More companies are publishing diversity, equity and inclusion reports in a bid to show movement on pledges to increase the representation of historically under-represented minorities (URMs) and create more inclusive workplaces, but the lack of progress with DEI is largely hidden.
When data is presented in a way that creates a false sense of progress, white employees and leaders feel a sense of accomplishment while URMs remain skeptical, or worse, disenfranchised for calling out a lack of progress they experience.
The impact is palpable.
Three years after pledging $50B to Black Lives Matter, and over $8B annual spending on diversity, equity, and inclusion, Washington Post reports there has been little actual progress in improving the workplace experience for historically underrepresented groups (URMs), along with slow movement to the C-Suite and corner offices.
I spoke to a people analytics leader, Rob Hadley, to uncover common ways in which organizations misrepresent progress and what they can do differently.
Reading dozens of DEI reports, what stands out is that the majority of these progress reports are essentially marketing – a recent 46 page DEI report from data giant Qualtrics had a handful of pages with high-level data, wrapped in pages of quotes, vision statements and often repeated commitments.
We think companies can do better, and workers should advocate for greater transparency.
Demographic Data and URM Categories
The US government requires companies with over 100 employees to submit an EEO-1 report each year, detailing employees’ job categories, ethnicity, race, and gender to EEOC and the U.S. Department of Labor every year.
But how many make this information public?
The overall percentage of all companies sharing demographic information is not available, however, according to JUST Capital’s 2023 analysis of trends in demographic data disclosure revealed that one-third of Russell 1000 companies currently disclose their EEO-1. In the social impact sector a recent study found just 3.7% of the over 12,000 nonprofits divulged diversity data.
Hadley shares that even when companies share demographic data, they are unwilling to disclose breakdowns by race and ethnicity, gender, and intersectionality by job title and department or function. In the majority of DEI reports, we noticed all historically underrepresented groups being rolled-up under the URM category (perhaps aggregated as “BIPOC” or “Diverse”), without disclosing what URM means at the company with respect to industry and region in which they operate.
For instance, Reboot Representation’s data shows that while men of South Asian descent are underrepresented in the overall population, they are overrepresented in technology. Women may hold a significant proportion of senior leader roles in HR but be vastly underrepresented in product or finance roles.
When it comes to demographic data Rob suggests the following best practices –
- For race/ethnicity, stop aggregating all non-White and URM; the minimum standard is a representation breakdown by EEOC race/ethnicity categories, better yet, drill one level deeper into the sub-categories of race/ethnicity.
- Commit to developing a deeper understanding of identity and intersectionality in your organization through self-ID surveys and other tools to help understand disability, sexual orientation, caregiving, and other characteristics that will impact how they show up at work.
- Show representation percentages at all levels and across functions using splits meaningful to the organization and industry (hourly vs. full-time in retail, technical vs. non-technical in technology, etc.); ideally relative to workforce availability.
- Share trends over time, a minimum of 3 years.
Inclusion and Belonging Scores
“At Hooli, we are committed to DEI with 97% of our employees reporting a sense of belonging.”
We see boasts of DEI success daily. All you need to do is scan your LinkedIn feed to see mentions of a company’s “inclusion” scores.
According to Rob, these numbers often tell us very little. He says that while survey items developed by companies such as Lattice, Qualtrics, Tiny Pulse, and Culture Amp are well researched and evidence based, most HR and DEI practitioners don’t understand the survey science and may pick and choose questions to create their own inflated inclusion score, negating their reliability and validity.
He said, “For example, it’s entirely possible for an organization to show alignment between groups for one element of inclusion, such as ‘managers ability to integrate differences,’ while showing disparities elsewhere, such as ‘senior leaders’ commitment to diversity.’”
In order for surveys and scores to offer actionable insights, practitioners need to understand what goes into an inclusion score, adding “as an employee or candidate, if I see the organization understands the issues it needs to address, I have confidence they can address them.”
If a company is going to share results from an employee engagement or inclusion survey, best practices include:
- Share the model, dimensions, or survey items the inclusion score is based on and highlight areas of strength and opportunity within “inclusion”.
- Share participation rate and results breakdown by demographics.
- Share sentiment data and operational data together; e.g. “82% of Black/African American employees agree/strongly agree that promotion processes are fair and we have seen a 22% increase in the promotion rate for these employees over the past 3 years”.
Operational Data
Rob spoke about the over reliance on employee sentiment data as proxy for real operation data such as increases in the diversity in hiring, retention of URMs, and equity in people processes such as promotions and pay – metrics that signal actual change.
For instance, he says “sentiment data is important and it will tell you how employees feel about change and progress, but it can be either a leading or lagging indicator. It must be combined with operational data to measure impact.”
One example of this is pay equity. Companies will survey employees about compensation perceptions with items such as “I am fairly compensated for my work.” Relatively few employees have the data necessary to be able to make that assessment.
Yet when companies report pay equity information in their DEI reporting, we often see pride in statements saying “we are 99% pay equity compliant!” by either race and ethnicity, gender or both. To this point, Rob simply says, “pay disparities should be seen as HR malpractice and meeting traditional pay equity standards should be considered the lowest bar in promoting truly fair pay practices”.
In order for pay data to be meaningful in a report, consider the following:
- Look beyond base pay controlled for differences in tenures, job type, and region; consider total compensation – including benefits and we must account for variable pay – RSUs, bonus, equity and other variables.
- Understand how representation at seniority levels contributes to inequity and look at disparities in along with promotion velocity – there are huge gaps in most organizations and projections can be made to understand if the organization is on the right track.
- Analyze organizational equity through total compensation, by level, and by demographic splits.
Again, it’s not enough to ask how employees feel, we have to understand what is actually happening.
Learning and Training
According to Rob, a focus on training efforts is a clear indication that the investment in equity and inclusion is weak, in particular when organizations report percentages of employees who’ve taken training. Often, there are no details on what the training included, the length of time, cadence and outcomes.
Another expert, Aubrey Blanche, Head of DEI at Culture Amp, states, “Culture Amp’s research shows that the *majority* of DEI training that’s currently being provided is Unconscious Bias Training, despite very limited research to its effectiveness. What does that mean? Companies are doing the equivalent of lighting their budgets on fire, rather than investing in what might actually move the needle. Did you know it was largely developed by lawyers to limit liability, not DEI experts optimizing for sustainable behavior change?”
If including information about training, Rob suggests sharing why that training approach is needed and how the organization will measure impact. For example, an organization can state “Based on employee feedback that management was not committed to DEI, we set out to integrate DEI into our new management training program over the next 2 years. Our goal is that by 202X 90% of employees will agree their managers are committed to DEI”
Progress with DEI is not linear and transparency is critical to aligning on areas of improvement, building shared understanding of what it takes to make progress, and prioritizing accountability.
Read the full article here
