Useful Resources

Statistical Package Information

Melbourne University has licenses to a wide range of statistical packages that are available to students. Below is an outline of some of the statistical packages that may be relevant to Psychology researchers

AMOS

Amos is used to for structural equation models. It can implement Bayesian analysis. Users can choose between a graphical interface, or a non-graphic, programmatic interface. Amos is licenced with SPSS.

For more information, see SPSS below, and the SPSS-Amos website.

GENSTAT

GenStat is general statistical package for both data analysis and statistics – it is able to manage your data in addition to data display.  It is able to work with models, comparisons and various experimental designs including ANOVA, regression, and sample size calculations. There are some similarities to Excel particularly with shortcuts, however GenStat offers multiple windows for the output of results and graphics.

The university licences GenStat for computers owned by the university, employees, and postgraduate students. See the university's software support website for more.
For support and more information visit the GenStat tutorial site.

Latent GOLD

Latent GOLD 4.5 is a powerful latent class and finite mixture program.  It is a specialized program which contains separate modules for estimating three different model structures: Latent Class Cluster Models (and Latent Profile Cluster Models), Discrete Factor (DFactor) models, and Latent Class Regression models.

The program is relatively straightforward to learn, with full windows implementation (i.e. point and click).

It is useful for latent class analysis, particularly without model implementation. It is used in conjunction with a more general statistics program, such as SPSS.

There is no university license for LatentGOLD, although individual licences can be found around the department.

See the Latent GOLD website for further information.

MATLAB

MATLAB (MATrix LABoratory) is a general purpose programming language and associated integrated development environment with a particular emphasis on numerical computation.  MATLAB  is a script-based product; it includes an interactive command window (or a REPL) and built-in visualisation tools allows users to build scripts iteratively and inspect data structures dynamically.  It has been a popular product in both academia and more widely in commercial industry. In psychological sciences, MATLAB is advantageous because it can be used for stimulus presentation, data analysis, and the development and simulation of quantitative cognitive models.  It features a wide range of bundled libraries (called "toolboxes") that extend the base language, with its high performance matrix computation, to many specialised applications.
One of the more popular toolboxes for Psychological Sciences is Psychtoolbox (open source), which can be used for stimulus presentation.

The University of Melbourne offers all staff access to a site-wide license on a year-by-year basis, requiring a staff (...@unimelb.edu.au) email account. Information about this licensing programme can be found here.

A free software version of the language known as Octave is available from http://www.octave.org/. This version can be used for many major tasks requiring MATLAB, including developing and running experiments in Psychtoolbox.

MiniTab

Minitab is a powerful statistical analysis software. It can be used for learning about statistics as well as statistical research. Minitab is relatively easy to use once you know a few fundamentals.

Minitab is useful for ‘classical statistics’, particularly for beginners.

The university has a site license. More information can be found at the Minitab site, or the tutorial website.

MPlus

Mplus is a statistical modeling program that provides researchers with a flexible tool to analyse their data. The program is highly versatile, offering a wide choice of models estimators, and algorithms. It is script based (not point and click) and customisable, but there are numerous defaults, minimizing the amount of inputs required. There are many Mplus scripts published both on the Mplus website and in various journal articles. The Mplus creators also run a messageboard on their website, which is frequently monitored.

It is highly useful for SEM analysis, particularly databases with categorical outcome variables, and hierarchical data  By default it includes missing data cases. Mplus can also generate and analyse Monte Carlo simulation studies.

There are two computers in the 10th floor student computer lab with Mplus installed (located near the lockers).  Discounted student prices are available for purchase. Bi-annual Mplus courses are run within the university at a cost.

See the Mplus website for further information.

R

R is a general purpose, free software programming language with an emphasis on statistical and numerical applications. It is script based, and customizable.  It offers a set of libraries and syntactic features for handling and analysing data.  However R requires a steep learning curve.

The set of built-in libraries that are installed and immediately accessible in R cover a wide range of common statistical analyses, from basic descriptive statistics, power analyses, non-parametric tests, t-tests, and (generalised) linear models/ANOVA, to cluster analyses, time-series analysis, tree-based modelling, discriminant function analysis, exploratory factor analysis and principal components analysis, and bootstrapping. There is also extensive built-in support for model-fitting using various optimisation techniques.

A strength of R, beyond its availability, is exhibited in the wide variety of freely available open source packages which extend the core functionality, allowing more complex and contemporary analyses such as multiple imputation, structural equation modelling/confirmatory factor analysis, mixed-effects linear and non-linear analysis, self-organising mapping, complex Bayesian modelling, and more.

R can be downloaded for free at http://www.r-project.org/ and all toolboxes can be installed from within R for free

SAS: Statistical Analysis Software

SAS is a broad statistical and analytical program. It allows you to manage data, data mine, graphics and statistical analysis.  The user interface is designed for researchers and has several components that deal with analysis and data reporting.  This program is well established and has a lot of online support, as well as consultants at the university. For beginners the user interface is not the easiest to work with but SAS does have a large range of capabilities. It is a popular alternative to SPSS, and is considered better for handling large databases.

The university holds a licence for this program which you can access. If you are downloading this for your personal computer there is an extra form to download and email to the service desk (details on the website).  This program is free through the university for all operating systems on computers owned by the University or you personal computer.

For more information and support, see: SAS website, tutorials, and eLearning courses (free access) for which the university service desk (ITS) can give you an activation key.

SPSS

SPSS is a powerful statistical and data management software package. It is a point and click interface, with script options. SPSS is widely used, and taught to in the undergraduate psychology courses. The program is useful for data entry, data cleaning, analysis, and basic graphics (but not APA style!).

The university has a licence for staff, and is available on all computers in the department computer labs.

Support is available at the SPSS site, and with various user guides (e.g., Pallant, 2011).

STATA

Stata is a complete, integrated statistical package that provides tools for data analysis, data management, and graphics.

Stata can be used for standard methods and advanced techniques, such as survival models with frailty, dynamic panel data (DPD) regressions, generalized estimating equations (GEE), multilevel mixed models, models with sample selection, multiple imputation, ARCH, and estimation with complex survey samples. The generate publication-quality graphs, including regression fit graphs, time-series graphs, survival plots, and contour plots.

Stata may be useful for anyone looking to carry out more advanced modelling. There is plenty of user support on Stata website and quarterly journal.

There is no university wide license, but it is available on two computers in the 10th floor computer lab (near the lockers). Annual or perpetual licenses are available. For more information, see the stata website.

For more statistical package information, the University of Texas has a good overview and introduction.

Other useful programs:

Referencing Packages

  • Papers: like iTunes for journal articles
  • Mendeley: like Papers, but free up to 1GB
  • Zotero: works with your web browser

Notes, Planning, & Writing

  • OneNote: planner and note taking software with Microsoft office
  • Scrivener: allows you to concentrate on composing and structuring long and difficult documents

We also have a database covering a broad range of statisical information that you can request access to, and contribute to: Get to GRIPS with stats!