Standardized Regression Coefficients: A Further Critique and Review of Some Alternatives

摘要: Standardized regression coefficients continue to be used, no doubt because they appear simplify the interpreta tion of results: Unitless quantities such as standardized seem easier describe than that depend on units, and, precisely lack comparable. Nevertheless, previously explained,1 this apparent comparability is illusory. Now a defense has been mounted,2 we can offer some new illustrations logical fallacies inherent in their use. We begin by repeating main argument against comparing standard ized coefficients: comparisons are fact confounded, rather standardized.1 Next, modification traditionally allows them compared across studies when simple logistic model adequate, and more complex procedures necessary effects under study cannot modeled single coefficient, or one wants compare public health different factors. Finally, note all our arguments apply with equal force use correlation measures effect.