The lecture emphasizes the necessity of mastering fundamental concepts in research design, regression analysis, and factor analysis before pursuing complex techniques like structural equation modeling (SEM). By drawing parallels to marathon training, it emphasizes the importance of gradually building foundational skills before progressing to advanced levels in any field.
The ease of modern software for SEM analysis does not equate to the production of good research or a comprehensive understanding of the technique. A proper understanding of how to specify models, justify the choices made, and accurate interpretation of results are crucial for effective research. The focus needs to shift from looking purely at the p value to considering the size of the effect.
Incorrect application of advanced techniques can potentially lead to deceptive results. Hence, it's preferable to apply basic techniques correctly rather than misuse the advanced ones. The approach should be learning and building on the basics first, and then progressing toward the complex, resulting in quality research.
Slides: https://osf.io/hrc3d