Are Diverse Groups Bad At Predicting The Future?

The benefits of thought diversity are fairly well established, and indeed something I’ve touched upon many times before.  Groups with diverse perspectives tend to forge more innovative and successful ideas.

A new study, from the University of Michigan, suggests that whilst this is largely the case, diverse groups can be poor when it comes to predicting the future.

The study found that whilst crowds are often wiser than individuals, the diversity of that crowd didn’t seem to matter so much.  The team looked at a range of social diversity characteristics, including gender, race, religion and interests, and combined these with cognitive diversity (ie how we think).

They wanted to test the hypothesis that social diversity boosts the cognitive diversity of the group.  The results suggest however that you can achieve significant cognitive diversity even in demographically homogeneous groups.

Wisdom of crowds

Participants were asked to complete nine judgement tasks, with groups created based upon the estimates in these tasks.  The groups consisted of eight randomly selected participants who were designed to consist of both diverse and homogeneous groups.

Interestingly, the homogeneous groups produced very similar results as the more diverse groups on numerical tasks.  The authors suggest that the belief that identity diversity will automatically improve cognitive performance may be down to erroneous stereotyping, with the actual differences between various social groups much smaller than we might expect.

“In other words, not all women think alike, not all liberals think alike, and so forth,” they explain.

The final word?

As we perhaps might expect however, this isn’t the final word, as a recent study from North Carolina State University tested a similar hypothesis of whether diverse groups were more creative and innovative.

The study suggests that making an organization more diverse can also make it more innovative, whether that innovation is measured in terms of new products, more patents or the influence those patents have (measured by citations).

“We wanted to know whether companies with policies encouraging the promotion and retention of a diverse workforce – in terms of gender, race and sexual orientation – also perform better at developing innovative products and services,” the authors say. “The short answer is that they do.”

The researchers crawled the MSCI ESG STATS dataset, which provides the diversity policies of the 3,000 largest publicly traded companies in the US.  They then cross-referenced this with patent data from the US patent office, whilst also collecting data on product announcements from the Capital IQ Key Development database over a fifteen year period.

“Basically, once you get through all the statistical jargon, we found that a company that checks all of the diversity boxes would see about two new additional product announcements over 10 years,” the team explain. “Given that most firms produce an average of two new product announcement per year, that’s significant. On the other hand, it shows that improving diversity policies won’t lead to overnight improvements in innovation. It may take time to reap the benefits.”

So, the message would appear to be that it’s not just thought diversity that matters for innovation, and the more your organization supports diversity on all levels, the better they’re likely to be at innovating.