Advanced Methods: Structural Equation Modeling
The course is designed to provide scholars with a basic understanding of structural equation modeling (SEM). Special attention is given to the translation of theoretical expectations into SEM, the interpretation of results in SEM analyses and the general use and misuse of SEM in the social sciences. While the course is predominantly designed to give you the knowledge of SEM. Applications will include path models, factor analyses and structural equation models, and, if time allows, multilevel SEM. The goal of the course is to offer a basic introduction and the foundation for students to start using and critically assessing SEM and also have the ability to independently discover and master advanced SEM statistical topics. Upon completion the students will have a basic conceptual understanding of SEM and its statistical foundations. Students will be able to critically assess the appropriateness of such techniques in their own and other people’s research and conduct SEModeling themselves to the highest academic standards.
The course covers both inductive and deductive form of models. It is designed to advance scholar’s ability to comprehend and critique articles using structural equation models and to apply these approaches in research. Throughout the course students’ use computer software (Mplus and R) for conducting statistical SEM analysis. Understanding and application of the aforementioned techniques are applied autonomously in a creative way. The course advances the scholar’s ability to contribute to the development of political science in their home countries by applying the learned methods to analysis of their home countries (or any other place for that matter) by acquiring cutting-edge approaches and methods of SEModeling. Since SEM data structures are not localized to political science, the scholars emerged in the course will acquire the ability to expand both their own and their field’s horizons by extending the their inferential understanding beyond the boundaries of the discipline. Statistical inference is universal in all of empirical sciences and therefore this course will certainly expand the scholar’s view into an interdisciplinary vision of the world. The course will expand the scholar’s ability to design, implement and write up a good quality research in a thorough, rigorous and consistent manner. Methodologies, such as SEM covered by the course advance the scholar’s ability even going beyond writing research proposals and formulating research questions. The course will help students conduct actual empirical research. The final paper (article) for the course is designed to advance the student in mastery of academic writing style and argumentation in English: ability to use English grammar, vocabulary and style appropriate for written academic products and ability to construct academic arguments. And in this final paper (and also in other components of the course) we will use graphs and other visual communication of results. In writing and also sharing the findings of the final paper students will advance their ability to synthesize information, determine a focus point, discern the main line of argumentation and orally present these. Their ability to generate logical, plausible and persuasive arguments, connect, compare and contrast, ability to identify logical relations and mistakes of arguments, ability to make appropriate analytical distinctions, (and etc.) will be advanced through the course. Through this both students’ higher order thinking (such as to seeing patterns and generalizing from facts) reasoned judgment will grow. If this sounds all weird, that is because it is. We have some global learning outcomes we have to sync our classes with and I wanted to adhere to these principles. It is all true, but stated explicitly it does sound weird. Sorry about that.
Participation (engagement in class): 10% Preparedness (to make sure you read for class): 15% Presentation of Design and Analysis (date TBA): 26% Publishable Article (not required for auditors) due April 9, 11:55pm: 49% On a 100 point scale, the grading would be as follows: A 94.00 – 100 A- 87.00 – 93.99 B+ 80.00 – 86.99 B 73.00 – 79.99 B- 66.00 – 72.99 C+ 59.00 – 65.99 It is an absolute requirement that you come to class, you stay engaged in class and that you come to class prepared. If you will miss a class or come late for any reason, make sure I know about it before the class. (Even if it is a few minutes before class.) These two components make up a quarter of your grade. Additionally everyone will have to present their individual project (design and analysis) at the meeting outside of class in the presentations evening outside of class. Final paper will have to include original research using one of the methodologies covered in class. It has to be of publishable quality adhering to the submission guidelines of one peer reviewed ISI ranked journal selected by the student. The sophistication of the methodology used needs to be true to a PhD level statistical methods class. Topic needs not be political science but has to be publishable in a peer reviewed ISI ranked scientific journal. The manuscript will have to be appropriately anonymized if this is required by the journal only identifying you via the file name. If journal asks for multiple files, ignore this request and tag additional figures and tables on to the end of the manuscript. Journals generally ask for papers no longer than 4000-8000 words. For specific maximum paper length, please check the journal’s guidelines.