#20 - Journalist: Newspapers generally report on only

[Jonathan Evans]

15 Veries, I'm going to move this over to the other question section and respond to it there.

[ericau02]

Hi can you please explain why ac B is incorrect?

[Brook Miscoski]

Erica,

(B) is wrong because the author doesn't draw any conclusions about strength of evidence. The information about reliability and strength of evidence is not used, so it can't be part of the flaw. The author's argument is:

Premise 1. Mostly, Newspapers report on Dramatic Studies.
Premise 2. Mostly, Newspapers report on Small Studies.
Conclusion. Small Studies are more Dramatic than are Large Studies.

The flaw is that newspapers could select for drama and simply have thousands of small studies to pick from, answer choice (D). But reliability and strength are red herrings--they're not actually part of the logical structure of the argument. Be careful to identify the parts of the stimulus that play an argumentative role.

[cgs174]

hi!!

I found this question to be the most difficult on the Logical Reasoning in this test. I identified two premises:

1. Scientific studies published by newspapers generally have findings which sound dramatic
2. Scientific studies published by newspapers are more frequently small observational studies rather than large randomized trials.

With the conclusion being an invalid combination of both premises. In my analysis the flaw with this argument was that the author concluded that because two things were independently true, a conclusion could be made by combining the two. I quickly eliminated A) because it was irrelevant and not factually accurate and E) because it was counter to what was stated in the stimulus. I then eliminated C) because I couldn't identify a similar claim and those made were not conflated.

My problem was in my analysis of D) in relationship to the conclusion. As the conclusion states "a small observation study" and "a large randomized trial" I understood this to mean that if you picked at random one small observational study and one large randomized trial, you have a better chance of the small study having dramatic findings than the large trial. I saw this being a question that dealt with percentages not raw numbers. Because of that, I thought small observational studies being more common would not have an impact on the percentage total of dramatic findings, making D) useless. I thought B) was referencing the large randomized trials when it referred to studies with strong findings. In this case, I interpreted it as saying that the large randomized trials reported, while a smaller number of studies reported on, could have 100% dramatic findings, while a smaller % of the newspaper stories had dramatic results. In this case the opposite of the conclusion made could be drawn. Studying it again however I'm concerned B) may be a mistaken reversal.

Sorry for the long thought explanation, I'm still not happy with D) and not sure exactly where my reasoning was incorrect.

[Adam Tyson]

The reason the two premises cannot be combined to get to the conclusion, cgs174, is because one premises is about percentages ("generally" means more than half the time, a percentages concept) and the other is about numbers ("more frequently" just means a larger number, without expressing any percentages or other comparisons to a total). Having more of one kind than the other does not support the claim that the first kind is more likely to have that characteristic. What if there are billions of small observational studies, and only 5% have dramatic findings, while there are just a dozen large randomized trials and all of them have dramatic findings? We could publish all the large randomized results and still publish far more results from small observational studies. Their higher frequency would tell you nothing about the odds of them being dramatic!

So, your reason for rejecting answer D based on there being nothing in the argument about raw numbers was where you went astray. Although no specific numbers were given, the concept of numbers (as opposed to the concept of percentages) WAS an essential element of the argument and was the source of the flaw.

Beware of arguments that try to combine claims about likelihood/probability/rate/incidence, etc. (percentages) with claims about frequency/amount (numbers). Without more data, like the relative sizes of the groups being compared, we may be unable to draw any valid conclusions!

[teddykim100]

Hello,

I get why D is correct, however, I have a hard time settling down on it and deciding it to be the correct answer because

1. the question asks us to describe a flaw, yet D seems to just provide ONE alternate explanation. Can someone break down how presenting one potential alternate explanation is equivalent to describing a general flaw?

2. Since this is an ultimately causal conclusion, erroneously made by the author, why wouldn't E be correct? I know this comes down to me not understanding the language of E, but at the same time the two ways to counter a CAUSAL argument is to either present an alternate explanation (like D), or say that B causes A (instead of A causing B)

isn't that what E does?

[Chandler H]

Hi Teddy,

With regard to your first question, we are trying to find a flaw in the reasoning that leads the journalist from the two premises to the conclusion. Here are our two premises:

Premise 1: Newspapers generally report on scientific studies whose findings sound dramatic, and
Premise 2: Newspapers write more frequently about small observational studies than large randomized trials, even though the small studies are more likely to be unreliable.

Based on those, the journalist makes this conclusion:

Conclusion: A small observational study must be more likely to have dramatic findings than a large randomized trial.

Our task is to identify the error in reasoning in that jump from premises to conclusion. One classic error of reasoning is failing to consider an alternate cause for an effect. (That falls under the Mistaken Cause and Effect category of errors.) That is what the journalist does here—they fail to consider one alternate cause (small studies simply being much more common than large ones) for the described effect (newspapers write more frequently about small studies). Therefore, the failure to consider this obvious alternate explanation IS a general flaw in the reasoning.

As for your second question about answer choice (E), the main issue with this answer choice is that it doesn't actually answer the question of why small studies are reported on more frequently than large trials, even though they're more unreliable. Even if it's true that a study's being reported on makes it more dramatic (as choice (E) states), there's still no given reason why newspapers should report on small studies and not large trials.

Does that help explain things?