software

Learn Python The Hard Way

Zed Shaw takes you through the process of learning Python, via an approach which applies to any programming language. Video/ pdf edition also available, but the html version is fine. Of course, it isn’t really hard, but his method puts the emphasis on you finding things out for yourself, rather than being fooled into thinking you have learned things when you haven’t.

Udacity

Udacity provide a range of online learning tutorials. In the computer science section there is a course which uses Python, and focuses on a structured approach to solving problems with programmes and structuring the project properly. The example used is the building of a search engine.

Visualising data

A website dedicated to collecting tools and resources for data visualisation and related matters. The company itself is a design consultancy which also provides training.

G*Power 3

An altogether more advanced tool for statistical power analysis, helping with required sample sizes etc. However, it is quite hard to understand, and a simpler version might be easier. Requires installation.

Learn Python!

Python is a relatively simple and widely used programming language. If you haven’t learned any yet, it is recommended, partly because of its simple syntax, and partly because of the range of applications it is used for. Statistically speaking is it valuable because it is the scripting language of choice for SPSS. The in-house syntax script is limited in some ways, and true control of SPSS comes through the use of Python via the Programmability Extension, or Python Integration Plug-in, for which there is much documentation.

Even if advanced use of SPSS – and a good example is the ability to run large numbers of cluster analysis in one go – is not relevant, Python is a good place to start for learning programming in general. A good place to start learning is Codecademy, where you can go through online lessons, without even needing to have Python installed on your machine. See also Udacity.

Python is an open source programme so that Python itself and all additional developed programmes are available free of charge, including the packages required for its use in SPSS. And, for instance, Raynald Lavesque’s spsstools site can be used for additional resources, although with nowhere near as much information as he has there for syntax, macros and scripts.

R

R is a powerful open source stats programme which can (with effort) produce very nice graphics, often used in academic papers. Command-line interface, which takes a while to get used to, but it seems to be very well regarded and is widely used by people working with statistics.

Might be useful for times where we don’t have a particular kind of analysis available in SPSS – because R is open-source, there are huge numbers of available functions.

On the other hand, there is a programmability extension for SPSS which accesses a number of (free) additional functions which are written in R (also the name of the scripting language used in R).

See also UCLA stats portal for tutorials.

Ozgrid forum for Excel etc.

Ozgrid discussion Excel forum – can get a wide range of questions answered pretty quickly about Excel, particularly when using VBA macro code in Excel. There are a lot of questions already answered.

It is also possible to register on the site and post your own questions (having checked that a similar question hasn’t already been posted). When all goes well, questions may be answered very quickly, sometimes within the hour.

The same site also covers other software, such as Word.

UCLA Statistics Portal

Online resources from UCLA on various elements of statistics, in particular training material for SPSS and R.