Written by Ben Baginsky, Programme Director
When you are presented with a complex challenge, what is your first response? Are you more likely to reach for a phone, a pen, or a computer? We each have preferred ways of receiving information, processing information, and communicating information. Some of us prefer to talk a problem through, others to create, and others to calculate. Too often, the talkers, the creators, and the computers don’t communicate well enough with each other.
I was recently in a consulting session with a group of senior leaders in a public sector organisation. We spent the morning working together on the struggle one member of the group was having getting their organisation to create a more dynamic system for managing clients without risking organisational reputation. They described to me and the group a system with multiple competing stakeholders. Each had different motivations and dug their heels in to protect their own interests. This was highly political, complex work, calling for smart, informed, skilful leadership.
With a boardmarker in my hand, capturing the challenge on a flipchart with the group of mainly technical experts gathered around me, it struck me particularly hard how each one of us was seeing the challenge through our own lens and how challenging it was to undo that. Me on one side with my Roffey Park person-centred approach, and the group on the other with their technical orientation. We did a good job of working the challenge, of bringing forth the best of our different worldviews; but genuinely bridging the divide required more than the commitment we’d all made at the start of the session to be open-minded and creative in working together.
Stepping Outside Familiar Thinking
Leaders often find it hard to identify strategies for avoiding narrow or biased thinking that actually work. I recently took part in an experiment I at first thought I could have done without. The experiment placed a people-centred organisational consultant in an advanced statistics class. As I emerged from this uncomfortable experience, I gained one insight into how leaders and consultants might step outside the familiar zones of their brains. They can then see challenges more vividly and make better choices in how to respond.
A framework we often use at Roffey Park is ‘Cynefin’, which details different types of work – clear, complicated and complex. Complex work demands the greatest part of our attention as organisational development consultants. Technical expertise can shift even the most complicated problems, but complexity requires different and more hard-to-acquire capability. Yet no work fits neatly into one Cynefin category. Even the most complex work contains clear and complicated elements.
My concern is that too often, especially when working with technical leaders, we jump to a definition of ‘complex’ – and stay there. We fail to attend to the other, adjacent types of work, and the overlaps that exist between them.
Seeing Beyond One Lane
Very often, leaders or groups will try to view complex work as complicated. This can become a tactic for avoiding hard work. I’m not in favour of that. I favour an approach that captures the full 360 degrees of a situation, rather than just the angle our brain finds most attractive.
This is beautifully captured by Hassard in his 1993 study of the British Fire Service. He examines the changes taking place within the service through research blending positivist, interpretivist and critical approaches. Organisations often encourage researchers, as with consultants and leaders, to stick to their lane. Numbers and words don’t mix. Hassard insists they should. His work is not pure numbers research nor pure words research; it is rounded and relevant in its holistic nature. What I am saying here is simple: let’s follow Hassard’s lead and do more of the same in our own work.
Opening a Path Between Worlds
With that aim in mind, I share here three core statistical concepts that word-oriented consultants and leaders like me might find intimidating, but could be illuminating. I present them as simply as possible while retaining their shape and logic. I offer them as one way to create a more open path between ways of seeing the world that often don’t meet. When they fail to meet, we lose opportunities to generate meaningful change.
Statistical Concept 1 – Mean, Median, Mode
The word ‘average’ is used a lot; though the different forms it can take are not always well understood. Recognising these distinctions starts with understanding the different types of ‘average’ and what each can reveal. The mean is all of the observed data added together and divided by the total number of datapoints. The median is the number that sits at the 50% mark when you divide a set of datapoints into two equal parts. The mode is the data point that shows up most frequently.
When you are working with a group, is your decision-making based on what most people are saying? Do you rely on where the exact middle of opinion sits, or on the loudest voices? Could making a more conscious choice about what counts as a valid ‘average’ help you support teams more effectively?
Statistical Concept 2 – Autoregression
This is the idea that a current piece of data relates to the last piece gathered about the same situation. Today’s price of a particular stock is likely to relate to the price of that stock the last time it was measured. The problem with autoregression is that you need to treat it cautiously. How certain are you that the data you’re basing your forecast on is reliable? Was a factor affecting one element true before but not now?
You’d expect, for example, to see different staff survey results when a stressor is present compared with when it has been removed. Many of us base what will happen now on what has just happened. We think of the last leader we worked with, or what the last person in a group said or did. Sometimes this helps, sometimes it doesn’t.
Statistical Concept 3 – The Null Hypothesis
Several statistical tests use this principle. You begin with a model and seek to disprove it. It is the statistical equivalent of being innocent until proven guilty. Your goal as a researcher is to prove that an idea is wrong, rather than that it is right. You need a significant amount of certainty before you can reject that something is the case.
Researchers need to prove this beyond reasonable doubt, and they can set that level at different points. In credible research the very least you need to do is disclose your degree of reasonable doubt. Are you 99%, 95% or 90% certain of what you assert? I don’t remember the last time I heard a leader or a consultant make a similar disclosure.
Making the Unapproachable Accessible
These remain fairly superficial ideas. We could go deeper. My statistics professor might cringe at how over-simply I have described them. But the idea feels important. Leaders and enablers of change have a lot to gain when they challenge the foundations of their thinking through available tools. These tools often sit just across the corridor in a different unit of the same business or school. They sometimes wear clothes that make them feel unapproachable. The opportunity lies in making them less so.
