Every little helps : an investigation of frequency biases in comparative judgments
Lewis, Richard, 1979- (2010) Every little helps : an investigation of frequency biases in comparative judgments. PhD thesis, University of Warwick.
WRAP_THESIS_Lewis_2010.pdf - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Official URL: http://webcat.warwick.ac.uk/record=b2341138~S15
Intuitive statistical inferential judgments involve the estimation of statistical properties of samples of information, such as the mean or variance. Prior research has shown that human judges are generally good at making unbiased estimates of sample properties. However, a series of recent applied consumer research experiments demonstrated a systematic bias in comparative judgments of item distributions in which the individual items are paired across those distributions, for example comparing the prices in two stores selling the same items. When the two distributions have the same mean, the distribution with the higher number of items that are smaller in magnitude than the equivalent item in the other distribution is typically judged to be the smaller of the two distributions: a frequency bias. In a series of experiments, the research in this thesis provides a robust demonstration of the frequency bias and explores possible explanations for the bias. A comparison between simultaneous and sequential presentation of information demonstrates that the frequency bias cannot solely be explained by the salience of the frequency cue. A novel web-based experiment, in which information was sampled incidentally from the environment and a naturalistic task was used to elicit comparative judgments, showed that the frequency effect persists in an ecologically-valid context. A systematic comparison between alternative cognitive models of the judgment process supports an explanation in which items are recalled from memory and compared in a pair-wise fashion, meaning the frequency bias may be found in a wide range of other judgment tasks and domains, which would have significant implications for our understanding of intuitive comparative judgments.
|Item Type:||Thesis or Dissertation (PhD)|
|Subjects:||B Philosophy. Psychology. Religion > BF Psychology|
|Library of Congress Subject Headings (LCSH):||Prejudices, Judgment -- Testing, Prices|
|Institution:||University of Warwick|
|Theses Department:||Department of Psychology|
|Sponsors:||University of Warwick. Dept. of Psychology|
|Extent:||xxx, 330 leaves : ill., charts|
Actions (login required)