FRHD 3070 Chapter 9 - Experimental Research Exam
What kinds of questions are appropriate for using an experimental logic? What kinds of
limitations do researchers employing experimental methods face? What sorts of
research topics are best suited for experiments?
The logic of experiments is that experimenters intervene or induce a change in some
focused part of social life and then examine the outcomes that result from the
intervention.
The experiment is limited to research questions in which a researcher is able to
manipulate, control, or simulate conditions.
The experiment is usually best for issues that have a narrow scope or scale.
The experiment is better suited for micro-level than for macro-level questions.
Experimental research is the strongest for testing causal relationships because the
three conditions for causality (temporal order, association, and no alternative
explanations) are best met in experimental designs.
Experiments are not good at addressing questions that require looking at conditions
across an entire society or across decades.
Distinguish between random assignment and random selection and then describe how
the two are connected. How does random assignment work in practice?
Random assignment is a mechanical and automatic method (the researcher’s
personal preferences do not play a factor) for assigning cases such as individuals or
organizations into two or more groups for the purpose of increasing one’s
confidence that the groups do not differ in some systematic way.
Random selection lets a researcher calculate the odds that a specific case will be
sorted into one group over another.
When a researcher randomly assigns, he or she sorts a collection of cases into two or
more groups using a random process; in random sampling, he or she selects a smaller
subset of cases from a larger pool of cases.
Ideally, a researcher will both randomly sample and randomly assign. He or she can first
sample to obtain a smaller set of cases (e.g., 150 people out of 20 000) and then use
random assignment to divide the sample into groups (e.g., divide the 150 people into
,three groups of 50).
Identify the two groups in a simple experiment and specify which one receives the
treatment. What determines what the independent and dependent variables are?
The experimental group is the group that receives the treatment or in which the
treatment is present and the group that does not receive the treatment is called the
control group.
The independent variable is the treatment. Such as, for example, “degree of fear
or anxiety.”
The dependent variable is the outcome. It might consist of the physical conditions,
social behaviours, attitudes, feelings, or beliefs of subjects that change in response to
or as a result of the treatment.
Why is control a necessary and crucial feature in experiments? How is deception a
feature of control in experiments? In what circumstances is it ethical to use deception in
experiments?
A researcher wants to control all aspects of the experimental situation to isolate the
effects of the treatment and eliminate alternative explanations.
Through deception, the researcher tries to control what the subjects see and hear and
what they believe is occurring.
For realistic deception, researchers may invent false treatments and dependent
variable measures to keep subjects unaware of the true ones.
Deception is dishonesty and thus is usually not condoned; it is acceptable only as the
means of achieving a goal that could not be achieved otherwise.
The amount and type of deception should not go beyond what is minimally necessary,
and research participants should be debriefed.
Compare the classic experimental design with each of the three pre-experimental
designs. What are the problems with each of the pre-experimental designs?
Classical experimental design has random assignment, a pretest and a post-test, an
,experimental group, and a control group.
The one-shot case study design has only one group, a treatment, and a post-test; since
there is only one group, there is no random assignment.
The one-group pretest-post-test design has only one group, a pretest, a treatment, and
a post-test; it lacks a control group and random assignment.
The static group comparison design has two groups, a post-test, and treatment but it
lacks random assignment and a pretest.
Compare and contrast interrupted and equivalent time series designs. What is common
to both of them? What is unique to each of them? How do interrupted and equivalent
time series designs differ from pre-experimental designs?
In an interrupted time series design, a researcher uses one group and makes multiple
pretest measures before and after the treatment.
An equivalent time series design is a one-group design that extends over a time period
so that instead of one treatment, it has a pretest and then a treatment and post-test,
followed by another treatment and post-test, followed by yet another treatment and
post-test, and so on.
Interrupted and equivalent time series designs differ from pre-experimental designs in
that identifying a causal relationship is more certain in these designs than in
pre-experimental designs.
Diagram the design notation for a classical experimental design. What do each of the
symbols represent?
O = observation of dependent variable; X = treatment, independent variable; R = random
assignment.
Symbols are in time order from left to right: The R is first, followed by the pretest, the
treatment, and then the post-test.
Symbols are arranged in rows, with each row representing a group of subjects; the rows
are on top of each other because the pretest, treatment, and post-test occur in each
group at about the same time.
, In experimental research, what does internal validity refer to? Provide and briefly
describe three examples where internal validity is an issue.
Internal validity refers to the ability to eliminate alternative explanations of the
dependent variable—variables, other than the treatment variable(s), that affect the
dependent variable are threats to internal validity. These variables threaten the
researcher’s ability to say that the treatment was the true causal factor
producing change in the dependent variable.
Students can discuss any three of the following instances where internal validity is an
issue:<ol type="a">
Selection bias is the threat that research participants will not form equivalent groups.
History effects refer to the threat that an event unrelated to the treatment will occur
during the experiment and influence the dependent variable.
Maturation is the threat that some biological, psychological, or emotional process within
the subjects and separate from the treatment will change over time.
Testing effect threatens internal validity because more than the treatment alone affects
the dependent variable.
Instrumentation occurs when the dependent variable measures changes during the
experiment.
Statistical regression refers to a problem of extreme values or a tendency for random
errors to move group results toward the average.
Mortality arises when some subjects in the experiment leave the experiment before it is
concluded.
