It's not possible. Several of the questions have multiple valid answers. It's pretty obvious what the scheme is.
It's not possible to know the right answers because there never were any. This means the test has no predictive power, not that it's impossible, and again, since some Whites unable to prove education did have to contend with this, it was designed that way intentionally.
I feel "near impossible literacy test" is a terrible description. The "intentionally ambiguous literacy test" would be more apt.
More worrying is I am unable to find a definitive provenance for this document. It suggests it was used in the early 1900s but the print quality and format seems unusual in several ways to me. Which is why I attempted to reduce it in favor of considering the rest of the system.
Most notably, pilot examinations for aviation and maritime certifications.
I think they use these types of questions to exclude rule memorisation and test the ability to reason about the intention or relevant effects of rules and principles of the art involved.
It seems like the intention is also to penalise the inability to reason about ambiguous situations, ensure that the subject can effectively divide attention (if you spend too much time focusing on these ambiguous situations trying to find a nonexistent perfect answer you will fail the test), and to filter out low cognitive ability in general.
I’m not a test design expert, however, so ymmv.
"Impossible" also refers to how the test administrators used it: in order to make voter registration impossible for some people.
> That comment is a reflection of my pedantry
Stop with this sort of thing, please. It's just noise, and doesn't add to the discussion.
> 28. Divide a vertical line in two equal parts by bisecting it with a curved horizontal line that is only straight at its spot bisection of the vertical.
I have no idea what a curved horizontal line is. A horizontal line is parallel to the X axis of the XY plane and has no curvature.
I’d be interested to know which ones you saw as ambiguous?
FWIW the test is obviously mostly about tricking the test taker, and not that much about literacy. Along with one question that seems possibly designed to filter out people with a non-Christian interpretation of the cross as a geometric figure.
Could be answered with:
right
right from the left to the right
right from the left to the right as you see it spelled here
That one is tricky but not actually ambiguous.
For instance, "We need to make sure you agree to the conditions. Write I agree on the line below." is pretty unambiguous in its intention if you assume the writer is communicating in good faith, even if it could be quoted to make it clearer.
In the case of the original test, the lack of good faith is the entire point, which is why the sentence is considered ambiguous in this case. In a different context where you could assume good faith, you would be correct.
Any other interpretation fails to produce a consistent, sensible rule; therefore the only logical conclusion is to assume the single word case. The single word interpretation is clearly the less wrong answer.
Many tests accepted as being legitimate and foundational to regulating aviation, marine navigation, and other important occupations include questions that have no “right” answer, only a less wrong one. This is intentional and useful to judge certain aspects of understanding and judgment in those safety critical industries.
I don’t think this type of question is relevant to a “literacy” test, though.
I agree that the test is clearly given in bad faith, and is largely not a literacy test but rather a series of trick questions that require much more than literacy to analyse - but I reject the premise that this question in particular has no correct answer.
The question is in effect a multiple choice question with a few specific granular possible answers, with one being clearly less wrong than the others by process of elimination referencing the epistemological content contained in the question, with the operative assumption being that there is a correct answer.
The question would be more at home in a test to probe advanced reasoning or logic skills, perhaps even philosophy, but still it has exactly one arguably correct answer and therefore fails the bar for being irreducible in its ambiguity.
If you can provide a convincing argument based on a logical premise that supports an alternative answer other than the single word interpretation, I will be forced to reconsider my opinion. Until then, I maintain that this question has a correct answer.
The fact that the test is administered in bad faith and the answers may be judged in bad faith is immaterial; the test Could be perfect and still be judged in bad faith, so there is no argument about the technical validity of the test material anchored to the good faith of the examiner.
It can only be said that the process of the test and judgement is ambiguous, not the test itself, if the test itself is solvable.
It can also be said that the test was not designed to serve its stated purpose, or was poorly designed for its purpose, and that it was not meant to be given in good faith, but none of these statements has bearing on the solvability of the test questions.
That being said, I would expect to fail this test.
Both "backwards" and "forwards" could be correctly interpreted as the adverb in this one. It could be asking you to "Spell the word backwards, in a forwards manner" or "Spell in a backwards manner, the word forwards".
It's ambiguous enough that someone grading the test who wanted the disqualify you could make the case you got it wrong, no matter if you wrote "backwards" or "sdrawrof".
1. Draw a line around the number or letter of this sentence.
I have no idea what "the number of this sentence" or "the letter of this sentence" even means.
Regardless of which you chose, if the examiner wished to disqualify you, they could simply say it's the opposite.
Most people would say "The number _preceding_ this sentence". "of this" sounds like it's part of the sentence.
However I argue that the question by itself is fine: it is well defined and has only one reasonable answer. No one presented any other sensible answer so far.
My problem with the whole discussion here is that I actually fail to come up with an ambiguous answer to this question that I subjectively find reasonable. Can you positivity contribute by providing an alternative interpretation that you personally would find more plausible? Otherwise complains about subjectively are hollow. Everything is subjective but it only matters in cases people actually disagree.
> ...attempts to fully remove ambiguity are often harmful.
Really? I'd like to see some real-life examples that are not just further examples of bad question-writing. Regardless, this is beside the point here, where there is a seemingly insurmountably high bar for making the case that removing ambiguity would make the situation worse (to be clear, the problem with tests like this is not just that an examiner might screw you, but that they are intended to be used that way.)
I don't claim this test is useful, but as a matter of fact the first question is not hard.
But then you can't believe it has two answers. If you don't understand the question you can't make any claims about it except maybe that you believe it is incoherent. As alexey-salmin says, the only reasonable interpretation is to circle the 1. to left of the question.
To claim that has 2 answers is similar to claiming a "Find x" style question has two answers (solving an equation vs. circling x). No, the question only has one reasonable interpretation. If the examiner is being unreasonable it is not a problem with the question and not something that can be deduced by examining the question.
It's "Paris in the the spring", with two the's!