Students are given an introduction to hypothesis testing in Linguistics and Cognitive Science. The course gives a basis for understanding statistical hypothesis formulation, research design, and analysis of the results with statistical tests. The course will provide an understanding of the normal distribution, z-values, and parametric and non-parametric tests. Students will become familiar with concepts such as variance, standard deviation, central tendency measures, statistical significance, relevance, and type I- and II-errors.

Students will be taught to read and evaluate scientific literature with statistical argumentation. The course will also give a basis for evaluating statistical significance and relevance. Teaching is designed so that students will see how statistics is relevant for linguistics and cognitive science through relevant examples. They will also be able to choose a valid statistical test and carry out an analysis with the help of a suitable statistics program, such as R.

Knowledge

The candidate has knowledge about...

• statistical techniques and models
• good practice for statistical analysis of experimental data in linguistics and cognitive science
• theories and conventions for statistical analysis
• various ways to communicate statistical results as well as possible
• analytical models

Skills

The candidate can...

• use statistics software, for instance, R
• use theories and conventions to create analytic models
• report statistical results correctly in text and graphs.

General competence

The candidate can...

• work independently with empirical materials
• choose correct tests for fairly straightforward experimental designs
• have a critical eye for one's own and others' analyses
• put statistical analysis in a relevant scientific context
• prepare experimental design which is suitable for statistical analysis.

Lectures, independent study and exercises with statistics software on a computer.

Teaching includes lectures with examples from language related data. Demonstrations will be given on formatting data and automatic processing in statistics software. Active participation is a requirement for this course. Deeper investigations and case studies are done through exercises with a computer, in which students get a practical understanding of statistical processing. It is not given that all aspects of the course goals will be covered by the lectures and seminars; a certain level of self-study is expected through reading the literature and active participation in exercises.