Social influences on food choice and dietary change: a sociological attitude

Abstract

For someone with a training in social anthropology or sociology a title 'Psycho-social influences on food choice: implications for dietary change' immediately looks a bit odd. The title may seem obvious and perfectly straightforward to some, and in certain respects it is. However, others who look at it from another disciplinary point of view will hesitate; it does not get things quite right. So what I have tried to do in this short paper is describe the sociological attitude which gives rise to this hesitation. I begin by picking out the underlying methodological and theoretical model that the symposium title immediately calls to mind. I then describe an alternative model, before going on to use the main part of the paper to illustrate this alternative. I stress that neither of the models I discuss is superior or inferior to the other. Both have strengths as well as drawbacks. The underlying model implicit in the title treats food choice as a dependent variable and social factors as independent variables. The relationship between dependent and independent variables is indicated by the word 'influences' which suggests more than mere correlation but less than cause and an assessment of the strength of that correlation, if there is one. The model provides for a mode of scientific procedure that requires operational decisions about the instances of food to be studied, the definition of choice to be agreed, the social factors to be selected and the means of measuring them all to be devised. And, somewhere at the core of the model is the assumption that food choice involves 'the individual' who is subject to one or more of the (psycho-) social influences in question. The rest of the title, 'implications for dietary change', suggests that we are to use the results of investigation conducted according to the scientific model just described in order to help decide one of two things. Either we shall know better how dietary patterns, habits, or customs are or are not changing, assuming we have data for the past; or, for the future, the results will tell us how better to manipulate behaviour successfully to get people to eat what is best advised for their health. I shall, incidentally, say no more in the present paper about the use to which results may be put, and instead reserve that discussion for another occasion.