Facilitate Open Science Training for European Research 
 
RESEARCH DATA MANAGEMENT AND OPEN DATA 
 
6-7 October 2015 
University of Manchester, UK 
 
ANONYMISATION 
Veerle Van den Eynden, UK Data Service 
Options for sharing research data that may 
contain confidential information  
 
 
• Obtain informed consent, also for data sharing and 
preservation / curation 
 
• Protect identities e.g. anonymisation, not collecting 
personal data 
 
• Regulate access where needed (all or part of data) e.g. 
by group, use, time period 
 
 
 
Why anonymise research data? 
Ethical reasons  
• protect people’s identity (sensitive, illegal, confidential info) 
• disguise research location  
 
Legal reasons  
• not disclose personal data  
 
Commercial reasons 
 
Discuss with your research participants 
 
 
Identity disclosure 
 
A person’s identity can be disclosed through: 
•  direct identifiers 
e.g. name, address, postcode, telephone number, voice, 
picture 
 
often NOT essential research information (administrative) 
• indirect identifiers – possible disclosure in combination with 
other information  
e.g. occupation, geography, unique or exceptional values 
(outliers) or characteristics 
Anonymising quantitative data 
• Remove direct identifiers (or replace with pseudonyms) 
e.g. names, address, institution, photo 
• Reduce precision/detail through aggregation 
e.g. birth year vs. date of birth, occupational categories, 
area rather than village 
• Generalise meaning of detailed text 
e.g. occupational expertise 
• Restrict upper lower ranges to hide outliers 
e.g. income, age 
 
Anonymising qualitative data 
• Remove direct identifiers, or replace with pseudonyms – often 
not essential research info 
• Avoid blanking out; use pseudonyms or replacements 
• Identify replacements, e.g. with [brackets] 
• Plan or apply editing at time of transcription  
• Avoid over-anonymising – removing information in text can 
distort data, make them unusable, unreliable or misleading; so 
balance anonymisation with the need to preserve context 
• Consistency within research team and throughout project.  
• Keep anonymisation log of replacements or removals made – 
keep separate from anonymised data files 
 
 
 
Anonymisation log 
Example: Anonymisation log interview transcripts
Interview / Page Original Changed to
Int1
p1 Spain European 
p1 E-print Ltd Printing 
p2
20
th
 June
June
p2 Amy Moira
Int2
p1 Francis my friend
P31. Joan  Mary 
P97. Carol  {Mother} 
P34. Colchester  {Town in S.E.England} 
P65. Welshpool High School  @@##High School##@@ 
Audio-visual data 
 Digital manipulation of audio and image files can remove personal 
identifiers 
 e.g. voice alteration, image blurring (e.g. of faces)  
 
 Labour intensive, expensive, may damage research potential of data 
 
Better: 
• obtain consent to use and share data unaltered for research 
purposes 
• avoid mentioning disclosing information during audio recordings 
In practice: example anonymisation 
In practice: example anonymisation 
What if anonymising is impossible? 
• obtain consent for sharing non-anonymised data 
• regulating/restricting user access, e.g. at UK Data Archive: 
• archived data NOT in public domain 
• use of data for specific purposes only after user registration 
• data users sign legally binding End User Licence – e.g. not identify 
any potentially identifiable individuals 
• stricter access regulations for confidential data (case by case 
basis):  
• access to approved researchers only 
• requiring data access authorisation from data owner prior to 
data release 
• confidential data under embargo for given time period 
• secure access to data 
 
• researchers - consider access to data and safe storage 
Managing access to data 
• available for download/online access 
under open licence without any 
registration 
Open 
• available for download/online access 
to logged-in users who have 
registered and agreed to an End 
User Licence 
Safeguarded 
• available for remote or safe room 
access to authorised and 
authenticated users whose research 
proposal has been and who have 
received training 
Controlled 
Can such research data be open ? 
• ESRC research data policy: 
• Publicly-funded research data are a public good, produced in the 
public interest, which shall be made openly available and 
accessible with as few restrictions as possible in a timely and 
responsible manner that meets a high ethical standard and does 
not violate privacy or harm intellectual property. 
 
• Openly available research data, with as few restrictions as 
possible, means in the ESRC context that research data will be 
made available for re-use free of charge, as open data, 
safeguarded data or controlled data; the access category being 
selected to minimise the risk of disclosing personal information 
 
 
Open about data with restricted access 
Publish: 
• Which data exist 
• Where data are kept, e.g. which repository 
• Who can access them 
• For which purpose can they be used 
• Under which conditions 
In practice: data with access conditions 
Health and Social Consequences of the Foot and Mouth Disease 
Epidemic in North Cumbria, 2001-2003 (study 5407 in UK Data Archive 
collection) by M. Mort, Lancaster University, Institute for Health 
Research. 
• Interviews (audio + transcript) and written diaries with 54 people 
• 40 interview and diary transcripts are archived and available for re-
use by registered users 
• 3 interviews and 5 diaries are embargoed until 2015 
• audio files archived and only available by permission from 
researchers 
 
discover.ukdataservice.ac.uk/catalogue/?sn=5407  
doc.ukdataservice.ac.uk/doc/5407/mrdoc/pdf/q5407userguide.pdf  
 
In practice: access conditions ReShare 
In practice: access conditions ReShare 
In practice: access conditions ReShare 
In practice: access conditions ReShare 
Questions ? 
 
 
 
 
 
 
 
 
• Veerle Van den Eynden 
• veerle@essex.ac.uk