The Superiority of Simple Alternatives to Regression for Social Science Predictions

132 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Correlation Weights in Multiple Regression (2010, background): “Although prior work (Campbell, 1974; Claudy, 1972; Dana & Dawes, 2004; Davis & Sauser, 1991; Goldberg, 1972; Marks, 1966) supported the effectiveness of correlation weights in sample data, much of this work relied on Monte Carlo simulations or reanalyses of existing data sets.”
Findings from Evidence-Based Forecasting: Methods for Reducing Forecast Error (2006, unknown): “Promising methods for cross-sectional data Damped causality In a large-scale study of cross-sectional data, Dana and Dawes (2004) showed that the gain from equal weights is larger when sample sizes are smaller and predictability is poor.”
Illusions in Regression Analysis (2012, unknown): “Einhorn and Hogarth (1975) and Dana and Dawes (2004) show the conditions under which regression is and is not effective relative to equal weights.”
Golden Rule of Forecasting: Be conservative (2015, unknown): “Dana and Dawes (2004) analyze the relative predictive performance of regression and equal-weights models for five real non-experimental social science datasets and a large number of synthetic datasets.”
Partial least squares path modeling: Time for some serious second thoughts (2016, methodology): “Commonly used prediction weights include regression, correlation, and equal weights (Dana and Dawes, 2004).”