Course Catalog

Familiarizes students with several types of multivariate statistics methods with respect primarily to applications and interpretations in the area of social sciences. This course will cover the data-analysis concepts and procedures used in applied and experimental psychology, economics, business and in general in social sciences. Emphasis will be given to the qualitative interpretation and manipulation of mathematical and statistical concepts, showing the students their effectiveness through concrete applications. Students will use appropriate software packages for labs and projects.

The 21st century has seen a big increase in the amount of data which is made accessible. Social media such as Facebook, online shops such as Amazon and many others, are all gathering raw data. But what can be done about this data? Data Science covers tools and methods around the extraction of knowledge from data. Such tools cover its collection, storing, processing and analysis. In this course we will learn about several of the most important tools in the above flow and will apply them to real-world examples.

We will explore how to understand environmental systems from data science angles. The course encourages students to think critically and reason quantitatively about an environmental problem rather than just focusing on getting a specific answer. This course will have hands-on practical work with real data, R or Python, and statistical or machine learning software packages.

This course is designed to extend the statistical analysis of environmental and social science data: it will highlight the building blocks of multivariate analysis from the definition of the research problem to the interpretation of the results. Both dependence methods (that is in which one or several variables can be expressed in terms of the others – for instance Multivariate Analysis of Variance or Discriminant Analysis) and interdependence methods (where all the variables are analysed simultaneously – for instance Factor & Cluster Analyses or Multidimensional Scaling) will be studied.
Significant applications will be analysed and discussed so as to develop new insights.
Projects (individual or with peers), will allow the students to apply the multivariate models, thereby enhancing the importance of work and knowledge sharing.
Statistical software package: SPSS.
Prerequisite: MA 1020