A Debreceni Déri Múzeum Évkönyve 1985 (Debrecen, 1986)

Természettudomány - Szathmáry László: A Factor Analysis on the Cranial Finds of the Early Neolithic Körös-Cris-Starčevo Civilization

Thresholds Figure 6. The cluster dendrogram of the 12 measurements on the basis of the loadings of the 4 extracted factors

—Females In the case of the females (Table 7) 5 variables (measurements) could be linked to the first factor absolutely certainly: the same 5 face measurements as with the males. In the second factor there was only 1 variable expressed (M20). None of the rest of the measurements could be definitely attached either to the third or to the fourth factor. As for the communalities, they were high. Thereupon a Varimax rotation was performed, which promised to be advantageous from the point of view of differentiation. The factor loadings gained in this way are consisted of by Tables 8 and 9 in sorted form. The two tables show expressively that face measurements gathered into the first factors of both sexes in a great extent. In the second factor of the males breadth measurements concentrated (M9, 8, 45), while the maximum length (Ml) dominated in the third factor and the upper facial height (M48) in the fourth factor. With the females the upper facial height was also summed up in the facial factor, while the breadth measurements (M45, 9) were primarily expressed in the third factor. As for the further factors, the second was mainly characterised by length, the fourth by the cranial height of low representation (M17) and cranial breadth (M8). Figures 3 and 4 indicate the Varimax rotated factor loadings in the first two dimensions. These figures clearly suggest the discreteness of the facial skeleton. Relying upon the four extracted factor loadings Figures 5 and 6 demonstrate hierarchic connexions of the measurements. Individual analysis could be carried out on the basis of the factor scores. Figure

7 consists of the male finds represented according to the scores of the first two factors. Here the finds of the two cultures dissevered by the first (facial) factor, in fact: the scores of the Starcevo finds were higher. With the females (Figure 8) it was rather the second (breadth) factor, by which the finds of the two culture groups' dissevered, while by the first (facial) factor both finds groups form 2 groupings separately. It is very hard to judge the separation of the finds of the two culture groups by illustrating the factor pairs (altogether six and six figures). Orientation was made easier by a culture analysis performed on the basis of the scores of the four extracted factors of the finds (Figures 9 and 10). Figure 9 shows an astonishingly regular arrangement of the males' dendrogram. Except for two small mixed groups the finds of the two culture groups entirely kept separate. The situation was similar in the case of the females: almost each groupings were homogeneous (Figure 10). 26