PREDICTION BY REGRESSION AND INTRARANGE DATA SCATTER IN SURFACE-PROCESS STUDIES

Modeling is a major component of contemporary earth science, and regre ssion analysis occupies a central position in the parameterization, ca libration, and validation of geomorphic and hydrologic models. Althoug h this methodology can be used in many ways, we are primarily concerne d with the prediction of values for one variable from another variable . Examination of the literature reveals considerable inconsistency in the presentation of the results of regression analysis and the occurre nce of patterns in the scatter of data points about the regression lin e. Both circumstances confound utilization and evaluation of the model s. Statisticians are well aware of various problems associated with th e use of regression analysis and offer improved practices; often, howe ver, their guidelines are not followed. After a review of the aforemen tioned circumstances and until standard criteria for model evaluation become established, we recommend, as a minimum, inclusion of scatter d iagrams, the standard error of the estimate, and sample size in report ing the results of regression analyses for most surface-process studie s.