This paper discusses the issue of lateral collinearity in variance-based structural equation modeling (SEM), which is often overlooked in the field of information systems research. Lateral collinearity refers to the relationship between predictor variables and the criterion variable, where two variables may measure the same construct. Unlike vertical collinearity, which is assessed between predictor variables, lateral collinearity is rarely explicitly tested in multivariate analyses. The authors illustrate the problem of lateral collinearity through an example using WarpPLS 2.0 software, demonstrating that standard validity and reliability tests do not effectively capture lateral collinearity. They propose a new approach to assess both vertical and lateral collinearity, which involves creating multiple "dummy" blocks of latent variables or performing a "full" collinearity test. The study emphasizes the importance of conducting comprehensive collinearity tests to avoid misleading conclusions and recommends that researchers use these tests in future empirical research. The findings highlight the need for more rigorous methods to address collinearity issues in SEM, particularly in the context of information systems research.This paper discusses the issue of lateral collinearity in variance-based structural equation modeling (SEM), which is often overlooked in the field of information systems research. Lateral collinearity refers to the relationship between predictor variables and the criterion variable, where two variables may measure the same construct. Unlike vertical collinearity, which is assessed between predictor variables, lateral collinearity is rarely explicitly tested in multivariate analyses. The authors illustrate the problem of lateral collinearity through an example using WarpPLS 2.0 software, demonstrating that standard validity and reliability tests do not effectively capture lateral collinearity. They propose a new approach to assess both vertical and lateral collinearity, which involves creating multiple "dummy" blocks of latent variables or performing a "full" collinearity test. The study emphasizes the importance of conducting comprehensive collinearity tests to avoid misleading conclusions and recommends that researchers use these tests in future empirical research. The findings highlight the need for more rigorous methods to address collinearity issues in SEM, particularly in the context of information systems research.