Problems understanding PT 45, Section 1, Number 24 Forum

[sdwarrior403]

I cannot grasp what is going on with the DNA tests.

How does the DNA test fail to distinguish DNA samples from different individuals? I have read on LSATBlog that this is basically doing a false positive instead of a false negative. I can fully understand the question if I knew that what this faulty test does is a form of false positive.

I think what would help me is a specific example of what is going on. Thanks in advance.

[sdwarrior403]

It seems to me that the first part of the AC does not describe the test given in the premise.

"It confuses a test that incorrectly identifies DNA samples as coming from the same person with a test that incorrectly shows as coming from different persons samples that come from a single person."

[bitsy]

This one's a little hard to explain.

The argument says that DNA tests can be faulty because they can fail to distinguish between two individuals. For example, DNA from the crime scene might match both Bob and Rico. Because of this failing, the argument goes on to say that DNA tests shouldn't be used to exonerate those whose DNA doesn't match samples from the crime scene. It says that we shouldn't exonerate Cindy, even though her DNA doesn't match that found at the scene.

This conclusion would require a different sort of error than has been previously established-- a mixup where two samples from the same person don't match. B is right, because the author wrongly interprets the sort of error involved.

[sdwarrior403]

I apologize but i still do not understand. Under timed conditions, i would have selected B through POE and moved on, but right now i would like to get this straight.

I guess i would like to have B reworded in a different sense. Maybe that will help me understand.

[ws81086n]

The issue is false positives (false conviction), yet the author's conclusion understands it to be false negatives (false exoneration). If the flaw of DNA tests is that they "often fail to distinguish...distinct individuals", you're not within your rights to conclude, on that basis, that someone the DNA test clears can't be exonerated, because the risk associated with using the test is the other way around.

[sdwarrior403]

Yes, that is my exact issue: what does it mean to fail to distinguish from distinct individuals. How does that mean it is a false positive?

[ws81086n]

Let's say you have two people, Jim and Rick. Rick committed the crime, and his DNA is at the scene. They test Jim's against the sample of DNA from the scene, however, and because the DNA test is prone to failure in distinguishing between the DNA of different individuals, it says that the DNA from the scene is a match with Jim's, and he is convicted. You can see from this that if the DNA test cleared Jim, there would no reason to be skeptical of the result, because the DNA test's problem is failing distinguishing between two different DNA samples, and not failing to provide a match.

[sdwarrior403]

Let's say you have two people, Jim and Rick. Rick committed the crime, and his DNA is at the scene. They test Jim's against the sample of DNA from the scene, however, and because the DNA test is prone to failure in distinguishing between the DNA of different individuals, it says that the DNA from the scene is a match with Jim's, and he is convicted. You can see from this that if the DNA test cleared Jim, there would no reason to be skeptical of the result, because the DNA test's problem is failing distinguishing between two different DNA samples, and not failing to provide a match.

This part stumps me. Wouldn't this system allow for someone, Person A, to submit their DNA, then have their DNA labeled as being from Different Individual 1, while Different Individual 1's DNA is labeled as being from Person A. This situation could certainly lead to an exonaration that is false in reality.

Of course, this hypothetical situation, I assume, is not occurring in the DNA test in the stimulus. I simply still fail to see how this stuff works.

[sdwarrior403]

So I have had even more time to think about this question. I completely understand that the DNA test can fail to distinguish among individuals, even though each person's DNA is unique. So this means that we could have 3 people: Adam, Bob, and Carol. Even though all three have different DNA, the DNA test could show Adam's DNA being Bob's. The only common sense thing to assume is that Bob has to have some kind of DNA associated with his name for his test. It would not make sense for Bob to also have his own DNA associated with himself as well since Adam already is given that one. So if Bob is given the distinction of having Adam's DNA, then obviously an exonaration is not justified.

How does the above situation not consistent with the argument? It seems that situation I present is allowing for the failure of distinguishing individuals.

[05062014]

I thought I understood this question until I read this thread in my current hungover state of mind. Thanks

[bitsy]

So I have had even more time to think about this question. I completely understand that the DNA test can fail to distinguish among individuals, even though each person's DNA is unique. So this means that we could have 3 people: Adam, Bob, and Carol. Even though all three have different DNA, the DNA test could show Adam's DNA being Bob's. The only common sense thing to assume is that Bob has to have some kind of DNA associated with his name for his test. It would not make sense for Bob to also have his own DNA associated with himself as well since Adam already is given that one. So if Bob is given the distinction of having Adam's DNA, then obviously an exonaration is not justified.

How does the above situation not consistent with the argument? It seems that situation I present is allowing for the failure of distinguishing individuals.

Oh no, I don't think that's right. The exoneration in the conclusion does not refer to people whose DNA is indistinguishable.

Let's say that the three suspects-- Adam, Bob, and Carol-- have their DNA compared to a DNA sample found at the crime scene. Assume only one sample from the scene, and the sample is from only one human. In a hypothetical situation, the sample matches both Adam and Bob, but not Carol. We would know that Adam and Bob could have committed the crime (though we're unsure which one is guilty, since the test doesn't tell us), but there is no evidence that Carol was ever there. So we should keep Adam and Bob as suspects, and exonerate Carol.

This is the mistake that the conclusion makes-- it says that we couldn't exonerate Carol if such a hypothetical situation were to occur.

[sdwarrior403]

Thank you. I understand it now.