Christina's LIS Rant
Monday, October 31, 2005
ASIST: Surveys of Scientists and Engineers
Surveys of Scientists and Engineers: Ensuring Reliable Research Evidence for Good Practice
Tenopir and King, The Critical Incident Technique
Cecelia Brown, Information Flow in Civil & Environmental Engineering in Oklahoma and Thailand
Surveyed a few Thai and Oklahoma grad students and faculty on their information use. Found some differences.
Brad Hemminger, Information Seeking Behavior of Scientists
Tenopir and King, The Critical Incident Technique
- Types of questions determine what you will be able to say from the results
- Demographic questions – the standards, plus academic discipline, questions related to productivity (articles authored), and awards received
- Recollection of behaviors – they try to stay away from this because memory is faulty. The farthest back they will go is about four weeks
- Opinions – reactions to statements on a scale, valuing services on a scale, they do very little of this
- Critical incident technique – this is the key type for them. A specific event – memory is better
- Incident of last reading –
- Variation of critical incident
- Two stage random sample 1) readers 2)readings
- Last readings are random in time
- Why?
- Specific
- Easier to remember
- Includes all reading (print, web, library, personal copies)
- Details of readings (how located time spent)
- Purpose motivation outcomes
- Their experience from 1970s, 30k responses, 100 orgs.
- An example, 10% of the last readings were done to support writing (they mainly came from library electronic collections)
- Summary
- Saves time
- Gives more information than logs
- See http://web.utk.edu/~tenopir/research/survey_instruments.html (feel free to use and share data)
Cecelia Brown, Information Flow in Civil & Environmental Engineering in Oklahoma and Thailand
Surveyed a few Thai and Oklahoma grad students and faculty on their information use. Found some differences.
Brad Hemminger, Information Seeking Behavior of Scientists
- Hopes to reach all scientists at UNC, then the US. His program is a survey and interviews.
- Sampling census vs. random sample – he did census, less expensive, easier to cover a lot more people
- picked questions of interest to lis and library and to correlate to previous findings
- subjects are UNC faculty, grad students, and research staff
- got support from chairs, then contacted, followed up. 27% (904) response rate
- Correlations
- Some results
- Which of these do you use weekly or more freq -- Journals, online db 65%, web pages 65%, personal communication 50%
- How do you search 97% electronic
- Which interface do you prefer to start search google 50%, library 50%
- Often did you visit the library last few months like once, and when they did, it was to photocopy
- He’d like to get participants from sites all over the U.S. (volunteer now!)
Comments:
Post a Comment
