Information Relevance, Content and Source Credibility in the Revision of Opinions
Abstract
In the first of two experiments participants assessed either (1) the probability of success or (2) the acceptability of hypothetical job candidates. These judgements were made on the basis of information about particular traits or abilities possessed by the candidate. The traits themselves varied in relevance to the job and the sources of information differed in credibility. Results showed that the information was discounted (had reduced influence on the judgment) when its relevance decreased and when its sources credibility decreased. Four discounting models were examined and rejected, although by estimating parameters from part of the data and fitting the remainder, one model could be acceptably modified to account for the data. Discounting was a multiplicative rather than a subtractive process and the same process apparently accounted for both the acceptability judgments and the probability assessments, which were virtually identical. In the second experiment, participants received three pieces of information about the potential negative effects of constructing nuclear power plants on a number of potential sites. For each site, participants revised their judgments about the favorableness of the site after each of the three pieces of negative information. Again, discounting was obtained as a function of the credibility of the information sources and a form of 'conservatism' was obtained; regardless of the source and the content of the message, the amount of revision decreased from the first to the second to the third revision.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1976
- Accession Number
- ADA030945
Entities
People
- Joyce Prothero
- Lee Roy Beach
- Marcia D. Deaton
- Terence Mitchell
Organizations
- University of Washington