Introduction to multivariate statistics for usage-based linguistics
Stage de formation à l’analyse statistique avec R
Organisatrice du stage : Ilse Depraetere en collaboration avec le service formation de la Direction des Ressources Humaines de l’Université Lille 3.
Formateur : Dylan Glynn, Université Paris 8.
Deux journées de formation sont prévues, les
The use of statistics in empirical science is universal and fundamental. Linguistics is possibly the last field to adopt these tools, some 30-40 years behind our neighbouring fields of sociology and psychology. That has now changed and, for many within the field, it has already become a necessity. This trend is set to continue, not because it is fashionable, but because, ultimately, if linguistics is to be an empirical science, it is necessary.
Statistics should not be confused with, on the on hand, counting the number of occurrences of a given phenomenon, or on the other hand, blind high-tech studies devoid of actual language analysis. Statistics is simply a tool that helps any empirical scientist in three ways:
– identifies complex structure
– calculates probability that the structure represents ‘reality’
– determines predictive accuracy of that structure.
Regardless of the research question, regardless of the theory, these three things are incredibly useful to any empirical research.
This entirely practical course will introduce:
– statistical techniques and the tools to do them
– for the study of langauge from a Cognitive and / or Functional linguistic perspective
– using natural contextualised langauge (corpora).
voir le détail de la formation