A liberal organization contacts a simple random sample of 500 Manitoba voters, asking them the following question: “Given the failures of the current Conservative government on health care and education, will you support the Conservative Party in the next election?” 34% of those responding answered yes. Do you suspect an underestimate or overestimate to result from this survey? Why?
Underestimate because the manner in which question is formed is going to create a negative bias in the respondent's bias towards the present government due to its ineffectiveness in health care and education area's failure. Hence a given respondent is more likely to give answer "NO" to the stated question than "Yes" ; hence it is going to be underestimate.
Comments
Leave a comment