On Testing for Assumption Recognition (SR-97-01)
Gilbert E. Plumer

Executive Summary

On the LSAT there is an item subtype within the Logical Reasoning sections called the “Assumption” subtype. Questions of this subtype are designed to examine the test taker’s ability to analyze an argument and, on logical grounds, determine which unstated assumptions are—together with the premises stated in the passage—necessary to establish the conclusion of the argument or sufficient to establish it. Sufficient-assumption test items have proven to be relatively unproblematic. Necessary-assumption test items are more controversial. The following remarks by Norris and Ennis are representative of a view that is not uncommon: “Although it is tempting to think that certain [unstated] assumptions are logically necessary for an argument or position, they are not. So do not ask for them ... no significant assumptions are logically necessarily made.” Nevertheless, numerous writers of introductory logic texts, as well as various highly visible standardized tests such as the LSAT and GRE, presume without giving much (or any) justification that the Norris-Ennis view is wrong.

This paper proposes criteria for determining necessary assumptions of arguments, criteria that I think are correct in themselves in addition to being appropriate for writers and reviewers of the kind of test that the LSAT represents: a measure of advanced yet nontechnical reasoning and reading abilities using the format of multiple-choice questions. The main criterion that is advocated and explained is called the “gap-filling criterion.” The idea is that a necessary assumption fills a gap in an argument's reasoning or inferential structure such that if the assumption is presumed to be false the logical cogency of the argument is undercut. The criteria presented are applied to actual disclosed questions from the LSAT.

The criteria that I propose are then defended against the kind of objections that Norris and Ennis raise, among other objections. One important objection is that there is always a logically viable narrower alternative (e.g., “all dogs whose name begins with ‘M’ are animals”) to any putatively necessary assumption ("all dogs are animals") that renders it unnecessary (for the stated argument “since Mike is a dog, Mike is an animal”). My response, in essence, is that the specificity of the context and explicitly given argument determines the specificity of the necessary assumption; so if the former is in certain respects general, the necessary assumption is accordingly general.

The paper closes by canvassing empirical evidence that necessary-assumption test items that are constructed in keeping with the theoretical principles advocated are not distinctively problematic. The main evidence is that assumption items pretested on the LSAT are not rejected in content or statistical review, or identified as marginal, at a rate significantly different than that of items of other comparable Logical Reasoning subtypes that themselves are relatively uncontroversial.


On Testing for Assumption Recognition (SR-97-01)

Research Report Index