1. Feed A Cold and
Starve A Fever
A thorough experiment of sickness and health
by Nick Arbaugh and Matthew Allen
2. Subjects (Experimental Units)
❏ 250 male volunteers and 250 female
volunteers who have a doctor proven cold,
and 250 male volunteers and 250 female
volunteers who have a doctor proven fever.
❏ All subjects will be required to undergo an
examination by a doctor to determine if they
have either a cold or a fever.
3. Explanatory Variables/ Factors
❏ Categorical
❏ A change in the amount of food one eats
when one has a fever or a cold. In this case,
a lack of food when you have a fever, and an
increase in food when you have a cold.
4. Treatments (Factor Levels)
The varied amounts of food consumed by
volunteers with the different afflictions.
Cold -
1. Eat significantly more than usually
2. Eat normally
Fever -
1. Eat significantly less than usually
2. Eat normally
5. Response Variables/Factors are
❏ Quantitative
❏ The amount of time it takes in days from
when the treatment is given until the subject
is given a clean bill of health from a
physician.
6. Experiment Explanation(Cold) with
Diagram
Explanatory
Variable -
Amount of food
intake
Males
Females
Eat significantly more
than normal
Eat normally
Eat significantly more
than normal
Eat normally
Response
Variable - How
fast the cold is
over?
R
A
N
D
O
M
A
S
S
I
G
N
M
E
N
T
7. Experiment Explanation(Fever) with
Diagram
Response
Variable - How
fast the fever is
over?
Males
Females
Eat significantly less than
normal
Eat normally
Eat significantly less than
normal
Eat normally
R
A
N
D
O
M
A
S
S
I
G
N
M
E
N
T
Explanatory
Variable -
Amount of food
intake
8. Experimental Design Principals
❏ Control - 1 group given no treatment in each part of the
experiment (control groups). Volunteers will be utilized which
means they are more willing to participate in the experiment,
consequently decreasing confounding variables.
❏ Randomization - to determine if they were going to be in the
control or experimental group was done by randomization,
the use of chance to assign subjects to treatments.
❏ Replication - we will do the experiment several times to
increase validity, as well as having a large number (500)
people in each part of the experiment
9. Blocking?
We would break each separate group of
subjects into male and female subgroups in
order to examine the effects our methods have
on each gender. This way we will be able to
determine if the different gender has an impact
on the response variable, being how fast they
get better.
10. Blinding?
This experiment cannot be blinded to the
volunteers because they will have to know how
much (or little) food they are consuming.
However, the researchers can be blinded
allowing for less bias results and an increase in
the validity of the experiment.
11. Experiment concerns
❏ High variability due to people’s different
immune systems and their different
tolerance levels to pain or discomfort.
❏ The amount of “significantly more than
normal” and “significantly less than normal”
can mean completely different things for
different people, therefore skewing results.