This preview shows pages 1–3. Sign up to view the full content.
DIFFERENCE IN POPULATION MEANS
Determining the Model Type
1.
Ask, "Do you know the standard deviations of the populations?"
YES – Select zTest: Two sample For Means
NO
 Go to Step 2
2.
Ask, "Is the data paired?"
That is, "Is there something in common
between the first, the second, the third, etc. data points in the first
sample and the first data point in the second sample?"
YES  Select tTest: Paired Two Sample for Means
NO
 Go to Step 3
Examples:
Scores of Women and Men on a Test
The ordering is irrelevant.
The first woman's score of 77
has no relation to the firs man's score of 88.
Data
Not
Paired!
Attendance at Lakers and Clippers Basketball Games
The observations occur on different dates
Data
Not
Paired!
Sales on
Same Date
at Two Different Stores
Each pair of observations occur on the same date
Data
IS
Paired!
(By date)
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document3.
For data that is not paired, ask, "Can I assume the standard deviations
are equal?"
This is an
FTest
of H
0
:
σ
1
2
/
σ
2
2
= 1 vs. H
1
:
σ
1
2
/
σ
2
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '11
 Gupta
 Standard Deviation

Click to edit the document details