Archiving Nursing Research: The Challenges of Building and Electronic Data Library

1
Archiving Nursing Research The Challenges of
Building and Electronic Data Library
Chris Miller Knowledge Utilization Studies
Program Faculty of Nursing, University of
Alberta Netspeed, 2006
2
Thanks toCarole Estabrooks, PhD, RNCharles
Humphrey
3
(No Transcript)
4
The General Accounting Office recently audited
NASA and reported that hundreds of thousands of
NASAs tapes are currently kept under deplorable
conditions. The audit stated that NASA has not
performed an agency-wide inventory and does not
know what data are retained and what may have
been lost. Some tapes were found stacked on
pallets in warehouses where the temperature
reached in excess of 100 degrees. -- Johnson,
P. (1990). Who you gonna call?Technicalities 10
(4) 6-8.
5
Some of the original tapes are so fragile that
the layer of oxidized fiber that contains the
data flakes off as the tape is read. These tapes
are destroyed in the copying process and if
something goes wrong the data they contain are
lost forever.-- Blakeslee, S. (1990, March 20).
Lost on Earth Wealth of Data Found in Space. New
York Times. C1.
6
NASA scientists often grab data and put it
aside, intending to process it later. But
sometimes the mission financing runs out and the
processing never gets finished. -- Blakeslee, S.
(1990, March 20). Lost on Earth Wealth of Data
Found in Space. New York Times. C1.
7
(No Transcript)
8
Data becomes inaccessible because
Technology changes over time
Different hardware
Data was recorded with a machine that ran tapes
both forward and backward and which wasnt in use
in 1990
Data was read with an Ampex tape reader that had
been out of production for 25 years by 1990
Different storage media
Data was stored on 7 tapes, not the 9 tapes
used in 1990
Different software
Instructions for encoding the data were written
in technical jargon that was incomprehensible in
1990
Source code needed to process images was lost and
a new program had to be written to extract the
images
9
Data becomes inaccessible because
Poor intellectual management and storage practices
Data was not catalogued
Data was stored in deplorable conditions that
caused storage media to degrade
10
Hardware changes over time
State of the art
in 1950?
11
Hardware is still changing
12
Different Storage Media
13
Different Storage Media
Amal Graafstra and his girlfriend Jennifer
Tomblin never have to worry about forgetting the
keys to her Vancouver home or locking themselves
out of Graafstra's Volkswagen GT. ... Graafstra
asked a surgeon he knew to implant a tiny tag in
his left hand. -- Shaw, G. (2006, January 7).
With a wave of their hands, techies are opening
doors. Edmonton Journal. A8. Retrieved July 13,
2006, from Canadian Newsstand database.
14
How much research data has been archived?
About this much
15
More than 97 of the research data produced by
SSHRC-funded projects between 1978 and 1981
cannot be located today.-- Humphrey, C. (2003).
Preserving research data A time for action. In
Preservation of electronic records new knowledge
and decision-making postprints of a conference,
symposium 2003. (pp. 83-89). Ottawa Canadian
Conservation Institute
16
The SSHRC Data Archiving Policy has not
achieved its objectives. In fact, over an
eleven-year study period only 10 data sets have
been deposited -- Social Sciences and
Humanities Research Council of Canada National
Archives of Canada. (2002). National Data Archive
Consultation Building Infrastructure for Access
to and Preservation of Research Data. P. 7
Retrieved August 2, 2006, from www.sshrc.ca/web/ab
out/publications/da_finalreport_e.pdf
17
In 1991, the Economic and Social Research
Council commissioned Paul Thompson to find out
what was happening to qualitative data from
projects that it had funded. The results of the
survey showed that 90 of social science
qualitative research material from projects
funded by the ESRC was at risk or already
lost. -- Corti, L. Backhouse, G. (2005).
Acquiring qualitative data for secondary
analysis. Forum Qualitative Social Research 6
(2).
18
Why dont researchers archive?
Academic environment doesnt encourage it
Lack of resources support for archiving
Lack of knowledge about archiving
Insecurity
Extra work and effort involved in preparing and
documenting data for archiving
19
Academic environment doesnt encourage archiving
In an environment where promotion, tenure and
publication are important motivating forces,
nurse researchers may be reluctant, as are many
social scientists, to share their data freely.
