My question - the stimulus provides death rates per thousand members for each group. With a bit of multiplication, we can determine death rate / thousand! Why is it useful?

I thought the same thing at first, but if you look at what the stimulus is actually saying, it is providing total number of deaths per year as a function of the entire population. Not just out of the members of each group.

So the chances of dying in an automobile are 1 in 19,000 this means that one out of every 19,000 people in the entire population will die while driving our riding in a car. Now there is a good chance that most people will ride in a car several times per year - even if it is just a taxi.

Now the figure of 1 out of every 73,000 people being killed on a motorcycle is a bit misleading. I have not been on a motorcycle this year and neither has any member of my family, so we have zero chance of dying on a motorcycle. In fact I doubt that 1 out of every 10 people ride a motorcycle even once in a given year. So it makes sense that a smaller percentage of the general population will die on a motorcycle.

Answer Choice D corrects this flaw by talking about the death rates as a proportion of the members of those groups. So if you take out the 90% of people who do not ride a motorcycle then suddenly the death rate jumps to like 1 person out of every 5000 motorcyclists will die each year.

Does that help?

David,
Thanks for your reply. If I understand it correctly,

You have said two things :

David@VeritasPrep wrote: #1-I thought the same thing at first, but if you look at what the stimulus is actually saying, it is providing total number of deaths per year as a function of the entire population. Not just out of the members of each group.
#2-Now the figure of 1 out of every 73,000 people being killed on a motorcycle is a bit misleading.

If I understand correctly, because the phrase "73000 people" could refer to general population or total motorcyclist, the statistics is misleading. D) clarifies this confusion by making it clear that we need % of individual groups.

Am I correct? Please let me know.

Thanks

You are correct. The stimulus talks about members of the population at large and this may be misleading if few people ride motorcycles.
D is more precise.

voodoo_child wrote:David,
Thanks for your reply. If I understand it correctly,

You have said two things :

David@VeritasPrep wrote: #1-I thought the same thing at first, but if you look at what the stimulus is actually saying, it is providing total number of deaths per year as a function of the entire population. Not just out of the members of each group.
#2-Now the figure of 1 out of every 73,000 people being killed on a motorcycle is a bit misleading.

If I understand correctly, because the phrase "73000 people" could refer to general population or total motorcyclist, the statistics is misleading. D) clarifies this confusion by making it clear that we need % of individual groups.

Am I correct? Please let me know.

Thanks

It's a sampling flaw, which states that the given figures are not representative of the whole population.

Thanks,
Ankit

Here is an analogy that i used. Someone please correct me if this is wrog

Assume there are 100 people in a county . There were 10 automoile deaths and 2 motorcycle detas in the county.Since 10>2 the argument concludes that driving an automobile is unsafe.

The reason this argument could be flawed is-- what if there were only 2 people in the county owning and driving motorcycles but 90 people owning and driving automobiles
% of people pwning and dying on a motorcycle--> 100%
% of people pwning and dying on a automobile --> 11%

Obviously automobiles are safer.

My question at this point is
The analysis that i had put in is a whole lot. I think it is extremely impossible for me to come up with such an analysis on a CR problem under 1 minute 50 seconds.
The analysis that i put in seems out of the box to me

What should i do to get there in 1 min 50 seconds other than practising these type of problems regularly ?
Can some xpert please advise?

My question at this point is
The analysis that i had put in is a whole lot. I think it is extremely impossible for me to come up with such an analysis on a CR problem under 1 minute 50 seconds.
The analysis that i put in seems out of the box to me Sad

What should i do to get there in 1 min 50 seconds other than practising these type of problems regularly ?
Can some xpert please advise?

First, I am not sure that this really would take you more than 2 minutes. And even if it did, there are other questions that will take less time so you could spend 3 minutes on a question and it would not be a big deal.

Second, this is a standard weighted average flaw. The comparison is not valid until you know what the population of each group contains. So if in the future you see a statistic that says "In the U.S. 500,000 more degrees are awarded to women than to men" and the conclusion says "therefore women in college are more likely to graduate than are men." You will think, "wait, what if there are lots more women in college." So you will not need to do "your analysis" each time.

Third, this is an example of type of problem that is becoming much more common - in fact it has two characteristics that are becoming more common. It is a "most useful to know in order to evaluate" question and it is a question on statistics. Both of these categories are increasing on the GMAT. So you will want to know how to do this question.

Third, this is an example of type of problem that is becoming much more common - in fact it has two characteristics that are becoming more common. It is a "most useful to know in order to evaluate" question and it is a question on statistics. Both of these categories are increasing on the GMAT. So you will want to know how to do this question.

Wow this is some news!-- Thanks David.

So the only way to improve speed on these type of problems is to keep doing more of them thereby getting adjusted to these problems?