Pundit: By passing the recent tax bill, the Republicans have likely

There seems to be a lot of discussion and confusion about how to tackle this question, so here is the full "GMAT Jujitsu" for this question:

Our first strategic item of business is to use leverage from the question stem to identify the problem type. This one isn't subtle; after all, the question asks for an "assumption on which the... argument relies." Assumption questions are a sub-type of traditional Strengthen questions, but for those of you studying for the GMAT, it might be worth noting that Assumption questions play by slightly different rules.

If a question asks you for an unstated fact (in other words, an "assumption") upon which an argument "relies", that is a pretty high bar. With regular Strengthen questions, any answer choice that helps the conclusion to be valid can strengthen the argument. However, when a problem asks for an "assumption on which the... argument relies," you are looking for an answer choice FUNDAMENTALLY CRITICAL to the viability of the conclusion. In other words, you might look at the correct answer of an Assumption question and say to yourself, "Self, if that answer isn't a reality, there NO WAY the conclusion could function." This is why the tactic I call my classes the "Assumption Negation Technique" is so useful. The basics of the Assumption Negation Technique is this: if the OPPOSITE of an answer choice makes it impossible for the original conclusion to be true, then the original (non-negated) answer choice must be necessary for the conclusion.

However, in my opinion, the Assumption Negation Technique can be both overused and misused. First of all, sometimes it is quite difficult to logically negate a complex sentence. Then, once you have done that, many test takers start trying to add in all sorts of extra information into their logic to "prove" that the negated statement is crucial. For me, I only use the Assumption Negation technique once I have easily eliminated most of the other answer choices using regular logic. It is much easier to use when you are down to a 50-50 choice between two answer choices. Trying to negate all five answer choices -- and then evaluating them to see if their logical opposites undermine the conclusion -- can be a time-consuming process. I don't want to work harder than I have to. I bet neither do you. So, let's evaluate this question, but let's do it intelligently.

There is a clear logical gap between the conclusion and the rest of the argument. The pundit seems to be arguing that the Republicans' passing of the tax bill will cause them to lose their majority in an upcoming election. You could drive a truck through this logical gap. In my classes, I call this type of error "Correlation is not Causation." Just because the Republicans passed a tax bill before an upcoming election doesn't mean the tax bill -- and the tax bill alone -- would cause them to lose the majority. The problem states that ever since the bill, the Republicans have been trailing their opponents. But could there be other factors that might cause this? Who says the tax bill is even a deciding factor in the election? We need to find an answer choice that connects the polling numbers to the tax bill. Anything else is a trap.

Answer choice "A" is a classic trap I call "Overkill." First of all, extreme statements that begin with "never", "always", or "none" are rarely NECESSARY for an argument. Even if the tax bill was a major sticking point in the election, is it easily possible that a few opponents on the other side of the aisle liked the bill. The argument doesn't require "none." Get rid of "A". (Incidentally, an extreme statement like "A" could be used in a regular Strengthen question. There is a reason why I said Assumption questions play by slightly different rules.)

Answer choice "B" also contains a classic trap of the GMAT, but for entirely different reasons. If you eliminated answer choice "B", the test totally used your own thinking momentum against you. Many people might read "B" and think: "But 'substantially different' could be either higher or lower!" Yes, "substantially different" could mean higher or lower. And yet, think about it: the primary assumption of the pundit is that there is a causal link between the bill and the polling numbers. If there was no substantial difference in the polls before and after the tax bill, then the tax bill clearly couldn't have been a cause of the polling numbers! Answer choice "B" is necessary for the argument. (By the way, can you see how I modeled the "Assumption Negation Technique" here? It isn't used haphazardly. You have to understand what the problem is asking, as well as understand the logical gap!)

Many test takers incorrectly eliminate answer choice "B" because they misunderstand the structure of the question. If this were a regular "Strengthen" question (where you are simply looking for information that helps the conclusion) then "B" would be a poor choice. After all, "substantially different" could mean lower or higher. But this is not a regular "Strengthen" question. The question stem explicitly asks for something upon which the argument "relies." The argument assumes that the tax bill caused a difference in the polling. If the tax bill didn't cause a difference, the argument fails. Answer choice "B" says there is a before-and-after difference. We don't know what that difference is, but the argument needs there to be a difference! "B" is it.

Answer choice "C" tries to sound complicated by inserting awkward negation. Don't fall for it. If you slow down and pick apart the answer, you should be able see that, if anything, "C" would actually help -- not hinder -- the Republicans. It says that the tax bill is not "less popular" with likely voters. If it is not less popular, then the bill could be more popular with likely voters. This would help the Republicans, not hurt them. "C" is easy to eliminate.

"D" is a very common wrong answer. When you contrast "D" with "C", it sounds like "D" isn't far from the mark. This answer choice strengthens the argument by implying that, on average, likely voters do not support the tax bill. But this is not a regular "Strengthen" question. We need to focus on what the question is actually asking: we are looking for an assumption necessary for the conclusion to function. While "D" helps the argument, it is not necessary for the argument to be valid. Think about it: what if 10% of the voters don't like the tax bill, 9% like it, and 81% don't care at all? A higher percentage of voters might not like the tax bill, but we still don't know if the tax bill even matters to the majority of the public! Additionally, while likely voters might not like the tax bill, it is equally possible that the polling results are due to something else. What if the Republicans in this question were already behind in the polls? What if they were going to lose the majority anyway? If this were the case, then the pundit's assumption that the reaction to the tax bill caused the poll results (and will cause problems with the election down the road) falls apart. "D" is useful, but isn't required by the argument. Get rid of it.

A parallel argument might be instructive here. Imagine if the problem said, "Candidate A has really bad facial acne. Most people don't like facial acne. Therefore, Candidate A is going to lose the election." Yes, it might be true that most people dislike zits. But we still can't assume a causal link between acne and elections. Ugly candidates get elected to office all the time! In like manner, just because people don't like the bill, this still doesn't mean that this is what is causing the polling numbers to shift. The argument needs something more... something to show that the passing of the tax bill truly changed how people feel about the candidates. That is what the argument requires, and answer choice "D" doesn't quite get there.

Finally, "E" can be eliminated for the same reasons as "A". It is "Overkill" (a common trap in Assumption questions.) It isn't necessary that the tax bill was the "most unpopular piece of legislation." While answer "E" might be a great option if this were a Strengthen question, we can't argue that the argument hinges on such an extreme statement.

Now, for those of you studying for the GMAT, here is a review of some of the major takeaways with this question: First, notice the difference in approach between Strengthen questions and Assumption questions. Yes, these question types are related. Both question types require you to recognize the logical gap between the conclusion and the rest of the argument. But Assumption questions have a higher standard. 3 of the 5 answer choices in this problem could arguably strengthen the argument (A, D, and E), but that is not sufficient. Answer choice C actually undermines the pundit. In the end, there is only one answer upon which the argument "relies" -- "B". This problem is a great reminder for making sure you know exactly what the question is asking.