Access to European Statistical System microdata – an overview · (CIS, SES, MMD) Scientific use...
Transcript of Access to European Statistical System microdata – an overview · (CIS, SES, MMD) Scientific use...
Access toEuropean Statistical Systemmicrodata – an overview
Aleksandra BujnowskaEurostat
Objective of the presentation
To present the ESS microdata access system:• Data• Modes of access• Procedures• Users• Outcomes• Work in progress
2
Outline
1. Microdata sets available2. Modes of access3. Microdata access workflow4. Research entities5. Research proposals6. Recent achievements and plans for 2017
3
Microdata access at the EU level
• New rules since July 2013• Objective: wider access to EU microdata
4
1. Microdata sets available
5
ESS microdata for scientific purposes1. European Community Household Panel2. European Union Labour Force Survey3. Community Innovation Survey4. European Union Statistics on Income and Living Conditions5. Structure of Earnings Survey6. Adult Education Survey7. European Road Freight Transport Survey8. European Health Interview Survey9. Continuing Vocational Training Survey10. Community Statistics on Information Society11. Micro-Moments Dataset12. Household Budget Survey 2010 (since September 2016)
6
Use of microdata sets
7
EU-SILC, 602, 38%
EU-LFS; 455; 29%
ECHP; 126; 8%
CIS; 117; 7%
SES; 95; 6%
AES; 50; 3%
EHIS; 55; 4%
CVTS; 20; 1%CSIS; 17; 1% MMD; 10; 1% ERFT; 15; 1% HBS; 9; 1%
Use of microdata sets
8
EU-SILC; 602; 38%
EU-LFS; 455; 29%
ECHP; 126; 8%
CIS; 117; 7%
SES; 95; 6%
AES; 50; 3%
EHIS; 55; 4%
CVTS; 20; 1%CSIS; 17; 1% MMD; 10; 1% ERFT; 15; 1% HBS; 9; 1%
Use of microdata sets
9
EU-SILC; 602; 38%
EU-LFS; 455; 29%
ECHP; 126; 8%
CIS; 117; 7%
SES; 95; 6%
AES; 50; 3%
EHIS; 55; 4%
CVTS; 20; 1%CSIS; 17; 1% MMD; 10; 1% ERFT; 15; 1% HBS; 9; 1%
Use of microdata sets
10
EU-SILC; 602; 38%
EU-LFS; 455; 29%
ECHP; 126; 8%
CIS; 117; 7%
SES; 95; 6%
AES; 50; 3%
EHIS; 55; 4%
CVTS; 20; 1%CSIS; 17; 1% MMD; 10; 1% ERFT; 15; 1% HBS; 9; 1%
Use of microdata sets
11
EU-SILC; 602; 38%
EU-LFS; 455; 29%
ECHP; 126; 8%
CIS; 117; 7%
SES; 95; 6%
AES; 50; 3%
EHIS; 55; 4%
CVTS; 20; 1%CSIS; 17; 1% MMD; 10; 1% ERFT; 15; 1% HBS; 9; 1%
2. Modes of access
12
Data anonymisation
13
Secure usefiles
Scientific usefiles
Public usefiles
Statisticaluse files
De-identification Partial anonymisation Full anonymisation
Anonymised datasets and modes ofaccess
14
Secure usefiles
Scientific usefiles)
Public usefiles
• On-siteaccess
• Remoteaccess
• Remoteexecution
• DARA
• Datatransmittedon CDs orDVDs orover theinternet
• Dataavailable onthe websitewith orwithoutsubscription
Statisticaluse files
• Access bystatisticaloffice only
De-identification Partial anonymisation Full anonymisation
Anonymised datasets and modes ofaccess
15
Secure usefiles
Scientific usefiles)
Public usefiles
• On-siteaccess
• Remoteaccess
• Remoteexecution
• DARA
• Datatransmittedon CDs orDVDs orover theinternet
• Dataavailable onthe websitewith orwithoutsubscription
Statisticaluse files
• Access bystatisticaloffice only
De-identification Partial anonymisation Full anonymisation
Types of microdata
16
Secure use files(CIS, SES, MMD)
Scientific usefiles (all except
MMD)
Public usefiles
(EU-SILC, EU-LFS)
Criteria Approved researchproposal
Approved researchproposal
_
Access In Eurostat safecentre
At researchers'workplace
Public
Datapreparation
Only direct identifiersremoved
Partialanonymisation(variables groupedtogether, rounded,swapped orsuppressed)
Data fullyanonymised
Identification Possible Possible but difficult Impossible
3. Microdata access workflow
17
Recognition ofthe research
entity(institutional
eligibility)
18
Researchproposals
Step 1Institutional level(1 month)
Step 2Researchers(2 months)
Eurostat
National statisticalauthorities(4 weeks)
Is the entitydoing
research?(eligible to
ask foraccess to
microdata?)
