s 2
4) Where , 2
s=
n1 +n2 −2

5) n1 +n 2−2 is the degree of freedom

6) Therefore , df = 24 (refer to figure 1)

7) The critical value is a Two – Tailed Test with df = 24 and α =

0.05 is 2.064

8) P – Value > Two – tailed T-value (0.05 > 0.10)

9) The T-value = 0.10 lies in the Acceptance region.

 T-value 0.10 (acceptance region)

Conclusion :-

Therefore , we come to a conclusion that –We fail to reject the

null hypothesis ( H 0) , as there is no significant difference in the
academic performance between the two groups.

T - Distribution Table
Literature Review
1. The T-test hypothesis is one of the most implemented statistical
hypothesis method. William Gosset , Karl Pearson and R Fisher
had formulated this technique in the early 1800s. It was later
improvised by them to become two-samle test which assesses the
independent samples drawn from two known populations. The
two-sample T-test is often used to ascertain whether one group or
dataset is better than the other.
2. There are two different types of independent sample means tests,
they are : non-parametric Wilcoxon-Mann-Whitney test and the
parametric Welch t-test. In the nonparametric, the empirical
likelihood testing method increases the Type 1 error in small
sample sizes.
3. The most important classification of the T-Test Hypothesis is The
Independent Samples t-Test and the Paired t-Test. The test for
difference of means is used to test the hypothesis of two
independent random samples which is drawn from two datasets.
Here , the population variance is equal but unknown.
4. A statistic T-test ascertains if two means are significantly different
from one another. While assessing the means may stipulate a
deviation, it does not provide any particular information about its
reliability. Comparing the means does not reveal any deviation.
5. The procedure of drawing conclusions and assumptions about a
dataset under the assessment involves the use of hypothesis testing.
 The basic parameters of statistical hypothesis are – the sample
mean and standard deviation, they act as a basis for the result.

6. Comparisons between the outcomes of the investigation and

supervision is a common occurrence in clinical research. The most
popularly used Statistical T-Test Hypothesis test is the test for
comparing the sample mean values between the two samples are
Paired T-test and The Independent sample test.
7. Social Media is a web based platform used by people to
communicate across nations and even countries. Social means
Human society , as a result of communities and word media is the
medium of expression. Social Media generates a large number of
innovations and is one of the greatest inventions of technology,
which is now widely used and preferred.
8. The use of Social media is rapidly increasing among the young
generation of the world. Students of all age groups are regularly
engaging themselves in Social Media, thus affecting the personal
and academic lives of the students. Thus , various studies are being
conducted to determine the effects of social media of the academic
performance of the students.
9. A T-test hypothesis is a statistical test, that is most widely used to
compare the sample averages of two populations. There are two
 most important types of statistical inference : parametric and
nonparametric methods. The T-test hypothesis is widely used in
various sectors of the ecomony.
10. The terms “significance level” or “level of significance” refer to
the likelihood that the random sample you choose (for example,
test scores) is not a representation of the population. The lower the
significance level,the more confident you can be in analyzing your
results.
 REFERENCE :

1. Stephanie D. Wilkerson, Mathematics „Application of the Paired t-test'

XULAneXUS: Xavier University of Louisiana‟s Undergraduate Research
Journal. Scholarly Note. Vol. 5, No. 1, April 2008

2. Ahmad, A. (2011, Feburary). Rising of Social Network Websites in India

Overview. IJCSNS International Journal of Computer Science and Network
Security, 11, 155-158

3. Carlos Alberto Bavastri, Euda Mara da S Ferreira, Jose Joao de Espindola,

Eduardo Marcio de O Lopes (2008), “Modeling of Dynamic Rotors with
Flexible Bearings due to the use of Viscoelastic Materials”, Journal of the
Brazil, Soc. of Mech. Sci. & Eng., Vol. 3, No. 1, pp. 23-29
4. Albert, J. (1997). Teaching Bayes’ rule: A data-oriented approach. The
American Statistician 51(3), pp. 247–253.

5. Espindola J J, Silva Neto J M and Lopes A (2005), “Generaliz Shabaneh N H

and Jean W (1999), “Dynamic Analysis of Rotor-Shaft Systems with
Viscoelastically Supported Bearings”, Mechanism and Machine Theory,
University of Toronto, Canada, Vol. 35, No. 1, pp. 1313-1330.

6. Ted Fractional Derivative Approach to Viscoelastic Material Properties

Measurement”, Applied Mathematics and Computation, Vol. 164, No. 2, pp.
493-506.

7. World Health Organization. The world health report 2001 – Mental Health:
New Understanding, New Hope; 2001 [Retrieved 2008-10-19].

8. Sharma Suresh K. Nursing research and statistics. 1st edition 2011,

a. Elsevier publisher, printed at Harayana, page number: 70-73

9. J.C.F. de Winter Delft University of Technology 'Using the Student‟s t -test

with extremely small sample sizes' Practical

10.Assessment, Research & Evaluation. Volume 18, Number 10,

August 2013, ISSN 1531-7714.
 Methodology

 Data Sampling –
I have conducted a survey on the subject – The impact
of The impact of social media on the academic performance of
student. The data was collected by conducting a survey among the
MBA students in Section G at PES University. I have taken a random
sample of 26 students.

 Data Source :-
I have collected data from a population of 26 students. By
conducting the survey and sending a survey form to MBA students
of Section G. My subject for the research was -The Impact of
Social Media on the Academic Performance of the Students. I was
able to collect primary data and interpret by using the Independent
T-test hypothesis technique to find the interference and form the
analysis.
 Sample Size :- 26 Students among the MBA students in Section G
at PES University
 Tools Used :- The tool used to calculate the T-test hypothesis for
Independent test is MS Excel.