The official answer has not been disclosed and I could not find this question anywhere else, but let me make a case for 2 or B as the answer.
The conclusion is "by wearing life jackets, boaters can reduce their risk of drowning if they are involved in a boating accident." The evidence in favor of this is "ninety percent of the victims were not wearing life jackets at the time of their accidents. "
So, what we know is that 90% of those who died did not have their life jackets on. However the conclusion is speaking of helping a person survive an accident. So we are really talking about the relative survival rates of those who are involved in an accident and are wearing life jackets and are not wearing life jackets.
This is a classic setup for a critical reasoning problem. Let me give you another example from the Veritas Critical Reasoning 2 book. There is a problem that talks about bicycling on the left side and on the right side of the road. The conclusion is that it is more dangerous to bicycle on the left side than on the right side. The evidence is that 25% of bicycle accidents involve bicycles on the left side of the road. Now the question is - what does this prove? If we do not know what percentage of people cycle on the left then we do not know if they are safer or not. For example - what if 75% of bicycling is on the left side and only 25% of the accidents? That would make the left side safer. Yet if only 10% of cyclists are on the left and they account for 25% of the accidents this is not safer.
So, let's apply this reasoning to this question. We know here that 90% of those who drowned in boat accidents were not wearing their life jackets. But what if 99% of those involved in accidents did not wear jackets? Then you can see that it is actually safer not to wear a life jacket because the percentage of those involved in accidents is greater than the percentage of those who drowned. Using these numbers you can see even more clearly that if 99% of those in accidents did not wear life jackets only 1% did wear them - yet 10% of those who drowned were wearing life jackets. So that means that - as a percentage - many more of those who drowned were wearing jackets. So in this case the evidence would fail.
Now answer 2 (or B) gives tells us that more than 10% of those who were involved in accidents were wearing the life jackets and since only 10% of those who drowned were wearing life jackets you can see that there are people surviving accidents at a greater rate if they are wearing life jackets.
What a tricky question!