There are several potential reasons for total non-response. In a household-based survey, the housing unit might be vacant or destroyed. If the housing unit is intact, the occupants might not be at home, or the interviewer might not be able to locate the housing unit. Assuming the interviewer locates the housing unit and at least one occupant is home, that occupant might be unable to answer due to illness, the need to care for children, or another reason. Finally, the occupant might refuse to participate in the survey all together.

Refusals may result from apathy, fear of invasion of privacy, fear of direct threat, from the police or others, or any number of reasons. Some refusals are partial, where the respondent will answer some questions but not all. The end result is item non-response. Others are temporary, where the respondent will answer on the second or third contact.

The consequences of non-response can vary. As non-response increases, the potential for a biased sample increases. This means that the obtained responses of a random sample may no longer be representative of the larger population. In short, response bias can reduce a random sample to what is essentially a convenience sample and, consequently, the conclusions are much weaker.

1. Modified from http://edis.ifas.ufl.edu/PD008 (accessed 29 December 2006).