You are on page 1of 35

# Hypothesis Testing;

## Z-Test, T-Test, F-Test

BY NARENDER SHARMA
2
Shakehand with Life
Corporate Training and Management Education

www.shakehandwithlife.in , www.shakehandwithlife.puzl.com

Leading Training, Coaching, Consulting services in Delhi NCR for Managers at all levels,
Future Managers and Engineers in MBA and B.E. / B. Tech.,
Training with MS-Excel for managerial decision making skills,
Working with MS-Excel to solve all mathematical and statistical problem.

## Call Now WhatsApp

9468267324, 8684861131 9468267324

E-mail: shakehandwithlife@gmail.com
www.shakehandwithlife.in
3
What is Hypothesis?

## Hypothesis is a predictive statement, capable of

being tested by scientific methods, that relates an
independent variables to some dependent
variable.
A hypothesis states what we are looking for and it is
a proportion which can be put to a test to
determine its validity
e.g.
Students who receive counseling will show a greater
increase in creativity than students not receiving
counseling

www.shakehandwithlife.in
4
Characteristics of Hypothesis

## Clear and precise.

Capable of being tested.
Stated relationship between variables.
limited in scope and must be specific.
Stated as far as possible in most simple terms so that the same is
easily understand by all concerned. But one must remember that
simplicity of hypothesis has nothing to do with its significance.
Consistent with most known facts.
Responsive to testing with in a reasonable time. One cant spend a
life time collecting data to test it.
Explain what it claims to explain; it should have empirical reference.

www.shakehandwithlife.in
5
Null Hypothesis

## It is an assertion that we hold as true unless we have

sufficient statistical evidence to conclude otherwise.
Null Hypothesis is denoted by 0
If a population mean is equal to hypothesised mean
then Null Hypothesis can be written as

0 : = 0

www.shakehandwithlife.in
6
Alternative Hypothesis

## The Alternative hypothesis is negation of null

hypothesis and is denoted by
If Null is given as 0 : = 0

## Then alternative Hypothesis can be written as

: 0
: > 0
: < 0

www.shakehandwithlife.in
7
Level of significance and
confidence

## Significance means the percentage risk to reject a

null hypothesis when it is true and it is denoted by .
Generally taken as 1%, 5%, 10%
(1 ) is the confidence interval in which the null
hypothesis will exist when it is true.

www.shakehandwithlife.in
8
Risk of rejecting a Null Hypothesis
when it is true

Risk Confidence
Designation Description

More than \$100 million
0.001 0.999
Supercritical (Large loss of life, e.g. nuclear
0.1% 99.9%
disaster
0.01 0.99 Less than \$100 million
Critical
1% 99% (A few lives lost)
0.05 0.95 Less than \$100 thousand
Important
5% 95% (No lives lost, injuries occur)
0.10 0.90 Less than \$500
Moderate
10% 90% (No injuries occur)

www.shakehandwithlife.in
9
Type I and Type II Error

Decision

## Situation Accept Null Reject Null

Null is true Correct Type I error
( )

## Null is false Type II error Correct

( )

www.shakehandwithlife.in
Two tailed test @ 10
5% Significance level

## Acceptance and Rejection

regions in case of a Two Suitable When 0 : = 0
tailed test : 0

/ /
( = 0.025 2.5%) ( = 0.025 2.5%)
(1 ) = 95%
0 : = 0

www.shakehandwithlife.in
Left tailed test @ 11
5% Significance level

## Acceptance and Rejection

regions in case of a left tailed Suitable When 0 : = 0
test : < 0

/
( = 0.05 5%) (1 ) = 95%

0 : = 0

www.shakehandwithlife.in
Right tailed test @ 12
5% Significance level

## Acceptance and Rejection

regions in case of a Right Suitable When 0 : = 0
tailed test : > 0

/
(1 ) = 95% ( = 0.05 5%)

0 : = 0

www.shakehandwithlife.in
13
Procedure for Hypothesis
Testing

## State the null State a Decide a test Calculate the

(Ho)and alternate significance level; statistics; z-test, t- value of test
(Ha) Hypothesis 1%, 5%, 10% etc. test, F-test. statistics

Compare
Calculate the p-
the p-value P-value >
value at given
with Calculated Accept Ho
significance level
calculated value
from the table
value

