CR Guide, explain an event or discrepancy example

[contact.sumeshn]

In a recent poll, 71% of respondents reported that they cast votes in the most recent national election. Voting records show, however, that only 60% of eligible voters actually voted in that election.

Which of the following pieces of evidence, if true, would provide the best explanation for the discrepancy?

A. The margin of error for the survey was plus or minus 5% points
B. 15% of the surveys respondents were living overseas the time of the election.
C. Prior research shows that people who actually do vote are also more likely to respond to polls than those who do not vote.
D. ......
E. Polls abt voting behavior typically have margins of error within +/- 3% points

Per explanation, A is ruled out 'coz it discusses about only one premise. E is ruled out coz the answer is categorized under "Wrong direction" type. Now my question is, don't both A and E have the same issue wherein the margin for error does not explain the discrepancy? If that is the case, why is that both are categorized under 2 different type of wrong answers?

Also, i don't completely comprehend how C addresses the discrepancy. Can someone help me in getting a better understanding on this?

Thanks.

[tim]

The difference between the categorizations of A and E relates to the wording of the answer choices. In A, we have information about the margin of error in this particular poll, which is not sufficient to explain the discrepancy. In E, we are told about typical margins of error, which as described in the book serves only to highlight the fact that the discrepancy is actually a discrepancy (the survey results fall outside typical margins of error) without actually explaining it. Ultimately of course when you're taking the test it's not nearly as important to attach a label to a wrong answer as to recognize that it's wrong, so the more important thing to recognize here is that both of these answers are wrong because they fail to explain the discrepancy.

C on the other hand helps us understand that there is really no discrepancy. We're getting different results because the percentages are drawn from different cross-sections of the population (eligible voters versus people inclined to respond to polls), so there is no reason why we should expect the numbers to match up.

[addy199]

This post is also addressed at
http://www.manhattangmat.com/forums/post44266.html#p44266

[tim]

thanks, addy