Flashcards in Unit 7 Deck (9)
List 4 sources of inaccuracy and bias in polls.
1. Sampling procedure.
2. People not answering the poll questions truthfully.
3. People might answer differently to different interviewers.
4. Stratification assumptions. When for example dividing the population into groups (strata) - the information about these groups can be biased or wrong.
List 3 reasons why a sample might be unrepresentative of its population.
1. Only the people that could be reached were asked the questions. This might exclude the group that does not have that specific method of communication.
2. Certain group of people might be more inclined to answer the questions asked in the poll.
3. A sample could have been selected from a formed group of people with a certain opinion. (Think a poll on a left-wing website - it is rarely visited by right-wingers)
Can you think of any sources of inaccuracy or bias in polls and surveys other than those mentioned by Huff?
A poll might be conducted during certain public events that might swing the public opinion.
What is a mean average?
Sum divided by the count.
What is median?
Is the number in a set where half the numbers are lower than it and half are higher.
What is mode?
Mode of a set of data is the number that appears most often in a set.
What are four reasons why figures might be impossible to interpret.
1. If the sampling procedure(size of a group) is not known.
2. The degree of significance is not given.
3. Deviation from average is not indicated.
Why is a degree of significance important?
The degree of significance shows the probability of the degree of precision. For example if the degree of precision of the study is +- 1 percent that is impressive. However, if the degree of significance is 5 out of 10 tests that indicates that only in 5 cases out of 10 this degree of precision will apply. In the rest of the cases this degree of precision cannot be guaranteed.