In addition to evaluating the reliability of evidence, one must ask whether the movement from evidence to claims or sub-claims is warranted. Certain tests for reasoning are especially useful in particular fields (e.g., tests of statistical reasoning). However, some questions about reasoning from evidence are widely helpful. Such questions include the following:
enough evidence provided to warrant
- Does the argument account for
all available evidence on the subject?
- Is evidence cited
relevant to the point being made?For
- If an example, does it represent an instance of the larger point?
- If an expert opinion, is the quotation directly related to the point being made?
- If a report of test results or other measurements, is what is measured relevant to the point being made?
- If findings for a sample, is the sample representative of
the larger population to which generalizations are made?
causal inferences from evidence warranted?
- If evidence of an effect is cited and the inference is that
a cause is or was present, is the cause necessary to produce
If evidence of a cause is cited and the inference is that an effect is or will be present, is the cause sufficient to produce the effect?
- If evidence of an effect is cited and the inference is that a cause is or was present, is the cause necessary to produce the effect?
- In an
analogy inferring what is true in one case from
what is known about another case, are the two cases similar in
all relevant respects?
Another approach to evaluating reasoning involves spotting
logical fallacieserrors of reasoning which occur so
frequently that they have been named. Fallacies often present
reliable evidence, but make flawed inferences from it.
Fallacies may or may not be intentional. Common fallacies
include the following:
- Attacks on character: Arguing that someones ideas should be rejected because he or she has a particular trait, even though the trait isnt relevant to the discussion.
- Begging the question: Circular reasoning in which the assumption that a claim is true serves as support for the claim.
- Hasty generalization: Arguing from too little evidence to a much broader conclusion.
- Over-simplification: Arguing from an analysis that ignores or understates the complexities of a situation. Includes falsely arguing that a single factor causes an effect, when the actual cause involves several factors.
- False dichotomy: Arguing that only two, incompatible choices are available, when in fact the choices are not incompatible or more choices exist.
- Appeal to tradition: Arguing for something on the grounds that it has always been done or believed.
- Irrelevant appeal to authority: Arguing that a statement must be true because a particular source made it, when the source isnt a legitimate authority on the subject.
- Appeal to popular belief: Arguing that because a view is widely held, it must be true.
- Post hoc, ergo propter hoc: Arguing that because X occurred before Y, X caused Y.
- Straw man: Attacking a distorted or misrepresented version of anothers position.
Activities and Assignments
(L) Can be done in large section courses
Whats the Point?
Have students read or listen to an argument and list each piece
of evidence they encounter on the following worksheet. If the
argument is lengthy, assign portions to individuals or groups of
students. Then ask students to complete the remaining
columns. Discuss students responses. If wished,
compare them to a key provided by the instructor.
Used to support point that...
|Does evidence support this point? If not, why?|
After students complete activity 1, ask them to write about or discuss whether the total amount of evidence provided in support of each point in the argument is adequate. (L)
Explain that an argument should account for all available evidence and should not ignore evidence contrary to the claim being made. Have students read or listen to an argument related to the course on a subject with which they are acquainted. Ask them to list evidence that the argument should have considered but did not, then discuss how this evidence might influence or change the outcome of the argument. (L)
Fallacy Scavenger Hunt
Give students a list of logical fallacies, then assign them to find an example of each fallacy on web sites related to the course. If this assignment is graded, consider awarding bonus points for examples not reported by other students. (L)