Statistics for industrial consumer studies12 June 2009
Garmt Dijksterhuis, Unilever12 June 2009, 15:30 – 16:15
Garmt Dijksterhuis, Unilever R&D Vlaardingen Abstract: As the field of Food Consumer Science is about understanding and predicting food related behaviour, its research methods should be valid with respect to behaviour. Choosing the best research method requires a clearly formulated research question. Without a clear goal a method cannot be judged do be good or bad. Measuring the behaviour itself is, per definition, the most valid method. Psychological theory amply illustrates that one is mostly not aware of ones own motivations to show a certain behaviour. It is concluded from this that methods that rely on consumers' reflecting on their own behaviour, or on their hypothesised behaviour, can not be very good in predicting actual behaviour. Three criteria to judge methods on are introduced and used to illustrate that a large number of methods that are often applied in Food Consumer Science may be invalid with respect to actual food related behaviour. It may even be impossible to gain insight about future behaviour without ever studying that very behaviour in an experiment.