UW Madison UW School of Medicine and Public Health
UW Biiostatistics & Medical Informatics UW Biostatistics & Medical Informatics UW Madison UW Biostatistics & Medical Informatics Site Map


General Departmental Seminar Series

Do Americans or the British have more national pride? :
Analyzing Likert Attitude Data with Varying Response Styles

Kristin Javaras
Waisman Laboratory for Brain Imaging and Behavior

Friday, November 30
12:00 pm
5275 MSC


Researchers often collect Likert attitude data to measure
individuals' underlying attitudes towards entities such as medication
or the individuals' own countries. As an example, Likert data
measuring attitudes towards one's own country consist of responses to statements such as "I am ashamed of my country" and "My country is better than all others"; responses are selected from ordered
categories ranging from total agreement to total disagreement.
Individuals' responses are affected by their underlying attitude, and
also by their response style, which is defined as a consistent and
content-independent pattern of response category selection, such as a
tendency to agree with all statements. Models that ignore response
style differences can produce biased inferences, especially when
respondents are demographically diverse and the proportions of
favorable and unfavorable statements are not equal. This talk
introduces a new latent variable model that allows response style, as
well as one or more underlying attitudes, to affect the data. The
model can be used to examine the effects of background covariates on
underlying attitudes. As an illustration, data from the 1995 National
Identity Survey (NIS) are used to investigate whether Americans or
the British have more favorable attitudes towards their respective
countries. The results are adjusted for American and British
differences in response style; to better measure these differences,
additional NIS data on immigration attitudes are included in the

Return to seminar list