Exam 6: Causation and Experimental Design

The process through which the independent variable creates changes in a dependent variable is known as a(n):

Is it possible to determine whether a treatment led to a change in the value of cases on the dependent variable in an experiment in which there was no pretest? Explain your answer.

In an experiment, measurement of the dependent variable after the experimental treatment is known as:

It is never ethical to randomly assign treatment to subjects in experiment just because the effect of the treatment is unknown.

While researchers changed the work environment to see how it affected worker productivity, they got surprising results. No matter what conditions the researchers changed (such as increasing the lighting), the workers seemed to work harder simply because they were part of a special experiment. This is the origins of a concern about internal invalidity in experiments known as:

In what situations is it ethnical to deceive participants in an experiment? Provide an example that would ethically allow deception and an example in which the ethics of deception are more ambiguous.

A politician campaigns on a platform of eliminating welfare benefits. One ad used by politician says: "Receipt of welfare benefits leads to poor school performance: People who receive welfare benefits have, on average, lower grades than people who do not receive welfare benefits." Construct an argument to challenge this causal reasoning. First, give three examples that drive home the point that association does not establish causation. Then, construct a plausible alterative explanation for low grades that does not involve an effect of welfare benefits on grades, but which does allow for their association.

Why aren't experiments very generalizable? What can be done to improve the external invalidity of experiments?

Use the following to answer questions 16-18: In each set, match a concept from Group A to a definition or example from Group B. -Experimental Designs Group A 1. True Experiment 2. Non-Equivalent Control Group 3. Before-and-After 4. Ex Post Facto Control Group 5. Solomon Four Group Group B a. Randomized assignment into control and experimental groups b. Lacks a control group c. Contains two sets of control groups and two sets of experimental groups d. Experimental and control groups are designated after experimental treatment e. Lacks a posttest f. Control group is created not by randomization, but to be as similar as possible to the experimental group

Why aren't context and mechanism as critical as time order, association, and nonspuriousness for establishing causation?

In terms of validity, experiments are weakest in terms of establishing:

To say that variables have an association is to say that:

Quasi-experimental designs are like true experiments, except:

Which of the following is not an example of endogenous change, in which internal validity is compromised because of natural developments in the subjects, independent of experimental treatment?

A true experiment allows the establishment of nonspuriousness through:

Aggregate matching is when individual cases in the treatment group are matched with similar individuals in the comparison group.

A nonspurious relationship between variables is a false relationship.

Use the following to answer questions 16-18: In each set, match a concept from Group A to a definition or example from Group B. -Criterion of Causality Group A 1. Association 2. Context 3. Mechanism 4. Nonspuriousness 5. Time Order Group B a. Identification of other variables that allow for a relationship between independent and dependent variables b. The process that creates the connection between the independent and dependent variables c. The degree to which the relationship can be modeled in other experiments d. The relationship between independent and dependent variables is not due to a third variable e. Observed correlation between the dependent and independent variables f. The temporal priority of the independent variable

Discuss the multiple forms of internal validity that create problems in experiments. (There are more than ten!) Describe each in your own words and provide an example that demonstrates why each potentially limits validity.