Attributing the Right Cause

Apr 18 2011

“There comes a time in the affairs of man when he must take the bull by the tail and face the situation.”  - W.C. Fields

Fundamentally flawed

In What the Dog Saw, Malcolm Gladwell discusses the effects of the fundamental attribution error (FAE) in several different articles.  For example, in one chapter he discusses the Challenger disaster and the impossibility of having complete control of complex technologies and systems, arguing that attempts to find causes and scapegoats in such situations are futile.  Without acknowledging the FAE, his argument in the chapter touches on the desire of all of us to attribute outcomes, and especially bad outcomes, to specific traits of the people involved rather than the situation they are in.

In a later chapter which specifically references and talks about the fundamental attribution error, he discusses job interviews and some of the research conducted on the FAE showing that our behaviour is much more dependent on context than we would like to admit (unless we want to find an excuse).  His argument is that job interviews are merely ‘samples’ of someone’s behaviour in a very specific context, and not at all predictive of their behaviour in other situations.  Many research studies have shown this effect, that our behaviours in a specific context are very good predictors of our behaviour in a similar context, but very poor predictors of how we behave when the situation changes.

Social biases

Psychologists use the term fundamental attribution error, to describe the human tendency to fixate on supposedly stable character traits and overlook the influence of context in determining behaviour.  When asked about Stanley Milgram’s famous experiment on authority and obedience (or other social psychology experiments such as the Stanford Prison experiment), most of us attribute behaviour to the characteristics of those involved (‘a few bad apples’) and invariably underestimate the likelihood that others (or ourselves) would behave in such terrible ways.

We often know others through one main social role, and do not see them in other contexts which reveal very different sides of their personalities.  While personality traits can predict some things (people who are very confident may be confident across a range of situations), they may also fail to predict behaviour in some situations (where someone’s confidence may fade away in a different social context).

A job interview is a very specific situation, and provides a snapshot of someone’s behaviour in a very artificial environment.  Much research has shown that the judgements we make in such situations (and when judging a new teacher or work colleague) are in large part formed within the first few seconds (and handshake) of meeting them.  Our behaviour in the longer term and real environment may be different.  Indeed, even exams and performance tests arguably (and with a lot of evidence), only measure our ‘peak’ performance and are very poor indicators of our day-to-day performance on the job.

Attributional biases

The FAE is only one of several attributional biases which are socially based.  We tend to see the causes of behaviour differently when we describe ourselves as opposed to when we describe others (our behaviours are much more down to situation whereas other’s behaviour is due to their personality, something known as the actor-observer bias).  And for ourselves we tend to focus much on personal traits when describing successes and much more on situations when we describe our failures.

Marketing and research biases?

We tend to ignore these important effects in marketing and research in two important ways. Firstly, we assume that the behaviour of participants in a research setting (which is often very artificial and only a small sample) will be typical of their behaviour in other contexts (which is clearly not supported by the evidence). Secondly, we focus on asking questions about the opinions and beliefs of participants, believing they can predict behaviour.  All the evidence shows that contextual cues are far more powerful than disposition in driving behavioural outcomes.

If we want to understand consumer behaviour we need to understand the person in the situation.  This means developing a greater understanding of consumers across a broader sample of different environments, and ensuring that some of those environments reflect real situations (or mentally simulated ones at least).

If marketers are to avoid the fundamental attribution error, they need to think more about context and real behaviours, and much less about opinions.

The Social Animal (7th Edition) by Elliot Aronson  (1994)

What the Dog Saw: And Other Adventures by Malcolm Gladwell (2010)

No responses yet

Leave a Reply