from the data
Categories are formed in the human mind: they do not exist out there in the external world for us to find. Students, even the art school students I teach, are often worried about the notion of 'subjectivity' because they feel that research should be 'objective' and 'scientific' ('Isn't that just a subjective judgement?'). In the sense of being a function of human intelligence all judgements are subjective: they could not be anything else. The 'objectivity' lies in making explicit the criteria for the judgement; but the process remains a matter of interpretation and opinion to a greater or lesser degree, and particularly in relation to human behaviour, feelings, opinions, and the like.
In physics and chemistry one comes closest to the notion of objective categories, for example, the periodic table in which the elements are arranged in terms of increasing atomic number. First devised in 1869, the formation of this classification has been progressive, the product of human discoveries and justified revisions. Its value is that it provides a universal basis for understanding and predicting the behaviour of naturally occurring elements. Agreed categories are fundamental to these processes of shared knowledge.
There are established procedures for identifying and classifying natural elements; in that sense there is not much scope for 'interpretation'. But if we move to human beings, the picture becomes quite different, even when one is dealing with what would seem to be an objective phenomenon.
The World Health Organization (WHO) has established an agreed classification of the causes of death (the ICD index): there are some 900 main causes of death in this classification. These are used to categorize the mortality patterns of different countries and groups within it (by age, gender, social class, etc.). Death, we can agree, is a relatively objective phenomenon; but deciding the cause(s) of death is another matter. Quite apart from mistakes in post-mortem