|
Author |
Message |
|
TAGS:
|
|
|
Magoosh GMAT Instructor
Joined: 28 Dec 2011
Posts: 836
Followers: 213
Kudos [?]:
585
[0], given: 14
|
Re: Industry experts expect improvements in job safety training [#permalink]
 20 Dec 2012, 18:47
SravnaTestPrep wrote: Dear Mike,
I just need to place my argument for the benefit of the readers.
You need to spend more time understanding the gist of the argument before you harp on the details. I might not have been consistent in the use of terminology but I never thought that would have hindered the understanding of the argument. To be honest, I am not sure that your argument benefits readers. After all, you still have not demonstrated that you understand how to reach the correct answer of this single CR question. Furthermore, your arguments are all over the place, more about defending yourself and your points than supporting a particular answer to this question. In this post and the previous post, you didn't even quote the original CR argument. You are off arguing about something else, something not directly connected to the GMAT CR question explored in this post. If anything, this sets an extremely poor example for other readers, and I very much want reader to understand how important it is to stay close to the text of the CR arguments. My complaints about the inconsistency of your arguments were not merely that it made your arguments harder to follow but, much more importantly, you were using terms and ideas that have no relationship to the individual question. As I have said above, careful reading of the prompt is absolutely essential for the GMAT CR. SravnaTestPrep wrote: My point was that we cannot come to the conclusion that the increase in the size of the workforce resulted in the increase in the number of accidents. I am glad you admit that. Answer me this simple point. If it is something that could have been either way, remember that there was no mention that the workforce increased sharply such as 100% or so, how is it better than randomly selecting one of the two possibilities? That is I can toss a coin and if a heads comes up say that the increase in the size of the workforce increased the number of accidents.
Is that how we are supposed to answer such questions? Right there, you demonstrate --- you don't know how to determine the correct answer to this CR. You demonstrate that you need a great deal of help in your understanding of how GMAT CR work. There is absolutely no issue of choosing randomly between two "equally good" options. Answer (B) is the correct option, for a variety of reasons, and answer (E), when understood correctly, doesn't have a the least connection to what would constitute a correct answer. Here's the original question again. Industry experts expect improvements in job safety training to lead to safer work environments. A recent survey indicated, however, that for manufacturers who improved job safety training during the 1980s, the number of on-the-job accidents tended to increase in the months immediately following the changes in the training programs.
Which one of the following, if true, most helps to resolve the apparent discrepancy in the passage above? Experts expect, as the rest of us would, that improvement to safety training would create fewer accidents. For many companies in the 1980s, they improved safety training, but the number of accidents increased, at least in the short term. This is a discrepancy between what we would have expected and what actually happened. Which of the follow best resolves this discrepancy? (A) A similar survey found that the number of on the-job accidents remained constant after job safety training in the transportation sector was improved. This is an entirely different business sector --- irrelevant. This is incorrect. (B) Manufacturers tend to improve their job safety training only when they are increasing the size of their workforce. If the workforce increased substantially, then a slight decrease in the percent of accidents would still result in a greater number. Also, new recruits may have more accidents in their adjustment period. This answer opens the door to plausible explanations. (C) Manufacturers tend to improve job safety training only after they have noticed that the number of on-the-job accidents has increased. Irrelevant --- we don't care about what happened before the training. Why did the number of accidents increase after the training? This is incorrect. (D) It is likely that the increase in the number of on-the-job accidents experienced by many companies was not merely a random fluctuation. First of all, this is jaw-droppingly obvious. More important, it doesn't explain anything. This is incorrect. (E) Significant safety measures, such as protective equipment and government safety inspections, were in place well before the improvements in job safety training.As I explained a few posts back, these are all things that are "exterior" to the experience of the workers --- probably many of them are outside the workers' awareness. Furthermore, all this stuff was in place and didn't change, so it in no way explains an increase in the number of accidents. This is incorrect. One answer, (B), leads to plausible, possible explanations, and the other four lead nowhere. Therefore, (B) is the only possible answer. As I said several posts ago, the answers to CR are not always logically airtight. We are not guaranteed that (B) leads to a cogent explanation, but we can imagine a plausible chain of facts that would lead from (B) to an explanation, and we can do that with no other answer. (B) is not the best possible logically conceivable answer --- it is merely the best among the five given. The fact that we have to make a few plausible assumptions to go from the text of (B) to a full explanation does not give us license to go on flights of fanciful imagination wildly justifying any of the other four answer. Once again, GMAT CR are hard. They are not as precise as mathematics, but they are designed so one answer is quite clearly correct. It takes considerable skill to read closely enough, to recognize the subtle implications, and not to be trapped by superficial similarities in wording. It is apparent to me that you need considerable practice with all these skills. The fact that you characterize it as a random coin flip indicates that you are not seeing the subtle signposts that these questions always contain. Let me be very clear. This question has a clear right answer. You missed that right answer and chose a wrong answer. You fell into a trap, a very predictable trap for folks who don't read carefully enough. This is a very common trap-type on the GMAT CR. Your arguments justifying why you chose a wrong answer are both irrelevant and not helpful. This is a GMAT forum, designed to help readers perform well on the GMAT. Arguments for answers other than OA, no matter how sophisticated, are entirely besides the point. They help no one. The entire point of all the instruction we give on this site is helping students to get to the answer of questions on the GMAT. You have not yet demonstrated the ability to arrive at the one correct answer for this question, and this concerns me, because it doesn't at all help the students who simply want to understand how to isolate the correct answer. The concerns of such students are paramount on this site. Mike
_________________
Mike McGarry
Magoosh Test Prep
|
|
|
|
|
|
|
|
|
Manager
Joined: 17 Dec 2012
Posts: 170
Followers: 2
Kudos [?]:
21
[0], given: 5
|
Re: Industry experts expect improvements in job safety training [#permalink]
20 Dec 2012, 19:12
mikemcgarry wrote: SravnaTestPrep wrote: Dear Mike,
I just need to place my argument for the benefit of the readers.