Is therequest
for accesswell
justified?Researchproposals
Eurostat
National statisticalauthorities(4 weeks)
Recognition ofthe research
entity(institutional
eligibility)
19
Step 1Institutional level(1 month)
Step 2Researchers(2 months)
National statisticalauthorities(4 weeks)
Researchproposals
Eurostat
Is the entitydoing
research?(eligible to
ask foraccess to
microdata?)
Is therequest
for accesswell
justified?
Researchproposals
Eurostat
National statisticalauthorities(4 weeks)
Recognition ofthe research
entity(institutional
eligibility)
20
Step 1Institutional level(1 month)
Step 2Researchers(2 months)
National statisticalauthorities(4 weeks)
Researchproposals
Eurostat
Is the entitydoing
research?(eligible to
ask foraccess to
microdata?)
Is therequest
for accesswell
justified?
Researchproposals
Eurostat
National statisticalauthorities(4 weeks)
Researchproposals
Eurostat
National statisticalauthorities(4 weeks)
4. Research entities
21
Eligible institutions are:
• Doing research• Making their results public• Independent• Safe
• 620 entities recognized on 27/03/2017 (779applications received since 2013)
22
Recognised research entities, byentity type
23
Universities/Schools
60%
Researchorganisations
26%
(Central) banks2%
Internationalorganisations
2%
EuropeanDGs/Agencies
2%
Privatecompanies
4%
Governmentalorganisations
4%
5. Research proposals
24
Eligible research proposal
• Scientific purpose well described• Need for microdata justified• Research results made public• Data security measures in place
25
Assessment of research projectproposal (RPP)
• Eurostat:• Microdata access team: initial check• Technical units: is the microdata appropriate for
the planned research• MS: 4 weeks
26
Number of research proposalsreceived (2013-2016)
27
91
310352 356
640
50
100
150
200
250
300
350
400
450
2013 2014 2015 2016 2017
forecast
RPP number
Research subjects 2016: LFS
• Labour market studies• Impact of migration• Skills• Gender studies• Hours worked• Inequalities• Employment• Job satisfaction
28
Research subjects 2016: EU-SILC
• Economic studies• Impact of crisis on households situation• (income, ethnic, geographical) inequalities• Labour market studies• (in work) poverty• Mobility in Europe• Euromod / tax-benefits models• Wages, incomes• Well-being, housing• Youth
29
6. Recent achievements and plansfor 2017
30
Achievements 2016 (1): new data
• Public use files• for EU-SILC (2012-2013) and EU-LFS (2013)• for DE, FI, HU, NL, SI• https://ec.europa.eu/eurostat/cros/
Topics, A-Z, Access to microdata, Public use files for Eurostat microdata)
• Household Budget Survey 2010 available forscientific purposes
31
Achievements 2016 (2): processimprovements
• Self-study material for microdata users• http://ec.europa.eu/eurostat/web/microdata/overvi
ew/self-study-material-for-microdata-users• New contract templates for non-EU entities
32
• On CROS portal:• Newsletters• Database with
publicationswritten usingESS microdata
Achievements 2016 (3): IT
• Workflow tool and webforms replacing Wordmicrodata applications in (since January 2017)
• Pilot on SFTP (secure file transfer protocol)transmission of LFS 2015 data
33
Workflow tool
34https://webgate.ec.europa.eu/multisite/microdata/
Plans for 2017
• PUFs: all countries EU-SILC and EU-LFS• First meeting of the microdata access network
group – June 2017• Collaboration with Council of European Social
Science Data Archives - CESSDA (entry point fornational microdata access systems)
• Table builder allowing tailored tabulations(confidentiality on the fly)
35