Bias
Bias is an unintended influence on a sampling method.
Sampling Bias
Sampling bias occurs when the chosen sample does not accurately represents the population
Household Bias
Household bias (a type of sampling bias) occurs when one type of respondent is overrepresented, usually because groupings of different sizes are polled equally
Response Bias
Response bias occurs when factors in the survey design influence the result
Leading Questions
A leading question is designed to make a respondent answer in a certain way
"Our company offers the best sandwiches, right??"
Loaded Questions
Loaded questions contain an assumption of guilt or controversial information
"Have you stopped being a violent felon?" or "have you stopped beating your wife?"
Non-Response Bias
Non-response bias occurs when results are influenced due to lack of response
Only those who care about an issue enough will answer a survey, which leaves out the less extreme opinions.
Funding Bias
Funding bias occurs when a study's outcome is likey to support the organization funding the study
Measurement Bias
Measurement bias occurs when the data-collection method consistently either under or over-estimates a characteristic of the population.
Often, measurement bias results from a sampling process that affects the variable it is measuring.
A police officer watches a stop sign from their car to observe driver behaviour.
In this case, drivers are more likely to drive safely with the obvious presence of the police officer.