Evidence-Based Management Blog

Correlation Is Not Causation (Part II)

Nathan Green — Thu, 02 Feb 2012 11:19:57 +0000

Repeat after me: correlation is not causation, correlation is not causation, correlation is not causation …

At times during my statistics studies I felt like Jack Nicholson in the film The Shining, in which we witness his descent into madness as he types the same sentence over and over again, “All work and no play makes Jack a dull boy. All work and no play makes Jack a dull boy … ”

“Correlation is not causation” is a statistics mantra. It is drilled, military school-style, into every budding statistician. But what does it actually mean? Well, correlation is a measure of how closely related two things are. Think of it as a number describing the relative change in one thing when there is a change in the other, with 1 being a strong positive relationship between two sets of numbers, –1 being a strong negative relationship and 0 being no relationship whatsoever.

“Correlation is not causation” means that just because two things correlate does not necessarily mean that one causes the other. As a seasonal example, just because people in the UK tend to spend more in the shops when it’s cold and less when it’s hot doesn’t mean cold weather causes frenzied high-street spending. A more plausible explanation would be that cold weather tends to coincide with Christmas and the new year sales.

Ask For Evidence!

senseaboutscience.org — Tue, 03 Jan 2012 10:33:56 +0000

Evidence matters in many of the decisions we make – as professionals, patients, consumers, voters and citizens. If you want to know whether a claim made in a policy, newspaper article, advert or product is backed by scientific evidence, ask the people making the claim to provide it. You can ask penetrating questions about evidence, whatever your experience. You just need an inquisitive mind and a desire to stand up for science in public life.

How do you ask for evidence?

If claims are not supported by evidence get in touch with the company. Companies are often unprepared to be questioned about the claims they make suggesting not many people do ask.

Questions to ask

Ask them about the science behind the claim: What kind of testing has been done (controlled, blinded tests; a clinical trial; lab studies on an ingredient)? What is the mechanism behind the science?
Ask about the status of evidence for the claim: Has the research been peer reviewed and published? Has it been replicated?

Need help?

You don’t need to be a scientist to ask for evidence. Anyone can pick up the phone or send an email asking for evidence behind any claim. But if you need more help, Sense About Science has hundreds of organizations and 5,000 scientists standing by.

How mixing work incentives put us on the horns of a dilemma

Alex Fradera — Wed, 14 Dec 2011 20:51:05 +0000

To encourage collaboration, many organisations structure incentives so that whole groups are rewarded – or not – based on their collective output. However, the groups-eye view allows for social loafing, where people shirk duties and assume team-mates will carry their load, so it’s tempting to keep everyone accountable by adding incentives to individual performance too. Christopher Barnes and his colleagues set out to see just how these mixed incentives turn out in practice.

Elective Surgery and Leadership

Wally Bock — Wed, 14 Dec 2011 20:42:29 +0000

I’m blessed with a deep, distinctive voice. I had nothing to do with it, all the credit goes to God and my ancestors. My children refer to it as “The Voice.” Two of them have it as well.

I thought about this after reading a post on Jobmouse with the title “Deep Male Voices are Indicative of Good Leaders.” It would be easy to take this as proof that I’m a good leader. It would also be dumb.

It would be even dumber to decide that because “deep male voices are indicative of good leaders” you should try to make your voice deeper. Trust me on this one, what you say and do will determine whether you’re a good leader, not your voice.

Consider Patton. He was a legendary combat commander even though he had a high pitched voice. It was what he did and what he said in that voice that mattered more than the tone.

Bottom Line:

When it comes to leadership, if it requires elective surgery or a personality change, it probably won’t work anyway.

Cynicism is bad for business

Alex Fradera — Thu, 08 Dec 2011 20:06:53 +0000

When someone we trust takes us for a ride, the bump back to earth is something we’re unlikely to forget. But when we suspiciously reject an offer from someone else, we may never know what we’ve missed out on due to too little trust. Over time, such asymmetries in feedback can tip us toward an unwarranted cynical stance. It’s clear that cynicism is as unhelpful a bias as naivety: it leads to guarded communication, reduced sharing, and more self-serving biases, all of which may cause interactions to nosedive. A recent review by Chia-Jung Tsay and his team from Harvard Business School may help us understand cynicism and how it develops.

The review identifies some key triggers that enhance cynicism, including:

Being new to negotiation – novices are more likely to believe that negotiation is always competitive;
Thinking about the power of influence; for instance, knowledge that another party is a sales expert leads negotiators to suspect their offers more;
Inclusion of a shady character – negotiating groups take the least trustworthy individual in the other group as the best indicator of group trustworthiness;
Clear power asymmetries – people expect more misrepresentations from authorities with access to hidden information.

Bad Science: Can head shape determine chances of business success?

Marc Abrahams — Thu, 08 Dec 2011 19:26:32 +0000

Research suggests that the shape of a chief executive’s head can show whether he will be successful

A new line of American-British research suggests that the shape of a chief executive officer’s head can indicate how well his firm will prosper. The shape also predicts whether the chief executive will act immorally.

