Chapter 13: Experiments and Observational Studies

What is an observational study?
-a study based on data in which no manipulation of factors has been employed
What is a retrospective study?
-an observational study in which subjects are selected and then their previous conditions or behaviors are determined
Do retrospective studies need to be based on random samples?
-no
In what fields are observational studies widely used?
-public health
-marketing
What do retrospective studies usually focus on?
-estimating differences among groups or associations between variables
What is a prospective study?
-an observational study in which subjects are followed to observe future outcomes
We will write a custom essay sample on
Any topic specifically for you
For only $13.90/page
Order Now
Is a prospective study an experiment? Why or why not?
-because no treatments are deliberately applied, a prospective study is not an experiment
What do experiments study?
-the relationship between two or more variables
What is a factor?
-a variable whose levels are manipulated by an experimenter
What is a response variable?
-a variable whose values are compared across different treatments
What do experiments attempt to discover?
-the effects that differences in factor levels may have on the responses of the experimental units
What is one important requirement of an experiment?
-a random assignment of subjects to treatments
What are subjects/participants?
-the individuals who participate in an experiment, especially when they are human
What is a more general term that is often used instead of “subject” or “participant?”
-experimental unit
What distinguishes an experiment from other types of investigation?
-that 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
What are the specific values that an experimenter chooses for a factor called?
-levels
What is the “treatment” that an experimenta unit/subject receives?
-the combination of specific levels from all of the factors that an experimental unit receives
What are the four principles of experimental design?
1. Control
2. Randomize
3. Replicate
4. Block
What does the principle of “control” specify?
-we must CONTROL aspects of the experiment that we know may have an effect on the response, but that are not factors being studied
Why is it advantageous to control extraneous sources of variation?
-it makes it easier to detect differences among the treatment groups
What does the principle of “randomize” specify?
-we must RANDOMIZE subjects to treatments to even out effects that we cannot control
Randomization does not eliminate the effect of these effects. What positive impact does it have?
-it should spread the effects across treatment levels so that we can see past them
What does the principle of “replicate” specify?
-we must REPLICATE over as many subjects as possible
What does the principle of “block” specify?
-we must BLOCK/reduce the effects of identifiable attributes of the subjects that cannot be controlled
What is blocking an important compromise between?
-randomization and control
Which of the four principles of experimental design are required?
-control, randomize, and replicate
-(blocking is not required in an experimental design)
When do we consider differences to be “statistically significant?”
-when an observed difference is too large for us to believe that it is likely to have occurred naturally
What can be used to confirm statistical significance?
-a variety of different statistical tests
What is a control group?
-the experimental units assigned to a baseline treatment level
What are two typical “baseline treatment” levels that control groups are assigned to?
1. the default treatment, which is well understood
2. a null, placebo treatment
What two main classes of individuals can affect the outcome of an experiment?
1. those who could influence the results
2. those who evaluate the results
What type of participants can influence results?
-subjects
-treatment administrators
-technicians
What type of participants evaluate results?
-judges
-treating physicians
What is a single-blind experiment?
-an experiment in which individuals in either one of the two classes (those who influence and those who evaluate) are blinded
What is a double-blind experiment?
-an experiment in which individuals in both classes (those who influence and those who evaluate) are blinded
What is a placebo?
-a treatment known to have no effect
Why is a placebo administered?
-so that all groups experience the same conditions
What is the “placebo effect?”
-the tendency of many human subjects to show a response even when administered a placebo
What does the placebo effect highlight?
-the importance of effective blinding and the importance of comparing treatments with a control
What four traits characterize the best experiments?
They are:
1. Randomized
2. Comparative
3. Double-blind
4. Placebo-controlled
When is blocking used?
-when groups of experimental units are similar
What is blocking used to accomplish?
-to isolate the variability attributable to differences between the blocks so that we can see the differences caused by the treatments more clearly
Why is it often important to include several factors in the same experiment?
-to see what happens when the factor levels are applied in different combinations
When does “confounding” occur?
-when the levels of one factor are associated with the levels of another factor in such a way that their effects cannot be separated
What two terms are synonymous with “explanatory variable?”
-factor
-independent variable
What is an explanatory variable?
-any variable that explains the response variable
What term is synonymous with “response variable?”
-dependent variable
What is a response variable?
-the outcome of a study
-a variable you would be interested in predicting or forecasting
What is an experiment?
-a study design that allows us to show a cause-and-effect relationship
In general what are the individuals on whom or which we experiment called?
-experimental units
What are the four principles of experimental design?
1. Control
2. Randomize
3. Replicate
4. Block
When is a difference statistically significant?
-if we don’t believe that it’s likely to have occurred only by chance
What is a baseline measurement called?
-a control treatment
What is a control group?
-the experimental units to whom a control treatment is applied
What is a placebo?
-a fake treatment that looks just like the treatment being tested
What are the two main classes of individuals who can affect the outcome of an experiment?
1. those who could influence the results
2. those who evaluate the results
What is a confounding variable?
-a variable that is related to the explanatory variable that affects the response variable