Stats Project Chapter 9

Stats Project Chapter
9
By: Rebecca Scott and Sarah
Musgrave
Objectives:
● Subsets
● Extrapolation
● Outlier
● Leverage
● Influential Point
● Lurking Variables
Subsets
● Subsets are used when the data consists of two or
more groups that have been thrown together.
● Best to fit different linear models of each group.
● Displays of the residual can help find subsets.
•The top shelf is different.
Therefore, two regressions
should be reported, one for
the top shelf and one for
the two bottom shelves.
Extrapolation
● A prediction made by extending the model
or graph.
● This is a VERY DANGEROUS tool due to
the questionable answers.
● This is only to be used to make
predictions of the future outcome.
Outliers
● Any data point that is away from the
others
● Outliers can have a large residual or have
high leverage
Leverage
● Data points whose xvalues are from the
mean of x
● High leverage points pull
the line close making a
large effect on the slope
Influential point
● If omitting the point from the data results
in a very different model.
Problem 17
Suppose a researcher studying health issues
measures blood pressure and the percentage of body
fat for several adult males, finding a strong positive
association. Describe three different possible causeand-effect relationships that might be present.
1) blood pressure can cause high body fat
2) high body fat can cause high blood pressure
3) Both can be caused by lurking variables such as
genetics.
Problem 19
The table summarizes the rates of
assault and injury for these employees
for 5 years, 1995-1999. Can the assault
rate be used to predict injuries or
deaths?
Problem 19 cont.
How to find solution
1) Type data into the calculator.
2) Go to “stat” then “cal” and select “LinReg”
3) This will give you r and r 2
This gives you the answer of no. Due to
the high leverage outlier, the National
Park Service, r 2 is 46.2%. However,
without this point the r 2 is 0.