You need to spend more time understanding the gist of the argument before you harp on the details. I might not have been consistent in the use of terminology but I never thought that would have hindered the understanding of the argument. To be honest, I am not sure that your argument benefits readers. After all, you still have not demonstrated that you understand how to reach the correct answer of this single CR question. Furthermore, your arguments are all over the place, more about defending yourself and your points than supporting a particular answer to this question. In this post and the previous post, you didn't even quote the original CR argument. You are off arguing about something else, something not directly connected to the GMAT CR question explored in this post. If anything, this sets an extremely poor example for other readers, and I very much want reader to understand how important it is to stay close to the text of the CR arguments. My complaints about the inconsistency of your arguments were not merely that it made your arguments harder to follow but, much more importantly, you were using terms and ideas that have no relationship to the individual question. As I have said above, careful reading of the prompt is absolutely essential for the GMAT CR. SravnaTestPrep wrote: My point was that we cannot come to the conclusion that the increase in the size of the workforce resulted in the increase in the number of accidents. I am glad you admit that. Answer me this simple point. If it is something that could have been either way, remember that there was no mention that the workforce increased sharply such as 100% or so, how is it better than randomly selecting one of the two possibilities? That is I can toss a coin and if a heads comes up say that the increase in the size of the workforce increased the number of accidents.
Is that how we are supposed to answer such questions? Right there, you demonstrate --- you don't know how to determine the correct answer to this CR. You demonstrate that you need a great deal of help in your understanding of how GMAT CR work. There is absolutely no issue of choosing randomly between two "equally good" options. Answer (B) is the correct option, for a variety of reasons, and answer (E), when understood correctly, doesn't have a the least connection to what would constitute a correct answer. Here's the original question again. Industry experts expect improvements in job safety training to lead to safer work environments. A recent survey indicated, however, that for manufacturers who improved job safety training during the 1980s, the number of on-the-job accidents tended to increase in the months immediately following the changes in the training programs.
Which one of the following, if true, most helps to resolve the apparent discrepancy in the passage above? Experts expect, as the rest of us would, that improvement to safety training would create fewer accidents. For many companies in the 1980s, they improved safety training, but the number of accidents increased, at least in the short term. This is a discrepancy between what we would have expected and what actually happened. Which of the follow best resolves this discrepancy? (A) A similar survey found that the number of on the-job accidents remained constant after job safety training in the transportation sector was improved. This is an entirely different business sector --- irrelevant. This is incorrect. (B) Manufacturers tend to improve their job safety training only when they are increasing the size of their workforce. If the workforce increased substantially, then a slight decrease in the percent of accidents would still result in a greater number. Also, new recruits may have more accidents in their adjustment period. This answer opens the door to plausible explanations. (C) Manufacturers tend to improve job safety training only after they have noticed that the number of on-the-job accidents has increased. Irrelevant --- we don't care about what happened before the training. Why did the number of accidents increase after the training? This is incorrect. (D) It is likely that the increase in the number of on-the-job accidents experienced by many companies was not merely a random fluctuation. First of all, this is jaw-droppingly obvious. More important, it doesn't explain anything. This is incorrect. (E) Significant safety measures, such as protective equipment and government safety inspections, were in place well before the improvements in job safety training.As I explained a few posts back, these are all things that are "exterior" to the experience of the workers --- probably many of them are outside the workers' awareness. Furthermore, all this stuff was in place and didn't change, so it in no way explains an increase in the number of accidents. This is incorrect. One answer, (B), leads to plausible, possible explanations, and the other four lead nowhere. Therefore, (B) is the only possible answer. As I said several posts ago, the answers to CR are not always logically airtight. We are not guaranteed that (B) leads to a cogent explanation, but we can imagine a plausible chain of facts that would lead from (B) to an explanation, and we can do that with no other answer. (B) is not the best possible logically conceivable answer --- it is merely the best among the five given. The fact that we have to make a few plausible assumptions to go from the text of (B) to a full explanation does not give us license to go on flights of fanciful imagination wildly justifying any of the other four answer. Once again, GMAT CR are hard. They are not as precise as mathematics, but they are designed so one answer is quite clearly correct. It takes considerable skill to read closely enough, to recognize the subtle implications, and not to be trapped by superficial similarities in wording. It is apparent to me that you need considerable practice with all these skills. The fact that you characterize it as a random coin flip indicates that you are not seeing the subtle signposts that these questions always contain. Let me be very clear. This question has a clear right answer. You missed that right answer and chose a wrong answer. You fell into a trap, a very predictable trap for folks who don't read carefully enough. This is a very common trap-type on the GMAT CR. Your arguments justifying why you chose a wrong answer are both irrelevant and not helpful. This is a GMAT forum, designed to help readers perform well on the GMAT. Arguments for answers other than OA, no matter how sophisticated, are entirely besides the point. They help no one. The entire point of all the instruction we give on this site is helping students to get to the answer of questions on the GMAT. You have not yet demonstrated the ability to arrive at the one correct answer for this question, and this concerns me, because it doesn't at all help the students who simply want to understand how to isolate the correct answer. The concerns of such students are paramount on this site. Mike You are again missing the point. It is not the question of not being able to choose the right answer. But when a choice that is given as the official answer is open to different interpretations and is a dubious one, I think one is justified in pointing it out. My point in selecting E as the answer is only to show that one can be creative if needed and come up with that choice as the answer . It was not my intention to provide an airtight case for E. You need to understand that before you try to demolish my reasoning for choice E. I will leave it to the readers to decide what really benefits them, whether pointing out an error in a question in the GMAT or just ignore it under the assumption that errors in a GMAT test should not be pointed out because it might be hindering their performance in the actual exam.
