Scribd
Upload
81 views

ANOVA-Single Factor - Tukey Test For Differences

This document describes how to use a single factor ANOVA and Tukey test to analyze differences in average battery life between 5 types of notebook computer batteries (Types A, B, C, D, and E). An ANOVA analysis found significant differences in battery life. A Tukey test then determined which battery types specifically differed, finding Types E and D had significantly longer battery life than Types A, B, and C, while the life of Types C and B did not differ.

Uploaded by

aaandik123
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
81 views

ANOVA-Single Factor - Tukey Test For Differences

This document describes how to use a single factor ANOVA and Tukey test to analyze differences in average battery life between 5 types of notebook computer batteries (Types A, B, C, D, and E). An ANOVA analysis found significant differences in battery life. A Tukey test then determined which battery types specifically differed, finding Types E and D had significantly longer battery life than Types A, B, and C, while the life of Types C and B did not differ.

Uploaded by

aaandik123
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 5

ANOVA

SINGLE FACTOR
Tukey Test for Differences
SITUATION:
Sampling From 3 or more Populations Based on a Single Factor
No common link between the k-th observation from Population 1
and the k-th observation of Population 2, or 3, etc.
Must have same sample sizes from each population
Have determined that there ARE differences in the means
QUESTION:
Which means differ from the others?
Example:
Can we conclude that there are differences in average battery life
of 5 different types of notebook computer batteries? And if so, which
ones differ from the others?
Battery lives on the same type of computer
6 were of Type A
6 were of Type B
6 were of Type C
6 were of Type D
6 were of Type E

The Statistical F-Test:


H0 :
H1 :

A = B = C = D = E
At least one of these j's does not equal the others

Go to:
TOOLS
Data Analysis
Anova: Single Factor

Highlight the block of cells that contains all


data -- the 6 data rows (plus the label row)
and the five columns (A1:E7).

Check Labels in first row


Designate beginning
of output area

Low p-value
(.000000000108)
Can conclude there are differences in
the average battery lives

So which means differ from the others?


Since there are the same number from each population (6) -- we can use
the TUKEY approach.

The TUKEY Approach


The Tukey approach states that any two sample means ( x 's) that differs
by "a lot" indicates the corresponding population means ('s) differ.
The Tukey approach defines "a lot" as:

MSE
ng

Here:
MSE = Mean Square Error (found on Excel printout)
ng
= Number in each group (in this case 6)
q
= studentized range value -- a number from a table (7(a))
that depends on:
-- the probability of drawing a wrong conclusion (let's use .05)
k -- the number of populations (5 in this case)
-- the number of degrees of freedom for the error term (25 in this
case)
The table we use does not have a value for k = 5, = 25, but does have a
value for k = 5, = 24 which is q = 4.17 -- we will use this value and
put it on the Excel spreadsheet!
By Excel: First Copy and Paste the Averages ( x 's) and the Groups, in that
order to another set of cells on the spreadsheet.

Now type in the q-value and calculate (omega).

MSE
ng

4.17

=H6*SQRT(D23/6)

Thus any two sample means that differ by more than 16.51002 indicates
that the corresponding population means differ.
Battery E:
150 -140 = 10 --- cannot conclude a difference between E and D
150 - 121.6667 = 28.3333 -- can conclude E differs from A (and C and B).
Battery D:
140 - 121.6667 = 18.3333 -- can conclude D differs from A (and C and B).
Battery A:
121.6667 - 100 = 21.667 -- can conclude A differs from C (and B).
Battery C:
100 - 90 = 10 --- cannot conclude a difference between C and B

You might also like