Mar 8, 2012
Spelling Suggestions and know title searching were the most often selected auto-complete suggestions
The rate for using autocomplete suggestions on subject searches was much lower (no use in one test case, half used it in another test case). Initial research they didn’t use it, but when they got farther into the research project, they would use the suggestions to refine their focus (“extra brainstorming”). If they felt they already were knowledgeable about the topic, they were less likely to use any of the suggestions.
When asked where the suggestion were coming from, they usually thought it was from items we held or matching results that would come up (from those suggestion) although some said it was Google-like (it is using the Google API).
Bill let the group look at the raw data for UGL Easy Search usage for January-August 2011
72,300 total searches during the period performed from the UGL home page (or that originated from UGL page and continued on the Easy Search results/refinement page)
46,353 total searches resulted in an autocomplete choice being offered
17,327 times people choose one of the autocomplete options
37.38 % of the time chose one of the (Google Powered) auto-complete suggestions.
Jenny sent us a revised spreadsheet thinned down to the 23 EBSCO databases we subscribe to. We are limited to including 15 in Primo (by EBSCO rules limiting how Primo is allowed to interact with the EBSCO API).
Josh wants to know if we would be allowed to create our own EBSCO gateway/interface and then use that instead of letting Primo manage EBSCO.
The “E-journal” collection doesn’t appear to actually be a full-text or A&I database, but a service that provides links to EBSCO full text items in one collection in results in non-full text EBSCO databases.
We wonder why MLA, Business Source Complete, American History & Life, [women’s study database], [LGBT not there].
The meeting adjourned at 5:04, and we’ll pick up with the EBSCO database discussion next week