Description and contextualization of the subject

Introduction to Matlab, Introduction to R; Formulating hypotheses and making predictions; Probability theory; Summarizing and displaying data, graphical methods, Experimental design:; Programming for stimulus presentation (DMDX, Presentation, cogent)

Temary

The Quantitative Methods course is divided in three sections.



This course will teach the core concepts of scientific statistical analysis using the programming language R in order to put into practice the statistical concepts that are acquired. Guidelines to install R and Rstudio as IDE will be sent before the start of the course. There will be an extra class for introduction to programming (in R) during the first week.



In the first part, you will get acquainted with the types of variables and data distributions that we use and observe in research related to the cognitive neuroscience of language. Core concepts of probability and distributions will be discussed and you will learn how to describe statistical distributions graphically as well as numerically through simulation in the programming language R. In addition to descriptive statistics, we will also cover basic statistical comparisons (including t-tests), and procedures for deciding when conclusions can be drawn from comparisons. The relationship between variables (using R and ggplot2). The content of the lectures will focus on the statistical part. Demonstrations of how to perform the statistical analyses programmatically in R will be given during the lectures, but most of the hands-on practice using R will be homework.

The second section focuses on analysis of variance (ANOVA), the mostly commonly used statistical technique in much of the field. We will cover single-factor and two-factor designs, and the necessary follow-up tests for each of these. A single-factor ANOVA is essentially an expanded version of a t-test, with the expansion allowing you to run experiments that compare more than two groups. The two-factor ANOVA expands on the single-factor case, allowing you to look at the effects of two independent variables simultaneously. The follow-up tests allow you to pinpoint exactly where any observed effects in an ANOVA are coming from. We will see how each of these statistical tests is built to do what it is designed to do, and we will cover the assumptions that one makes when using these tests. Demonstrations of how to perform ANOVAs using R will also be given during the lectures and combined with textbook exercises to further understand these concepts.

The third (final) section will focus on linear regression, one of the most general and widely-used statistical models throughout science. Mimicking the structure of the second ANOVA section, we will cover univariate (single variable) and multiple regression, as well as hierarchical regression and in the final part of the course general linear model (GLM). This material should build up to a firm understanding of the GLM, which is a common tool of neuroimaging data analysis.

Bibliography

Compulsory materials

Part 1: Learning statistics with R: A tutorial for psychology students and other beginners. (Danielle Navarro): https://learningstatisticswithr.com/book/



Part 2: Design and Analysis- a researcher’s handbook (G.Keppel and T.D. Wickens). BCBL would provide the textbook to each student at the beginning of the course.



Part 3: Discovering Statistics Using IBM SPSS Statistics (Andy Field)



https://uk.sagepub.com/en-gb/eur/discovering-statistics-using-ibm-spss-statistics/book257672

Basic bibliography

Matlab for Neuroscientists by Wallisch et al. (selected chapters)

Linear Models with R by Julian Faraway (selected chapters)