By ‘cross-cutting’ we mean that these areas of practice don’t fit within just one pillar of the Code but feature in each of the three pillars of Trustworthiness, Quality and Value.
There are three themes: collaboration, coherence and transparency. These are not mutually exclusive – Code practices can bring together aspects of each theme. For example, practice Q2.2 calls for coherent statistics and transparency in explaining the impact of inconsistencies.
1. Collaboration
Collaboration involves working jointly.
Under Trustworthiness, the Code requires statistical leaders to promote collaboration both within government and beyond, including across professional groups (T2.4). This goal is now being supported by the development of the Analysis Function within the Civil Service. We are looking for producers to seek out and encourage opportunities for collaboration.
Regarding Quality, working with others helps ensure that the data sources are suitable, the methods sound, and data effectively quality assured. The Code requires statistics producers to maintain constructive relationships with those involved in collecting, recording, supplying, linking and quality assurance (Q1.2). It also encourages producers to work with topic and method experts and other statistics producers (Q2.6).
Value is all about ensuring that statistics are useful. Working with users and other stakeholders can help make sure the statistics deliver what is needed. It can also mean that effective improvements can be achieved. The Code encourages collaboration with other partners, such as policy colleagues, when conducting public engagement – recognising your common goals can lead to new and mutually beneficial ways of gathering and sharing insight (V1.4).
We also encourage statistics producers to work with experts and other producers to provide a comprehensive and coherent narrative for the statistical topic (V3.5), to provide a clearer understanding of the meaning behind the figures. Bringing together the big picture can help users gain a fuller insight and make best use of related data.
Value is also concerned with driving improvements to the data and statistics. Working jointly with experts, users and other statistics producers can ensure relevant experience and expertise informs the statistics development (V4.4).
2.Coherence
Coherence reflects the degree of similarity between related statistics and the fuller insight achieved by drawing them together.
Producers need to understand and explain the consistency and comparability of their statistics with relevant standards and other related statistics, as well as providing a sufficient understanding of inconsistencies. Producers must demonstrate that they do not simply publish a set of numbers, but that they explain how they relate to other data on the topic, and how they combine with other statistics to better explain the part of the world they describe.
In demonstrating Trustworthiness, statistical leaders should promote adherence to appropriate standards, harmonisation with methods reflecting established good practice, and collaboration with other producers to achieve coherent statistics. (T2.1, T2.4, T4.4)
Under Quality, information is useful when it leads to understanding; when patterns can be found and examined. Effective analysis requires data that are coherent across different levels of aggregation (internally coherent), consistent over time, and comparable between geographical areas. Users need to understand the reasons for any lack of consistency and any implications for their use of the data and statistics.
Methods need to be sound and in line with scientific principles. To understand the importance of changes in statistics, the data need to be consistent over time – changes to the way the data are collected or defined need explaining. Harmonisation is essential to support comparisons and create opportunities for data linkage – for example, ensuring common definitions are used, asking the same questions and using the same codes in data recording.
Triangulation is helpful when analysts can look at the relationship over time with other similar statistics. Monitoring coherence and comparability helps alert producers to changes that may impact the interpretation of the statistics. (Q1.4, Q1.7, Q2.1, Q2.2, Q2.5, Q3.3)
The Value of the statistics is best achieved through ensuring their ongoing usefulness, including asking users about their satisfaction with coherence, providing appropriate comparisons to support interpretation and signposting to other related statistics. Bringing together a coherent narrative by drawing on related statistics delivers greater insight about the world. (V1.3, V3.3, V3.5, V4.5, V5.1)
3.Transparency
Transparency means being clear and open, for example, about the choices being made – and not holding back or being opaque about decisions.
Transparency is needed about processes, methodology and content. The theme occurs in each of the Code’s 14 principles, and particularly in T3: Orderly release, T4: Transparent processes and management, each of the three Quality principles, as well as in V2: Accessibility, V3: Clarity and insight and V4: Innovation and improvement.
Trustworthiness comes from the organisation that produces statistics and data being open and transparent: in their release of statistics; about their plans, priorities and progress; about their commitment to quality and approach to quality management; about identified areas for improvement; and how information collected for statistics will be used and protected.
Within Quality, producers should explain what judgements they have made about the data and methods, and their strengths and limitations. These explanations are as important as the numbers themselves.
Being open in providing the data and clear about the meaning of the statistics helps ensure they achieve their Value. Achieving relevant statistics needs a good understanding of users’ needs and an openness to hear their experiences. Confidence in the producers’ response comes from being transparent about the feedback and their subsequent actions, including their decisions and reasons for them.