Vous êtes sur la page 1sur 22

Statistics presentation

group b2
USE OF STATISTICAL METHODS TO 2

PROVE HYPOTHESIS
MEMBERS:
 ATUL ANAND (19PGPM081)
 CYNTHIA MAJHI (19PGPM084)
 ISHANT(19PGPM093)
 KUSUM (19PGPM097)
 PRASUN KUMAR (19PGPM107)
 PUNEET SHUKLA (19PGPM112)

MENTOR: Dr. Amrita Sengupta


3

INTRODUCTION

WHAT IS HYPOTHESIS TESTING ?


• a statistical inference method
• a technique used to compare two datasets, or a sample from a
dataset
• We infer something about the characteristics of what we're
comparing
4

HYPOTHESIS
TESTING
1. What kind of test should you use?
a. Mean
b. Median
c. Mode
d. Standard deviation
Already used these?

what shape does your data take?


6

Bell Curve – the normal distribution


a probability distribution that is
symmetric about the mean. • most values are centered
around zero.
• those sides of the curve are
moving away from the mean
in increments of 1 unit.

data near the mean are more


frequent in occurrence than data
far from the mean
7

Steps to follow:
1. Define your hypothesis
2. Set the significance level of the statistical
test.
8

Define your hypothesis:


HYPOTHESIS

Null Hypothesis (Ho)


• a type of conjecture used in
statistics that proposes that no
statistical significance exists in a Alternate Hypothesis (H1)
set of given observations.
• states that there is statistical
• attempts to show that no
significance between two variables.
variation exists between variables,
• The alternative hypothesis is the
or that a single variable is no
hypothesis that the researcher is
different than its mean.
trying to prove
9

Set the Significance Level of the Statistical Test

 It is described as a  it specifies the


probability. probability of
rejecting the Null
 The significance level is set Hypothesis when it
as a standard of quality for was actually true
your test, in order to be
able to draw accurate  represented by the
conclusions. Greek letter alpha
10

PROBLEM STATEMENT
 Is the average tuition fee of
other B-schools of “A”
category in the SBI Scholar
loan list less than the tuition
fee of MDI, Murshidabad?

 Is there any relationship


between fees paid and the
number of students attending
the college?
11

Hypothesis:

Hypothesis HO: the average cost of


I college tuition per 2 year
of other B school of ‘A’
category is less than the
cost of MDIM college
tuition per 2 year
12

Hypothesis the cost of


II tuition is
dependent on the
number of
students
13

METHODOLOGY

T- Test Chi- Square Test


14

DATA ANALYSIS

different
management
colleges in List ‘A’
of SBI
Scholar Loan
15

At α=.025,
if we can conclude that the average is less than Rs.11,
50,000 based on a sample of 26 colleges.

H0: µ< Rs.11, 50,000


H1: µ>= Rs.11, 50,000
16

T-test
Make the decision to
Compute the sample test Summarize the
Find the critical value reject or not reject
value
the null hypothesis results

Higher values of the t-


At α=.025 m= Rs.11, 50,000 value, also called t-score,
Standard deviation = Since the T sample indicate that a large
and d.f. = 25, 575155.7746 test value doesn’t lie difference exists between
the critical s=944515.3846 in the Critical value the two sample sets
Null hypothesis is A large t-score
value = -2.060 t= (944515.3846-11,
not rejected. indicates that the
50,000)/
groups are different.
(575155.7746/sqrt A small t-score
(26)) = -1.82 indicates that the
groups are similar
I 17

CHI-SQUARE TEST
18

Observed value
19

Make the decision to reject or


Compute the test value. Find
Find the critical value the expected value.
not to reject the null
hypothesis.

Degree of Freedom
= (3 – 1) (5 – 1) 𝑥 2 < 𝑥𝑐2
=8 16.433 <17.535
Failed to reject
the null hypothesis
test value is x² = ∑ (O-E) ²/E
Critical value = Test value = 16.433
17.535
20

Summarize
your results
There is not enough evidence to
support the claim that the cost
of tuition is dependent on the
number of students that attend
the college.
21
CONCLUSION

 In our first hypothesis some of the costs of tuition may include


other fees and some may not. Some fees were listed separately and
some were not. This could have led to a Type I error where the
null hypothesis was true and it was rejected.
 In our second hypothesis we only included the number of seats of
PGPM students and no other courses like PGPX courses. So we
would have involved these and it may have affected the result a
little, however it was not enough to change the hypothesis.
22

Thanks!

Vous aimerez peut-être aussi