Wednesday, October 2, 2013

Finding statistics during the government shut-down...

closed
photo credit: flickr user Nick Papakyriazis

On October 1, 2013 The government shut down (again).

I've been working with students to two sociology classes who's assignments were to gather statistics on a country or state assigned to them. Of course, a good deal of the statistics they need to gather come from government websites like the U.S. Census Bureau and the Bureau of Economic Analysis, whose websites now post shutdown notices rather than providing access to the data usually available.

Frantic students in those classes are now looking for help finding alternative sources for the information they need.  Here are a few suggestions for students:

1. Try using government sites that appear unaffected (at least so far) by the shutdown (i.e. bjs.gov and bls.gov are both still accessible at the time of this posting).

2.  Search for state government sites that contain federal data (http://www.statelocalgov.net, http://www.globalcomputing.com/StatesContent.htm, or http://www.50states.com might be good places to start).

3. Try a Google search for the state and statistic for which you are looking (ie. GDP and California).  Look for state websites among the results.  If your professor will not allow you to use wikipedia, you may wish to see if wikipedia cites a source you CAN use.

4. Try finding an article (scholarly or news) that incorporate the statistics you seek.

Tuesday, October 1, 2013

Skilling Up for Data Curation Infographic

My latest infographic, Skilling Up for Data Curation, using Piktochart examines the skills and tools I'll need for data curation at my campus.  The infographic was used for a poster session on the topic for the Fall 2013 conference of the Western New York/Ontario Chapter of the Association of College Research Libraries (ACRL).

There has been a lot of discussion over the last several year about the role that libraries should play in data curation efforts at their institutions.  Technical advances have made it possible for the creation of larger and larger amounts of information/data/research/scholarship.  How best to manage and preserve this influx is under debate, especially given the challenges; the sheer volume, different media types, intellectual property issues, obsolescence of formats/software and lack of metadata to name a few...

[caption id="attachment_190" align="aligncenter" width="300"]DataCurationLifecycle-DCC DCC Curation Lifecycle Model[/caption]

What we do know is that data curation must be a collaborative effort between librarians and data creators  for two important reasons: to have the metadata necessary for curation later in the data lifecycle and to education data creators about the need for standardization of metadata.  Consistent standards used by researchers within a discipline, or better yet across disciplines, will allow for the opportunity to automate some (perhaps all) of the curation process and the possibility for adding smaller datasets to the corpus of curated data outside our own small institution for reuse by others, attaining an even greater return on investment.

Digital curation is far removed from the the institutional repository of the past.  Reappraisal and providing access in ways (and formats) that the data can be readily reused is key so that our digital collection don't wind up looking like an old attic where we've abandoned our institutional data.

Through a series of discussions with faculty on our campus this summer, we found that, as yet, there is no great demand for data curation with respect to faculty research.  However, we have begun developing skill sets in this area so we'll be prepared with technology and infrastructure options in anticipation of future needs.

During these discussions, we found several areas where the library immediately could serve:

  • Provide assistance locating discipline specific repositories for finding and publishing research data.

  • Provide instruction or workshops for undergraduate students to improve skills in managing both laboratory and their own data.

  • Provide assistance in developing data management plans on funding applications.

  • Identify faculty work or research projects that could/should be digitally curated.


Education and ongoing discussion with faculty about developing standards for metadata and opportunities and benefits for open data sharing will be key.  We also have the opportunity to to be selective about the projects we pursue and to pace our digital initiatives in ways that are practicable in terms of resources, time and funding.

References:
Bird, C., Willoughby, C., Coles, S., and Frey, J. (2013). Data curation: Issues in the chemical sciences. Information Standards Quarterly, 25(3): 7-12.

Digital Curation Centre. (n.d.). Data curation lifecycle model.

DataCite: Helping you find access, and reuse data. (n.d.). Why cite data?

Lee, C. A., Tibbo, H., & Schaefer, J. C. (2007a). DigCCurr: Building an international digital curation curriculum & the Carolina Digital Curation Fellowship Program.

Schirrwagen, Jochen, Paolo Manghi, Natalia Manola, Lukasz Bolikowski, Najla Rettberg, Birgit Schmidt. (2013). Data Curation in the OpenAIRE scholarly communication infrastructure. Information Standards Quarterly, 25(3): 13-19.

Smith, K. (2009). All universities should have an institutional repository. Bulletin of the American Society for Information Science and Technology (Online), 35(4), 11-31.