_________________
Srinivasan Vaidyaraman
sravna@gmail.com
Sravna Test Prep
http://www.sravna.com
|
|
|
|
|
|
GMAT Instructor
Joined: 24 Jun 2008
Posts: 973
Location: Toronto
Followers: 167
Kudos [?]:
443
[0], given: 3
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 10:42
Fun debate, guys!  In the question, we learn that something changed (training improved). Yet that change produced an unexpected result (accidents increased). We need to explain that result. The only answers that could possibly be correct here are ones which either: * explain why the change might have had unexpected consequences * describe a second change that might have taken place at the same time as the first one, and which might explain the result So to find the right answer here, we need to focus on what happened during the change itself. Answer E is patently wrong. It does not tell us anything about *changes* in company safety after the training measures were introduced, so it does nothing to explain why the new safety measures might have increased the number of accidents. B is the only answer that could conceivably be correct here, because it's the only answer that focuses on the right thing: what happened during the change in training practices. And it does explain perfectly why the number of accidents could increase, even if safety improved: if you have a lot more workers, you can have more accidents even if you successfully lower the rate of accidents per worker. Finally, as Mike points out, we are not looking for logical 'proofs' in these types of CR questions. We just need an answer that most plausibly explains why the unexpected result *could* happen, and not an answer which explains why it *must* happen. B does precisely that. There is no 'creative' way to justify the other answer choices here.
_________________
Nov 2011: After years of development, I am now making my advanced Quant books and high-level problem sets available for sale. Contact me at ianstewartgmat at gmail.com for details.
Private GMAT Tutor based in Toronto
|
|
|
|
|
|
Manager
Joined: 17 Dec 2012
Posts: 170
Followers: 2
Kudos [?]:
21
[0], given: 5
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 16:46
IanStewart wrote: Fun debate, guys! In the question, we learn that something changed (training improved). Yet that change produced an unexpected result (accidents increased). We need to explain that result. The only answers that could possibly be correct here are ones which either: * explain why the change might have had unexpected consequences * describe a second change that might have taken place at the same time as the first one, and which might explain the result So to find the right answer here, we need to focus on what happened during the change itself. Answer E is patently wrong. It does not tell us anything about *changes* in company safety after the training measures were introduced, so it does nothing to explain why the new safety measures might have increased the number of accidents. B is the only answer that could conceivably be correct here, because it's the only answer that focuses on the right thing: what happened during the change in training practices. And it does explain perfectly why the number of accidents could increase, even if safety improved: if you have a lot more workers, you can have more accidents even if you successfully lower the rate of accidents per worker. Finally, as Mike points out, we are not looking for logical 'proofs' in these types of CR questions. We just need an answer that most plausibly explains why the unexpected result *could* happen, and not an answer which explains why it *must* happen. B does precisely that. There is no 'creative' way to justify the other answer choices here. Let me say why I am not convinced with B even as a "could be right" answer. It is mentioned that "Industry experts expect improvements in job safety training to lead to safer work environments." But it is considered a paradox as "job accidents tended to increase in the months immediately following the changes in the training programs." So the training program is supposed to give immediate results. So increase in the size of the workforce should have negligible effect as all are equally well trained. It is something like taking a vaccination for a disease. It doesn't matter whether 10 take it or 100 take it. Almost none will contract the disease The argument of size doesn't work in such cases. It works only when you the element of chance is predominant. Thus when you toss a coin twice the number of times heads is likely to come up in once whereas if you toss it fifty times, the number of times the heads is likely to come up is 25. Here you are doing nothing to prevent heads from coming up. If you do something effective about preventing that from happening the number of times heads comes up reduces drastically. and if it is 100% effective , the number of times heads comes up doesn't change. A similar argument holds in this case. The argument of size in principle is not the right argument
_________________
Srinivasan Vaidyaraman
sravna@gmail.com
Sravna Test Prep
http://www.sravna.com
|
|
|
|
|
|
GMAT Instructor
Joined: 24 Jun 2008
Posts: 973
Location: Toronto
Followers: 167
Kudos [?]:
443
[0], given: 3
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 17:26
I really don't understand what you're saying; safety training is not analogous to vaccination. With training you can still have an accident. If you decrease the monthly rate of accidents per worker from, say, 2% to 1%, but you increase your workforce from 100 employees to 1000 employees, you'll have more accidents. That's what B is getting at.
