Search: onr:"swepub:oai:DiVA.org:su-180158" >
Improving data shar...
Improving data sharing in research with context-free encoded missing data
-
Hoevenaar-Blom, Marieke P. (author)
-
Guillemont, Juliette (author)
-
- Ngandu, Tiia (author)
- Karolinska Institutet
-
show more...
-
Beishuizen, Cathrien R. L. (author)
-
Coley, Nicola (author)
-
van Charante, Eric P. Moll (author)
-
Andrieu, Sandrine (author)
-
- Kivipelto, Miia (author)
- Karolinska Institutet,Stockholms universitet,Centrum för forskning om äldre och åldrande (ARC), (tills m KI),National Institute for Health and Welfare, Finland; University of Eastern Finland, Finland; Karolinska Institutet Center for Alzheimer Research, Sweden
-
Soininen, Hilkka (author)
-
Brayne, Carol (author)
-
Meiller, Yannick (author)
-
Richard, Edo (author)
-
show less...
-
(creator_code:org_t)
- 2017-09-12
- 2017
- English.
-
In: PLOS ONE. - : Public Library of Science (PLoS). - 1932-6203. ; 12:9
- Related links:
-
https://doi.org/10.1...
-
show more...
-
https://journals.plo...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
http://kipublication...
-
show less...
Abstract
Subject headings
Close
- Lack of attention to missing data in research may result in biased results, loss of power and reduced generalizability. Registering reasons for missing values at the time of data collection, or-in the case of sharing existing data-before making data available to other teams, can save time and efforts, improve scientific value and help to prevent erroneous assumptions and biased results. To ensure that encoding of missing data is sufficient to understand the reason why data are missing, it should ideally be context-free. Therefore, 11 context-free codes of missing data were carefully designed based on three completed randomized controlled clinical trials and tested in a new randomized controlled clinical trial by an international team consisting of clinical researchers and epidemiologists with extended experience in designing and conducting trials and an Information System expert. These codes can be divided into missing due to participant and/or participation characteristics (n = 6), missing by design (n = 4), and due to a procedural error (n = 1). Broad implementation of context-free missing data encoding may enhance the possibilities of data sharing and pooling, thus allowing more powerful analyses using existing data.
Subject headings
- MEDICIN OCH HÄLSOVETENSKAP -- Annan medicin och hälsovetenskap -- Övrig annan medicin och hälsovetenskap (hsv//swe)
- MEDICAL AND HEALTH SCIENCES -- Other Medical and Health Sciences -- Other Medical and Health Sciences not elsewhere specified (hsv//eng)
Publication and Content Type
- ref (subject category)
- art (subject category)
Find in a library
-
PLOS ONE
(Search for host publication in LIBRIS)
To the university's database
- By the author/editor
-
Hoevenaar-Blom, ...
-
Guillemont, Juli ...
-
Ngandu, Tiia
-
Beishuizen, Cath ...
-
Coley, Nicola
-
van Charante, Er ...
-
show more...
-
Andrieu, Sandrin ...
-
Kivipelto, Miia
-
Soininen, Hilkka
-
Brayne, Carol
-
Meiller, Yannick
-
Richard, Edo
-
show less...
- About the subject
-
- MEDICAL AND HEALTH SCIENCES
-
MEDICAL AND HEAL ...
-
and Other Medical an ...
-
and Other Medical an ...
- Articles in the publication
-
PLOS ONE
- By the university
-
Stockholm University
-
Karolinska Institutet