Originally Posted by
Phaedrus
No. The percentage that get inaccurate results is totally irrelevant. The only thing we need is that a small percentage gets a correct result and that the cause of their correct results is that they understand what type they are for the right reasons.
How do you know which results are correct and which are incorrect? I'll illustrate my point with a story problem:Jan did a survey of the students in her class in order to create statistics regarding pie preferences. Eight of the students reported that they prefer cherry pie. Six of the students reported that they prefer banana cream pie. Six of the students reported that they prefer key lime pie. Four of the students reported that they prefer apple pie. Two of the students reported that they prefer lemon meringue pie.
A small percentage of the students answered the survey truthfully. What percentage of the students actually do prefer banana cream pie?