Chasing Insights: A question of causality

13/10/2013 19:04 BST | Updated 23/01/2014 23:58 GMT

As researchers, we all love that 'eureka' moment; when we catch a glimpse of an underlying truth beyond the reams of data, the endless transcripts and the lengthy debates. Numbers take on meaning and context, isolated opinions start to connect, and we are struck by a shared respondent experience, which looms larger than the enormity of individual differences. We see our sample as united by common beliefs or motivations, borne out of a shared consumer or business landscape. We've found a pattern!

Truth be told, it's not just the researchers who get a kick out of patterns, it's human nature to search for trends and short cuts, to rationalise, generalise, sort, filter, or do anything at all really to get past the details and minutia that surround us. The idea that our social and physical world can be explained and even predicted is comforting, and indeed we must assume that this is the case if we are ever to feel any sense of control, or confidence in our decisions.

But the psychologists will tell us that our short cuts and generalisations are not always reliable. We often make poor decisions based on our own past experiences, because we group together choices and situations that in fact have little in common. It worked last time, so why shouldn't it work again? Because the time, the place, the people involved and any number of other details are different.

In our eagerness to see even the truly random as a potential pattern, we can leap to some very unlikely conclusions. In the context of our insight-hungry research industry, we must be wary of this temptation. Under pressure to 'add value' and in our search for that holy-grail of an insight, we need to remember that not every trend holds the answer, and that there is such a thing as a random pattern.

Insight or Oversight?

The laws of cause and effect are a common stumbling ground when it comes to connecting the unrelated. Two measurements, behaviours or events may appear to be connected, always occurring at the same time or at the same rate, however, this does not necessarily mean that one is the cause of the other. For example, there may be a third, hidden variable that links the two occurrences.

A strong positive correlation has been observed between the annual number of nesting storks, and the annual number of babies born in the city of Copenhagen. From this you could infer that there was some truth to the legend that storks bring babies, or if you don't go in for legend, it could even be the babies bringing the storks. But a little scepticism, or perhaps common sense will quickly reveal that as the population grows, so too does the building construction to accommodate it, thus providing more nesting opportunities for storks who like to make their homes in the rooftops. Though the construction is linked to both the new born babies and the nesting storks, sadly the storks and the babies have no direct causal relationship.

So, before we start shouting our latest wild interpretation from the rooftops, we should consider the possibility that it may be just a red herring, or indeed a nesting stork, posing as a friendly insight.

Written by Nichola Kent-Lemon, Research Manager at Northstar Research Partners. For more posts like this, or to get in touch with Nichola, please visit

Follow Northstar on Twitter: