On the previous pages you learned about sampling bias and how simple random sampling methods can be used to avoid sampling bias. Here, we will discuss two other sources of bias: non-response bias and response bias. These are both problems that should be prevented in the design of a research study.
- Non-Response Bias
- Systematic favoring of certain outcomes that occurs when the individuals who choose participate in a study differ from the individuals who choose to not participate
- Response Bias
- Systematic favoring of certain outcomes that occurs when participants do not respond truthfully; they may do so to align with social norms or to appease the researcher
Example: Restaurant Experience Survey Section
A restaurant invited their recent customers to complete an online survey. Customers who had really strong feelings about their experience, either positive or negative, were very likely to complete the survey while customers who had a neutral experience were much less likely to complete the survey. This is an example of non-response bias because the individuals who chose to participate differed from those who chose to not participate.
Example: Retail Store Hours Section
A retail store was considering expanding their operating hours. To determine if this was a need perceived by their customers, they conducted a survey over the telephone to obtain data. Research assistants called the phone numbers of customers who were randomly selected to participate between the hours of 9AM and 4PM. Individuals who were at work were less likely to answer their phone call or agree to participate in the study than individuals who were at home at that time. This is an example of non-response bias because the individuals who responded to the survey were different from individuals who did not respond in terms of their work schedule.
Example: Sexual Activity Survey Section
A psychologist is conducting a research study concerning sexual activities. The survey is administered over the phone and many of the questions are personal. Some participants feel uncomfortable and do not answer honestly. This is an example of response bias because the participants are not responding truthfully; instead their responses are biased toward what they perceive as being socially acceptable.
Example: Cheating in Class Section
Using an anonymous online survey, a professor asks his students “Have you cheated on an exam in my class?” Many of the students who have cheated still answered “no.” This is an example of response bias because the participants are not responding truthfully; instead their responses are biased toward responses that are less likely to get them in trouble.