2-6: Analysis of classification

In our classification procedure, multiple assignments were often made concerning a single drawing since, in a single drawing, different drawing aspects or models can occur. Moreover, the range of age of the children or adolescents related to the drawings of a specific aspect or model proved to be very large.

On the basis of the documented drawings of Gilles Porte and their qualities and models related to early human figure drawing, and in order to answer to the question of whether these qualities and models, and potentially their developmental tendencies, prove to be entirely dependent on the concrete context of their production, or whether they are at least in part unrelated to that context, we conducted an analysis of our classification as described below.

First, for each of the two precursory graphic aspects and for the drawing models of the types I to III, drawings were selected in which the aspect or model of the category in question is manifest, but no "higher" aspect or model–"higher" according to the classification catalogue–can be observed in the same drawing. In doing so, the hierarchy of subgroups of a type were also considered.

Types IV and V as well as the manifestation of stick-figures were considered in parallel. However, they were also excluded from the selections of drawings showing non-figurative aspects, pre-forms, or type I.

In consequence, with regard to the precursory manifestations and manifestations of types I to III, the selection procedure was:

• Selection of drawings showing non-figurative aspects, exluding drawings with parallel assignments of pre-forms and of all types of human figure drawing
  
• Selection of drawings showing pre-forms, exluding drawings with parallel assignments of all types of human figure drawing
• Selection of drawings showing type 1, exluding drawings with parallel assignments of types II to V
  
• Selection of drawings showing transitions from type I to type II, exluding drawings with parallel assignments of explicit variants of type II and of types III
  
• Selection of drawings showing explicit variants of type II, exluding drawings with parallel assignments of type III
• Selection of drawings showing transitions from type II to type III, exluding drawings with parallel assignments of explicit variants of type III
• Selection of drawings showing type III

Secondly, for all selections of drawings, we determined

• the total number of drawings
• the age range of picture production
• the countries represented

Thirdly, sorting the pictures according to the age of production, we reduced each of the the selections to their first 2/3 and determined correspondingly

• the age range of picture production
• the average age of picture production (mean)
  
• the countries represented

According to such a procedure, the effect of multiple assignments as well as the effect of persistence of a graphic aspect or model in the graphic development (manifestations in higher ages) can be limited when questioning the developmental tendencies.