DEI Training Activities That Work in a Real Workplace

The DEI training activities that work best in a real workplace have people practice a specific job skill, such as scoring an interview, handling a complaint or stepping in when a colleague is mistreated, and then apply it to a real task within weeks. Activities that only raise awareness tend to fade, and the research on training is mixed enough that design matters.

This guide covers what the evidence says, eight practical activities with instructions for running them, the exercises worth avoiding, and how to tell whether an activity changed anything. Each activity is built to be open to every employee in the relevant role, which keeps it within anti-discrimination law.

What the research says about which approaches work

The evidence on diversity training is genuinely mixed, and an honest practitioner says so. The largest review, a meta-analysis of 260 samples by Bezrukova and colleagues, found that training improved knowledge in a way that lasted, while effects on attitudes faded over time. Effects were stronger when training built skills, ran over a longer period and sat alongside other initiatives.

Field experiments point the same way. In a study of 3,016 employees at a global organization, Chang and colleagues tested a one-hour online diversity training. It changed some attitudes but produced only limited behavior change, and the authors concluded that one-off sessions are unlikely to be stand-alone solutions.

How training is framed matters too. In Why Diversity Programs Fail, Frank Dobbin and Alexandra Kalev reported that five years after companies made training mandatory for managers, there was no improvement in several groups' share of management, while voluntary training was followed by better results.

Taken together, those findings suggest a few working rules, which the table sets out.

None of these rules guarantees a result. They shift the odds, which is why every activity below ends by connecting the skill to a real process the organization already runs.

Eight activities and how to run them

The eight activities fall into three kinds. Some practice a decision, some rehearse a hard moment, and some build a team habit. Each takes between 30 and 90 minutes and works with groups of six to twenty.

Activities that practice a decision

These three suit hiring managers, recruiters and anyone who writes performance reviews. They work because the skill is concrete and the output can be compared across the room.

• 1. Structured interview practice. Give each pair a short role description, three written questions and a scoring guide. One person interviews, one answers, and an observer scores. Compare scores across pairs and discuss where they differed and why.
• 2. Job posting rewrite. Each participant brings one real posting and strips out any requirement the job does not need. Partners check each other's edits against the actual duties.
• 3. Review calibration. Hand out three anonymized sample reviews for similar performance. Participants rate each one, then compare ratings and underline vague phrases such as "not a culture fit" that carry no evidence.

Activities that rehearse a hard moment

Rehearsal builds the confidence to act when a real situation arrives, which a slide cannot do. These activities suit supervisors and customer-facing staff.

• 4. Complaint-handling rehearsal. Work in threes, with a speaker, a manager and an observer. The speaker raises a scripted concern, the manager listens, records and explains next steps, and the observer notes what was missed. Rotate roles so everyone plays the manager once.
• 5. Bystander practice. Present short written scenarios and have small groups choose a response, such as speaking up directly, creating a distraction, asking someone else for help or checking on the person afterward. Groups then act out one choice.

Activities that build a team habit

The last three work best with an intact team, because the habit has to survive after the session ends. Each one produces something the team keeps using.

• 6. Meeting audit. For two weeks, a rotating volunteer notes who spoke, who was interrupted and whose ideas were credited. The team reviews the notes together and agrees one change, such as a round where everyone speaks before open discussion.
• 7. Perspective-taking writing. Each person writes a short paragraph imagining a workday from the point of view of a colleague in a different role or schedule, then compares it with what that colleague actually reports.
• 8. Accessibility check. Participants run a screen reader and a color contrast checker on one of their own documents, then fix what they find.

How to run any activity in four steps

The activities differ, but the structure that makes them stick is the same. Each step depends on the one before, so a facilitator who skips the debrief loses most of the value of the practice.

The fourth step is where the research findings come in. A real task within a month, such as running the next interview with the scoring guide, links the session to the process change that the evidence says makes training more effective.

Framing matters as well. Here at the DEI Excellence Institute, practitioners we have trained over the past two years report that roughly one in ten more managers complete the follow-up task when a session is offered as an opt-in skills workshop rather than a required course. Where training must be required, as harassment training is in some states, presenting it as skill practice helps.

Activities to avoid or redesign

Some popular exercises carry legal risk, tend to backfire, or both. The EEOC's guidance on DEI-related discrimination at work warns that separating employees by a protected characteristic for any training can be unlawful, and that training discriminatory in content or delivery can support a hostile work environment claim.

With that in mind, these are the exercises to drop or rebuild.

• Exercises that sort people by identity, such as lining participants up or splitting them into groups by race or sex. Rebuild them around roles or scenarios instead.
• Confessional exercises that ask people to disclose personal experiences or admit bias in front of colleagues. Many people find these intrusive, and they rarely teach a skill.
• Using a bias test as a diagnosis. Self-assessment tools can start a discussion, but treating a score as proof of how someone behaves at work overstates what the tools show.
• A single annual lecture as the whole program. The research above suggests this is the least likely format to change behavior.

How to tell whether an activity worked

Measurement starts before the session, not after it. A short knowledge check at the start and end shows whether people learned the content, which the Bezrukova review suggests is the most durable effect. A satisfaction score alone says little, because people can enjoy a session that changes nothing.

The stronger test is behavior a month or more later. For interview practice, check whether the scoring guide is being used. For complaint handling, check whether concerns are being recorded and escalated on time. The employee DEI survey template can track how staff experience these changes over a year.

Here is how that looks in an illustrative case. A hospital runs interview practice for 24 hiring managers. Two months later, HR checks the next 30 interview files and finds a completed scoring guide in 26 of them, against 9 of 30 before the session. That count is the result worth reporting, far more than a feedback score.

Planning the sequence in advance makes all of this easier. The DEI training plan template lays out which activities go to which audience and when, and the list of twenty DEI training topics suggests what each session can cover. For managers in particular, the guide to DEI training for managers explains why their layer matters most.

Where the certification goes further

Designing and evaluating activities like these is the focus of Unit 6, DEI Education and Training Development, in the Professional DEI Certification Course. It covers needs analysis, building workshops and e-learning, choosing outside vendors and evaluating training with pre and post assessments.

Unit 2, Building Inclusive Organizational Culture, covers facilitating the conversations most managers avoid, including setting the purpose and the ground rules before anyone speaks. That is the skill a facilitator needs to run the rehearsal activities above well.