#22 - In a recent study, one group of participants watched

[hadimadi]

Hi,

I feel like the assumptions to make (D) the right choice are too far. What if you only consistently report a higher amount of exercise than your peers only if you know you see yourself, and not someone very similar to yourself?

If we can assume that twins are nearly similar, or similar in the relevant aspects (looks, etc.), then we inherently assume that knowing that you really see yourself, and not someone similar in the mentioned aspects, is irrelevant to this study.

And this is where for me, it goes too far from being accurrate...

I chose (E), as it is additional evidence that people who watch themselves doing something tend to report in general higher average hours of exercise than their peers, and it is unlikely that they always are really doing that activity for longer than their peers on average.

Please help me to understand where I went wrong

Thanks Guys!

[Robert Carroll]

hadimadi,

But if the stimulus has a problem that there is a difference between people actually exercising more and people merely reporting that they exercised more, answer choice (E) has that exact same problem. It doesn't weaken the argument - it provides another situation where watching themselves doing an activity can cause people to report doing that activity more. But, as in the stimulus, if they're actually doing the activity more, that's what the argument is trying to prove! So the exact same question arises for answer choice (E) as for the stimulus - was it an increase in reporting only or in reporting and doing? Answer choice (E) doesn't address that, and nothing in it makes it more plausible that they overestimated than that they correctly reported!

In fact, only answer choice (D) involves that overreporting issue we want!

Robert Carroll

[hadimadi]

Hi Robert,

thanks for your answer, but I don't think it went into what I pointed out:

(E) is, just as you said, just another case of people tending to report higher amounts of an activity when watching themselves doing it than peers. So yes, it is nothing new, however, if we assume this to be the consistent pattern, meaning that people always report higher hours of a specific exercise when they watch themselves doing it as compared to peers, we can conclude that that it is highly unlikely that they also REALLY do more of that exercise than their peers

Also, your answer doesn't explain the deficiencies of (D) that I pointed out, namely that we have to make these hard assumptions as I mentioned before.

I get that when making these assumptions, (D) is perfect because it talks about overreporting, which is a strong counter.

Thanks for always taking the time to reply!

[Adam Tyson]

As I see it, hadimadi, answer E doesn't weaken the argument because it shows a similar outcome as in the stimulus. If anything, that might strengthen the argument, although it seems too vague to be of any real help. The problem with that answer is that it raises no doubts about the original conclusion. Were these people (in answer E) more sedentary than the others, or did they just claim to be? Answer E fails to answer to that question, and so we cannot say whether this answer weakens the argument.

Answer D requires no assumptions about similarities between watching oneself and watching your identical twin. As annoying as that difference may be, it is, in my view, irrelevant to the core issue, which is that answer D indicates that in at least some cases, self-reporting is not reliable. If self-reporting cannot be trusted in that case, why should we trust it in the case described in the stimulus, or in any other case? Pointing out ANY case where self-reporting was inaccurate should be enough, by itself, to at least raise the spectre of bad data in the instant case. It would force the author of the argument to either admit that there could be a problem with their data, or else defend their argument by giving us some reason to believe that there was no similar problem in their study. If the author is forced to defend themselves or else concede, and they cannot simply shrug it off and say "so what, who cares?", then the answer is an attack on their argument. It weakens it, even if only a little.

This is a killer question, primarily because of the "twins/self" difference between the two cases, but that difference is, in my view, nothing but a distraction intended to confuse us. It does that extremely well, as evidenced by how many questions we continue to get about it, and it comes up with every student I speak to who has taken this practice test. This will likely go down as one of the most hated LR questions of all time, and rightfully so!

[Cmoeckel]

I've read through all the responses here and I think what they're missing is the real question which is between weighing ways to attack casualty.

C. This is a classic cause without the effect. They watch a record of themselves exercising but they are not motivated to exercise more.
D. Data attack. Calling into question the self reporting data the author bases their casual connection on.

I was tight on time and so immediately chose C (I know big mistake) and didn't even look at D. In looking back, I see why the data attack might be stronger here ("overreported by a significant amount") vs. the cause without the effect, especially because causal relationship in stimulus is a "can". So my question for you guys...if the "can" was absent from last sentence (i.e. just watching a recording motivates participants to exercise more), they would most probably not have both of these answers right?