The research offers a mathematical tool that financial analysts can add to their professional kit bag: the chief executive officer’s facial width-to-height ratio. The “chief executive facial WHR”, for short.

The research and its financial implications are outlined in a study called A Face Only an Investor Could Love: Chief Executive Facial Structure Predicts Firm Financial Performance, to be published in the journal Psychological Science.

The authors, Elaine Wong and Michael Haselhuhn at the University of Wisconsin-Milwaukee, and Margaret Ormiston at London Business School, explain the significance of their work. Prior researchers, they say, failed “to empirically identify physical traits that predict leadership success” or predict “the ability of leaders to achieve organisational goals”.

A Nobel Prize for Management?

Eric Barends — Thu, 01 Dec 2011 20:48:02 +0000

Research suggests traditional promotion methods are as good as promoting people at random.

Last month’s award of the ‘Nobel Prize’ in Economic Sciences to Thomas Sargent and Christopher Sims brings to an end the 2011 Nobel season. Alfred Nobel specified only five prizes: physics, chemistry, physiology and medicine, literature and peace. However, in 1968, the Swedish central bank backed a new award for economists. Although it is not an official Nobel Prize, its announcements and presentations are made along with the other prizes. But is there room for one more: a Nobel prize for management?

Well, that’s not likely to happen, so I’m afraid we have to settle with the somewhat less prestigious Ig Nobel Awards. The Ig Nobel Prizes “honor achievements that first make people laugh, and then make them think”. In 2010, it was the first time that the prize was awarded to research in the field of management.

Three Italian researchers were awarded the prize for their research on how to avoid the Peter Principle. This famous Principle states that employees of a hierarchical organization climb the hierarchy until they reach a position at which they are incompetent. Alessandro Pluchino and his team tested this by simulating different promotion strategies.

To their surprise, they found that that companies are more efficient when employees are promoted randomly, rather than being promoted because of competence. Even promoting incompetent persons produced a more effective organization than promoting employees based on their performance and skills.

Read more about this study on the Guardian site or watch this video on YouTube

Evidence-Based Management has Issues

Wally Block — Thu, 01 Dec 2011 19:40:04 +0000

I was raised in a family where spirited discussions were common. Sometimes when we were in the middle of a good debate, my mother would break in with one of her classic lines. “It’s a fact,” she would remind us, “We don’t have to argue about it. We can look it up.”

At that point we’d get up from the dinner table and begin pulling out reference books. We might call the library (you could do that then) or call a family friend who would know the answer to our question. We were on a quest for facts.

That’s the whole idea behind Evidence-Based Management. Get the facts about what works. Forget about management folklore. Forget about management fads. Forget about what you think is true. Get the facts.

I got excited when I first started reading about Evidence-Based Management. It seemed perfect for a time when new “studies” of dubious quality flow across the news wires hourly. I liked the idea of applying the scientific method to business.

This is scientific method in its basic and universal form, not the “physics-envy” form where the only good data is acquired using quantitative methods in carefully controlled situations. That might be good for some science, but it won’t work for any human endeavour like management which involves making decisions on the fly in a complex and changing environment.

Employee Engagement: What to Make of Survey Findings

Stephen J. Gill — Tue, 22 Nov 2011 21:48:15 +0000

In reference to a recent Gallup survey of employee engagement that shows that 71% of employees are not engaged in their work, Wally Bock, in his Three Star Leadership Blog, offers these provocative observations:

“After a full decade of high-priced experts and “engagement partners” laboring away, engagement scores have not changed significantly. Maybe it’s time to send the consultants packing and see if we can find a better use for the money we’ve been paying them…Perhaps we can do better with homebrew methods of improving engagement. Perhaps we should quit worrying about engagement per se and turn our attention to something where our investment can make a difference…Remember this common definition of insanity: Doing the same thing over and over and expecting a different result.”

I think there are at least two other interpretations of the Gallup findings. One is that Gallup is not measuring what many organizational leaders think is engagement.

Evidence-Based Management: What Else Is There?

Stephen J. Gill — Tue, 15 Nov 2011 02:15:25 +0000

Jeffrey Pfeffer and Robert Sutton, in a recent article in the New York Times (check out the “uncut” version), write…

…the growing pile of studies on the human and financial costs of employee disengagement, management distrust, poor group dynamics, faulty incentive schemes and other preventable damage suggests a need for an evidence-based management movement.

I admire these two Stanford University thought-leaders, but my immediate reaction to their thesis is, “Duh!” What have companies been doing for the past hundred years of modern management…managing by the-seat-of-their-pants? I suppose the answer is, “Yes.” However, I don’t think it’s a conscious choice or because they didn’t listen during their MBA classes. If you ask corporate leaders if they use research to make their decisions, I’m sure most, if not all, would say, “Of course I do.” They already believe that they practice evidence-based management.

However, they, like most human beings, have a “confirmation bias”. They tend to look for data that confirms what they already believe to be true and disregard the rest.