L-scaling is introduced as a technique for determining the weights in weighted averages or scaled scores for T joint observations on K variables. The technique is so named because of its formal resemblance to the Leontief matrix of mathematical economics. L-scaling is compared to several widely-used procedures for data reduction, and the discussion proceeds in terms of descriptive statistics since the various techniques have sampling properties that are either unknown or intractable. In principle, a researcher should choose a scaling method by proposing a model that explains how the discrepancies arise. This inferential approach is difficult in cases where the X data do not satisfy such requirements as multivariate normality and the statistical independence of the observations. In these cases, a researcher may choose to apply a kind of sensitivity analysis by comparing the outcomes of several scaling methods, including L-scaling. Four tables present analysis results. (SLD)