Chapter 12: Glossary
This alphabetical glossary contains short definitions for all the important terms and concepts from the chapter. You will also find hyperlinks to Websites relevant to the study of these terms and concepts. You should employ good critical thinking when evaluating the merit of any information you find on the World Wide Web, including what you find by following these links.
Assuming a common cause: An error in causal reasoning that claims a link between two effects where none exists; a bad form of reasoning on the basis of a common thread.
C: In studies of causation in a population, an abbreviation for the suspected causal factor.
Causal argument: An argument that offers support for a causal claim, that is, for a claim that one thing causes another. the form of such arguments depends on whether the causation concerns specific events or populations.
- Mission: Critical (Introduction to Causal Arguments)
- Causal Arguments: Introduction to Logic site gives a listing and discussion of terms having to do with Causal Argument and Causal Explanation.
Causation among specific events: A link between one event and another, such that the first brought the second about. We usually establish such causation on the grounds of a difference or a common thread.
Causation in populations: a kind of causal effect relevant to a population. To say that C causes E in a population P is to say that C is on factor producing P. Such causation is determined through a controlled cause-to-effect experiment, a nonexperimental cause-to-effect study, or a nonexperimental effect-to-cause study.
Common thread: In arguments about specific events, a reason for seeing X as the cause of Y. X must be the only factor common to more than one example of Y; and the examples of Y should not be linked by chance.
Control group: In studies of causation in a population, a sample of the target population whose members are not exposed to C (or known not to display E). In all other respects, the members of the control group are treated exactly like the members of the experimental group.
Controlled cause-to-effect experiment: a study intended to demonstrate causation in populations, by means of the exposure of an experimental group, but not a control group, to a suspected causal agent C.
- Welcome to the Ammonia Factory: This interesting site offers a simulation to help you understand the principles of experimental design.
- Therapeutic Touch Study Data Good site which explores controlled cause to effect experiments.
- Skeptic's Dictionary: More good information on cause to effect experiments.
- Skeptic's Dictionary: General information on cause to effect experiments: control groups, double-blind and randomized tests, etc.
- Research Design Slide Show from Kansas State University
- Research Studies: General terminology definitions.
d: In studies of causation in a population, an abbreviation for the difference between frequency of effect (or cause) in the control group.
Difference: In arguments about specific events, a reason for seeing X as the cause of Y. X must be the only relevant difference between a situation in which Y occurs and situations in which Y does not occur.
E: In studies of causation in a population, and abbreviation for the given or suspected effect.
Experimental group: In studies of causation in a population, a sample of the target population whose members are exposed to C (or know to display E).
Frequency of effect: The portion of the experimental or control group (measured as a percentage) that displays E.
Ignoring a possible common cause: An error in causal reasoning that makes X the cause of Y, when some other event W might have caused both X and Y.
Nonexperimental cause-to-effect study: A study intended to demonstrated causation in a population, by means of an experimental group that has already been exposed to suspected causal agent C, and a control group that has not.
Nonexperimental effect-to-cause study: A study intended to demonstrated causation in a population, by means of an experimental group that has already displayed effect E, and a control group that has not.
Post hoc, ergo propter hoc: An error in causal reasoning that makes the mere appearance of Y after X a reason for calling X the cause of Y.
Reversing causation: An error in causal reasoning that correctly attributes a causal link to two events, but calls X the cause of Y when really Y has caused X.
Statistical significance: The sign that a measure of d could not have arisen by chance. We speak of d's being statistically significant given a certain confidence level; so if the level is 95 percent, to call d statistically significant is to say there is only a 5 percent chance that d could have been a random result.
