The evaluation of corrosion data by statistical methods often is desirable in order to obtain maximum useful information from the data. This discussion is concerned with a graphical method for a multiple correlation analysis involving several variables. The graphical procedure is easier to use than mathematical procedures. Also, it provides visual evidence of the relationship between variables as the work progresses. The method is not limited to simple linear functions. For purpose of illustration, weight loss data from an atmospheric exposure of low alloy steels are studied. The method is described step by step and the effect of concentration of nickel, manganese, phosphorus and silicon on weight loss is shown. The final results of the graphical analysis are compared with a mathematical analysis of the same data.
No takes yet. Share an insight, caveat, or question.
O. B. ELLIS (1953) studied this question.