Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
). For the simple 2W2V design, where two variables (x, y) are measured at two points in time (1, 2), the purpose of the cross-lagged panel procedure is to infer the direction of causality by comparing the cross-lagged correlations [i.e., r(xl, y2) vs. r(yl, x2)]. In path analysis, a model is constructed which hypothesizes the causal relationships among a set of variables. A multiple linear regression equation is derived for each dependent variable in the model, and consists of those variables believed to influence that particular dependent variable. The estimates of these influences are known as standardized path coefficients. Structural equation modeling is a form of path analysis where there are multiple measures or indicators of each unmeasured latent variable, instead of equating indicator variables with latent variables. Even with the recent popularity of such techniques, some researchers still believe that the intent of causal modeling is to make causal inferences from correlational data. In actuality, the intent is to test the plausibility of alternative theoretical models by gathering evidence useful for evaluating the possible causal relationships among the relevant and important variables.Given these introductory remarks, four points will be discussed which relate to the use of these specific methods, particularly with respect to the modeling of reading acquisition. The first point concerns the causality interpretation of cross-lagged panel correlation (CLPC) analyses. Rogosa (1979) has shown that even when a strong causal relationship between two variables is known, the cross-lagged correlations may not be different or may indicate the opposite direction of the causal influence. Thus the CLPC method may not uncover a "true" causal relationship and instead may indicate either that there is no relationship or that there is an inverse relationship. The method merely compares sample bivariate correlations and does not yield estimates of the causal parameters that we are really interested in. For this reason the CLPC technique is not recommended.The second point concerns the measurement error assumption in the path analysis (PA) method. An implicit assumption of PA is that the variables are measured without error, that is, the variables are perfectly or infallibly measured. Considerable measurement error in a path model will have an effect upon the path coefficients. When the variables are based on fallible measure-
). For the simple 2W2V design, where two variables (x, y) are measured at two points in time (1, 2), the purpose of the cross-lagged panel procedure is to infer the direction of causality by comparing the cross-lagged correlations [i.e., r(xl, y2) vs. r(yl, x2)]. In path analysis, a model is constructed which hypothesizes the causal relationships among a set of variables. A multiple linear regression equation is derived for each dependent variable in the model, and consists of those variables believed to influence that particular dependent variable. The estimates of these influences are known as standardized path coefficients. Structural equation modeling is a form of path analysis where there are multiple measures or indicators of each unmeasured latent variable, instead of equating indicator variables with latent variables. Even with the recent popularity of such techniques, some researchers still believe that the intent of causal modeling is to make causal inferences from correlational data. In actuality, the intent is to test the plausibility of alternative theoretical models by gathering evidence useful for evaluating the possible causal relationships among the relevant and important variables.Given these introductory remarks, four points will be discussed which relate to the use of these specific methods, particularly with respect to the modeling of reading acquisition. The first point concerns the causality interpretation of cross-lagged panel correlation (CLPC) analyses. Rogosa (1979) has shown that even when a strong causal relationship between two variables is known, the cross-lagged correlations may not be different or may indicate the opposite direction of the causal influence. Thus the CLPC method may not uncover a "true" causal relationship and instead may indicate either that there is no relationship or that there is an inverse relationship. The method merely compares sample bivariate correlations and does not yield estimates of the causal parameters that we are really interested in. For this reason the CLPC technique is not recommended.The second point concerns the measurement error assumption in the path analysis (PA) method. An implicit assumption of PA is that the variables are measured without error, that is, the variables are perfectly or infallibly measured. Considerable measurement error in a path model will have an effect upon the path coefficients. When the variables are based on fallible measure-
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.