Fleur Hughes spoke data management for the Welfare Conditionality research project at the first Open Data in Practice event at the University of York on 15 November 2018.

1. Data Management in the Welfare
Conditionality Research Project
Open Data in Practice
15 November 2018

2. Twin aims:
 To consider the ethics and efficacy of welfare
conditionality
 Fieldwork with three sets of respondents:
1. Semi-structured interviews with 52 policy stakeholders
2. 27 focus groups with frontline welfare practitioners
3. Three rounds of repeat qualitative longitudinal interviews with a
diverse sample of welfare recipients who are subject to welfare
conditionality
Funded by ESRC grant ES/K002163​/2
Welfare conditionality: sanctions, support and behaviour
change (2013-2018)
2

3.  Exploring welfare conditionality across a range of
policy domains and groups
Recipients of social security benefits (unemployed people, lone
parents, disabled people, Universal Credit), homeless people, social
tenants, individuals/families subject to anti-social behaviour
orders/family intervention projects, offenders and migrants
 11 Locations in England and Scotland
Bath, Bristol, Edinburgh, Glasgow, Inverness, London, Manchester,
Peterborough, Salford, Sheffield, Warrington
Welfare conditionality: sanctions, support and behaviour
change (2013-2018)
3

4. Challenge #1
Dataset size
4

5.  Standard qualitative project sample:
30
 Large qualitative project sample:
60
 Welfare Conditionality sample:
481
 Total interviews:
1082
Dataset size 5

6. Solution #1
Planning and structure
6

7.  File structure
 File naming protocol
 Code number protocol
 Key database of participants
 Data validation in spreadsheets
Planning and Structure 7

8. Challenge #2
Anonymisation
8

9.  Two stage process
 Stage One: names replaced with descriptor
Eg. [BROTHER NAME]
 Stage Two: all identifying information replaced with
descriptor
Eg. [RESPONDENT SCHOOL NAME]
 Stage Two required additional staffing
Grade 6 post, 6 months @ FT
Anonymisation 9

10. Solution #2
Budgeting
10

11.  Initial budget for archiving:
£500
 Actual amount spent on archiving:
£20,835
Budgeting 11

12. Challenge #3
Archiving
12

13. Archiving 13

14. Archiving 14

15. Solution #3
Archive preparation from the
start
15

16.  Checking the archive’s ingestion forms
 Discussing the data set and file formats
 Sending test data before formatting the final dataset
 Building in time to review errors after sending
Archive preparation from the start 16

17. www.welfareconditionality.ac.uk
Project Manager: fleur.hughes@york.ac.uk