What data should organisations be collecting?

Section 4

What data should organisations be collecting?

Dimensions of diversity

As part of the scoping process, it is important to be clear on which diversity data points (referred to here as dimensions of diversity) are going to be collected. Most organisations will already have data on sex (but not necessarily gender identity) and age, as this is usually collected from employees on joining. However, other dimensions of data will likely need to be identified. Identifying which dimensions to collect will be an individual organisation’s decision, but will often include consideration of:

Data required to meet D&I goals and overall strategy;
External data requirements (e.g. under local reporting requirements / requests from clients);
Legal permissions and restrictions;
Local cultural considerations; and
Market practice.

Globally, this can become more complex as local law may prevent the collection of data on some dimensions, while local market practice and cultural considerations will also vary.

Given this, most organisations identify any priority dimensions globally, but then allow for local variation on a country by country basis. This allows a balance between the need for globally consistent data and compliance with local law and awareness of local cultural norms.

Market Practice

Typically collected in the UK:

Sex
Age
Ethnicity
Sexual orientation
Gender identity
Disability

Increasingly collected in the UK:

Socioeconomic status / background (‘social mobility’)
Faith
Caring responsibilities
Further detail in relation to disability (e.g. neurodivergence)

Diversity data categories

Once the relevant dimensions have been identified, it is important to then think carefully about the way questions will be asked and the options provided for answering each question (referred to here as diversity data categories). Organisations should seek a reputable local source to determine the categories and question wording where this exists. Where possible, a national government census can provide a good initial source, which may then be supplemented by best practice guidance from specialist groups where organisations wish to collect more granular data, or data not within the census. Using any census or government source will also make benchmarking against any central demographic data easier. In the UK, market practice is to use the UK Census as an initial source for diversity data questions and categories, particularly as these questions have recently been reviewed for the 2021 census.

It is important for global organisations to note the complexities involved in categorising diversity data globally as often there is no ‘one size fits all’ approach. You must ensure that

your categories are tailored to each territory to reflect the diversity of the country and not cause offence. For example, the individual categorisations of ethnicity data would differ depending on the jurisdiction. It is important to tailor your approach to data collection and how you communicate to suit each country.

More generally, to recognise that diversity categories may not capture everyone’s identity, it is also considered best practice to include an option to allow individuals to self-describe. This should be in addition to a “prefer not to say” option, which is also advisable and may be required depending on local laws.

Example diversity dimension - Ethnicity categories in the UK (source: PwC)

Type your content here…

What categories of diversity data are you planning to collect and analyse?

Age
Sex
Ethnicity
Sexual orientation
Gender identity
Disability
Socioeconomic background
Religion/ Faith
Caring responsibilities