by Sharon L. Baker
The Library Quarterly
Vol. 56, #3, July 1986, pp 237-257.
Hypothesis 2 accepted: "Fiction books marked as "recommended" will circulate more than their counterparts in the prime display area, in the nonprime display area, or on the regular shelves because they (i.e., recommended books] serve to narrow the browser's choice of materials, thus providing a type of selection guidance that overcomes effects of information overload."
Experiment supporting hypothesis 2
Location:
- Mahomet (Illinois) Township Library: 10, 230 volumes
- Monticello (Illinois) Township Library: 31,214 volumes
- Experimental group at prime location: 38 of the 75 adult fiction books placed spine out on bookcart near circulation desk were marked with red dots. The display is under the sign, "RED DOTS ARE RECOMMENDED BY AN ILLINOIS LIBRARY COUNCIL." This claim is actually false.
- Experimental group at nonprime location: 38 of the 75 adult fiction books placed behind stacks at the back of the library were marked with red dots. The display is under the sign, "RED DOTS ARE RECOMMENDED BY AN ILLINOIS LIBRARY COUNCIL." This claim is actually false.
- Control group: 38 of the 75 adult fiction books remained on regular shelves. were marked with red dots.
- Pre-test period: July 1, 1984 to September 30, 1984, circulation measured for all groups while on regular shelves.
- Test period #1: October 1, 1984 to December 31, 1984, circulation measured for all groups in experimental condition and control group on regular shelves.
- Test period #2: January 7, 1985 to April 6, 1985, red dots were moved to books without red dots in test period #1 and then circulation was measured.
Patron interviews revealed that slightly more that half of those selecting recommended books, did so because of the recommendation. Ten percent choosing a recommended book claimed the recommendation did not effect their choice.
In Mahomet Library with only 10,230 volumes, using red dots to recommend books increased their circulation 108% to 160%. This increase was not statistically significant. This means the increase could be mere random variation of the measurement rather than the result of the red dots.
In Monticello Library with 31,214 volumes, using red dots to recommend books increased their circulation 257% to 367%. This increase was statistically significant, meaning that we can be pretty sure that the increase is due to the red dots.
Red dots had the greatest effect in the regular shelves where they increased circulation 471% in test #1 to 795% in test #2. Since regular shelves would cause greater information overload than displays, the dramatic increase of circulation in the regular shelves strongly supports Sharon Baker's hypothesis that red dot recommendations "narrow the browser's choice of materials, thus providing a type of selection guidance that overcomes effects of information overload."
