Guidance

Quality management approach

Published 27 August 2026

Introduction

The Code of Practice for Official Statistics requires all producers of Official Statistics to publish quality guidelines. Quality is a core principle of the Code of Practice, which states that statistics producers should prioritise quality, be rigorous and be open about quality.

This statement describes the quality guidelines used within the Ministry of Housing, Communities and Local Government (MHCLG) to ensure our statistics adhere to the quality principles set out in the Code.

In MHCLG, quality is managed throughout the full statistical production process, and we encourage and promote a culture of continuous improvement, self-assessment and quality reviews.

Quality has three essential areas of practice:

• prioritise quality
• be rigorous
• be open about quality

The Dimensions of statistical quality, set out below, show how we define statistical quality within MHCLG and the practices that our statistics producers adhere to.

Quality processes during statistics production

Within the overall dimensions of quality in the table below, a key aspect of ensuring accuracy and reliability is preventing errors during the processing and production of statistics.

Given the wide range of different statistics produced by MHCLG, we don’t mandate a specific process for all statistics producers to follow, but, as is proportionate, all our production processes draw on some or all of the approaches below to help ensure the highest quality management standards:

• Peer review: Statistics production processes are reviewed by other team members to detect errors and ensure they are able to run the process if needed.

• Review: Every stage of the statistics production process is followed by a review of the outputs, either manually or using error detecting code.

• Automation: Implementing the principles of Reproducible Analytical Pipelines (RAP) to automate complex and repeatable processes, to increase efficiency and reduce errors by minimizing manual work. Many of MHCLG’s statistical production processes are automated, or partially automated, with more automation work planned in future.

• Dual running: Production processes are run more than once, to look for discrepancies (e.g. using different software or dummy data).

• Documentation and resilience: Comments are written and documentation is created so that others can understand the process, pick up any errors and avoid over-reliance on the knowledge of single individuals during the production process.

Continuous improvement

An in-house statistics assurance exercise in 2026 suggested there was scope to continue strengthening these quality processes and developing these will be a priority for our statistical leaders over the next year.

We are developing a ‘Statistics Play Book’, intended to set clearer direction for good statistical practice across the department, including on the use of AI and RAP, Quality Assurance, and wider expectations of how we design and deliver high-quality statistical work.

More broadly quality considerations are embedded across MHCLG’s wider analytical community, including for statistics producers. MHCLG has a network of ‘Quality Champions’ who are linked into wider cross-government quality work. MHCLG’s Quality Champions work across analyst teams to help raise awareness of good data quality practices, the Aqua Book and signpost to helpful resources. The Champions’ Network has delivered content on quality for teams to use for staff induction.

Some MHCLG statistics are derived from, or used within, analytical models. We have templates for building models in Excel and guidance for creating models in code and follow the Duck Book for coding best practice.

We also seek continuous improvement in statistical processes by releasing statistical work in progress, as Official Statistics in Development, such as the recently developed MHCLG enabled spend statistics. These will be published in order to involve users and stakeholders in their development and as a means to build in quality at an early stage.

Further detail on the quality of individual statistical products is included in the relevant statistical release under the data quality section of the technical notes.

The dimensions of statistical quality

MHCLG use 5 dimensions of statistical quality, consistent with those used by the Office for National Statistics and the European Quality Standards. These are:

• relevance
• accuracy and reliability
• timeliness and punctuality
• accessibility and clarity
• coherence and comparability

Relevance

The degree to which the statistical output meets user needs for both coverage and content.

Key aspects

Any assessment of relevance needs to consider:

• who are the users of the statistics
• what are their needs
• how well does the output meet these needs

Users can expect

To be appropriately consulted about their needs and MHCLG will seek to review data collections and statistical outputs on an ongoing basis to ensure that they continue to meet user needs.

See details on our User engagement strategy.

Accuracy and reliability

For survey data: the closeness between an estimated result and the (unknown) true value.

For all data sources: how well the information is recorded and transmitted.

Key aspects

Accuracy can be split into sampling error and nonsampling error, where nonsampling error includes:

• coverage error
• non-response error
• measurement error
• processing error
• model assumption error
• completeness
• timeliness of recording and transmission
• accuracy of recording of data items
• correct use of coding
• correct interpretation

Users can expect

Survey data in MHCLG will be presented with full information on:

• sampling variability
• confidence intervals
• response rates
• other relevant criteria to allow users to make informed judgements on quality

All statistical publications will:

• include details of how the underlying data are collected to allow users to understand the strengths and limitations
• contain a description of data quality issues; and any impact this may have on analysing changes over time
• comparisons between different groups will be transparent to both lay and expert audiences.
• be compliant with the published MHCLG revisions policy or the specific policy for that output

Timeliness and punctuality

Timeliness refers to the lapse of time between publication and the period to which the data refers.

Punctuality refers to the time lag between the actual and planned dates of publication.

Key aspects

An assessment of timeliness and punctuality should consider the following:

• production time
• frequency of release
• punctuality of release

Users can expect

Publications will comply with the Code of Practice on pre-announcing dates or will state clearly at the time of pre-announcement any reasons why this has not been followed.

We will comply with the Protocol 2 in the Code of Practice.

We will publish statistical releases as soon as possible after the relevant time period.

Accessibility and clarity

Accessibility is the ease with which users are able to access the data. It also relates to the format in which the data are available and the availability of supporting information.

Clarity refers to the quality and sufficiency of the metadata, illustrations and accompanying advice.

Key aspects

Specific areas where accessibility and clarity may be addressed include:

• needs of expert and nonexpert users
• consistency of standard in relation to revisions, rounding, data suppression and spreadsheet type
• assistance to locate information
• clarity
• dissemination

Users can expect

Statistical publications will be published in line with the MHCLG website accessibility policy.

All publications will use plain English wherever possible.

Coherence and comparability

Coherence is the degree to which data derived from different sources or methods, but which refer to the same phenomenon are similar.

Comparability is the degree to which data can be compared over time and domain.

Key aspects

Coherence should be addressed in terms of:

• data produced at different frequencies
• other statistics in the same domain
• sources and outputs
• coverage of different databases
• data published at different geographic levels
• definitions and coding used for different data sources

Comparability should be addressed in terms of comparability over:

• time
• spatial domains e.g. subnational, national, international
• domain or sub-population e.g. crime/offence type, ethnicity

Users can expect

As standard practice, we will release related statistical publications on the same day in order to aid user understanding unless:

• this would mean significant delay to one set of figures in order to present the coherent set of releases
• user engagement suggests that separate releases on separate days would be preferable

Where related measures are published across several publications, we will make it clear to users where the related information can be found.

We will use harmonised concepts and definitions in statistical publications wherever they are available. Any statistical publication which does not use harmonised definitions will explain why this has not been used and any plans to move it onto a harmonised basis.