_________________
Nov 2011: After years of development, I am now making my advanced Quant books and high-level problem sets available for sale. Contact me at ianstewartgmat at gmail.com for details.
Private GMAT Tutor based in Toronto
|
|
|
|
|
|
Manager
Joined: 17 Dec 2012
Posts: 170
Followers: 2
Kudos [?]:
21
[0], given: 5
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 17:46
IanStewart wrote: I really don't understand what you're saying; safety training is not analogous to vaccination. With training you can still have an accident.
If you decrease the monthly rate of accidents per worker from, say, 2% to 1%, but you increase your workforce from 100 employees to 1000 employees, you'll have more accidents. That's what B is getting at. What I am saying is that the argument of size can in principle be used only when events are left to chance. In this case the event, which is the accidents is not left to chance. The safety measures are used to decrease the rate of accidents. So as you increase the number of workers the number of accidents is supposed to not increase. Of course you can use the numbers to your advantage and try to prove your point. But the argument in principle is not sound.
_________________
Srinivasan Vaidyaraman
sravna@gmail.com
Sravna Test Prep
http://www.sravna.com
|
|
|
|
|
|
Magoosh GMAT Instructor
Joined: 28 Dec 2011
Posts: 836
Followers: 213
Kudos [?]:
585
[0], given: 14
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 19:52
SravnaTestPrep wrote: You are again missing the point. It is not the question of not being able to choose the right answer. That sentence I highlighted is the core of my disagreement with you. This is a GMAT forum, a forum for readers who want to perform well on the GMAT. Learning how to identify the right answer on a GMAT question is everything. That is precisely what this forum is about. Concerns that are orthogonal to the concern of finding the correct answer really have no place in this forum and are of no help to the vast majority of readers who come here. If you have a point that has to do with something other than finding the correct answer to the question, it is of no use to readers, and frankly, I am not interested in it either. SravnaTestPrep wrote: But when a choice that is given as the official answer is open to different interpretations and is a dubious one, I think one is justified in pointing it out. Here's another crucial distinction you are missing. The OA is not airtight certain, but it is plausible. That is quite different from being dubious. GMAT CR is not about absolutely black vs. white. We need to deal with shades of grade, and you are essentially saying all shades of gray are equally doubtful. That attitude would be absolutely disastrous on the GMAT CR. SravnaTestPrep wrote: My point in selecting E as the answer is only to show that one can be creative if needed and come up with that choice as the answer . It was not my intention to provide an airtight case for E. You need to understand that before you try to demolish my reasoning for choice E. Learning how to isolate single answer is precisely the skill folks need to learn. Learning how to think one answer may be right but creative justify another answer --- that is precisely what gets people into trouble on the GMAT. Creativity, while immensely valuable in other endeavors, is a positive detriment on the GMAT CR. You need to focus, very un-creatively, on exactly what is said, exactly what is implied, and exactly what is omitted SravnaTestPrep wrote: I will leave it to the readers to decide what really benefits them, whether pointing out an error in a question in the GMAT or just ignore it under the assumption that errors in a GMAT test should not be pointed out because it might be hindering their performance in the actual exam. First of all, there is absolutely no error in this question. This is a question, like questions on the GMAT, in which the OA allows for some interpretations, and some plausible interpretations lead to scenarios that do exactly what the question asks (here, helps to explain the apparent discrepancy). That is not an error. That is not a flawed question. Once again, you are equating all shades of gray, saying that if there's any uncertainty at all, the question is doubtful and prevents you from making a clear decision: once again, I cannot underscore enough: this attitude is pure poison on the GMAT CR. The GMAT CR is all about drawing inferences and conclusions under conditions that are less than 100% certain ---- after all, when folks get their MBAs and are working in management, they will also have to make crucial decisions in conditions of less than 100% certainty. More important, you are putting your faulty logic in front of the readers of this forum as if you are an authority, and you are not. You are someone still struggling with the very basics of what the GMAT CR is about. What most benefits the readers of this forum is knowing how to isolate the right answer, and that seems almost irrelevant to your concerns. SravnaTestPrep wrote: What I am saying is that the argument of size can in principle be used only when events are left to chance. In this case the event, which is the accidents is not left to chance. The safety measures are used to decrease the rate of accidents. So as you increase the number of workers the number of accidents is supposed to not increase. Of course you can use the numbers to your advantage and try to prove your point. But the argument in principle is not sound. Notice that you are still confusing the most basic terminology of the question. The prompt is about improvements in the "safety training." Only answer (E), an incorrect answer, addressed "safety measures", something very different. Throughout this whole discussion, you have been confusing those two to the point of indistinction. This does make crystal clear why you fell for the trap answer (E), an answer choice designed to ensnare folks who don't read carefully, but unfortunately, this adamant insouciance with respect to the precise vocabulary sets the most abysmal example for anyone who wants to be successful on the GMAT CR. Also, notice you are thinking in terms of absolutes --- either the safety training results in absolutely no accidents, or it's purely a random chance process. In fact, there are probabilities involved ---- in probability theory, these are known as "conditional probabilities." Let A = working having an accident, and T = working has safety training. Then we suspect, and what the industry analysts expected at the beginning of the prompt, was that P(A|T) < P(A|not T). That was precisely the assumption the industry analysis made, and it is precisely the assumption that almost anyone familiar with real world scenarios like this would make. We are not assuming, as you are, that P(A|T) = 0 ---- that is entirely unrealistic. Only if P(A|T) = 0 would the increase in number make no difference, but again, that is a completely unwarranted assumption. As IanStewart said, in a brilliant comment you apparently did not understand and totally missed, " safety training is not analogous to vaccination." In algebraic form, he is saying P(A|T) is not equal to zero. IanStewart, I thank you very much for your constructive and thoughtful comments in this exchange. Mike
