Forthcoming methods@manchester workshops

January 2013

An Introduction to Multimodal Methods for Researching Digital Data

methods@manchester and MODE

Thursday 17 January 2013, 10am – 5.30pm

Room 1.69/1.70, Humanities Bridgeford Street building, University of Manchester

More details

Digital data can provide a rich but also challenging empirical basis for social science research. Multimodality offers theoretical and methodological frameworks for analyzing these data. Focusing on the social interaction captured and/or shaped by digital technologies multimodality attends to all forms of representation and communication, including language, body posture, gaze, and moving image.

 

Generalized Linear Models: a basic statistics course – Graeme Hutcheson

28, 29 & 30 January

Basement Computer Lab, Humanities Bridgeford Street building, University of Manchester

This workshop is now fully booked. Please email methods@manchester.ac.uk to be placed on the waiting list.

More details

This 3-day course introduces Generalized Linear Models (GLMs) and shows how they can be applied to numeric and categorical data. The course makes extensive use of the open-source statistical environment “R” that is accessed through the Rcommander (Rcmdr) graphical interface (this software can be installed on any platform – Unix, linux, MacOS and Windows). A major advantage of GLMs is that they are based on a consistent underlying theory that can be easily understood and applied. This course uses simple terminology to define and input models and does not require participants to have a detailed mathematical knowledge.

 

February 2013

R and the R-commander (Rcmdr): tools for statistical analysis
Wednesday 6 February, 10 – 12am
Room LG9, Coupland 3 Building, University of Manchester
(Building number 47 on the campus map)

Booking form

This workshop introduces the R statistical analysis software and the R-commander graphical user interface (Rcmdr). These packages provide a simple, free and powerful solution for analyzing data. A demonstration of Rcmdr’s statistical and graphical capabilities is provided along with the possibilities provided by the many add-on libraries that are available for R and the Rcmdr.

This is an introductory session aimed at new users of R or the Rcmdr. Participants are not required to have any previous statistical or computing experience.

Software
http://www.r-project.org/
http://cran.r-project.org/web/packages/Rcmdr/index.html

Reference
Open-source statistical software: R and the R Commander. The Journal of Modelling in management, Volume 5, issue 3, 2010.

Data coding and management
Wednesday 13 February, 10 – 12am
Room LG9, Coupland 3 Building, University of Manchester (Building number 47 on the campus map)

Booking form

Data coding and management are essential skills for analysts, but are rarely taught explicitly in statistical courses. This session discusses the importance of measurement scaling and also provides a method for coding different types of data using readily available software and a standard format. This session also deals with issues of data re-coding and provides techniques for sharing data saved in SPSS format.

Participants are not required to have any previous statistical or computing experience.

Reference
Data coding, management and manipulation. The Journal of Modelling in management, Volume 6, issue 1, 2011.

Measures of Deprivation and Area Type

Monday 18 February, 9.30am – 4.30pm

Basement Computer Lab, Humanities Bridgeford Street, University of Manchester

More details

This workshop is now fully booked. Please email methods@manchester.ac.uk to be placed on the waiting list.

In this course we explain the basis of deprivation measures and consider their potential drawbacks in some detail. Alternative measures of small area deprivation will be considered and we will discuss methods for the construction of alternative measures from first principles.

Generalized linear models: a system for data analysis and test selection

Wednesday 20 February, 10 – 12am

Room LG9, Coupland 3 Building, University of Manchester (Building number 47 on the campus map)

Booking form

This session demonstrates how a regression-based analytical system can replace the more traditional cook-book approach that still commonly forms the basis of many introductory statistical courses in the social sciences. Analyses based on generalized linear models (GLMs) offer significant advantages in terms of their flexibility and power and also offer a much simpler system for teaching; one that is based on a solid theoretical base. The huge selection of hypothesis tests that are available may be represented using, essentially, a single model, which can be applied to different types of data and study design.

A system of data analysis based on GLMs is presented along with a discussion of how a ‘simple’ linear model can be applied to continuous and categorical data. Participants are not required to have any previous statistical or computing experience.

Reference
Test selection in the 21st century. The Journal of Modelling in management, Volume 7, issue 3, 2012.

 

Categorical explanatory variables: Contrast coding, dummy codes and reference categories

Wednesday 27 February, 10 – 12am

Room LG9, Coupland 3 Building, University of Manchester (Building number 47 on the campus map)

Booking form

 

Categorical explanatory variables can be easily included in statistical models using contrast coding techniques. Although there are a number of techniques available, many users only apply the default option provided by the statistical software.

This is a shame, as the different coding schemes are very useful and can, for example, provide different group comparisons, test for ordered effects, illustrate differences between individual categories and identify non-linear trends. This session shows how to use and interpret a number of contrast coding techniques and also manipulate the reference category to provide specific comparisons.

Although this session is illutrated using R and the Rcmdr, the techniques described can be easily applied using other packages.

Reference
Categorical explanatory variables. The Journal of Modelling in management, Volume 6, issue 2, 2011.

 

A hands-on introduction to Agent-Based Modelling for Social Scientists:
Exploring Complex and Dynamic Social Processes

Thursday 28 February – Friday 1 March 2013

Basement Computer Lab, Humanities Bridgeford Street, University of Manchester

More details and booking instructions

The Institute for Social Change, the Cathie Marsh Centre for Census and Survey Research and the Centre for Policy Modelling will be running a 2-day introduction to agent-based simulation workshop, funded by methods@manchester. The course is free to attend and will be suitable for people without any experience of computer programming but is particularly aimed at postgraduate social researchers, e.g. PhD students, post-docs, lecturers.

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