Benchmark Statistics Service BEST

1
Benchmark Statistics Service(BEST)

• Whats going on with Australian repositories?

2
What are repository statistics?
3
Who cares?

• Policy-makers and meta-researchers
• What products of research funding appear in
  Australian repositories, and how are they used?
  Gaps, growth areas, hot topics etc.
• Repository managers and administrators
• How do the contents and usage of my repository
  compare with others?
• Authors, depositors, users
• Item/author usage, deposit patterns, etc.

4
All repositories can tell you
What theyve got
5
Many also show
How theyre used
6
But how does this
Compare with this?
7
At a grand scale

• Numerous discovery services and directories
  (OpenDOAR, OAI-ster, Repository 66)
• Also some aggregators of content information
  (ROAR)
• Many industry-specific standards, followed to
  some extent by some of community (OAI-PMH, SUSHI,
  authority lists)

8
At a grand scale

• Other relevant statistics projects
• JISC IRS has examined ways of using OAI-PMH to
  harvest event information, and has recently
  developed IRStats, a package to process EPrints
  and DSpace logs at the repository level.
• MESUR and precursor activities have examined
  aggregation using link servers.

9
Whats missing

• The ability to combine and compare repository
  usage statistics across Aust. repositories.
• Broadly applicable standards for defining and
  representing contents and events
• Means of adjusting for variation between
  repositories, communities, disciplines,
  industries
• Service for harvesting, aggregating and
  presenting repository statistics

10
Whats the approach?

• Short project to design pilot service,
  demonstrating the possibilities of aggregated
  usage statistics
• Limited participants in first round primarily
  research institutions
• Build on existing work (IRS, MESUR, ROAR) and
  standards (DC, METS, OAI-PMH)

11
Whats the approach?

• Work with related groups (MACAR, PILIN, People
  Australia)
• Interoperable (use OAI-PMH software, platform
  and repository agnostic)
• Extensible (future needs of new users, other
  repository types, new applications)

12
Steps to an aggregator service

• Identify a group of repositories for pilot
• Identify priority use cases
• Define scope, guiding principles, needs
• Define data elements, schema
• Assess and build repositories ability to expose
  standard reports as defined
• Design and build aggregator
• Present to users
• Feedback, refine, extend

13
Future possibilities

• Integration with other systems (ORCA, AONS)
• Use of OpenURL
• Alternative to citation indices for assessment of
  research impact
• Examine user behaviour threads, use of related
  subject/author links