Uploaded January 2023 | Updated September 2026, 2 weeks ago
This video demonstrates how to perform a path analysis using latent variables based on an example provided by Kline (2016) in his text, Principles and Practice of Structural Equation Modeling (4th ed). In this video, I begin by demonstrating how to create a full covariance matrix for a set of variables based on summary correlations and standard deviations provided by Kline (2016) in his example. Next, I demonstrate how to test a measurement model to assess how well the indicator variables are reflecting their latent constructs. Following, I test two full structural equation models, where proposed structural relations among the latent factors are specified.
A copy of the TEXT FILE I referenced at the end the video can be downloaded here:
drive.google.com/file/d/1_m-rV5d5ignBQf1ff8jT2yBkiMfxw8mV/view
NOTE: I have created a video as an amendment to the this presentation that can be viewed in a shorter video here: youtu.be/KHAIUARdA8Q . This video describes how to compute and test specific indirect effects within this model.
More videos and resources on SEM with Lavaan can be found here:
sites.google.com/view/statistics-for-the-real-world/contents/structural-equation-modeling
Be sure to check out my online sites at:
drmcstats.net
sites.google.com/view/statistics-for-the-real-world/home
This video demonstrates how to perform a path analysis using latent variables based on an example provided by Kline (2016) in his text, Principles and Practice of Structural Equation Modeling (4th ed). In this video, I begin by demonstrating how to create a full covariance matrix for a set of variables based on summary correlations and standard deviations provided by Kline (2016) in his example. Next, I demonstrate how to test a measurement model to assess how well the indicator variables are reflecting their latent constructs. Following, I test two full structural equation models, where proposed structural relations among the latent factors are specified.
A copy of the TEXT FILE I referenced at the end the video can be downloaded here:
drive.google.com/file/d/1_m-rV5d5ignBQf1ff8jT2yBkiMfxw8mV/view
NOTE: I have created a video as an amendment to the this presentation that can be viewed in a shorter video here: youtu.be/KHAIUARdA8Q . This video describes how to compute and test specific indirect effects within this model.
More videos and resources on SEM with Lavaan can be found here:
sites.google.com/view/statistics-for-the-real-world/contents/structural-equation-modeling
Be sure to check out my online sites at:
drmcstats.net
sites.google.com/view/statistics-for-the-real-world/home

![A super-easy effect size for evaluating the fit of a binary logistic regression using SPSS
This video provides a short demo of an easy-to-generate effect size measure to assess global model fit for your binary logistic regression. The procedure is described by Tabachnick & Fidell (2013) and Menard (2000). It also a mathematical component of Menards (2011) approach to standardizing regression coefficients using his formula.
IMPORTANT: The approach discussed in this video is one of many possible pseudo-R-square approaches approaches (see Menard, 2000) and tests (e.g., likelihood ratio chi-square test; Hosmer & Lemeshow chi-square test) for evaluating and reporting on model fit. Although this is a very easy index to compute, I strongly encourage its use conjunction with other measures of global model fit (such as those just noted) when evaluating the fit of your model.
[Side note: In terms of alternative pseudo R-squares, my preference is McFaddens (which seems to be slightly preferred by (Menard, 2000), as it is more intuitive as a proportional reduction in error variance (p. 24). I review this measure here: https://youtu.be/vY3L_6myOc8]
Articles citing the procedure described in this video:
Menard, S. (2000). Coefficients of determination for multiple logistic regression analysis. The American Statistician , 54, 17-24.
Menard, S. (2011). Standards for standardizing logistic regression coefficients. Social Forces, 89, 1409-1428. (See above comment on this article)
Tabachnick, B.G., & Fidell, L.S. (2013). Using multivariate statistics (6th ed). Pearson: Upper Saddle River, NJ A super-easy effect size for evaluating the fit of a binary logistic regression using SPSS](https://i.ytimg.com/vi/LA5gWgfpHkY/mqdefault.jpg)








