An assessment of the trustworthiness, quality and public value of statistics from the Welsh Housing Conditions Survey, produced by the Welsh Government.
Judgement on National Statistics status
We judge that the statistics from the Welsh Housing Conditions Survey (WHCS) can be designated as new National Statistics once the Welsh Government demonstrates to us that it has enhanced these statistics in the ways described in chapters one to three of this report. This report includes four Requirements.
The WHCS is the first national housing conditions survey in Wales since the 2008 Living in Wales survey. The Welsh Government will publish the Welsh Housing Conditions Survey 2017-18: Headline report (Headline report) in November 2018 followed by more detailed topic reports and analysis from early 2019. Together, these statistical outputs are expected to provide a detailed picture of housing conditions in Wales.
Housing conditions is an important policy area, within the UK and internationally, with the quality of housing affecting health and the wider well-being of the population. The Welsh Government has a number of legislative obligations and strategic commitments that require evidence on housing conditions, including the Well-being of Future Generations (Wales) Act 2015 and The Fuel Poverty Strategy (2010). As such, the WHCS statistics will fill an evidence gap; they will be used to monitor the changing condition of the housing stock in Wales, to measure work being undertaken to the stock and to evaluate the impact of Welsh Government policies.
The Welsh Government conducted extensive user engagement during the development of the survey and production of the statistics; the statistical team actively sought input from policy officials and external stakeholders throughout. The Welsh Government has taken on board users’ views in the design of the survey and the dissemination of the statistics, including those gathered via an online user survey. The users we spoke to were positive about the way Welsh Government statisticians engaged with them.
We were not able to review the Headline report in full or the data tables as these were still in development at the time of the assessment. However, we reviewed an early draft of the Headline report. In general, the statistical commentary is presented clearly, and, where possible, the Headline report makes comparisons between the Welsh housing stock and housing conditions and those of other UK countries, and over time within Wales, which enhances the value of the statistics. To better support users’ interpretation of the statistics, key messages should be drawn out, either through an executive summary in the Headline report or modifying the report so it is shorter.
The Welsh Government collaborates effectively with UK and international housing statisticians. It has established links with housing statisticians in the Republic of Ireland; it is providing support and advice on building a business case for an Irish housing conditions survey. It meets regularly (every 2-3 months) with the housing condition leads in England, Scotland, Northern Ireland and the Republic of Ireland to keep abreast of housing condition survey developments and outputs, and to discuss methodological improvements.
The survey methodology is well-established and sound; it is similar to that of other UK housing conditions surveys and previous housing conditions surveys in Wales and therefore provides data that are comparable across the UK and consistent over time. The Welsh Government follows national and international good practice by adopting recommended and recognised systems and frameworks for modelling measures of housing condition, which ensures that the measures are harmonised.
The WHCS is a survey of just over 2,500 properties, with the sample drawn from the National Survey for Wales. The main limitation of the survey is its relatively small sample size; it only allows analysis at the national level. This limitation was highlighted to users during the survey topic development phase.
The Welsh Government is transparent about the methods used and the quality of the statistics. The Technical report contains detailed information about the sample design, data validation, data processing and data modelling. The Quality report is comprehensive, covering all aspects of the quality of the data and statistics. It clearly explains the main strengths and limitations of the WHCS and provides an overview of the data quality assurance arrangements used by the surveyors, Building Research Establishment (BRE; the survey contractors) and the Welsh Government. To increase users’ confidence in the methods and the robustness of the estimates, the Welsh Government should publish the pilot survey summary report and a summary of the steps it took to minimise surveyor variability and bias.
The Welsh Government established a close working relationship with BRE, and BRE maintained a constructive relationship with ONS, which provided the WHCS sample. The data collection and processing requirements were clearly defined from the outset and the Welsh Government and BRE told us this approach worked well.
The Welsh Government adopted a transparent approach to user engagement. Stakeholders receive regular updates about the progress of the WHCS, the online user survey about the statistical outputs was publicised widely, and the Welsh Government published a publication plan. The statistical team followed the Welsh Government’s Statistical Quality Management Strategy; adopting a transparent and consistent approach to quality gives users confidence in the quality of its statistical services and products.
Resource constraints negatively impacted the Welsh Government’s ability to publish timely summary reports and affected its plans for gathering external feedback on the Headline report. To provide assurance that future plans for housing conditions statistics and data will not be impacted by resource constraints, the Welsh Government needs to review and update its resourcing contingency plans.
The Welsh Government has a strong data governance process in place for the WHCS. The statistical team followed the Welsh Government’s Statement on Confidentiality and Data Access, which gives users confidence that the data are kept secure and that the team respects individuals’ rights to privacy. For the first time in Wales, data were recorded using digital pens, which ensures that personal information is kept safe and secure.
Confirmation as National Statistics (December 2018)
Ed Humpherson to Glyn Jones (October 2018)