Abstract
There has been an ongoing debate surrounding the use and interpretability of beta weights. At the end of each debate there seems to be apparent agreement between debators on each other's major points, but the final conclusion restates their disagreement about whether or not beta weights should be interpreted. The two viewpoints being expressed below are representative of this controversy. It appears that the disagreement is not whether beta weights are interpretable, but rather, under what conditions they should be interpreted. Harris' position is that beta weights should be interpreted and that they have very clear mathematical meaning, while cautioning that beta weights can be inappropriately used as a good indice of "overall importance" of individual predictors. McNeil presents three examples and cautions how beta weights have been and can be easily misinterpreted. However, both agree that while beta weights are often misinterpreted, they do have very useful and clear interpretations under appropriate conditions. These positions provide the applied researchers with excellent guidelines for the use and interpretation of the controversial beta weight.
Recommended Citation
Newman, Isadore; Harris, Richard J.; and McNeil, Keith
(1993)
"Both Sides Now: Interpreting Beta Weights,"
Mid-Western Educational Researcher: Vol. 6:
Iss.
1, Article 4.
Available at:
https://scholarworks.bgsu.edu/mwer/vol6/iss1/4
