Workshop on random effect structures in the modelling of language data

Language@Leeds is organising a two-day workshop on random effects in mixed-effect models of linguistic data, in partnership with the White Rose Doctoral Training Centre for linguistics.

The workshop will be taught by Harald Baayen (Linguistics, University of Tuebingen) and Arief Gusnanto (Statistics, University of Leeds).

The topics covered will include the following:

  • conceptual definition of random effects and their role in mixed models
  • practical guidance to estimate the optimal random effect structure in linear mixed models
  • the use of random effects in the case of interaction between variables
  • the use of random effects in the context of non-linearity of (a) predictor(s)
  • considerations regarding best practice when reporting mixed-effect models
  • considerations regarding best practice in the use of mixed-effect models for exploratory vs confirmatory analysis

The workshop will include demonstrations, hands-on exercises and discussions.

Participants are invited to submit general questions in advance of the workshop, by emailing them to statisticsworkshop at gmail dot_com by the 29th of May at the latest. We regret that specific questions associated with a particular data set cannot be considered, for practical reasons.

Some level of familiarity with mixed-effect modelling will be expected, but less experienced participants wishing to gain a conceptual understanding are also welcome to attend.

To reserve a place, please get a ticket through the Eventbrite website. You will need to get a ticket for each day (which will require you to go back to the main page). You will then be contacted by email for a modest, partial contribution towards the catering costs (refreshments and lunch on both days). The workshop is otherwise funded by L@L and the WRDTC.