Statistics

Statistics programs and packages

R
Free software for Win/Mac/Linux using the language S, whose website includes the following introduction: ‘R is a language and environment for statistical computing and graphics. […] R provides a wide variety of statistical (linear and nonlinear modelling, classical statistical tests, time-series analysis, classification, clustering, …) and graphical techniques, and is highly extensible. […] One of R’s strengths is the ease with which well-designed publication-quality plots can be produced, including mathematical symbols and formulae where needed. Great care has been taken over the defaults for the minor design choices in graphics, but the user retains full control.’ R is demanding but is increasingly the preferred environment for serious stats-in-linguistics. A moderately recent version of R is available on staff and student desktops.

SPSS
A suite of Windows (and Mac) programs, available on the network. There is some useful-looking documentation at Newcastle, including one entitled ‘Which statistics program?’, and at UCLA.

Goldvarb
A free program for variable rule analysis, based on the early Varbrul program.

Rbrul
Rbrul is “a program for analyzing linguistic data using – but without quite having to use – R … Rbrul’s first goal is to do everything Goldvarb can do, but do it better (and faster).”

Goldsearch
A Unix or DOS database program for searching, collating and sorting linguistic and social data for input to Goldvarb/Varbrul: info needed

DataDesk
Highly recommended by Maciej Baranowski for multivariate analysis. It has a graphics interface and can export data to Excel for the production of graphs and tables. The department may acquire a licence.

Corpus Frequency Wizard
Convenient website offering ‘the most commonly needed statistical procedures in a clear and uncluttered interface’ (Hoffmann et al. 2008: 80, Corpus linguistics with BNCweb), though only two tests seem to be available at present.

quantpsy.org (Kristopher J. Preacher, Vanderbilt University)
Useful site offering several statistical tests (such as Chi-square, Fisher’s exact test) on-line, with succinct explanations.

VassarStats (Richard Lowry, Vassar College)
Another useful-looking ‘web site for statistical computation’ with a good range of tests, utilities and information.

Sample size calculator (Raosoft)
How much of an excessively large dataset you need to sample for a given confidence level and margin of error.

Page last updated 10th January 2017.