This series of papers summarises the methodological approaches we have taken in carrying out innovative or novel analysis. Such analysis is conducted under our statutory authority to examine and report to Parliament on the economy, efficiency and effectiveness with which government departments and other bodies have used their resources.
The Audit insights papers are methodological summaries and do not make new observations about value for money.
Investigating the volume and readability of guidance on the government’s GOV.UK website
This paper sets out how we used a technique called “webscraping” to harvest and analyse textual content from the government’s GOV.UK website.
Work and pensions
Jobcentre journey time model
This paper sets out how we created this interactive data visualisation of our jobcentre journey time model, which provides an overview of planned changes to the jobcentre estate in England and allows exploration of the estimated impact that these changes may have on the time it takes for people to travel to their nearest jobcentre when using public transport.
Estimating the impact of benefit sanctions on the employment outcomes of sanctioned individuals
This paper sets out how we used instrumental variables to estimate the impact of benefit sanctions on the employment outcomes of sanctioned individuals in the Work Programme – a large welfare-to-work programme for the long-term unemployed.
Estimating the impact of benefit sanctions on the employment outcomes of sanctioned individuals pdf – 111KB)
See also, our report: Benefits sanctions (Nov 2016)
Investigating the potential impacts of cost of debt indexation in the water sector
This paper sets out how we constructed a counterfactual to establish how much water customers in England and Wales could have benefited from an alternative charging system between 2010/11 and 2014/15.
Exploring the association between A&E performance and bed occupancy using Hospital Episode Statistics
This paper sets out how we used Hospital Episode Statistics (HES) to explore measures of A&E departments’ performance and, in particular, their association with measures of bed pressure.
Investigating the characteristics of A&E attendances by mental health service users
This paper sets out how we used NHS administrative data to compare the characteristics of A&E attendances by mental health service users to those of the rest of the population.
Driving improvements in NHS stroke care: evaluating the impact
This paper sets out how we used existing data to estimate the impact of our value for money work on stroke care.
Investigating the impact of out-of-hours GP services on A&E attendance rates: multilevel regression analysis
This paper sets out how we used an analytical technique called multilevel regression modelling to investigate the factors affecting levels of attendance at accident and emergency (A&E) departments by patients registered at a GP practice.
Mapping between different commissioners of healthcare
This paper sets out how we created look-up files to map between different commissioning geographies. This enabled us, for example, to estimate overall health funding at an area-level despite the different geographical bases of the various commissioners. We provide links to these look-up files.
Modelling of maternity services in England
This paper sets out how we used a modelling technique called ‘discrete event simulation’ to investigate a local maternity service.