Digital Resource Log Analysis
- UC Berkeley
- Berkeley, CA
- January, 2005
Using data collected from several surveys (previously mentioned), combined with web server transaction logs, the challenge was to identify patterns and report exploratory results. A secondary challenge was to explore the feasibility of merging survey data with transaction log data, to see if usage patterns could predict survey responses. The results of this work are published as:
Harley, Henke, Lawrence, Miller, Perciali, and Nasatir. (2006) "Use and Users of Digital Resources: A Focus on Undergraduate Education in the Humanities and Social Sciences."
Highlights
- managed large data sets in the 5GB range
- created IP address anonymization procedure
- created Postgresql interface for SAS
- implemented web analysis algorithms in SAS
- prepared summary document for stakeholders
- hosted review meeting
- wrote relevant sections of final publication