@osirus raises a good point about the ratings. i'm mostly worried about whether the ratings will be meaningful since many judgment calls would be required. also, not all questions are even necessarily a good gmat question and so don't even merit a rating.
in other user review sites like amazon or yelp where you rate how much you like items with 1-5 stars, the "wisdom of the crowds" is useful because each vote represents a personal opinion. so if i see that 500 people gave a restaurant 4 stars on yelp, then i'm pretty confident that the restaurant will be good.
but i don't think that the same logic applies to the difficulty rating for gmat practice questions. just because 500 people think that a problem is a 700+ question doesn't mean that the question is actually a 700+ question since less than 10% of test takers get over 700.
even if we said that only members who actually got 700+ on the gmat could rate questions, i don't think that these people would always be able to differentiate the 700+ questions from 600-700 questions.
imo it'd be best for a set of very experienced instructors to apply their best judgement to rate questions difficulty. but then this might not work because the experts may not have incentive/time to rate question difficulty.
please let me know what you think.
in other user review sites like amazon or yelp where you rate how much you like items with 1-5 stars, the "wisdom of the crowds" is useful because each vote represents a personal opinion. so if i see that 500 people gave a restaurant 4 stars on yelp, then i'm pretty confident that the restaurant will be good.
but i don't think that the same logic applies to the difficulty rating for gmat practice questions. just because 500 people think that a problem is a 700+ question doesn't mean that the question is actually a 700+ question since less than 10% of test takers get over 700.
even if we said that only members who actually got 700+ on the gmat could rate questions, i don't think that these people would always be able to differentiate the 700+ questions from 600-700 questions.
imo it'd be best for a set of very experienced instructors to apply their best judgement to rate questions difficulty. but then this might not work because the experts may not have incentive/time to rate question difficulty.
please let me know what you think.
