Its making a conclusion about a population from a sample that is not likely to be representative. In the stimulus, the last week of the theater's operation drew sellout crowds, likely because everyone knows it is the last week and wants to go before it is gone forever and not because the owner's claim of not being able to regularly attract large enough crowds was false and, implicitly, that the theater can actually draw large crowds regularly.
AC C a meeting is held about proposed cuts in library funding. Given the subject of the meeting its likely that students who have strong feelings about the cuts to library funding would attend to make their voice heard, and those people would be in higher proportion relative to the proportion they make up of the entire population. This AC presents the sample of students who attended a meeting about a subject as evidence that all enrolled students have the same opinion.
There is a certain type of bias, I think it is called nonresponse bias, that describes the poor methodology that is analogous to the reasoning in AC C. Think of like a yelp review. Generally people don't bother leaving a review unless they were really upset about something, because they don't have to. Therefore you're likely to see a higher proportion of bad reviews from angry customers than what is truly representative of all customers. Basically, if something is not mandatory (attending a meeting about library funding) you can expect that the people not attending/responding don't really care about the issue, and that the people who have strong feelings about the issue are more likely to respond.
Comments
Its making a conclusion about a population from a sample that is not likely to be representative. In the stimulus, the last week of the theater's operation drew sellout crowds, likely because everyone knows it is the last week and wants to go before it is gone forever and not because the owner's claim of not being able to regularly attract large enough crowds was false and, implicitly, that the theater can actually draw large crowds regularly.
AC C a meeting is held about proposed cuts in library funding. Given the subject of the meeting its likely that students who have strong feelings about the cuts to library funding would attend to make their voice heard, and those people would be in higher proportion relative to the proportion they make up of the entire population. This AC presents the sample of students who attended a meeting about a subject as evidence that all enrolled students have the same opinion.
There is a certain type of bias, I think it is called nonresponse bias, that describes the poor methodology that is analogous to the reasoning in AC C. Think of like a yelp review. Generally people don't bother leaving a review unless they were really upset about something, because they don't have to. Therefore you're likely to see a higher proportion of bad reviews from angry customers than what is truly representative of all customers. Basically, if something is not mandatory (attending a meeting about library funding) you can expect that the people not attending/responding don't really care about the issue, and that the people who have strong feelings about the issue are more likely to respond.