Answering Social Research Questions with Statistical Models

SOST20131 Answering Social Research Questions with Statistical Models 20 Semester 1 Undergraduate Level 2 Tatjana Kecojevic Unit coordinator Social Statistics 1 FHEQ level (Framework for Higher Education Qualifications) ' Last part of a Bachelors ' European Credit Transfer & Accumulation System Rating : 10.0 This course will equip students with the knowledge and skills to answer theoretically-driven research questions involving causality in the social sciences. Specifically, student will use modern causal theory to specify and fit linear and binary logistic regression models using the R software platform. Many if not most social research questions are concerned with questions of causality, e.g. what are the causes of good and bad things in society? Only if we understand the causes can we hope to modify the good/bad effects.Much if not most of social research is observational, i.e. correlational; we can observe and measure things, ask people questions etc., but it’s not easy to run experiments. This means that often we only have correlational data with which to evaluate and test our causal research questions.Taken together, the two conditions above present a problem, because as we all know, correlation does not equal causation.Recently developed theories of causation challenge these limitations. We will use the theory of Directed Acyclic Graphs (DAGs) to understand how causality translates into correlations among variables. We will use this knowledge to help us specify statistical models that may allow us evaluate our causal theories.The statistical models we will use are varieties of Generalized Linear Models (GLMs), specifically Linear Regression and Logistic Regression. We will use the R software package to estimate these models using data. We will evaluate some existing social research studies using our knowledge of DAGs and GLMs.  The unit aims to:(i). Give students an introduction to the causal theory of Directed Acyclic Graphs (DAGs) (ii) Show students how DAGs can be seen as representations of theories in social science and other domains. (iii) Show how DAGs and causal theory can be used to guide the specification of quantitative statistical models, specifically linear and binary logistic regression models. (iv) Give students an introduction in how to use the R software package to specify and fit linear and binary logistic regression models to real-world social data, based upon prior causal analysis of DAGs. (v) Show students how to interpret the results of the regression models, and make inferences from them to the wider population. Student should/will be able to: Knowledge and Understanding:Understand the causal theory of DAGs;Distinguish potentially causal relationships from spurious ones.Understand the statistical formulation of regression models.Understand the basis of inference from samples to populations Intellectual skills:Distinguish between levels of measurement of variables, and use models and variables appropriately.Appreciate different types of functional relationship among variables, and use this to specify models appropriately.Evaluate correlations and consider to what extent they may represent causal as opposed to spurious, non-causal processes. Practical skills:Use the R software package to fit linear and binary logistic regression models. Transferable skills and personal qualities:Use the R software package.Critically evaluate claims of causal effects, e.g. those presented in the media and in research papers. Each week except the first, students will be given homework activities (something to read, and/or watch, and/or do).During the following 2-hour session we will review and explore those activities to check our understanding. It is imperative that students carry out the homework activities before the session, as the sessions will not be purely lectures as such; they will be a chance for us to ask each other questions to check our understanding of the material.The sessions will feature presentations/lecturettes, demonstrations using R software, causal analysis and data analysis tasks. Students will need to register with the UK data service, to gain access to real-world datasets that will be used extensively throughout the course. 0 Other 100% Formative assignment (0%): 500 words Written assessment (40%): 1200 words Written assessment (60%): Up to 2000 words All Social Statistics courses include both formative feedback - which lets you know how you’re getting on and what you could do to improve - and summative feedback - which gives you a mark for your assessed work.<div> </div> Y Recommended readingAgresti, A. (2018). Statistical methods for the social sciences, Global Edition. Pearson/ Prentice Hall.McShane, B. B., Gal, D., Gelman, A., Robert, C., & Tackett, J. L. (2019). Abandon Statistical Significance. American Statistician, 73(sup1), 235–245. <a href="https://doi.org/10.1080/00031305.2018.1527253">https://doi.org/10.1080/00031305.2018.1527253</a>Rohrer, J. M. (2018). Thinking Clearly About Correlations and Causation: Graphical Causal Models for Observational Data. Advances in Methods and Practices in Psychological Science, 1(1), 27-42. <a href="https://doi.org/10.1177/2515245917745629">https://doi.org/10.1177/2515245917745629</a> Verzani, J. (2001). SimpleR: Using R for Introductory Statistics.<a href="https://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf">https://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf</a>  On-line ResourcesAbsolute basic introduction to R: <a href="http://stats.idre.ucla.edu/stat/data/intro_r/intro_r_interactive.html#(1)">http://stats.idre.ucla.edu/stat/data/intro_r/intro_r_interactive.html#(1)</a>Analyze Survey Data for Free: <a href="http://asdfree.com/">http://asdfree.com/</a> Lectures 20 Practical classes & workshops 10 0 170