Show all articles in this section

Groundbreaking new organisational theory

AUTHOR: Lachlan Colquhoun   DATE: 26.11.06   ISSUE 2, 2006

Professor Donaldson’s Statistico-Organisational theory is the first major new organisational theory since the 1970s and is set to challenge the academic community around the world with its innovative approach.

The theory was first outlined in a paper Professor Donaldson presented at the University of Hawaii last year, and the full theory will be expounded in a book due for publication in 2008 by leading US academic publisher M.E. Sharpe.

In his original paper, Professor Donaldson says the key idea of the theory “is that organisations are statistical machines.”

Statistico-organisational theory draws on ideas from statistics but applies them in the context of organisational behaviour and decision making.

Illustration: Gregory Baldwin

“Their workings are governed by the universal principles of statistics, which pervade all of their operating processes,” he writes.

Statistico-organisational theory draws on ideas from statistics, many of them already familiar to social scientists versed in modern methodologies, but applies them in the context of organisational behaviour and decision making.

“A lot of people are trying to build theories of organisations from economics, or from sociology, but I’m going to a very different source and drawing ideas from the well of statistics,” says Professor Donaldson.

“So this is very much a new intellectual venture, because no-one has tried to build a theory of organisations from statistics before.”

The result, believes Professor Donaldson, will be a sounder and more scientific understanding of organisational behaviour, and more accurate and predictive decision making by organisations which apply the theory.

“People have been using statistics and other bits of methodology to test theories for some time now,” says Professor Donaldson. “But what I’m doing here is using statistics to generate the theory, and I think that is what is radically new about this and what nobody has done before.

“I’ve had this dissonance ever since I started doing research, and it has always struck me as odd that the formulas I would follow to test theories were very coherent and much more rigorous and precise than the theories we were testing.”

Much of the basis for statistico-organisational theory is around the accurate interpretation of the data on which organisations base their decisions.

By being aware of these common mistakes, managers can avoid pitfalls of interpretation which result in inefficiency and poor decision-making. One of the main sources of error, says Professor Donaldson, is known as the Law of Small Numbers.

Professor Lex Donaldson’s Statistico-Organisational theory is the first major new organisational theory since the 1970s.

Photo: Professor Lex Donaldson

“Take an example of somebody running a bank, and beginning as a manager of a small branch in Hobart,” he says.

“If they do well, the bank will say you are a good guy and eventually you will run a bigger branch, but are we really able to tell from the performance at the Hobart branch if this person is a good manager?

If, for example, a key performance criterion is the number of commercial mortgages written in a year, then the numbers for the small Hobart branch are likely to be distorted.

“You might do ten mortgages one year and three the next, so really what you’ve been given is a lottery ticket because this data is much more prone to bias or variation,” says Professor Donaldson.

Much of the basis for statistico-organisational theory is around the accurate interpretation of the data on which organisations base their decisions.

A way of avoiding this error, he says, is to pool or centralise all of this information to deliver a more accurate picture, which is likely to be a much better predictor of the true situation.

“The point of centralising information is that instead of having all the little bitty branches you have the data for the whole thing and so you can arrive at a much better average,” says Professor Donaldson. “So you have an average which is a much better predictor of what is actually going on.”

Another common statistical error many organisations make is measurement error, which is particularly common in situations where a number is calculated as a “difference score” – for example in the case of profits which are calculated as sales minus costs.

As many organisations use profit measurements as a key metric for assessing their performance, a distorted profit figure can often unwittingly be at the core of erroneous strategic decision-making.

“In determining profit, there might be uncertainty of around 10 per cent in the sales figure, and 10 per cent in costs, so when you take profit as the difference between those two things you can get a massive error,” says Professor Donaldson.

“These difference scores tend to have much more error in them than a single variable, so even if it seems a very innocent thing to subtract one number from another it can inherently deliver a lot more error.”

Statistico-organisational theory is a "logical step" in the development of organisational theory.

Professor Donaldson says statistico-organisational theory is grounded “very much in the tradition of rationalism” and describes it as a “logical step” in the development of organisational theory.

“Social scientists and positivists like me look to social science as a role model for our inspiration,” he says.

“And I’ve used that tradition as the ‘light on the hill’ in all my work, and with this theory I’m giving it a bit of a kick along by doing it in a way that is a bit more extreme.”

Professor Donaldson says that already his theory is prompting some interest among the world academic community.

“I’m inviting people to apply it to their favourite methodological principles, and I’ve found that if I talk about these ideas with other social scientists they can begin to see how they would construct their own theories around this,” he says.

“So I think you might see a rapid proliferation of variations of this root idea, and it might stimulate some more empirical research and find its way back into executive teaching programs.”

In this way, he says, statistico-organisational theory will begin to have an impact on the business world with AGSM making a “distinct contribution” to intellectual change.