JEDI1.2: Selecting a meaningful "additional social identity" for Large companies

Modified on Wed, 1 Apr at 10:51 AM

How to collect and disaggregate your data
Using stakeholder feedback to choose the additional social identity
Ensuring feedback is voluntary and anonymous
Case Study

Under the Justice, Equity, Diversity and Inclusion (JEDI) Impact Topic, you are expected to collect data and feedback (JEDI1) and use that information to help you choose and implement targeted JEDI actions (JEDI 2).

JEDI 1 focuses on data collection: understanding how worker experiences and conditions vary across groups through surveys, Human Resources metrics and other people‑related measurements.

If you are a company with 250 workers or more, from year 3, the JEDI 1.2 requires you to go further by selecting at least one additional social identity beyond gender identity or sex at birth. This could include race, ethnicity, disability status, age, or any other identity relevant to your specific context. You are required to:

Collecting data on that additional social identity for at least five worker-related measurements (such as hiring, promotion, training, wages, or perceptions of workplace culture).
Using stakeholder feedback to choose the additional social identity.
Ensure workers can provide their social identity voluntarily, and can choose to remain anonymous.
Disaggregate data to report internally on your chosen worker-related measurements.

While the standard sets a minimum of one additional identity, large firms may decide to track more than one. However, for each identity you choose, you need to track it across at least five worker‑related measurements. This quickly becomes complex, so it is reasonable to adopt a multi‑year approach, adding identities over time as systems and trust mature.

How to collect and disaggregate your data

JEDI 1 emphasises that companies can use multiple approaches to collect data, such as surveys or feedback discussions, provided they are designed to protect anonymity and allow voluntary participation.

For many large firms, a standalone anonymous survey is the most practical way to meet JEDI 1.2, especially when adding an additional social identity. Whatever method you choose, it should reflect core principles from JEDI 1.2:

Voluntary participation and clear communication about purpose
Anonymity and protection of worker identities in both collecting and reporting
Disaggregated analysis that remains robust and does not risk “outing” individuals

The anonymous survey method

The process

Where feasible, use a third‑party or internal survey tool that allows truly anonymous responses in order to protect workers’ identities, such as avoiding collecting employee personal identifiers and not linking surveys to employee accounts.This aligns with JEDI 1’s expectation that companies protect anonymity and clearly communicate the process and the option not to disclose.

Explain the purpose and safeguards

In the survey invitation and introduction, clearly state:

Why you are collecting data on the additional social identity (to understand inequities and inform JEDI actions).
That participation is entirely voluntary and that “Prefer not to say” is always an option.
That responses will only be used in aggregate for JEDI analysis and internal reporting, not for individual performance or disciplinary decisions.

This helps shift the perception from “surveillance” to shared problem‑solving and is consistent with the JEDI intent to address systemic inequities.

Disaggregate data while protecting anonymity

When analysing and reporting your survey results:

Apply a minimum threshold (often called a “rule of 5”) before reporting a disaggregated group. For example, if fewer than five workers are in a given cell (e.g. a region‑team‑identity combination), roll those results up to a larger unit (department, country, or global level).
Be cautious when applying multiple filters at once (e.g. gender + additional identity + small team + specific country). Check that your cell sizes remain large enough to protect anonymity.
Document your aggregation rules and thresholds so that your approach is transparent during the audit.

Maintaining trust over time

Trust is critical to achieving high participation, especially for sensitive social identities (e.g. disability, race/ethnicity, sexual orientation, religion). To build and maintain trust:

Reiterate in internal communications that data will be used only for JEDI analysis and to inform targeted actions under JEDI 2.
Share anonymised insights and resulting actions with workers so they can see how their input leads to change.

Under JEDI, companies are expected to use feedback from workers and other stakeholders, alongside data, to decide which JEDI actions to implement. Indeed, seeking worker feedback to select the additional social identity has a strategic advantage. This is because:

Leadership has a high‑level view of the business; workers have insight into everyday barriers that human resources metrics alone may not reveal (e.g. who gets invited to informal networks, which teams feel less psychologically safe or who feels safe raising concerns).
Workers may engage more readily with data collection exercises involving sensitive identities (such as disability, caste, migrant status, religion or sexual orientation) when they understand and shape the purpose.

In global firms, priorities may differ significantly by region because contexts vary. For example:

The company chooses ethnicity as a social identity because its workforce does not have the same ethnic diversity as its country, according to the most recent census, or because the country it is based in has a history of colonialism or slavery. This might be a priority in some countries, for example in the US, UK and Brazil, but not as much in Japan.
The company chooses LGBTQIA+ as a social identity because formal complaints from the past year show an increasing amount of anti-trans and homophobic behaviors. This might be done in Canada and the Netherlands, but not in countries where collecting this data is not possible due to the legal or cultural context.

