272 Fundamentals of Statistics
b. State your null and alternative
hypotheses.
c. Perform a two-sample t test compar-
ing the value of the Resistance vari-
able between the male and female
subjects, broken down by day. You do
not have to summarize your results
across days.
d. On what day or days is there a sig-
nifi cant difference between the two
groups? Do your results change if
you use an unpooled rather than
a pooled estimate of the standard
deviation?
e. Create a scatterplot of Resistance ver-
sus Days. Break the scatterplot down
by gender using the StatPlus com-
mand shown in Chapter 3. Describe
the effect displayed in the scatter
plot (you may want to change the
scatter plot scales to view the data
better).
f. Redo your analysis using the Mann-
Whitney test. Do your conclusions
from part b change any with the non-
parametric procedure?
g. Save your workbook and write a
report summarizing your fi ndings.
Explain how (if at all) the male and
female subjects differed in their re-
sponse to the study. Include in your
discussion the various parts of the
study (control period, bed-rest period,
etc.) and how the patients responded
during those specifi c intervals.
Include any pertinent statistics.- The Math workbook contains data
from a study analyzing two methods of
teaching mathematics. Students were
randomly assigned to two groups: a con-
trol group that was taught in the usual
way with a relaxed homework and quiz
schedule, and an experimental group
that was regularly assigned homework
and given frequent quizzes. Students in
the experimental group were allowed to
retake their exams to raise their grades
(though a different exam was given for
the retake). The fi nal exam scores of the
two groups were recorded. Investigate
whether there is compelling evidence
that students in the experimental group
had higher scores than those in the con-
trol group.
a. Open the Math workbook from the
Chapter06 folder and save it as Math
Scores Analysis.
b. State your null and alternative hy-
potheses. Is this a one-sided test or a
two-sided test? Why?
c. Perform a two-sample t test on the fi nal
exam score. Use a pooled estimate of
the standard deviation. What is the
95% confi dence interval for the differ-
ence in scores? What is the p value?
Do you accept or reject the null hy-
pothesis? Do your conclusions change
if you use an unpooled test?
d. Chart the distribution of the fi nal
exam scores for the two groups.
What do the charts tell about the
distributions? Do the charts cast any
doubt on your conclusions in
part c? Why?
e. Do a second analysis of the data using
the Mann-Whitney Rank test. How do
these results compare to the two-sample
t? Are your conclusions the same?
f. Save your changes to the workbook
and write a report summarizing your
fi ndings and reporting your conclu-
sion. Is there a signifi cant change in
the exam scores under the experimen-
tal approach?- The Voting workbook contains the per-
centage of the presidential vote that the
Democratic candidate received in 1980
and 1984, broken down by state and
region. You’ve been asked to investigate