I've posted solutions for the practice final exam I handed out in class.
Data sets for the problems below are available on the web in text format. The data sets are also available on the I drive in .mtw format.
Analyze each of the datasets referred to below as completely, accurately, and cogently as you can, with the goal in each case being to answer the question in italics. In each case, your job is to (i) determine which statistical procedure(s) might be relevant; (ii) list any assumptions to be made in the analysis and attempt to justify them; (iii) analyze the data, then interpret the results of this analysis; and (iv) reflect on what types of inferences may be made, based on what you know about the data in question. Use graphics when appropriate, particularly in step (ii) above. In addition, you should be familiar with where you could look up in the textbook any necessary formulas and critical values, since you may be asked on your final exam to perform some computations that Minitab performs automatically.
Problem 1.
The effects on the total life length of rats on a restricted diet
versus an ad libitum (free eating) diet were studied. Treatments
were begun after an initial weaning period.
Data are in days.
Does diet restriction
promote longevity?
Data
Problem 2.
The data come from a study in which the level of an enzyme, creatinine
kinase (CK), was measured in patients who were suspected of having
had a myocardial infarction (heart attack).
The enzyme was measured in 360 patients on admission, and
later an expert clinician reviewed the records of these patients to decide
with hindsight which of them had actually had a heart attack.
Was measuring the level of this enzyme on admission to the hospital
a useful diagnostic indicator for whether patients had really had
a heart attack?
Data
Problem 3.
In this study, 25 hospitalized schizophrenic patients were treated with
antiseptic medication, and after a period of time
were classified as psychotic or nonpsychotic by hospital staff.
Samples of cerebrospinal fluid were taken from each patient and
assayed for the dopamine b-hydroxylase (DBH) activity.
The data are given below; units are nmol/ml(h)/mg of protein.
How does DBH activity differ between the two groups?
Data
Problem 4. These data were collected as part of a design study for a
device to generate electricity from wave power at sea. The study was carried
out on scale models in a wave tank. The wave tank could simulate a wide
range of sea states. The model system was subjected to the same sample of 18
sea states with each of two mooring methods, one of which was considerably
cheaper than the other. The resulting data (root mean square bending moment
in Newton meters) are given.
How does the choice of mooring method for the system affect the
bending stress produced in the measured part of the device?
Data
Problem 5.
These data come from a clinical trial and show the effect of the
drug sulphinpyrazone on deaths after myocardial infarction.
Quantify the effect of the drug in reducing mortality.
Data
Problem 6.
A randomized block experiment was carried out to investigate
a drug added to the feed of chicks in an attempt to promote
growth. The comparison is among three treatments: standard
feed (control), standard feed plus low dose of drug, standard
feed plus high dose of drug. The experimental unit is a group of chicks,
reared and fed together in the birdhouse. The experimental units
are grouped three to a block, with physically adjacent units
going in the same block. The response is the average
weight per bird at maturity for the group of birds in each experimental
unit.
How does the drug appear to affect growth?
Data
Problem 7.
These data come from an experiment to study the best way of forming ground
cover in an apple plantation. Treatment O represents what was the usual
treatment, keeping the land clear during the growing season but letting
the weeds grow up towards the end. Treatments A, B, C, D, and E represent
the growing of various permanent crops under the trees. There
were four randomized blocks. The response was the total crop
weight in pounds over a four-year period after the treatments
were begun.
How do treatments A through E compare with treatment O and with each
other?
Data
Problem 8.
These are data on the number of failures of piston-rings in each of three legs
in each of four steam-driven compressors located in the same building.
The compressors have identical design and are oriented in the same way.
Is the pattern
of the location of failures different for different compressors?
Data
Problem 9.
A double-blind experiment was carried out to investigate the effect of
caffeine on performance on a simple physical task. Thirty male college
students were trained in finger tapping. They were then divided at random
into three groups of 10 and the groups received different doses of caffeine
(0, 100 and 200 mg). Two hours after treatment, each man was required
to do finger tapping and the number of taps per minute was recorded.
Does caffeine affect performance on this task? If so, can you describe
the effect?
Data
Problem 10.
These data come from a study comparing the health of juvenile
delinquent boys and a non-delinquent control group. They relate to the subset
of boys who failed a vision test, and show the numbers who did and did
not wear glasses.
Are delinquents with poor eyesight more or less
likely to wear glasses than are non-delinquents with poor eyesight?
Data
Problem 11.
The analgesic effect of iontophoretic treatment with the nerve
conduction-inhibiting chemical vincristine was studied on elderly
patients complaining of post-herpetic neuralgia. Eighteen patients
were interviewed six weeks after undergoing
treatment to see whether any improvement had occurred. In the dataset,
the response Y takes the value 1 if pain cessation was recorded.
The variable X1 indicates treatment; X2 is the patient's age;
X3 is the patient's sex; X4 is the pretreatment duration of symptoms
(in months).
What factors appear to influence cessation of pain? In particular,
what can we say about the treatment?
Data