Dissertation Information for Robert Jay Belvin
- Robert Jay Belvin
- Library and Information Science
- Rutgers University (USA) (2006)
- Betty Turlock
- Lea Stewart
- Tefko Saracevic
MPACT Status: Complete - Except Indecipherables
Title: An exploration of satisficing within the FSCS-NCES Public Library survey data set utilizing forensic analysis
Abstract: This exploratory methodological study examined a library-related establishment survey to determine if forensic analysis in a nonprofit governmental environment had potential as a tool for the identification of possible satisficing behavior in responding to an establishment survey. Building on a theoretical base from the works of Tomaskovic-Devey, Krosnick and others, and translating the techniques of forensic analysis from accounting to a nonprofit governmental environment, the theoretical base and techniques were applied to the Federal State Cooperative System (FSCS) for Public Library Data from the United States Department of Education's National Center for Educational Statistics (NCES).
Forensic analysis located 31,340 red flag values indicating possible satisficing behavior in the approximately 450,000 responses from 9,048 public libraries. Four or more values were red flagged for 3,043 (34%) of the libraries. These red flags were concentrated within certain measures with 2,682 (30%) of the responses for reference transactions and 3,153 (35%) of the responses for library visits indicating possible satisficing behavior. They were also concentrated in the responses of the least affluent libraries that serve the smallest populations from single outlets with the least affluent libraries having a mean for red flags of 5.2 while the remainder of the libraries had a mean of 3.3.
The conclusions are that forensic analysis was a useful tool in identifying satisficing behavior in large establishment surveys and that significant additional effort is needed to ensure the collection of quality data from smaller and less affluent public libraries. The research findings indicate that policy decisions relying on information from establishment surveys must take into account the effect of satisficing behavior upon the information. Further research is needed to find if satisfying behavior occurs in surveys of other types of libraries and in surveys in other fields.