Dr Kavandeep Aujla, Department for Work and Pensions (DWP) will discuss three aspects of the Employment Data Lab which is a DWP initiative. The DWP Employment Data Lab (or Data Lab for short) was designed to evaluate local employment programmes.
The event will cover why the Data Lab was needed, some of the challenges, the statistical techniques that underpin the Data Lab evaluations, and some of the key labour market policy challenges.
Dr Kavandeep Aujla from Department for Work and Pensions (DWP) will discuss three aspects of the Employment Data Lab which is a DWP initiative.
The DWP Employment Data Lab (or Data Lab for short) was designed to evaluate local employment programmes.
Kav will start with some background on why the Data Lab was needed, its long history to inception and how this free service supports charities, local authorities and other organisations. At this point Kav will discuss some of the challenges of using personal, sensitive participant data and how this data is linked to large Government administrative datasets.
The second part of the presentation will focus on the statistical techniques that underpin the Data Lab evaluations. In particular, how propensity score matching is used to identify a reliable comparison pool and hence attempt to mimic a randomised control trial. Then going onto show how we can use this approach to determine the true impact of a programme.
The final part of the presentation will focus on some of the key labour market policy challenges and how the Data Lab impact evaluations can help support the Government’s ambitious target to increase the employment rate from 75% to 80%.
Dr Kavandeep Aujla
The Employment Data Lab Team, Central Analysis & Science Strategy Unit, Department for Work and Pensions, London
Biography of Dr Kavandeep Aujla
After completing his doctorate in Applied Mathematics Kav joined the Government Operational Research Service at the Department for Work and Pensions (DWP). Over the last twenty years Kav has worked in many areas of DWP policy, including: developing statistical methodology to measure benefit fraud and error, using bootstrapping to calculate confidence intervals for National Statistics, applying logistic regression techniques to identify claimants with more chance of error, working with Office for National Statistics to answer key questions about the UK labour market. After a short sabbatical to focus on academic interests, Kav began his latest adventure leading the DWP Employment Data Lab.