2- Danat Valizade Research Data Management

download report

Transcript 2- Danat Valizade Research Data Management

Research Data Management
Dr Danat Valizade
Lecturer in Quantitative Methods
University of Leeds
‘Without data you are just another person with an opinion’
William Edwards Deming
•
•
The quality of research depends increasingly on the quality of
data
This poses a number of challenges: selectivity, robustness
(statistics ought to be treated with caution); computational power

Started with a small primary survey of trade union representatives (400
observations)
•
Data storage was not problematic (university server, memory stick, laptop etc.)
•
Data were fully anonymous and easy to handle and conceptualise

Secondary data: the Workplace Employment Relations Study (WERS)
•
Available at UK Data Service
•
A much larger dataset with a more complex structure (over 20000 observations
nested on nearly 2000 firms; around 200 variables)
•
RIP – SPSS, memory sticks, external HDDs etc.
•
Storage: university server, statistical package synchronised with the cloud
account
•
An entirely different approach to data mining
 Currently working with large longitudinal datasets
•
Myriad of variables
•
Often sensitive data - for example, data on the diversity in the
legal profession (over 190 000 solicitors observed over the
past 20 years)
•
Secure access through VPN network
•
Data mining and number crunching become immensely
complicated
Good luck with your research!