_________________
Mike McGarry
Magoosh Test Prep
|
|
|
|
|
|
Manager
Joined: 17 Dec 2012
Posts: 170
Followers: 2
Kudos [?]:
21
[0], given: 5
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 20:16
mikemcgarry wrote: SravnaTestPrep wrote: You are again missing the point. It is not the question of not being able to choose the right answer. That sentence I highlighted is the core of my disagreement with you. This is a GMAT forum, a forum for readers who want to perform well on the GMAT. Learning how to identify the right answer on a GMAT question is everything. That is precisely what this forum is about. Concerns that are orthogonal to the concern of finding the correct answer really have no place in this forum and are of no help to the vast majority of readers who come here. If you have a point that has to do with something other than finding the correct answer to the question, it is of no use to readers, and frankly, I am not interested in it either. SravnaTestPrep wrote: But when a choice that is given as the official answer is open to different interpretations and is a dubious one, I think one is justified in pointing it out. Here's another crucial distinction you are missing. The OA is not airtight certain, but it is plausible. That is quite different from being dubious. GMAT CR is not about absolutely black vs. white. We need to deal with shades of grade, and you are essentially saying all shades of gray are equally doubtful. That attitude would be absolutely disastrous on the GMAT CR. SravnaTestPrep wrote: My point in selecting E as the answer is only to show that one can be creative if needed and come up with that choice as the answer . It was not my intention to provide an airtight case for E. You need to understand that before you try to demolish my reasoning for choice E. Learning how to isolate single answer is precisely the skill folks need to learn. Learning how to think one answer may be right but creative justify another answer --- that is precisely what gets people into trouble on the GMAT. Creativity, while immensely valuable in other endeavors, is a positive detriment on the GMAT CR. You need to focus, very un-creatively, on exactly what is said, exactly what is implied, and exactly what is omitted SravnaTestPrep wrote: I will leave it to the readers to decide what really benefits them, whether pointing out an error in a question in the GMAT or just ignore it under the assumption that errors in a GMAT test should not be pointed out because it might be hindering their performance in the actual exam. First of all, there is absolutely no error in this question. This is a question, like questions on the GMAT, in which the OA allows for some interpretations, and some plausible interpretations lead to scenarios that do exactly what the question asks (here, helps to explain the apparent discrepancy). That is not an error. That is not a flawed question. Once again, you are equating all shades of gray, saying that if there's any uncertainty at all, the question is doubtful and prevents you from making a clear decision: once again, I cannot underscore enough: this attitude is pure poison on the GMAT CR. The GMAT CR is all about drawing inferences and conclusions under conditions that are less than 100% certain ---- after all, when folks get their MBAs and are working in management, they will also have to make crucial decisions in conditions of less than 100% certainty. More important, you are putting your faulty logic in front of the readers of this forum as if you are an authority, and you are not. You are someone still struggling with the very basics of what the GMAT CR is about. What most benefits the readers of this forum is knowing how to isolate the right answer, and that seems almost irrelevant to your concerns. SravnaTestPrep wrote: What I am saying is that the argument of size can in principle be used only when events are left to chance. In this case the event, which is the accidents is not left to chance. The safety measures are used to decrease the rate of accidents. So as you increase the number of workers the number of accidents is supposed to not increase. Of course you can use the numbers to your advantage and try to prove your point. But the argument in principle is not sound. Notice that you are still confusing the most basic terminology of the question. The prompt is about improvements in the "safety training." Only answer (E), an incorrect answer, addressed "safety measures", something very different. Throughout this whole discussion, you have been confusing those two to the point of indistinction. This does make crystal clear why you fell for the trap answer (E), an answer choice designed to ensnare folks who don't read carefully, but unfortunately, this adamant insouciance with respect to the precise vocabulary sets the most abysmal example for anyone who wants to be successful on the GMAT CR. Also, notice you are thinking in terms of absolutes --- either the safety training results in absolutely no accidents, or it's purely a random chance process. In fact, there are probabilities involved ---- in probability theory, these are known as "conditional probabilities." Let A = working having an accident, and T = working has safety training. Then we suspect, and what the industry analysts expected at the beginning of the prompt, was that P(A|T) < P(A|not T). That was precisely the assumption the industry analysis made, and it is precisely the assumption that almost anyone familiar with real world scenarios like this would make. We are not assuming, as you are, that P(A|T) = 0 ---- that is entirely unrealistic. Only if P(A|T) = 0 would the increase in number make no difference, but again, that is a completely unwarranted assumption. As IanStewart said, in a brilliant comment you apparently did not understand and totally missed, " safety training is not analogous to vaccination." In algebraic form, he is saying P(A|T) is not equal to zero. IanStewart, I thank you very much for your constructive and thoughtful comments in this exchange. Mike I suggest you refrain from making statements which imply that you are authoritative enough to comment on other's competence. Just by making big noises you cannot prove your point. So kindly stick only to the logic. My point is based on simple and straightforward logic and not on conditional probabilities. This is my point. You put a training program in place to decrease the rate of accidents. That is you expect the number of accidents to not be affected as the size goes up. So to use the argument of size itself to resolve the discrepancy is untenable. You reasons as follows: Something that would counter the effect of increase in size but apparently does not counter it, is resolved by saying it is not countered because of increase in size. Some authority!
