This guy is really something. He crunched statistics of a huge sample of test questions to get some secret code of scoring well for MCQs. The underlying theory is that humans are exceedingly bad at making answers seem random. So, if you can crack the code of a test setter's 'random' answers, you can get a higher probability of scoring higher marks. So what's the secret techniques to scoring MCQs? |
To summarise:
Nothing beats good ol' hardwork and preparation for your test, of course. But if you're stuck, try using these empirically tested tips.
- For True or False questions, go through the entire test marking the answers you know before you attempt to try to agar agar.
- Look at the known correct answers of the items before and after the ones that you left blank.
- If the two neighboring answers are the same (T _ T or F _ F), guess the opposite (T F T or F T F). Humans can't stand 3 of a kind type of answers. It somehow seems non-random.
- If the two neighboring answers are different (T _ F or F _ T), guess T. Most answers are true since it's more difficult to think of something that is wrong that right.
- For 3 choices MCQ (A, B or C), all options are equally likely to be correct
- For 4 choices MCQ (A, B, C or D), choose B if you don't know what choice to use. It gives you about 28% of the time to be correct, compared to 25% chances. Just a little edge better.
- For 5 choices MCQ (A, B, C, D or E), choose E (23% compared to 20% chance). Don't choose C! (17% compared to 20%)
Nothing beats good ol' hardwork and preparation for your test, of course. But if you're stuck, try using these empirically tested tips.