UNSW UNSW Business School UNSW Business School

The quality of what we know: Denise Rousseau on evidence-based management

February 20, 2018

Denise Rousseau is a professor of organisational behaviour and public policy at Carnegie Mellon University and a leader in the movement to make management practice more evidence-based. Rousseau spoke to Peter Murmann for BusinessThink.

An edited transcript of the conversation follows.

Murmann: In 1993 you wrote a famous article, 'Making teams more effective'. You stated: "The task conflict in teams is good for team performance but relationship conflict in teams is bad for team performance." I have taught this idea to thousands of MBA students but you have changed your view. Why?

Denise Rousseau: First, I think I should say mea culpa, mea culpa. But we should get used to that. The issue was, we believed for many years that relationship conflict in groups was bad but task conflict functional. But then a meta analysis was done by Carsten de Dreu and Laurie Weingart in which they pull all the studies on conflict together and examine the combined effect across all these different studies.

And it indicated both kinds of conflict are dysfunctional for group performance. And that was my first heads-up in the need to be more critical about even beliefs that management professors hold with regard to what their research says.

Murmann: So what does evidence-based management mean and why should businesses care about this?

Rousseau: Well, the key idea in evidence-based management is the notion that the quality of the evidence we use makes a difference to how well our decisions work and to the kind of understandings we have of organisational problems.

And there's been really little attention to the quality of what we know or the facts that we use and, I would say in that case also in terms of management research, to the quality of the management research on which a lot of teaching and conclusions are based.

'We don’t really have good evidence that performance reviews are effective for achieving organisational goals'


Murmann: General Electric and Deloitte have dumped their annual performance reviews and instead are doing continuous, more informal performance feedback. Does the scientific evidence support this change?

Rousseau: It's a very interesting trend to question the use of performance appraisals. Basically, I think it's accurate to say my own reading of the business research literature and some recent meta-analyses critiquing the research would say we don't really have good evidence that performance reviews are effective for achieving organisational goals.

What the research would say is that how the reviews are done is important when they affect what employees think of as organisational justice or fairness. If the process is fair, organisational performance improves. If the process is not fair, the effect is adverse.

The movement to abandoning performance reviews, I think, comes from a couple of things. They're hard to make fair because they're not very consistent and people aren't well trained in making reviews and giving feedback to their subordinates. Another issue is that, actually, if the goals are to improve the quality of people's behaviour in the organisation and improve the cooperation and business outcomes, coaching and ongoing feedback work better than relying heavily on the annual performance review.

So, I think Deloitte and other organisations are recognising that there's been sort of a false promise in the way performance appraisals and performance reviews have been presented and talked about and that it's time to explore other processes that we know work better. And coaching and regular feedback are very well supported in the research literature in terms of improving performance outcomes.

Murmann: But is there such a thing as potentially giving people too much feedback?

Rousseau: Oh, no, that's a different question. All right. Can we give too much feedback? The basic issue I think is the goals, and the context in which feedback is given, and the nature of the feedback. So giving people feedback about their task performance is really important and really useful, especially when people are new to the job.

But once somebody has acquired skill, too much emphasis on feedback from an external source – one's boss or others – could be distracting to people who really should be focusing on learning, creating their own task strategies to work better. So intermittent feedback would be better.

I think the notion is that continuous feedback is valuable for new learners but not later. I don't think what Deloitte and others are talking about is continuous feedback but more regular feedback. And let me say, feedback plus coaching, that's not necessarily always focused on performance.

Coaching could also be focused on issues that the performer wants to learn about or would like to improve on. So, there could be a much more of a learning focus, rather than a results focus. And so in that process, expanding ... the quality of information that comes to employees from just results to actually developmental content could be really, really valuable.

Murmann: Who are the leaders in evidence-based management? Google springs to mind. Who are the other companies that have adopted this style of management?

Rousseau: Well, I think when we talk about evidence-based management it's a family of approaches. But I would say organisations that are fact-based, that look to multiple sources of information, from their stakeholders, their employees, business information and from the research literature, those are the organisations that are sort of having the full meal of evidence-based practice.

And one organisation that I think is exemplary in that regard is actually the US Army. The army has, as a function of issues of manpower, issues of complexity, not only being a military force but also a peacekeeping force, had to focus on learning.

And if you want to learn and scale up your capability, the idea of vetting the information that you use and evaluating the actions you take to try to make improvements becomes very important. So there's really, I think, a much more diverse use of evidence sources in the army than are used by many companies.

'And of course, we know with people, if we have a hammer a lot of things look like a nail'


Murmann: Big data is the new buzzword, following closely on the heels of innovation and entrepreneurship. So many companies are investing time and money to figure out how to harness big data. So how does big data analytics differ from evidence-based management, or do they go hand-in-hand?

Rousseau: Well, big data – the whole movement towards big data – has lots of facets. Partly it is a function of the information explosion and also a function of the sophistication of machine learning and analytic capability. And of course, we know with people, if we have a hammer a lot of things look like a nail.

The overlap between big data and evidence-based management basically is this: organisational data or data on the market are one big potential source of evidence that could inform management decisions. But big data isn't very helpful to us unless we have a good understanding of what it is we want to learn from the data and many companies really aren't so clear yet as to its role.

Murmann: Do you have an example of a company that had a very clear model of what they wanted to achieve and then used big data? Is there a good template for this?

Rousseau: Well, I think the organisations that I've seen – being an organisational psychologist, organisations that I've done consulting and coaching with – have mostly been HR practices in banks and in healthcare organisations. And in this case, the interesting thing is they often hire people from marketing who are very good at analytics into HR to help to try to understand the factors that create high-potential employees and the factors that lead to their success.

I'm thinking about the Royal Bank of Canada as an example, which has a big data analytics project dealing with high-potential workers. And they've hired PhD level psychologists and statisticians to do that work but they've taken their time to understand ... the career cycle and career processes and how the organisation's long-term strategy is going to need to be populated by talent.

And in that kind of context you can ask some pretty intelligent questions about what experiences lead people to be better positioned for management succession. What kinds of personal skills need to be developed and coached in order to realise full potential.

Murmann: If I'm not at a business school – if I've graduated, or if I'm a consultant, or if I'm working in a large corporation and I'm presently not doing evidence-based management – how could I learn more about this? Where would I go?

Rousseau: Well, that's the other reason I'm here – to talk about how we can reach people more broadly who want to improve the quality of their professional practice as managers. And what I would refer you to is the Centre for Evidence Based Management (CEBMa).

It has a website that has free and available training modules with regard to what is evidence-based management, other modules explaining how to make sense of the scientific literature, and if you're interested you can even access the scientific literature on your practice question through CEBMa.

And I think that's an invitation for people to begin exploring some of the ways in which they could expand the kinds of information they're using in decisions and their ability to vet the quality of it. So CEBMa would be happy to help and it's free of charge. Look for us on Google.

Peter Murmann is a professor of management and AGSM fellow at UNSW Business School.

comments powered by Disqus

Subscribe now

BusinessThink is a free online publication. By subscribing, the latest edition will be delivered to your inbox once a month.

​ ​
Print PDF