ICPSR Collection Development: Lessons Learned

1
ICPSR Collection Development Lessons Learned
Moving Forward

• NDIIPP Partner Meeting
• Collecting Digital Content Going Forward
• Lessons Learned and New Initiatives
• Washington, D.C.
• 10 July 2008

2
Overview

• We have been successful at preserving important
  social science data for decades.
• Yet, important social science data have been lost
  despite our best efforts.
• We want to leverage the processes we developed in
  the Data-PASS project to avoid future loss.

3
Social Science Data Preservation Successes
4
(No Transcript)
5
ICPSRs Data-PASS Efforts
6
Collection Development Successes

• High response rate (44)
• Most PIs indicated that they wanted to be
  Good Citizens and help.
• This sounds like an exciting project.
• I hope your project is successful because I
  think that it is important.
• Good luck with the project.

7
Collection Development Successes

• Many of the data we have received are
  important mid-sized collections, which might not
  otherwise be collected.
• Around half of PIs still have access to data
  (49)

8
57 Commitments to Deposit

• U.S.-Sweden Cooperative Science Demographic
  Change in Nineteenth-Century Sweden (Katherine
  Lynch, 1989, NSF 8912792)
• Studies of the Economic Impacts of Global
  Environmental Change (Nordhaus, 1991, NSF
  9024860)
• Social Support and Mental Health in a Black
  Community (Dressler, 1980, NIH 1R01MH033943)

9
Recent Acquisitions

• Massachusetts Occupational Survey 1953-1954
  (Kahl)
• Hurricane Andrew Its Impact on Law and Social
  Control (Akers)
• Attitudes Toward Women and Work (Huber)

10
Challenges

• Good Citizens high willingness but no time,
  money, or resources to submit to us.

11
Challenges

• A good number (24) of PIs no longer could access
  their data.

12
Archival Status/Availability of Identified Data
(n1,668)
13

• Data Discarded
• It was just too long ago, I generally keep data
  for something like 10 years beyond the last time
  I do something with them.
• Data cards were destroyed long after several
  articles were published from them.
• Destroyed, in accord with APA 5-year
  post-publication rule.
• Hardware Problems
• Some data were collected, but the data file was
  lost in a technical malfunction.
• Electronic version was lost in a computer
  failure.

14
Destroyed for Confidentiality Reasons The
materialwas considered sensitive data.
Institutional review boards, and the state
registry board from which some data were derived
required us to promise to destroy the data after
a certain period of time, which was done a year
or so after publication. When the project
hibernated it was eventually decided to destroy
the records in order to protect the identities of
the subjects. I destroyed the records about a
year ago when the building that housed them was
torn down.Since I could no longer assure the
confidentiality of the records, I destroyed
them. Acts of Nature The data from the studies
were on punched cards that were destroyed in a
flood in the department in the early 80s.
15

• Discarded or Lost in a Move
• The data were on punched cards, which probably
  were discarded (after a decade or more of
  non-use) in one of several office moves.
• They were all published in journals, and when I
  moved from University A to University B, I
  discarded them.
• As I retired and gave up my office last year, I
  disposed of the data tape at that time.
  Unfortunately, I simply didnt have the room to
  store these data sets at my house.
• I lost track of them during several moves over
  the past 20 years.
• I had the data (ASCII and SPSS system files) on
  old floppy discs and they seem to have been lost
  in a move.

16
Obsolescence .a good deal ofdata from the
study were also entered (sometime in the late
1970s) onto the Universitys mainframe in an
early version of SPSS. I sincerely doubt that
they are still accessible in electronic formis
there even such a thing as a mainframe any
more? The data were backed up several times
using different media, but eventually, the media
became obsolete and I did not transfer it to
newer technologies. Speech recordings stored
on a LISP Machine, an experimental computer
which is long obsolete.
17
Simply Lost This was a long time ago, university
has changed its computer systems several times
since. I am not able to locate the data. I
really cannot recall. It was decades ago, and I
subsequently moved to new positions in various
universities. The records of the responses
obtained are forever lost. I dont honestly
know where they would be now. For all I know,
they are on a University server, but it has
been literally years and years since the research
was done, and my files are long gone.
18
Challenges

• Some PIs have data, but data are in obsolete
  formats, without sufficient documentation, or
  require massive clean-up.

19
Proprietary Formats of Data Not Archived
And Also .aiff, ArcView (ESRI), ARLEQUIN,
ATLAS.ti, AVI, Binary (UWAR format), BMPD,
Brainvoyager, Bvh, Coldfusion database, .DAT,
Dbase, DICOM, Digital Voice Files, .DSS, E-PRIME,
EQS, EVIEWS workfile, FLASH, FSL files, GAUSS,
HTML, iMovie, JPG, JMP databases, MAXQDA,
Molecular workbench database, MP3, MySQL, NEXUS,
.NSP, NVivo, Paradox data files, PHYLIP, QDS,
REFLEX, Rich Text Format, SALT transcription
format, Sawtooth WinCati, Sequencer, SQL,
Statistica, Statmost, STRUCTURE, Systat,
Teleform, Trimble Pathfinder spatial coordinate
raw data files, TSP, Viso, WordPerfect.
Last updated 8/17/2007
20
Storage Media of Data Not Archived
And Also cassette tape, DAT tape, digital video
tape, McBee cards, minidisks, mini digital video
tapes, network drive, QIC-80 tape, reel-to-reel
audio tape, reel-to-reel videotapes, Sun UNIX
network, 16mm film.
Last updated 8/17/2007
21
Collection Development Policy Update
22
Specific Updates

• Regular harvests of NIH NSF awards.
• Early, ongoing contact with PIs.
• Outreach Education.
• Peer Network.

23
(No Transcript)
24
(No Transcript)
25
Outreach Education
26
Peer Network
27
Collection Development Policy Update
28
Jared LyleICPSR(734)763.6075lyle_at_umich.edu