Candidate A was widely believed to be the favorite in her state’s gube

Official Explanation

STEP 1: IDENTIFY THE QUESTION TYPE

This question directs you to find an assumption on which the argument depends, so this is definitely an Assumption question.

STEP 2: UNTANGLE THE STIMULUS

The argument concludes that respondents to recent election exit polls and apreelection polls were not being honest when they claimed to have supported Candidate A for governor. The evidence for this is that despite a strong showing in these polls, Candidate A still lost the election.

STEP 3: PREDICT THE ANSWER

This conclusion is based in part on the results of two polls, so those polls need to have been conducted with representative samples in order for the conclusion to be valid. After all, what if the polls had both been conducted outside campaign rallies for Candidate A or in Candidate A’s hometown? The sample group for the polls needs to be an adequate cross section of the voting population, and since this argument stakes its conclusion on the polls, the author of the argument must be assuming that the sample is indeed representative.

This question shows you that Kaplan’s strategy for representativeness is not restricted to Strengthen and Weaken questions. On Test Day, if you’re asked an Assumption question and you notice that the stimulus focuses on a study, survey, poll, or experiment, know that a choice that essentially says, “The sample was representative” is likely to be correct.

STEP 4: EVALUATE THE CHOICES

(E) matches this prediction perfectly and is the correct answer. If you used the Denial Test to negate (E), by stating that the poll’s sample group was not representative, then the author’s conclusion that voters must have lied can no longer be valid. If the people who participated in the polls were not representative of the larger voting population, then there would be no particular reason to expect the poll and voting results to be similar. (E) is therefore a necessary assumption of the argument.

(A) is not necessary to the argument because the author doesn’t base her conclusion on a prediction drawn from the press coverage. Rather, the author bases her conclusion on a prediction drawn from the polling data.

(B), if true, would strengthen the argument, but this isn’t a Strengthen question; the right answer to an Assumption question must be something upon which the argument relies. While the argument asserts that people polled lied about whom they voted for, it does not depend on any particular reason why they
did so.

(C) is far too extreme; the argument’s point that the polls’ respondents lied is not undone if one or two people simply changed their minds.

As for (D), the author doesn’t necessarily assume anything about how Candidate B was able to eke out a victory.

Answer: E