[atierney]

At a fundamental level, your characterization of C is correct, my view of C is that we actually have the effect already present.

In other words, those individuals didn't increase their exercise time because they were already highly motivated. This is because the Cause and Effect here is really twofold. There's watching yourself causes you to become have higher motivation, and then because you were more highly motivated, you are now exercising more. In answer choice C, we are taking away the delta of the higher motivation, because that delta is limited in the sample size we're dealing with. These people are already highly motivated, therefore, we can assume (or at least I did when I read this) that there is a law of diminishing returns here at play. You are already at top motivation, so you're not getting substantially **more** motivated by watching yourself. When reading this answer choice, one can almost characterize it as one that strengthens the argument in a way, by demonstrating that motivation was indeed a key factor possibly in the change in running time.

Now in terms of answer choice D, here it implies, and I use this word loosely (this is not my favorite question or answer choice), that there is a motivation, upon viewing "yourself" perform an activity, to "report" subsequent progress on that activity whether or not you honestly engaged upon said progress. This calls into the question the validity of the conclusion by calling into the question the veracity of the reports. In other words, the proposed Cause and Effect relationship is destroyed because there is a potential problem with the data. Problems with data are relatively, but this is a good example of a sneaky one, and I think if you are clear on why D is correct, hopefully that helps with why C is not correct, to the extent that you have questions on the above.

And per usual, ask away!

[Delanoking1]

Hi, Bli,

Good question! Let me clarify.

Answer Choice (C) requires several additional (and unwarranted) assumptions to make it conceivably weaken the conclusion.For instance, to make the statement in (C) weaken the conclusion, we must assume without providing justification that:

• The first group with the longer exercisers comprised mostly "highly motivated" participants, an error in the integrity of the study.
• Within this first group, the causal factor that contributed to the longer exercise was not "watching the video" but instead their already high degree of motivation.
• Thus, watching the videos did not increase the motivation level. Instead, their existing higher level of motivation was the causal factor behind their increased amount of exercise.

Given these wholly unwarranted assumptions, we could imagine a way for (C) to weaken the conclusion. That's what I was referring to when I called it a "fill-in-the-blanks" answer: find a way to fit a square peg in a round hole by making unwarranted assumptions.

Do not do this.

Stick to the plain meaning of the text without trying to "tell a story" to make incorrect answers work. Instead, focus on improving predictions and recognizing the most significant flaw in questions like these so that you can anticipate the most likely way to harm or help a conclusion.

I hope this helps!

Even if we assumed that the highly motivated individuals were what comprised the first group, what we'd have at best is a dilemma that seemingly strengthens the argument. Because AC (C) says the highly motivated individuals do not report any increase, while the stimulus says an increase of 1 hour in average was observed after watching the video. Assuming that highly motivated individuals comprise this group and did not contribute to the average increase only makes it more likely that it was indeed the video that was responsible for the increase as an alternate cause has been eliminated. If we assume on the other hand that highly motivated individuals comprised the second group, then perhaps we could posit that their lack of increase was because they were already exercising at optimal levels. That still does not weaken the observation that a group of people did increase after watching the video. Finally, if we posit that they were proportionately distributed across both groups, then naturally their effects cancel out each other. All this to say, AC (C) cannot quite possibly do anything for you once you understand its implications fully, and was put there by the test creators in the likely chance that you overlook D for introducing a subject that is out of scope.

[Jeff Wren]

Hi Delanoking,

Excellent analysis!

The only point that I will add (not in response to your comment, but regarding this question more generally) is that this argument plays on the distinction between reported data/results versus actual data/results, which is a common theme that has been tested many times on the LSAT.

There have been LR questions focused on reported crimes rates versus actual crime rates, reported cow deaths versus actual cow deaths, etc..

In surveys in particular, the problem of participants answering the survey questions inaccurately (whether they are simply mistaken or outright lying) also comes up periodically.

Knowing this trend and focusing on the word "reported" in the argument can make this tricky question a bit easier.