Transcript Chapter 13: Experiments and Observational Studies

+
Chapter 13: Experiments
and Observational Studies
Madison Vincent
Period: 3
12/11/14
+
Observational Studies
■
In observational studies researchers don’t
assign choices; they simply observe them
■
Based on data in which no manipulation of
factors has been employed.
■
Valuable for discovering trends and possible
relationships
■ Not
possible for observational studies to
demonstrate a casual relationship
+
Retrospective and Prospective
Study
Retrospective
Subjects are selected and then their previous conditions or
behaviors are studied.
Are not based on random samples so they usually focus on
estimating differences between groups or associations between variables.
Prospective
Subjects are followed to observe future outcomes
No treatments are deliberately applied so it is not an experiment
Typically focus on estimating differences among groups that
might appear as the groups are followed during the course of the
study
+
Experiments
■
Study design that allows us to prove a cause-and-effect
relationship
Manipulates factor levels to create treatments, Randomly
assigns subjects to these treatment levels, Then compares
the responses of the subject groups across treatment
levels
Factors are variables whose levels are controlled by the experimenter.
Experiments attempts to discover the effects that differences in factor levels
may have on the responses of the experimental units
Response is variables whose values are compared across different
treatments. In a randomized experiment, large response differences can be
attributed to the effect of differences in treatment level
+
Experiments
■
In an experiment, the experimenter actively and
deliberately manipulates the factors to control the
details of the possible treatments, and assigns the
subjects to those treatments at random.
■
The experimenter then observes the response
variable and compares responses for different
groups of subjects who have been treated
differently.
■
The best experiments are usually randomized,
comparative, double blind, and placebo controlled.
+
Experiments
■
Experimental units are individuals on whom an experiment is
preformed. Usually called subjects or participants when they are human
■
Level is the specific values that the experimenter chooses for a factor are
called the levels of the factor
■
Treatment is the process, intervention, or other controlled circumstance
applied to randomly assigned experimental units. Treatments are the
different levels of a single factor or are made up of combinations of levels
of two or more factors
+
Four Principles of Experimental
Design
■
1) Control aspects of the experiment that we know may have an effect
on the response, but that are no the factors of being studied
■
2) Randomize subjects to treatments to even out effects that we cannot
control
■ 3)
Replicate over as many subjects as possible. Results for a single
subject are just anecdotes. If the subjects of the experiment are not a
representative sample from the population of interest, replicate the entire
study with a different part of the population
■ 4)
Block to reduce the effects of identifiable attributes of the subjects,
preferably controlled
+
Diagrams and Statistically
Significant
Diagrams
■
It is often helpful to diagram the procedure of an
experiment
Statistically Significant
When an observed difference is too large for us to believe that it is likely to
have occurred naturally, we consider the differences to be statistically
significant. Subsequent chapters will show specific calculations and give
rules, but the principle remains the same
+
Control Treatments
■
A baseline (“business as usual”) measurement is called a
control treatment
■
Control Group – The experimental units assigned to a
baseline treatment level, typically either the default treatment,
which is well understood, or a null, placebo treatment. Their
responses provide a basis for comparison
+
Blinding
■
Blinding – Any individual associated with an experiment who is not
aware of how subjects have been allocated to treatment groups is
said to be blind
■
There are 2 main classes of individuals who can affect the outcome
of an experiment: those who could influence the results and those
who evaluate the results
Single-Blind
When every individual in either of these cases is blinded
Double-Blind
When everyone in both classes is blinded
+
Placebos
■
Placebo – A treatment known to have no effect, administered
so that all groups experience the same conditions. Many
subjects respond to such treatment( a response known as the
placebo effect). Only by comparing with a placebo can we be
sure that the observed effect of a treatment is not due simply to
the placebo effect.
■
Placebo effect – The tendency of man human subjects (often
20% or more of experiment subjects) to show a response even
when administered a placebo
+
Blocking
■
Block – When groups of experimental units are similar, it is often a
good idea to gather them together into blocks. By blocking we isolate
the variability attributes to the differences between the blocks so that
we can see the differences caused by the variables of interest.
■
Designs – In a randomized block design, the randomization
occurs within blocks. In a completely randomized design, all
experimental units have an equal chance of receiving any treatment.
■
Matching – In a retrospective or prospective study, subjects who are
similar in ways not under study may be matched and then compared
with each other on the variables of interest. Matching, like blocking,
reduces unwanted variation
+
Confounding and Lurking Variable
■
Confounding – When the levels of one factor are
associated with the levels of another factor so their
effects cannot be separated, we say that these two
factors are confounded
■
A lurking variable creates an association between
two other variables that tempts us to think that one
may cause the other.
+
Question #29
■
Will listening to a Mozart piano sonata make you smarter? In a 1995 study,
Rauscher, Shaw, and Ky reported that when students were given a spatial
reasoning section of a standard IQ test, those who listened to Mozart for 10
minutes improved their scores more than those who simply sat quietly.
A) These researchers said the differences were statistically significant. Explain
what that means in this context.
The differences among Mozart and quiet groups were more than would
have been expected from ordinary sampling
B) Steele, Bass, and Crook tried to repicate the origional study. The subjects were
125 college students who participated in the experiment for course credit. Subjects
firsttook the test. Then they were assigned to one of the three groups: listening to
Mozart piano sonata, listening to music by Philip Glass, and sitting for 10 min in
sielnce. Three days after the treatments, they were retested. Draw a diagram
displaying the design of this experiment
+
#27
Music by
Glass
Pretest
Mozart piano
sonata
Post Test
Silence
C) The box plots on p.315 show the differences in score before and after treatment
for the three groups. Did the Mozart group show improvement?
The Mozart group seemed to have the smallest median difference and
thus the least improvement, but there does not appear to be a significant
difference.
D) Do you think the results prove that listening to Mozart is beneficial?
No, if anything there is less improvement, but the difference does not
seem significant compared with the usual variation
+
Question #29
■
The makers of Frumpies, “the breakfast of rug rats,” want to improve
their marketing, so they consult you:
A) They first want to know what fraction of children ages 10 to 13, like
their celery-flavored cereal. What kind of study should they preform
Observational. Randomly select a group of children, ages 10
to 13, have them taste the cereal, and ask if they like the cereal
B) They are thinking of introducing a new flavor, maple-marshmallow
Frumpies, and want to know whether children will prefer the new flavor
to the old one. Design a completely randomized experiment to
investigate this question
Get volunteers ages 10 to 13. Each volunteer will taste both
cereals, randomizing the order in which they taste them. Compare the
percentage of favorable ratings for each cereal.
+
#29
■
They suspect that children who regularly watch the Saturday
morning cartoon show starring Frump, the flying teenage
warrior rabbit who eats Frumpies in every episode, may
respond differently to the new flavor. How would you take that
into account in your design?
From volunteers, identify the children who watch Frump
and identify the children who do not watch Frump. Use a blocked
design to reduce variation in cereal preference that may be
associated with watching the Frump cartoon