A Comparison of Bias and Mean Squared Error in Parameter Estimates of Interaction Effects: Moderated Multiple Regression versus Error-in-Variables Regression.

The results of moderated multiple regression (MMR) are highly affected by the unreliability of the predictor variables (regressors). Errors-in-variables regression (EIVR) may remedy this problem as it corrects for measurement error in the regressors, and thus provides less biased parameter estimates. However, little is known about the properties of the EIVR estimators in the moderator variable context. The present study used simulation methods to compare the moderator variable detection capabilities of MMR and EIVR. Specifically, the study examined the bias and mean squared error of the MMR and EIVR estimates under varying conditions of sample size, reliability of the predictor variables, and intercorrelations among the predictor variables. Findings showed that EIVR estimates are superior to MMR estimates when sample size is high (i.e., at least 250) and the reliabilities of the predictors are high (i.e., rij ≥ .65). However, MMR appears to be the better strategy when reliabilities or sample size are low.

[1]  J. Tisak,et al.  A Theoretical Note on the Stochastics of Moderated Multiple Regression , 1994 .

[2]  T. Cleary.  TEST BIAS: PREDICTION OF GRADES OF NEGRO AND WHITE STUDENTS IN INTEGRATED COLLEGES , 1968 .

[3]  Richard J. Harris.  A primer of multivariate statistics , 1975 .

[4]  W. Fuller,et al.  An Errors-In-Variables Analysis of Managerial Role Performance , 1974 .

[5]  Edward R. Kemery,et al.  Effects of predictor intercorrelations and reliabilities on moderated multiple regression , 1988 .

[6]  M. Evans.  A Monte Carlo study of the effects of correlated method variance in moderated multiple regression analysis , 1985 .

[7]  A. Sockloff,et al.  The Analysis of Nonlinearity Via Linear Regression with Polynomial and Product Variables: An Examination , 1976 .

[8]  Philip Bobko,et al.  TESTING FOR FAIRNESS WITH A MODERATED MULTIPLE REGRESSION STRATEGY: AN ALTERNATIVE TO DIFFERENTIAL ANALYSIS , 1978 .

[9]  W. G. Cochran.  Errors of Measurement in Statistics , 1968 .

[10]  John E. Hunter,et al.  Halo in a multidimensional forced-choice performance evaluation scale. , 1980 .

[11]  Lynn Smith-Lovin,et al.  Impressions of Goodness, Powerfulness, and Liveliness from Discerned Social Events , 1981 .

[12]  E. F. Stone-Romero,et al.  Relative power of moderated multiple regression and the comparison of subgroup correlation coefficients for detecting moderating effects. , 1994 .

[13]  D. Cohen,et al.  REM sleep facilitation of adaptive waking behavior: a review of the literature. , 1978, Psychological bulletin.

[14]  J. Elashoff,et al.  Multiple Regression in Behavioral Research. , 1975 .

[15]  D. R. Saunders.  Moderator Variables in Prediction , 1956 .

[16]  Karl G. Jőreskog,et al.  LISREL: A General Computer Program for Estimating a Linear Structural Equation System Involving Multiple Indicators of Unmeasured Variables. , 1972 .

[17]  S. Zedeck.  Problems with the use of "moderator" variables. , 1971 .

[18]  Eugene F. Stone,et al.  Some issues associated with the use of moderated regression , 1984 .

[19]  H. F. Taylor,et al.  On a General Model for Social and Cognitive Consistency , 1979 .

[20]  L. Humphreys.  Statistical definitions of test validity for minority groups. , 1973 .

[21]  Michael A. Hidiroglou,et al.  Regression Estimation after Correcting for Attenuation , 1978 .

[22]  Franklin A. Graybill,et al.  Introduction to the Theory of Statistics, 3rd ed. , 1974 .

[23]  David R. Heise.  Estimating Nonlinear Models , 1986 .

[24]  Jacob Cohen.  Multiple regression as a general data-analytic system. , 1968 .

[25]  George W. Bohrnstedt,et al.  The Reliability of Products of Two Random Variables , 1978 .

[26]  L. E. Jones,et al.  Analysis of multiplicative combination rules when the causal variables are measured with error. , 1983 .

[27]  G. Bohrnstedt,et al.  Robustness in Regression Analysis , 1971 .