Q17 - Scientist: A number of errors can plague a

[ohthatpatrick]

Question Type:
Explain/Resolve

Stimulus Breakdown:
GIVEN THAT researchers correct collected data to detect errors
HOW IS IT THAT most of the corrections in the scientist's field end up making the data more in alignment with Jones's theory predictions?

Answer Anticipation:
What a weird, hard to understand paradox set up. My impulse is to resolve this by saying, "Jones's theory is a correct description of reality".

I mean, if we said "most of the corrections we do to our data make the data come more into alignment with quantum mechanics' predictions", I would think that this is possibly because quantum mechanics is a correct theory, so things that are not in accordance with it are probably flawed data points.

Another possible line of resolution though is to say that the people who are 'correcting' the data have a confirmation bias in favor of Jones's theory, and so they keep 'fudging' the data to make the theory look stronger.

Correct Answer:
B

Answer Choice Analysis:
(A) Doesn't help address why most corrections go in favor of Jones's theory.

(B) This could work. If we're less likely to scrutinize data points that already match Jones's theory, then we would rarely 'correct' a a data point so that it stopped conforming. i.e. If we're only analyzing stuff that starts off in conflict with Jones's, then corrections would almost always mean coming more into alignment with Jones's theory.

(C) This is about "lines of research", not individual data points.

(D) Super weak claim, so it will have almost no explanatory force. This doesn't explain an asymmetry in favor of Jones's theory.

(E) The mere existence of other theories doesn't matter, just explaining why corrected data points lean towards Jones's theory.

Takeaway/Pattern:
The correct answer is getting at how an uneven allocation of resources can lead to an uneven set of results. If we said "most of the proposed solutions to the economic recession relate to tax reform", one possible way to explain that tendency is "most of the time / money being spent on studying the economic recession has been focused on how taxes affect the economy".

#officialexplanation

[andrewgong01]

Question Type:
Explain/Resolve

Stimulus Breakdown:
GIVEN THAT researchers correct collected data to detect errors

(C) This is about "lines of research", not individual data points.



#officialexplanation

Adding onto the explanation for "C" I thought the reason we can get rid of ""C" is more so that it does not hit at the data collection correction process. The stem specifically said that this is something that occurs for corrected data, meaning that the lines of research has been carried out already. HEnce saying that we are more likely to pursue lines of research that supports Jone does not explain the tendency for CORRECTED data point.

However, if the argument had said that all the research results (uncorrected) point towards Jone then I think "C" would be valid because it goes back to a biased initial starting point where lines of research were being conducted because they were more likely to agree to Jone's theories

[mattop6]

With answer choice (C), we don't know that Jones's theory is accepted from the stimulus. So, couldn't we eliminate this answer choice because we cannot even be sure that it applies.

[seychelles1718]

I don't fully understand the explanation for why B is correct. Although I chose B as it's the best answer, for B to explain the paradox, shouldn't we assume that researchers correct the data in the way that makes the data closer to Jones' theory? How do we know whether the researchers like/accept Jones' theory from the stimulus?

[ohthatpatrick]

I agree that we don't know that, but why do we care whether scientists like / accept Jones's theory?

Let's use this analogous paradox:
A number of errors can plague the process of producing Skittles. By hand checking Skittles as they come off the production line, we can catch many of these flawed Skittles and correct them. However, there is a striking tendency for corrections to happen with orange Skittles; that is, the majority of corrected Skittles result in improving the quality of orange Skittles.

(B) would be saying, "the quality control people tend to give much greater scrutiny to orange Skittles than to all other colors".

If you mainly check orange Skittles, then you'll mainly fix orange Skittles.
If you mainly check data that conflicts with Jones, then you'll mainly be fixing things that conflict with Jones.

We don't have to think that these scientists like/prefer Jones's theory any more than we have to think that the Skittles quality control people like/prefer orange Skittles (it's unthinkable that anyone likes anything more than red Skittles).

Now, the answer isn't bulletproof by any means, but we don't need a perfect answer. We're just looking for the one that "most helps to explain".

If I'm mainly checking data points that deviate from Jones' theory, a correction could just as easily make the data point go FARTHER from the theory's prediction or CLOSER to the theory's prediction.

But the fact that we're paying more attention to something helps to explain why we find more problems there.