|
The Connection Between Survey Nonresponse
and Urbanicity
On Tuesday, December 4, 2007, Allyson Holbrook, Assistant Professor in the Public
Administration Program, Survey Research Laboratory, and the Department of Psychology at the
University of Illinois at Chicago presented a seminar entitled
The Relationship Between Survey Nonresponse
Bias and Urbanicity: Identifying Possible Mechanisms and Solutions. The focus of the lecture
was to review evidence regarding the relationship between urbanicity and nonresponse and
its effects on surveys as well as to address why improving response in urban areas is important.
Holbrook spoke about possible reasons for nonresponse in surveys categorizing them as either
non-contact or non-cooperation. Non-contact occurs when a respondent is unreachable, for example
when no one is home, or no one answers the phone. Non-cooperation happens when a potential
respondent refuses the survey for one reason or another. Nonresponse affects the quality of a
survey and is typically categorized as a nonresponse error in order to remove the assumption
from the results of a survey.
Allyson also addressed the issues associated with urbanicity and how that affects survey
nonresponse. She defined urbanicity to mean "the extent to which a particular geographical
location is urban." She stated that survey participation decreases the more urban the area.
There is more effort involved in getting people to participate in a survey in more urban
regions. This means that urban areas are typically underrepresented in surveys, which can
have implications for political surveys.
Holbrook made the case that it is important to understand why participation varies as a
function of urbanicity in order to reduce survey nonresponse bias by anticipating these
issues and developing strategies to correct for, or combat nonresponse, particularly in
urban areas.
Initial steps Holbrook is taking in her work are to measure urbanicity and survey nonresponse,
or define measures of them, and then test why urbanicity and survey nonresponse are related.
She believes it is important to look at strategies that increase response rates evenly across
groups rather than overall response rates. To "not just focus on getting more to participate,
but I want to get a representative sample of people to participate."
This lecture was part of the GCI Faculty Scholar Seminar Series, in which current Faculty Scholars
present the research they are conducting while at Great Cities Institute. Visit
the podcasts page to
download this presentation.
|
