Some Dangers in "Valid" Social Measurement

Abstract

This paper is a preliminary statement on the dysfunctions that social measurement may have for societal planning. Three problem areas associated with questions of internal validity are examined. The most general one is the area of fractional measurement, which concerns dysfunctions stemming from lack of coincidence between a social concept and its operational definition. Also examined are problems of indirect measurement and problems of formalistic-aggregative measurement of collective attributes. The area of indirect measurement concerns potential negative consequences of using data collected originally for other purposes as measures of social concepts. The area of formalistic-aggregative measure ment concerns dysfunctions flowing from imprecise measure ment of the states of social systems. Two broad classes of dysfunctions in these three areas are identified: (1) arriving at invalid conclusions which become the bases for erroneous policy decisions and (2) ignoring those dimensions and indicators of a concept that are most susceptible to social manipulation.