_________________
Srinivasan Vaidyaraman
sravna@gmail.com
Sravna Test Prep
http://www.sravna.com
|
|
|
|
|
|
Manager
Joined: 17 Dec 2012
Posts: 170
Followers: 2
Kudos [?]:
21
[0], given: 5
|
Re: Industry experts expect improvements in job safety training [#permalink]
21 Dec 2012, 21:34
Consider these arguments: A: This mouse trap will trap the mice very effectively (reduce accidents) and save you from the increasing number of mice in your house (offset the effect of increase in size) B: But the mouse trap has produced a counter effect (increase in accidents) A: That is because the number of mice has increased (increase in size) There is something wrong with the above argument. Also in our case, nothing suggests that the increase in size of the workforce has been sharp. So it is an additional assumption we have to make. The new workforce need not be previously untrained. More than anything else the argument lacks intellectual appeal because you are resolving a discrepancy by saying that the effect in the discrepancy (increase in accidents) is the result of the cause in the discrepancy (safety training program) or in other words the cause being not effective. Aren't we actually supposed to explain that, i.e., why the cause is not effective?
_________________
Srinivasan Vaidyaraman
sravna@gmail.com
Sravna Test Prep
http://www.sravna.com
|
|
|
|
|
|
Magoosh GMAT Instructor
Joined: 28 Dec 2011
Posts: 836
Followers: 213
Kudos [?]:
585
[0], given: 14
|
Re: Industry experts expect improvements in job safety training [#permalink]
22 Dec 2012, 16:00
SravnaTestPrep wrote: I suggest you refrain from making statements which imply that you are authoritative enough to comment on other's competence. Just by making big noises you cannot prove your point. So kindly stick only to the logic.
My point is based on simple and straightforward logic and not on conditional probabilities. This is my point. You put a training program in place to decrease the rate of accidents. That is you expect the number of accidents to not be affected as the size goes up. So to use the argument of size itself to resolve the discrepancy is untenable. My comments criticizing some of your advice was given, not with the intention of judging you, but rather making sure that readers of this site are exposed to best practices. Again, I regard the benefit to readers of this site, and their improved performance on the GMAT, as our paramount concern here. This is a very tricky misunderstanding you have. Both the analogy of vaccines (which are essentially 100% effective) and the analogy of mousetraps (single-used items) are deeply flawed for this scenario in a way that is hard to explain. I will try with an extended series of analogies. In all cases, "A" is some protective or preventive measure, and "B" is some unwelcome effect, the incidence of which people wish to reduce by employing A. I have tried to draw examples from as wide an array of endeavors as possible, to demonstrate the near universal applicability of the point I am trying to make. A = vitamin C, B = common cold A = flu shot, B = the flu A = seat belts, B = car accident injuries A = pilot training, B = plane crashes A = plane safety inspections, B = plane crashes A = law school, B = lawyers who don't know or don't obey the law A = baseball umpires, B = cheating in baseball A = school English stanards, B = adult illiteracy A = safety caps on medicines, B = children taking medicines they shouldn't A = condom, B = pregnancy A = free press, B = government corruption A = national armed forces, B = attacks on a nation's citizens A = freezing fruits & vegetables, B = the fruits & vegetables going bad A = safety training in a factory, B = accidents in the factory In all cases, P(B|not A) > P(B|A), and in a cases, P(B|A) > 0. That is, in all cases, the preventive measure A substantially lowers the risk of unwanted consequence B, but even with A in place, B still happens. Furthermore, in all cases, A causes a reduction in the rate of B, the ratio of cases with B, not in the absolute number of B, irrespective of changes in the population. In all cases, if we looked at a certain population with A in place, and then doubled the number of people (or planes or baseball games or whatever constitutes a "case"), then the ratio of B would remain about the same, but the absolute number of incidents of B would double. Let's think about this very specifically for the case in the question, where A = safety training, and B = accidents in the factory. Let's say for simplicity, there were originally 100 workers in the factory. Suppose somehow we know, without any safety