P-value <
Calculated Reject Ho
value
www.shakehandwithlife.in
14

Hypothesis
Testing of Z-TEST AND T-TEST

Means

www.shakehandwithlife.in
15
Z-Test for testing means

## Test Condition Test Statistics

Population normal and
infinite
Sample size large or small, 0
Population variance is
=
known
Ha may be one-sided or
two sided

www.shakehandwithlife.in
16
Z-Test for testing means

## Test Condition Test Statistics

Population normal and
finite,
Sample size large or small,
0
Population variance is =

known 1

Ha may be one-sided or
two sided

www.shakehandwithlife.in
17
Z-Test for testing means

## Test Condition Test Statistics

Population is infinite and
may not be normal, 0
Sample size is large, =
Population variance is
unknown
Ha may be one-sided or
two sided

www.shakehandwithlife.in
18
Z-Test for testing means

## Test Condition Test Statistics

Population is finite and may
not be normal, 0
=
Sample size is large, 1

Population variance is
unknown
Ha may be one-sided or
two sided

www.shakehandwithlife.in
19
T-Test for testing means

## Test Condition Test Statistics

Population is infinite and
normal, 0
Sample size is small,
=

Population variance is
. . = 1
unknown
Ha may be one-sided or 2
two sided =
( 1)

www.shakehandwithlife.in
20
T-Test for testing means

## Test Condition Test Statistics

Population is finite and 0
normal, =
1
Sample size is small,
Population variance is
. . = 1
unknown
Ha may be one-sided or 2
two sided =
( 1)

www.shakehandwithlife.in
21

Hypothesis
testing for
difference Z-TEST, T-TEST

between
means

www.shakehandwithlife.in
22
Z-Test for testing difference
between means

## Test Condition Test Statistics

Populations are normal
Samples happen to be
large, 1 2
=
Population variances are 2 2
known 1 2
+
Ha may be one-sided or 1 2
two sided

www.shakehandwithlife.in
23
Z-Test for testing difference
between means

## Test Condition Test Statistics

Populations are normal
Samples happen to be large,
Presumed to have been 1 2
drawn from the same
population =
1 1
Population variances are 2 +
known 1 2
Ha may be one-sided or two
sided

www.shakehandwithlife.in
24
T-Test for testing difference
between means

## Test Condition Test Statistics

Samples happen to be small,
Presumed to have been
drawn from the same 1 2
population =
2 2
1 1 1 + 2 1 2 1 1
Population variances are +
unknown but assumed to be
1 + 2 2 1 2
equal
Ha may be one-sided or two . . = (1 + 2 2)
sided

www.shakehandwithlife.in
25

Hypothesis
Testing for
comparing PAIRED T-TEST

two related
samples

www.shakehandwithlife.in
26
Paired T-Test for comparing
two related samples

## Test Condition Test Statistics

Samples happens to be 0
small =
.
Variances of the two
populations need not be ( 1) . .
equal
= Mean of differences
Populations are normal
. = Standard deviation of differences
Ha may be one sided or
two sided =

www.shakehandwithlife.in
27

Hypothesis
Testing of Z-TEST

proportions

www.shakehandwithlife.in
28
Z-test for testing of proportions

## Test Condition Test statistics

Use in case of qualitative
data
Sampling distribution may =
take the form of binomial .
probability distribution
Ha may be one sided or two
sided
=
= .
= . .

www.shakehandwithlife.in
29

Hypothesis
Testing for
difference Z-TEST

between
proportions

www.shakehandwithlife.in
30
Z-test for testing difference
between proportions

## Test Condition Test statistics

Sample drawn from two
1 2
different populations =
1 1 2 2
Test confirm, whether the +
1 2
difference between the
proportion of success is 1 = proportion of success in sample one
significant
2 = proportion of success in sample two
Ha may be one sided or
two sided

www.shakehandwithlife.in
31

Hypothesis
testing of
equality of F-TEST
variances of
two normal
populations

www.shakehandwithlife.in
F-Test for testing equality of 32
variances of two normal
populations

## Test conditions Test statistics

The populations are normal
2
Samples have been drawn 1
randomly
= 2
2
Observations are
independent; and 1 1 and 2 1 d. f.
There is no measurement
error
2
1 is the sample estimate for 1
2

## Ha may be one sided or two 2

2 is the sample estimate for 2
2
sided

www.shakehandwithlife.in
33
Limitations of the test of
Hypothesis

## Testing of hypothesis is not decision making itself; but help

for decision making
Test does not explain the reasons as why the difference
exist, it only indicate that the difference is due to
fluctuations of sampling or because of other reasons but
the tests do not tell about the reason causing the
difference.
Tests are based on the probabilities and as such cannot be
expressed with full certainty.
Statistical inferences based on the significance tests
cannot be said to be entirely correct evidences
concerning the truth of the hypothesis.
www.shakehandwithlife.in
34

Thank You
35
Shakehand with Life
Corporate Training and Management Education

www.shakehandwithlife.in , www.shakehandwithlife.puzl.com

Leading Training, Coaching, Consulting services in Delhi NCR for Managers at all levels,
Future Managers and Engineers in MBA and B.E. / B. Tech.,