Abstract
Bentler and Raykov (2000, Journal of Applied Psychology 85: 125–131), and Jöreskog (1999a, http://www.ssicentral.com/lisrel/column3.htm, 1999b http://www.ssicentral. com/lisrel/column5.htm) proposed procedures for calculating R 2 for dependent variables involved in loops or possessing correlated errors. This article demonstrates that Bentler and Raykov’s procedure can not be routinely interpreted as a “proportion” of explained variance, while Jöreskog’s reduced-form calculation is unnecessarily restrictive. The new blocked-error-R 2 (beR 2) uses a minimal hypothetical causal intervention to resolve the variance-partitioning ambiguities created by loops and correlated errors. Hayduk (1996) discussed how stabilising feedback models – models capable of counteracting external perturbations – can result in an acceptable error variance which exceeds the variance of the dependent variable to which that error is attached. For variables included within loops, whether stabilising or not, beR 2 provides the same value as Hayduk’s (1996) loop-adjusted-R 2. For variables not involved in loops and not displaying correlated residuals, beR 2 reports the same value as the traditional regression R 2. Thus, beR 2 provides a conceptualisation of the proportion of explained variance that spans both recursive and nonrecursive structural equation models. A procedure for calculating beR 2 in any SEM program is provided.
Similar content being viewed by others
References
P.M. Bentler (1992) EQS Structural Equations Program Manual BMDP Software Los Angeles
P.M. Bentler E.H. Freeman (1983) ArticleTitleTests for stability in linear structural equation systems Psychometrika 48 143–145 Occurrence Handle10.1007/BF02314682
P.M. Bentler T. Raykov (2000) ArticleTitleOn Measures of explained variance in nonrecursive structural equation models Journal of Applied Psychology 85 125–131 Occurrence Handle10.1037/0021-9010.85.1.125
L.A. Hayduk (1987) Structural Equation Modeling with LISREL: Essentials and Advances Johns Hopkins University Press Baltimore
L.A. Hayduk (1996) LISREL Issues, Debates and Strategies Johns Hopkins University Press Baltimore
Jöreskog, K. (1999a), http://www.ssicentral.com/lisrel/column3.htm dated October 2, 1999. On October 6, Stephen du Toit announced to “LISREL users” that this was available in “Karl’s Corner” of Scientific Software International’s (SSI’s) website.
Jöreskog, K. (1999b). http://ssicentral.com/lisrel/column5.htm dated December 30, 1999 on Scientific Software International’s website.
K. Jöreskog D. Sörbom (1996) LISREL 8: Users Reference Guide Scientific Software International Chicago
H.H. Pesaran R.J. Smith (1994) ArticleTitleA generalized R 2 criterion for regression models estimated by the instrumental variables method Econometrica 62 705–710 Occurrence Handle10.2307/2951666
D. Rindskopf (1984) ArticleTitleUsing phantom and imaginary latent variables to parameterize constraints in linear structural models Psychometrika 49 37–47 Occurrence Handle10.1007/BF02294204
J.E. Teel W.O. Bearden S. Sharma (1986) ArticleTitleInterpreting LISREL estimates of explained variance in nonrecursive structural equation models Journal of Marketing Research 23 164–168 Occurrence Handle10.2307/3151663
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hayduk, L.A. Blocked-Error-R 2: A Conceptually Improved Definition of the Proportion of Explained Variance in Models Containing Loops or Correlated Residuals. Qual Quant 40, 629–649 (2006). https://doi.org/10.1007/s11135-005-1095-4
Issue Date:
DOI: https://doi.org/10.1007/s11135-005-1095-4