training, these 100 workers will have, say 15 accidents a year, year after year. (We don't know whether that's 15 different individuals each year, or a few accident-prone individuals on whom accidents are concentrated --- this doesn't matter for this analysis.) Now, suppose we train these 100 workers, and with the benefit of the training, their accidents go down to 5 per year, year after year. All good, so far. Both the number and the percentage decline. Answer (B) suggests that the hiring and the safety training were simultaneous, but for clarity, let's suppose that these first 100 workers were trained and worked without new hires long enough that we could see the effect. Now, to be extreme, let's suppose the factory hires 400 new workers: we give them the safety training and set them to work. With the 100 old workers and the 400 new workers, how many accidents will the factory have in a year? I would say, the new workers, once trained, will probably be in every way comparable to the old workers, and all will have accidents at a rate of about (5 accidents per year)/(100 workers). This means that the total number would be 25 accidents in a year --- a lower accident rate than existed before the safety training, but a higher overall number because of the new hires. (I think a five-fold increase in the size of the factory is a little unreasonable in the context of the question itself, but I chose such a large increase in the population in this hypothetical example just for clarity.) Now consider the consequences of what you are arguing. Before the new hires, with just the 100 old employees, they had 5 accidents per year each year. Those 100 old employees will continue at this rate (improved compared to the rate before the safety training). Now, we hire 400 new employees. You would say --- the number would stay the same. This ineluctably implies --- if the old employees continuing having accidents at whatever rate they had, then in order for the number of accidents to stay the same, the 400 new employees all must be accident-free! That is, the quality of new workers is incommensurately superior to the quality of the old workers. That I would call a wildly flawed conclusion. If anything, we have to assume that new employees (with safety training) are, on average, about equally like to have an accident as the old employees (with safety training). In fact, the new workers, because they are new, might be slightly more likely than the old workers to have accidents: in that case, the number would increase even more. By assuming that number of accidents, rather than percent or ratio of accidents, stays the same, you are necessarily assuming that the quality of the new workers is of an order quite different from the quality of the old workers, and there's absolutely nothing in the prompt or in real-life experience to justify a mensch/ ubermensch split between old & new employees. Notice, for every A&B pair in the list above, we could construct a parallel argument demonstrating that, with a population increase, the rate would remain constant, not the number. This is precisely why places like insurance companies, the Center for Disease Control, the American Medical Association, government agencies, etc. keep track of rate & ratios & percentages, more than numbers. In thousands upon thousands of different data pools, it is understood that something, whether good or bad, has a fixed rate, and increasing the population size will increase the number even though the rate stays constant. This is absolutely standard throughout the social sciences, anything that uses statistical or epidemiological or actuarial data, and by insisting on fixed number rather than fixed rate, essentially you are directly assaulting the core methodologies of at least a dozen well-respected academic disciplines. This is precisely why conditional probabilities are crucial for understanding this problem, and to ignore conditional probabilities is not only to ignore something intrinsic to the situation, but to make vastly unwarranted assumptions about the difference between the old and new workers. What do you say to this? Mike
_________________
Mike McGarry
Magoosh Test Prep
|
|
|
|
|
|
Manager
Joined: 17 Dec 2012
Posts: 170
Followers: 2
Kudos [?]:
21
[0], given: 5
|
Re: Industry experts expect improvements in job safety training [#permalink]
22 Dec 2012, 17:57
mikemcgarry wrote: SravnaTestPrep wrote: I suggest you refrain from making statements which imply that you are authoritative enough to comment on other's competence. Just by making big noises you cannot prove your point. So kindly stick only to the logic.
