Home
PhD Research School in Linguistics and Philology

Statistics for linguists

Workshop on the use of statistical tools, 25 November 2010.

Main content

The workshop is aimed at students and PhD candidates with some basic grounding in statistics, e.g. as covered in the introductory courses given at the UiB. Prerequisites: some familiarity with concepts such as the t-test, the chi-square test, data levels, the normal distribution, and standard deviation.

The workshop will provide an introduction to statistical tools going beyond these basic concepts, aimed especially at corpus and survey / questionnaire data. The basic statistical methods (such as the t-test or the chi-square test) were developed for experiments and they are not always suitable for other areas of study, corpus linguistics being a particularly pertinent example.

The workshop will address three broad issues:

  1. How can statistical methods be used to go beyond the classical “significant / not-significant” question?
  2. How can we measure the effect of several variables at the same time and assess their relative importance?
  3. Visualization techniques as a method for analysis

Specifically, the workshop will give an introduction to the following (note that not all topics will be covered extensively):

  • Data collection
  • Data files
  • Software: R
  • Specific statistical methods: ordinary and logistic regression; visualization methods such as principal component analysis, multidimensional scaling, correspondence analysis .

Place: Room 265, HF-building

Time: Thursday 25 November, 0900–1600

Lecturer: Gard Buen Jenset (HiB)

Organised by the PhD Research school in linguistics and philology at the University of Bergen. As part of the course registration, participants are asked to submit a short (maximum 1000 words) description of their project, covering the following:

  1. Type(s) of data (i.e. data level(s))
  2. The amount of data
  3. Hypotheses (tentative or substantial)
  4. Overall aims of the project
  5. Any specific problems or challenges they foresee or struggle with

Programme

Thursday 25 November

09.00–12.00: Introduction and practical instruction in the computer lab

12.15–13.00: Lunch

13.15–16.00: Discussion of individual projects

Please sign up by sending an e-mail and the text to Martin Paulsen no later than 15 November, but be aware that we can only take a limited number of participants, and the first to sign up will be given priority.