Ways to gather worker feedback

Host a discussion or focus groups

Discussions are an excellent way for you to ask workers what social identities beyond gender they feel have the most significant disparities in your industry and workplace.
Where topics are sensitive, consider using external facilitators so workers can speak freely without fear of retaliation.

Consult with employee groups

If you have employee groups, for example established ERGs (e.g. for workers who identify as LGBTQIA+ or woman, for a specific race or ethnicity, neurodiversity, disability or different abilities, or for parents), you can hold a discussion or consult with them. These groups provide a safe space for workers to voice which identities face the most significant barriers.

Ensuring feedback is voluntary and anonymous

When you seek feedback, you must balance the need for deep insight and healthy participation while ensuring participation is voluntary, anonymous, and where participants feel psychologically safe to share their views.

Psychological safety and anonymity

Using independent or external facilitators, particularly where there is low trust in HR or leadership.
Applying “Chatham House Rules” where insights from the discussion can be used, but neither the identity nor the affiliation of the speakers, nor that of any other participant is revealed so that specific comments cannot be attributed to individuals.
Gathering collective feedback from ERGs (e.g. joint written statements) rather than attributing quotes to named individuals.
Using anonymous digital tools (Miro, Mural, anonymous forms) so people can submit input without speaking publicly.

Ensuring voluntary participation

To keep participation voluntary and avoid tokenism

Use open calls rather than hand‑picking individuals to represent identities.
Offer level‑specific sessions (e.g. separate sessions for entry-level workers, managers and executives) so workers can speak freely without line managers present.

Maximizing representation and inclusion

To avoid partial feedback that reflects only certain geographies,teams, or experiences

Clearly define the scope of the “additional social identity” with concrete examples (e.g. race, disability, LGBTQIA+, socioeconomic background, caste, migrant status), while allowing workers to propose alternatives relevant to their context. Sometimes people will also hold multiple identities and experiences.
Proactively seek input from all regions, divisions and job levels.
Schedule sessions at varied times to include shift workers, part‑time staff, and colleagues across time zones.

Case Study

Case study: Global logistics company with 1,200 workers

Step 1: The stakeholder feedback process (JEDI 1.2.2)

With 1,200 workers across multiple regions, the company knew a top-down approach would not be right. They decided to hold focus groups using an external facilitator to allow workers to discuss perceived barriers without social pressure. This included three focus groups in the Netherlands hub, made up of 8 workers each from across teams, who highlighted that migrant status and language barriers were significant barriers to internal promotion. In Brazil, the focus group highlighted colourism, noting that while more than half of the population identifies as Asian Black, Brown/Mixed, Indigenous, and White, they hold less than 10% of senior management and executive positions in the company. It also included focus groups with their Employee Resource Groups (ERGs). While the US-based Black Workers ERG highlighted racial equity, the European chapter of the ERG pointed toward socioeconomic background as a major barrier to progression to management roles. To help find a global consensus, they then ran a company-wide anonymous survey. The results showed that race/ethnicity was the most urgent priority for 65% of the global workforce, while socioeconomic background was a close second. They decided to track both to capture a true intersectional picture. The external facilitator compiled all insights and created a management report to share the analysis with the leadership team.

Step 2: Method of collection (JEDI 1.2.3)

To build trust and ensure participation was voluntary and anonymous, they decided not to use their internal HR system and used a standalone third-party survey tool with a generic link.

The survey included specific questions to understand experiences and sentiment across five worker-related measurements that the company used for JEDI 1.1: recruitment, promotion, training opportunities, wages, and perception of culture. At the end of the survey, participants had the option to share their race, ethnicity, socioeconomic background (with a question focused on parent’s educational attainment level), location, and department, and ensured there was a ‘Prefer not to say’ option for every question across the whole survey.

They did not include any survey fields that track personally identifiable markers such as employee names, IDs, and email addresses. They also issued a Data Privacy Pledge explaining that the data was to decide JEDI strategy and actions to improve workplace equity practices.

Step 3: Disaggregating and protecting identities (JEDI 1.2.4)

Once the data was collected, the HR team disaggregated the results by race, ethnicity, and socioeconomic background across the five five worker measurements. Where there were teams with less than five people responding, they rolled these team numbers up into the department or location level results.

Step 4: The outcome and JEDI action (JEDI 2)

The disaggregated reports revealed that while Black and Hispanic workers were being hired at higher rates, they were participating in Leadership Fast Track Training at a rate 45% lower when compared to their White peers. Using this evidence, the company:

Audited their training nomination process to remove subjective manager vouching and recommendation and replaced it with objective performance-based entry; and
Launched a targeted leadership mentorship program for all employees.