Tuesday, August 10, 2010

Logical Reasoning Zen Task Standard: LSAT Inference

Today we'll explain how we approach one of the tasks, infer, which is where the LSAT asks you to extrapolate a new conclusion by combining the fact patterns and evidence presented in a stimulus.

This task can be prompted by a wide variety of question stems, but most of them will mention "infer," "inference," or "[new] conclusion." On modern LSATs, one of the most common versions of this question stem is:
Which one of the following statements can be properly inferred from the statements above?
Two general structures that Zen of 180 students use to visually represent the extrapolation strand tasks

In the example on the left, the credited answer (C), is a necessary condition interpolated between two or more pieces of evidence presented in the stimulus (A and B), i.e. in order for B to be true in the context of A, C must also be true.  In the diagram on the right, the credited response is a combination of two or more pieces of evidence from the stimulus, that when combined lead to a new conclusion.

In comparison to must also be true, the infer task is more concerned with conclusions and is slightly less stringent in terms of logical deductions.  In order to correctly identify the inference, you may be required to read between the lines more than required for the more concrete task.  Conversely, most supported by [stimulus], is open to more loose interpretations, including analogies and applications to new situations.

The key to correctly answering an infer question is to clearly highlight the main pieces of evidence--the actors and their definitions--and how they can be combined in terms of topic, degree, certainty, and opinion.  As with depends upon assuming, the goal is not to describe the specific evidence--as the LSAT answers will invariably mention the "correct" parts--but rather the links in relationship and the degree of certainty between the pieces of the stimulus.

All the free Zen of 180 explanations are for the two sample PrepTests, from June 2007 and October 1996. The question stem for this task is fairly easy to identify, so be sure to actively read it to determine which task, zooming in our out, you're asked to perform. Marking the stimulus for actors and relationships with both highlighter and pencil will help you make a clean prephrase of the generalizations you'll look for in the answer choices.



Above is a gallery of the logical reasoning sections that have a greater than average density of the task, and below is the frequency that this task has been asked on modern LSATs and the percentage change in frequency from pre-2007 LSATs. It averages 2.2% of the points on a modern LSAT, and you can reasonably expect one point to be devoted to exclusively making an inference.
    10 times
    on LSATs since 2007
    2.2% of LR
    (1 per LSAT, range 0-2)
    -1.6% growth from pre-2007
This task appears less frequently on modern LSATs than in older PrepTests.  Be sure to clearly keep in mind how "far away" from the stimulus the task allows you to go while extrapolating.  When ordered by how limited the scope of the answer can be compared to the stimulus, infer tasks are in the middle of the extrapolation strand, with must also be true more stringent and most strongly supported by [stimulus] much less so.