The challenge for nursing is to demonstrate that
such academic rewards as tenure and publication
are better promoted through sharing than amassing
and storing data. -- Estabrooks, C.A. Romyn,
D.M. (1995). Data sharing in nursing research
advantages and challenges. Canadian Journal of
Nursing Research. 27 (1), 80.
20
Lack of resources and support for archiving
Few libraries have integrated the numerical data
base into their collections, although it can
logically be considered another information
resource. As a result, the traditional
abstracting and indexing services and other tools
and services which serve a document-based
community do not exist for numerical data.
--Robin, Alice. (1981). Strategies for improving
utilization of computerized statistical data by
the social scientific community. Social Science
Information Studies 1(1), 97.
21
Lack of knowledge about archiving
Most produces of machine readable data files
have neither the resources nor the expertise to
preserve, maintain and distribute their data.
Few data producers can supply adequate user
support services related to teaching, research,
coordination of multiple research projects,
computation and general information services
related to machine-readable data files--
Robin, Alice. (1981). Strategies for improving
utilization of computerized statistical data by
the social scientific community. Social Science
Information Studies 1(1), 97.
22
Insecurity
Researchers often feel Insecurity about exposure
of ones research practice, International
Property Rights (IPR) or threat of
misinterpretation. --Corti, L. (2005).
Qualitative archiving and data sharing Extending
the reach and impact of qualitative data.
Qualitative Researcher. December (1), 7.
23
Extra work and effort involved in preparing and
documenting data for archiving
Time spent anonymizing data
Preparation of documentation
Process of making sure documentation is clean and
corresponds to the data
24
The Result
25
Knowledge Utilization Studies Program
Carole A. Estabrooks, RN, PhD, Canada Research
Chair in Knowledge Translation
26
Use existing data to answer new research questions
27
Make better use of research funds
28
Reduce respondent burden
29
Allow replication, validation and sharing of
research data
30
Avoid duplication of research efforts
31
Allow readers to link publications to data
32
Permit longitudinal and retrospective studies
33
NESSTAR
Networked Social Science Tools and Resources
Converts metadata from plain text to
DDI-compliant XML
Imports quantitative data into the same record
that holds the metadata
Makes both the data and metadata available over
any web browser
Preserves the quantitative data in a
non-proprietary format, so the data will still be
useable in years to come, irrespective of
software changes
Creates the possibility of linking data
repositories through a common gateway interface
to create a union catalog of repositories
34
Data Documentation Initiative
Like AACR2 or Dublin Core, but for data
DDI is a library of XML tags, the descriptive
fields to which they are applied and the rules
that set out a standard format for creating
metadata for data
There are specific, pre-defined tags for each
section that allow data archivists/librarians to
describe the data
Once fully described, the data becomes machine
readable
35
(No Transcript)
36
(No Transcript)
37
Can respondents give informed consent when they
dont know how their information will be used in
future studies?
Problems with data archiving
38
Problems with data archiving
Of particular concern is the potential for
cross-referencing when data sets are combined.
For example, the identification of particular
individuals may occur when some respondent
characteristics are cross-referenced with
specific geographical and occupational data. --
Estabrooks, C.A. Romyn, D.M. (1995). Data
sharing in nursing research advantages and
challenges. Canadian Journal of Nursing Research.
27 (1).
39
Some of the things you can expect to be in this
sort of job.
Cataloger create metadata, writing abstracts
Archivist create fonds, arrange and describe
records, accession data
Reference librarian help people find data,
explain search strategies
Computer geek create profiles, grant user
access, ACU uploads
Unsupervised direct own work, determine
problems, solutions, goals
Unemployed funding for these sorts of positions
is never certain!
40
Library or Archive?
Library
Archive
unique
not unique
discrete
not discrete
not malleable
malleable
restricted
unrestricted
fonds classification
subject classification
41
Suggestions for project management
42
Expect tech problems
43
Steep learning curve
This is where I am now
Me back in fall, 2005
44
Early buy-in
DIE!
You want me to do WHAT?
N
45
One small step