Scribd
Upload
28 views11 pages

Section 5.8

Ukzn stat222

Uploaded by

lindokuhlenkosiyethu90
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
28 views11 pages

Section 5.8

Ukzn stat222

Uploaded by

lindokuhlenkosiyethu90
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 11

Section 5.

8:
Testing the Equality of Two
Population Variances
Recall:
• When testing there are various cases depending on the available
information.
• When the population variances and are unknown AND the sample
sizes are SMALL ( and/or ), there are two cases B and C to choose
from:
If the population If the population variances
variances are equal: i.e. are unequal: i.e.

Case B Case C
• Question: But how do we know if ?
• Answer: a) We either know from past studies or research.
b) Or we can perform a hypothesis test to determine if .
Therefore, this section presents methods for carrying out a hypothesis
test to determine if two population variances are equal.

Null
Hypothesis:

Alternative Hypothesis:
(two-tailed test)

Note: We can test the other one-sided alternatives or , however for this
course, we are only interested in the testing of population means. Thus,
we only use this test to determine if we use case B or case C to test the
difference in two population means. Therefore, we only use the two-
sided alternative above.
Recall: The general form of representing the null hypothesis is …

2
𝜎1
𝐻0 : 2
=1
𝜎 2

2
𝜎1
𝐻1 : 2
≠1
𝜎 2

parameter specified value


2
𝜎1
¿ 2
𝜎2

Due to the properties of variances, and in order to formulate a test statistic,


we use the ratio of and (instead of using their difference, as in the case
with means).
The test statistic for this hypothesis test is calculated as follows:

2
𝑆1
(on the formula sheet) 𝐹 𝑜= 2
𝐹 (𝑛1 − 1; 𝑛 2 −1)
𝑆 2

• Where and are the SAMPLE estimates (sample variances) for and ,
respectively.

• This test statistic now follows a new distribution called the F-


• distribution.
The F-distribution is defined by two degrees of freedom: that of the
numerator (for sample 1), and that of the denominator (for sample
2).
• Thus, the degrees of freedom for this test statistic will be and .
Tables for the F-
distribution:
• For each combination of the two degrees of freedom, there is a different
F-distribution.
• Only tables for the 95%, 97.5% and 99% points of the F-distribution are
given (tables F1 to F5). i.e. for areas to the left or .
• The F-distribution tables are slightly different to that for the t-
distribution where each page of tables represents F-values
corresponding to only one of the values of .
• The top row represents the numerator’s d.o.f. () and the first column
represents the denominator’s d.o.f. () .
• The values in the body of the table represent the F-value that has an
area of to its left.
Example
1:
If and :

This means we are


looking for a 99%
point go to table F5
(99% points of the
F-distribution) 3.7
0

Locate and in
the top row.

And in the first


column. ∴ 𝐹 𝑑𝑓 1 , 𝑑 𝑓 2 ;𝛼 =¿𝐹 7 ,20 ; 0.99¿ 3.70
• Only upper percentile values (those with large areas to the left) can be
found from the F-tables.
• Unlike the t-distribution, the F-distribution is not symmetric, therefore
we cannot use symmetry to find a lower F-value.
• Instead, we use the following property of the F-distribution:
We use the point F-
(NOT on the 1 tables to find this F-
formula sheet 𝐹 𝑑𝑓 ,𝑑𝑓 2 ;𝛼 =
1
𝐹𝑑 𝑓 , 𝑑𝑓 1 ;1 −𝛼 value in the
learn!) 2

denominator.
The degrees of
freedom are swopped
around:
and
Example
2:
If and :

27.0
5

We don’t have
tables for this value
of .
∴ 𝐹 3 ,1 2 ;0.01
1
¿
𝐹 12 , 3 ; 0.99
1
¿
2 7.05
¿ 0 .037 we use the 99% F-
tables
Back to testing …
• Since the test statistic follows the F-distribution, the critical values will
also follow the F-distribution with the same degrees of freedom and .
• However, the critical values for a two-tailed test (since ) are found in
the same way as those from the previous sections:

𝛼 𝛼
As usual for a two-tailed test, area 2 area 2
there are two rejection = =
regions:
Rejection Rule:
𝐹 𝛼 𝐹
Reject in favour of if OR if 𝑑 𝑓 1 ,𝑑 𝑓 2 ;
2 𝑑 𝑓 1 ,𝑑 𝑓 2 ; 1−
𝛼
2

(two critical values)


Important:
• N.B. the purpose of this test is to determine which case (B or C) is used
when testing where the population variances are unknown and the
sample sizes are small.
Use case B to carry out the test for :
If is NOT rejected the t-test statistic (using the pooled
variance to be calculated) with a
degrees of freedom of .

Use case C to carry out the test for :


If is rejected the t-test statistic with a degrees of
freedom of .

Section 5.8 Example on how this test is used.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy