SUBSTANTIVE METHODOLOGICAL SYNERGY
Substantive-methodological synergies (a term coined by Marsh & Hau, 2007) are joint ventures in which new methodological developments are applied to substantively important issues, or in which new methodological approaches are devised to provide better responses to substantive research questions.
We live in exciting times of fast-paced innovation in quantitative methods. This creates new research opportunities: methodological innovations enable researchers to address previously inaccessible research problems, to revisit classic unresolved issues, and to address new research questions. However, this also creates concerns for substantive researchers who fail to keep pace with new methodological developments, as well as for methodological experts who sometimes lose sight of the needs of applied researchers. Substantive-methodological synergies are specifically devised to address these issues by demonstrating the practical advantages of statistical innovations to substantive researchers, and to by presenting those innovations in a clear and replicable way to applied researchers who might lack the formal mathematical training necessary to be able to dig into pure statistical publications. A core objective of the Substantive-Methodological Synergy Research Laboratory goes beyond the simple application and illustration of new methodological tools. Rather, our goal is to actively create such synergistic joint ventures between teams of substantive and methodological experts. With this in mind, lab members always remain open to collaborative opportunities with substantive experts seeking better, clearer, or simply more precise answers to critical research questions, as well as with statistical experts seeking ideas aiming to provide practically useful responses to applied problems. A secondary objective of our lab is to promote and encourage secondary data analysis of the multiple incredibly rich data sets that remain underused in many research laboratories. Researchers are currently assessed by their ability to secure research funding, and research funding is usually conditioned on the collection of new data. This reality leads many to focus their energies on collecting data, analyzing it to respond to the key (yet limited) objectives of their proposal, and then to turn their attention to new funding opportunities. As new and evolving methodologies become available, it is often the occasion to capitalize anew on the wide array of incredibly rich data sets that unfortunately tend to become neglected as researchers move on to new projects. This ability to capitalize on already available information and to examine it with new and evolving methods is the essence of substantive methodological synergies, and a possibly “greener” approach to research. Human beings have started to realize that the capitalistic consumer-oriented approach that predominate modern life is slowly but surely destroying our own habitat. It seems doubly important for researchers - who have always been at the forefront of climate change warnings - to come to adopt their own recommendations to maximally capitalize on already existing resources rather than to maintain this planned obsolescence approach to research data. At the core of our focus on substantive methodological synergies is a desire to train the next generation of quantitatively-oriented researchers. Importantly, we do not expect students and postdoctoral fellows seeking to join us to be experts in statistics, or even to have a formal mathematical background (i.e., understanding calculus and matrix algebra, although useful for many purposes is not a requirement for the type of research that we do and indeed, many lab members do not have this type of expertise). The only requirement is a willingness to learn, and not being scared of statistics. After all, the goal our lab is to provide new members with mentoring from a a team of researchers that are passionate about statistics, theory, research, and life in general, and who enjoy introducing statistics in a "pain-free" manner to a wider audience. |
Methodological INTERESTS
There is a series of key statistical models that we particularly enjoy using, illustrating, and improving:
(a) Psychometric models (e.g., ESEM, Bifactor ESEM),
(b) Person-centered analyses (e.g., latent profiles analyses, latent transition analyses, mixture regression),
(c) Longitudinal analyses (e.g., latent growth models, latent change models, autoregressive models, autoregressive latent trajectories),
(d) Their combination (e.g., growth mixture models).
(a) Psychometric models (e.g., ESEM, Bifactor ESEM),
(b) Person-centered analyses (e.g., latent profiles analyses, latent transition analyses, mixture regression),
(c) Longitudinal analyses (e.g., latent growth models, latent change models, autoregressive models, autoregressive latent trajectories),
(d) Their combination (e.g., growth mixture models).