What kinds of questions are appropriate for using an experimental logic? What kinds of
limitations do researchers employing experimental methods face? What sorts of
research topics are best suited for experiments?
The logic of experiments is that experimenters intervene or induce a change in some
focused part of social life and then examine the outcomes that result from the
intervention.
The experiment is limited to research questions in which a researcher is able to
manipulate, control, or simulate conditions.
The experiment is usually best for issues that have a narrow scope or scale.
The experiment is better suited for micro-level than for macro-level questions.
Experimental research is the strongest for testing causal relationships because the
three conditions for causality (temporal order, association, and no alternative
explanations) are best met in experimental designs.
Experiments are not good at addressing questions that require looking at conditions
across an entire society or across decades.
Distinguish between random assignment and random selection and then describe how
the two are connected. How does random assignment work in practice?
Random assignment is a mechanical and automatic method (the researcher’s
personal preferences do not play a factor) for assigning cases such as individuals or
organizations into two or more groups for the purpose of increasing one’s
confidence that the groups do not differ in some systematic way.
Random selection lets a researcher calculate the odds that a specific case will be
sorted into one group over another.
When a researcher randomly assigns, he or she sorts a collection of cases into two or
more groups using a random process; in random sampling, he or she selects a smaller
subset of cases from a larger pool of cases.
Ideally, a researcher will both randomly sample and randomly assign. He or she can first
sample to obtain a smaller set of cases (e.g., 150 people out of 20 000) and then use
random assignment to divide the sample into groups (e.g., divide the 150 people into
,three groups of 50).
Identify the two groups in a simple experiment and specify which one receives the
treatment. What determines what the independent and dependent variables are?
The experimental group is the group that receives the treatment or in which the
treatment is present and the group that does not receive the treatment is called the
control group.
The independent variable is the treatment. Such as, for example, “degree of fear
or anxiety.”
The dependent variable is the outcome. It might consist of the physical conditions,
social behaviours, attitudes, feelings, or beliefs of subjects that change in response to
or as a result of the treatment.
Why is control a necessary and crucial feature in experiments? How is deception a
feature of control in experiments? In what circumstances is it ethical to use deception in
experiments?
A researcher wants to control all aspects of the experimental situation to isolate the
effects of the treatment and eliminate alternative explanations.
Through deception, the researcher tries to control what the subjects see and hear and
what they believe is occurring.
For realistic deception, researchers may invent false treatments and dependent
variable measures to keep subjects unaware of the true ones.
Deception is dishonesty and thus is usually not condoned; it is acceptable only as the
means of achieving a goal that could not be achieved otherwise.
The amount and type of deception should not go beyond what is minimally necessary,
and research participants should be debriefed.
Compare the classic experimental design with each of the three pre-experimental
designs. What are the problems with each of the pre-experimental designs?
Classical experimental design has random assignment, a pretest and a post-test, an
,experimental group, and a control group.
The one-shot case study design has only one group, a treatment, and a post-test; since
there is only one group, there is no random assignment.
The one-group pretest-post-test design has only one group, a pretest, a treatment, and
a post-test; it lacks a control group and random assignment.
The static group comparison design has two groups, a post-test, and treatment but it
lacks random assignment and a pretest.
Compare and contrast interrupted and equivalent time series designs. What is common
to both of them? What is unique to each of them? How do interrupted and equivalent
time series designs differ from pre-experimental designs?
In an interrupted time series design, a researcher uses one group and makes multiple
pretest measures before and after the treatment.
An equivalent time series design is a one-group design that extends over a time period
so that instead of one treatment, it has a pretest and then a treatment and post-test,
followed by another treatment and post-test, followed by yet another treatment and
post-test, and so on.
Interrupted and equivalent time series designs differ from pre-experimental designs in
that identifying a causal relationship is more certain in these designs than in
pre-experimental designs.
Diagram the design notation for a classical experimental design. What do each of the
symbols represent?
O = observation of dependent variable; X = treatment, independent variable; R = random
assignment.
Symbols are in time order from left to right: The R is first, followed by the pretest, the
treatment, and then the post-test.
Symbols are arranged in rows, with each row representing a group of subjects; the rows
are on top of each other because the pretest, treatment, and post-test occur in each
group at about the same time.
, In experimental research, what does internal validity refer to? Provide and briefly
describe three examples where internal validity is an issue.
Internal validity refers to the ability to eliminate alternative explanations of the
dependent variable—variables, other than the treatment variable(s), that affect the
dependent variable are threats to internal validity. These variables threaten the
researcher’s ability to say that the treatment was the true causal factor
producing change in the dependent variable.
Students can discuss any three of the following instances where internal validity is an
issue:<ol type="a">
Selection bias is the threat that research participants will not form equivalent groups.
History effects refer to the threat that an event unrelated to the treatment will occur
during the experiment and influence the dependent variable.
Maturation is the threat that some biological, psychological, or emotional process within
the subjects and separate from the treatment will change over time.
Testing effect threatens internal validity because more than the treatment alone affects
the dependent variable.
Instrumentation occurs when the dependent variable measures changes during the
experiment.
Statistical regression refers to a problem of extreme values or a tendency for random
errors to move group results toward the average.
Mortality arises when some subjects in the experiment leave the experiment before it is
concluded.
