Regression has been used to good effect in SPATIAL ANALYSIS for instance, Fulford and Hodder (1974) used regression analysis to study the distribution of late Roman pottery from the north Oxfordshire kilns. More complex types of regression are curvilinear regression (more complicated relationships between two variables) and multiple regression (relationship between one ‘dependent’ variable and several ‘independent’ ones).Ĭase-study: regression analysis and the distribution of Roman pottery.
The term correlation is also used more generally (but wrongly) to mean association ( see CONTINGENCY TABLE). that one variable can be exactly predicted from the other) is measured by the correlation coefficient, which can vary from –1 for an exact negative relationship through 0 for no linear relationship to +1 for an exact positive relationship. The extent to which the relationship can be described as a straight line (i.e. The simplest linear regression, seeks the best straightline relationship between two variables, and studies the differences (known as residuals) between the ideal and actual data. Regression analysis Techniques for examining the relationship between two (or more) continuous VARIABLES, each measured on a set of Trigger: A history of archaeological thought (Cambridge, 1989). Schiffer:įormation processes of the archaeological record Binford: In pursuit of the past (London, 1983) M.B. Cannon: ‘The corporate group as an archaeological unit’, JAA 1 (1982), 132–58 L.R. Hammond: ‘Child’s play: a distorting factor in archaeological distribution’, AA (1981), 634–6 B. Binford: ‘Behavioral archaeology and the “Pompeii premise”’, JAR 37/3 (1981), 195–208 G.
Schiffer: Behavioral archaeology (New York, 1976) L.R. Rathje: ‘The Garbage Project: a new way of looking at the problems of archaeology’, Archaeology 27 (1974), 236–41 M.B. Nevertheless it has proved difficult to validate even very general cross-cultural observations about refuse disposal. Are more likely to be abandoned in the localities where they were used in temporary hunter-gatherer sites than in larger and more sedentary ones, where the disposal of waste material was much more highly organized’.