My point is based on simple and straightforward logic and not on conditional probabilities. This is my point. You put a training program in place to decrease the rate of accidents. That is you expect the number of accidents to not be affected as the size goes up. So to use the argument of size itself to resolve the discrepancy is untenable. My comments criticizing some of your advice was given, not with the intention of judging you, but rather making sure that readers of this site are exposed to best practices. Again, I regard the benefit to readers of this site, and their improved performance on the GMAT, as our paramount concern here. This is a very tricky misunderstanding you have. Both the analogy of vaccines (which are essentially 100% effective) and the analogy of mousetraps (single-used items) are deeply flawed for this scenario in a way that is hard to explain. I will try with an extended series of analogies. In all cases, "A" is some protective or preventive measure, and "B" is some unwelcome effect, the incidence of which people wish to reduce by employing A. I have tried to draw examples from as wide an array of endeavors as possible, to demonstrate the near universal applicability of the point I am trying to make. A = vitamin C, B = common cold A = flu shot, B = the flu A = seat belts, B = car accident injuries A = pilot training, B = plane crashes A = plane safety inspections, B = plane crashes A = law school, B = lawyers who don't know or don't obey the law A = baseball umpires, B = cheating in baseball A = school English stanards, B = adult illiteracy A = safety caps on medicines, B = children taking medicines they shouldn't A = condom, B = pregnancy A = free press, B = government corruption A = national armed forces, B = attacks on a nation's citizens A = freezing fruits & vegetables, B = the fruits & vegetables going bad A = safety training in a factory, B = accidents in the factory In all cases, P(B|not A) > P(B|A), and in a cases, P(B|A) > 0. That is, in all cases, the preventive measure A substantially lowers the risk of unwanted consequence B, but even with A in place, B still happens. Furthermore, in all cases, A causes a reduction in the rate of B, the ratio of cases with B, not in the absolute number of B, irrespective of changes in the population. In all cases, if we looked at a certain population with A in place, and then doubled the number of people (or planes or baseball games or whatever constitutes a "case"), then the ratio of B would remain about the same, but the absolute number of incidents of B would double. Let's think about this very specifically for the case in the question, where A = safety training, and B = accidents in the factory. Let's say for simplicity, there were originally 100 workers in the factory. Suppose somehow we know, without any safety training, these 100 workers will have, say 15 accidents a year, year after year. (We don't know whether that's 15 different individuals each year, or a few accident-prone individuals on whom accidents are concentrated --- this doesn't matter for this analysis.) Now, suppose we train these 100 workers, and with the benefit of the training, their accidents go down to 5 per year, year after year. All good, so far. Both the number and the percentage decline. Answer (B) suggests that the hiring and the safety training were simultaneous, but for clarity, let's suppose that these first 100 workers were trained and worked without new hires long enough that we could see the effect. Now, to be extreme, let's suppose the factory hires 400 new workers: we give them the safety training and set them to work. With the 100 old workers and the 400 new workers, how many accidents will the factory have in a year? I would say, the new workers, once trained, will probably be in every way comparable to the old workers, and all will have accidents at a rate of about (5 accidents per year)/(100 workers). This means that the total number would be 25 accidents in a year --- a lower accident rate than existed before the safety training, but a higher overall number because of the new hires. (I think a five-fold increase in the size of the factory is a little unreasonable in the context of the question itself, but I chose such a large increase in the population in this hypothetical example just for clarity.) Now consider the consequences of what you are arguing. Before the new hires, with just the 100 old employees, they had 5 accidents per year each year. Those 100 old employees will continue at this rate (improved compared to the rate before the safety training). Now, we hire 400 new employees. You would say --- the number would stay the same. This ineluctably implies --- if the old employees continuing having accidents at whatever rate they had, then in order for the number of accidents to stay the same, the 400 new employees all must be accident-free! That is, the quality of new workers is incommensurately superior to the quality of the old workers. That I would call a wildly flawed conclusion. If anything, we have to assume that new employees (with safety training) are, on average, about equally like to have an accident as the old employees (with safety training). In fact, the new workers, because they are new, might be slightly more likely than the old workers to have accidents: in that case, the number would increase even more. By assuming that number of accidents, rather than percent or ratio of accidents, stays the same, you are necessarily assuming that the quality of the new workers is of an order quite different from the quality of the old workers, and there's absolutely nothing in the prompt or in real-life experience to justify a mensch/ ubermensch split between old & new employees. Notice, for every A&B pair in the list above, we could construct a parallel argument demonstrating that, with a population increase, the rate would remain constant, not the number. This is precisely why places like insurance companies, the Center for Disease Control, the American Medical Association, government agencies, etc. keep track of rate & ratios & percentages, more than numbers. In thousands upon thousands of different data pools, it is understood that something, whether good or bad, has a fixed rate, and increasing the population size will increase the number even though the rate stays constant. This is absolutely standard throughout the social sciences, anything that uses statistical or epidemiological or actuarial data, and by insisting on fixed number rather than fixed rate, essentially you are directly assaulting the core methodologies of at least a dozen well-respected academic disciplines. This is precisely why conditional probabilities are crucial for understanding this problem, and to ignore conditional probabilities is not only to ignore something intrinsic to the situation, but to make vastly unwarranted assumptions about the difference between the old and new workers. What do you say to this? Mike I do not have much to say on this further than the following: Safety programs are expected to decrease the rate of accidents. When the new workers are hired, it is reasonable to only assume in the absence of any other information, that the number of accidents is at least constant if not lower than before, So if 1000 workers were already working and the rate of accident is 5%, then 50 accidents occur. If the safety program brought down the rate to 3 %, then 30 accidents would occur.If 500 more are hired which is higher than one would normally assume, then the number of accidents among the new workforce is 15. The total number of accidents is 45. In a realistic situation, one would expect the number of accidents to remain at least constant. That may be one of the reasons for the need of a an improved safety program. If the number of accidents actually increased we need to look beyond the factor of size such as the effectiveness of the program. That is the reason I do not find choice B appealing though in extreme cases of significant increase in workforce it does seem to explain the reason for the discrepancy.
_________________
Srinivasan Vaidyaraman
sravna@gmail.com
Sravna Test Prep
http://www.sravna.com
|
|
|
|
|
|
|
Re: Industry experts expect improvements in job safety training
[#permalink]
22 Dec 2012, 17:57
|
|
|
|
|
|
|
|
|
|
Moderators:
metallicafan, rajeevrks27, souvik101990, PTK, MacFauz, noboru, kissthegmat, carcass, willigetmylifeback, mikemcgarry, doe007, Vercules, Legendaddy, tuanquang269, RaviChandra, Marcab, Narenn
|
close
