You are on page 1of 2

STAT 512

Lab #1

write-up due: Mon. 9/21/15

Pull up the Minitab worksheet called lab1.mtw found in the data sets folder of the Blackboard (eLearning)
site. This worksheet contains data sets for three different studies, as described below.
Data Sets
A. Bears Data
In this study, the researchers were interested in developing a formula or tables for hunters so that
they could estimate the weight of a bear based on other body measurements more easily recorded
in the forest. Although several measurements were taken, your data set has only two variables.
Weight is the weight of the bear, in pounds, and Neck.G is the neck girth (distance around the
neck), in inches, for 97 wild black bears captured, anesthetized, measured, and weighed before
being returned to the forest. Is there a linear relationship between weight and neck girth that
would be useful to hunters?
B. Crime Rate
A criminologist wants to study the relationship between the level of education and the crime rate.
The data set contains the Crime Rate recorded as crimes reported per 100,000 residents and the
percentage of residents of the county having at least a high school diploma (HSDiploma) for 84
randomly selected U.S. counties in a past year. Does this data give evidence that the crime rate is
related to level of education?
C. First Child
A demographic study was conducted for married couples with one or more children to determine
the effect of the husbands annual income at the time of marriage (in thousands of dollars) on the
time (in months) between the marriage and birth of the first child. Is there a linear relationship
between income and time until the first child?
During the lab session, use Minitab to help you answer the following questions. The objective is to
have responses for questions 1-7 by the end of the session.
1. Provide a detailed description of the relationship, if one exists, between the predictor and response
for each data set, based on what you see in a scatterplot.
In comparing the data sets, if asked which trend looks most linear, how would you respond?
Explain.
If asked which association looks weakest, how would you respond? Explain.
2. Fit the linear model Y i= 0 + 1 X i+ i for i = 1,, n to each set of data. What does b1 tell you
about the data set being studied (i.e., provide an interpretation of b1 in the context of the data set)?
How do the values of b1 compare between the three data sets? (e.g., Positive or negative? Value?
Units?)
3. Test whether or not 1=0 using a 5% level of significance. Be sure to state your hypotheses,
computed test statistic, p-value, and conclusion in each case.

4. Is the result of the hypothesis test what you expected from the scatterplot? Why or why not?
5. For each data set compute a 95% confidence interval for 1 and provide an interpretation in the
context of the problem. Which interval is widest? Narrowest? Why is this so?
6. For each data set, compute a 95% confidence interval for the mean response E[Y] when X= x
and provide an interpretation in the context of the problem. Which interval is widest? Narrowest?
Why is this so?
7. Based on your results, what would you tell the researcher for each data set?
Questions to consider after the lab:
8. What have you learned from this lab session? Be as specific and complete as possible in
answering this question.
9. What are the implications of todays lab session for analyzing a data set using regression?
10. What additional tools would be useful?
Next Monday, your group is to submit a write-up which includes a response to questions 1-7 for
each data set A-C, including the relevant output, as well as a response to questions 8, 9 and 10.

You might also like