Comment by Shenglong | Hacker News Reader

Comment by Shenglong | Hacker News Reader

Parent Full thread

It might not be that she's overthinking--it could just be that the test-writers don't have as much education/intelligence. I use ad-hominems sparingly, so let me offer a specific example for the beginner's license test in Halifax, NS, Canada:

In cold conditions, which of the following freeze before the others?

1. Bridges

2. Overpasses

3. Shaded Areas

4. All of the above

I wasn't entirely sure what a "shaded area" was, but I figured eventually that it must be some area under shade from the sun. Next, "all of the above" didn't make sense, because how can "all of the above" freeze before all of the above? I figured that water has a higher specific heat capacity than ground, and thus in cold conditions, the heat from the water would keep bridges warmer for slightly longer. I answered Overpasses.

Apparently the answer was "all of the above". The funny thing is, when I ask this question to people, I find a pretty consistent correlation between responses: for my non-university educated friends, they typically answer D. On the other hand, my university educated friends (whose majors vary significantly) tend to answer A, B, C, or "D doesn't make sense".

You would do well to assume the question isn't meant to be self-contradictory, and find a meaningful intepretation under this assumption. In this case, "before the others" could be extended to mean "before any other areas".

Actually the question should be rephrased to "Which of these can form precarious driving conditions in cold weather?" because that's what they actually want to know.

The problem as I see it is that the test-makers are trying to be overly clever and not ask the question they want the answer to. If I have to "extend" the language to say something it doesn't, the test maker has failed because the test question is the only source of truth.

This is the issue I see most often with these sorts of questions.

That rephrasing doesn't really work either. All roads can be precarious in cold weather; the point of the question is to determine whether or not you realise that some areas can become dangerous more quickly than others even though the rest of the road system has seemed "safe enough so far".

(And the "hidden question" is "did you know that shaded areas can freeze as quickly as a bridge or an overpass?" Most people who grow up around real seasons develop an intuitive understanding that thin, isolated objects cool off very quickly, but urban types are exposed to an artificial sense of ground temperature. The point is not so much to know D is correct before taking the test, but to realise that since both A and B are correct, then C must also be correct. It could have been better-phrased, yes, but a lot of questions on written driving tests are of that sort—they can teach as much as they test.)

* slight nitpick. I changed the basis of the question to encompass the need to reassess driving style in the presence of a change in weather conditions, as opposed to choosing specific instances of danger. That is not a rephrasing as I originally contended. My apologies.

If all roads can be precarious, the answer is all of the above, thus solving the dilema of the test taker understanding the need to be more perceptive in colder weather. We cannot control for all variables that would make a road slick or dangerous, and therefore cannot test them all individually, but must group them together into a situation (ex: road conditions changed). If we accept that driving in and of itself is an active activity, then we should not have to distinguish between specific changes in road conditions for any change in road conditions should be enough for the driver to reevaluate their driving.

This question on the test should then be changed completely to reflect the fact that a change in road conditions necessitates a reevaluation of driving style.

I would further contend that "hidden questions", such as the one described, test one's ability to take multiple choice tests. If I know A and B, I have no need for C and there is no need for me to even look at it. In fact, if there were 40 options, "All of the Above" still has to be the correct answer, even if I have no clue what the other 37 options are.

Teaching is not the intent of a test. Unless, of course, we are assuming that the test taker did not learn anything before taking the test; In which case the intent is not testing them at all but ensuring that they have some basic skills to get by.

However I agree with your intent, the creation of questions for a test is not a trivial pursuit.

I think the "meta-context" must be taken into account: it's a driving license test, if I got it correctly. The examiner wants to make sure you'll be a careful driver. From this perspective, the mysterious D answer totally makes sense and is obviously what they expect.

I really have a hard time understanding this lack of "meta-intelligence" (or should we say "street smarts", that'd explain the success of at least some of your non university educated friends)

While it's evident that D is nonsensical, it's just as obvious that it is indeed the intended answer.

I expect if you are smart enough to detect it is nonsensical, you are also smart enough to deduce what you are expected to answer. I fail to understand why people nitpick themselves to death by a million cuts.

Maybe I'm not smart enough!

A bridge doesn't have to be anywhere near water. I would have answered "Shaded Areas" - why wouldn't a bridge or an overpass that's lit by the sun keep warm longer than an area that's in the shade?

Reply on news.ycombinator.com