Two forthcoming methods@manchester Methods workshops

Friday 27 April 2012

Logit models: analysing categorical data

Presenter: Graeme Hutcheson

2 – 4pm, Hanson Room, Humanities Bridgeford Street

This session describes the theory behind logit models – a set of models that enable categorical data to be efficiently analysed. The session will look in detail at the binary logit model and also logit models for ordered and unordered categorical data (proportional-odds and multinomial logit models). Although this seminar makes use of R and the R-commander, the techniques described can be applied using any statistical software. It is recommended that participants are familiar with OLS regression. Please book a place at www.methods.manchester.ac.uk/events/workshops/

Monday 30 April 2012

Model selection: towards a restricted-set, multi-model procedure

Presenter: Graeme Hutcheson

2 – 4pm, Room 3.214, University Place

This session describes some of the problems commonly encountered by researchers dealing with survey and questionnaire data when selecting a model. A detailed explanation of multicollinearity is provided along with a critique of methods commonly used for model selection (eg., step-wise and all-subset selection). An alternative method for model selection is proposed that attempts to address some of the major issues. Although this seminar makes use of R and the R-commander, the techniques described can be applied using any statistical software. It is recommended that participants have some experience of regression techniques. Please book a place at www.methods.manchester.ac.uk/events/workshops/

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s