Submit: An Excel file with your answers.
Clearly label Part 1 and Part 2. For a lot of Part 2 answers, you might want to
create text boxes to write in.
Work in groups of two or three.
Part 1 - Correlation and Linear Trend
Goal: Investigate the relationship between smoking while
pregnant and the childs birthweight.
What is the correlation r between number of cigarettes smoked per day
and birth weight?
Form the scatterplot of birth weight (y-value) vs.
number of cigarettes smoked per day (x-value).
Add a linear trend line to the scatterplot in 3.
Also add the R2 and trend line equation to
the plot.
Does the relationship between x and y appear to be a linear
association?
What is the equation of the linear trend line?
What is the predicted birthweight for a mother that smokes
twenty cigarettes per day?
What is the predicted birthweight for a mother that does
not smoke at all?
Part 2 - False Positive and Negative Test Results
Goal: Investigate how false positives and negatives
affect the accuracy of medical tests.
The numbers for a particular type of cancer are as follows:
1 in 1000 tumors is malignant
A blood test is 90% accurate when given to 15,000 people with
this kind of tumor.
Create a table similar to the one in the
May 12 Lecture Notes,
which displays the number of true positives, false positives,
true negatives and false negatives. Make sure you label everything.
Compute the probability that a patient really has cancer,
given a positive blood test.
Compute the probability that a patient has cancer, even
with a negative blood test.