Blog Archives

JASP might finally be the SPSS replacement we’ve been waiting for

i-don-t-always-make-shity-tables-and-figures-from-my-data-but-whI use SPSS for statistical analysis, but I don’t like it. Every time I do, I feel like the victim in some kind of emotionally abusive relationship. The interface is deeply horrid, the outputs are butt-ugly, and it runs like a three-legged overweight sloth with a heavy suitcase. It’s an absolute bloated dog of an application, and IBM clearly don’t give a crap about it, other than making some cosmetic updates every now and again. Plus the licensing system is bat-shit insane, and very expensive.

So, why do I keep using it? Because a) It’s what I learned as an undergraduate/PhD student and I know it backwards, and b) there are few viable alternatives. Yes, I know I should learn R, but I actually don’t use ‘normal’ stats that often (I spend most of my analysis time in brain-imaging packages these days) and every time I learn how to do something in R, I try doing it again a month later, have forgotten it, and have to learn it all over again. At some point I hope to become an R master, but for occasional use, I find the learning curve to be too steep. I would also hesitate to try and use R to teach students; I find it generally pretty user-hostile.

So, for ages now, I’ve been looking for a good, user-friendly, open-source alternative to SPSS. One that isn’t a bloated monster, but has enough features to enable basic analyses. I was quite hopeful about PSPP for a while (free software that tries to replicate SPSS as closely as possible). However it lacks some relatively basic ANOVA features, and since one of the things I dislike about SPSS is the interface, trying to replicate it seems like a bit of a mistake. SOFA statistics was a contender too, and it does have a beautiful interface and produce very nice-looking results, but it only does one-way ANOVAS, so… fail.

So, I gave up and crawled miserably back to SPSS. However, fresh hope now burns within my chest, as the other day I came across JASP (which the developers insist, definitely does not stand for ‘Just Another Statistics Program’). The aim of JASP is to be ‘a low fat alternative to SPSS, a delicious alternative to R.’ Nice. It seems to cover all the analysis essentials (t-test, ANOVA, regression, correlation) plus also has some fancier Bayesian alternatives and a basic Structural Equation Modelling option. The interface is great, and the results tables update in real-time as you change the options in your analysis! Very nice. This demo video gives a good overview of the features and workflow:

It’s clearly very much a work-in-progress. One issue is that it doesn’t have any in-built tools for data manipulation. It will read .csv text files, but they basically have to be in a totally ready-to-analyse format, which means general data-cleaning/munging procedures have to be done in Excel/Matlab/R/whatever. Another major downside is that there appears to be no facility for saving or scripting analysis pipelines. Hopefully though, development will continue and other features will gradually appear… I’ll be keeping a close eye on it!

Open-Source software for psychology and neuroscience

microsoft-communismResearchers typically use a lot of different pieces of software in the course of their work; it’s part of what makes the job so varied. Separate packages might be used for creating experimental stimuli, programming an experiment, logging data, statistical analysis, and preparing work for publication or conferences. Until fairly recently there was little option but to use commercial software in at least some of these roles. For example, SPSS is the de facto analysis tool in many departments for statistics, and the viable alternatives were also commercial – there was little choice but to fork over the money. Fortunately, there are now pretty viable alternatives for cash-strapped departments and individual researchers. There’s a lot of politics around the open-source movement, but for most people the important aspect is that the software is provided for free, and (generally) it’s cross-platform (or can be compiled to be so). All that’s required is to throw off the shackles of the evil capitalist oppressors, or something. 

So, there’s a lot of software listed on my Links page but I thought I’d pick out my favourite bits of open-source software, that are most useful for researchers and students in psychology.

First up – general office-type software; there are a couple of good options here. The Open Office suite has been around for 20 years, and contains all the usual tools (word processor, presentation-maker, spreadsheet tool, and more). It’s a solid, well-designed system that can pretty seamlessly read and write the Microsoft Office XML-based (.docx, .pptx) file formats. The other option is Libre Office, which has the same roots as Open Office, and similar features. Plans are apparently underway to port Libre Office to iOS and Android – nice. The other free popular options for presentations is, of course, Prezi.

There are lots of options for graphics programs, however the two best in terms of features are without a doubt GIMP (designed to be a free alternative to Adobe Photoshop) and Inkscape (vector graphics editor – good replacement for Adobe Illustrator). There’s a bit of a steep learning curve for these, but that’s true of their commercial counterparts too.

Programming experiments – if you’re still using a paid system like E-Prime or Presentation, you should consider switching to PsychoPy – it’s user-friendly, genuinely cross-platform, and absolutely free. I briefly reviewed it before, here.  Another excellent option is Open Sesame.

For statistical analysis there are a couple of options. Firstly, if you’re a SPSS-user and pretty comfortable with it (but fed up of the constant hassles of the licensing system), you should check out PSPP; a free stats program designed to look and feel like SPSS, and replicate many of the functions. You can even use your SPSS syntax – awesome. The only serious issue is that it doesn’t contain the SPSS options for complex GLM models (repeated measures ANOVA, etc.). Hopefully these will be added at some future point. The other popular option is the R language for statistical computing. R is really gaining traction at the moment. The command-line interface is a bit of a hurdle for beginners, but that can be mitigated somewhat by IDEs like R-Commander or RStudio.

For neuroscience there’s the NeuroDebian project – not just a software package, but an entire operating system, bundled with a comprehensive suite of neuroscience tools, including FSL, AFNI and PyMVPA, plus lots of others. There really are too many bits of open-source neuro-software to list here, but a good place to find some is NITRC.org.

So, there you are people; go open-source. You have nothing to lose but your over-priced software subscriptions.

TTFN.

BPS Hackathon – 21st June; LaTeX, R, Python goodness

Very exciting news here: I’ve just been invited to the first British Psychological Society (Maths, Statistics and Computing Section) Psychology open textbook hackathon!

Inspired by this event (where people got together and wrote an open-source maths textbook in a weekend) the day aims to raise awareness and skills, as well as perhaps produce some usable output.

The organisers are Thom Baguley of Nottingham Trent University (and the Serious Stats blog and book) and Sol Nte of Manchester University. They’ve very kindly invited me as a guest, so I’ll be hanging out and learning some new tricks myself, I’m sure.

Here’s the flyer for the event, with sign-up details etc. It’s free, but strictly limited to 20 places – if you’re keen, best be quick… (click the pic below for a bigger version):

 

BPS_hackathon

Another miscellaneous grab-bag of goodies, links ‘n’ stuff

the-linksIn lieu of a ‘proper’ post (forgive me, dear readers, the vicious task-masters at my proper job have been wielding the whip with particular alacrity recently) I’m putting together a list of links to cool things that I’ve come across lately.

So, in no particular order:

Tal Yarkoni’s outstanding Neurosynth website has now gone modular and open-source, meaning you can embed the code for the brain-image viewer into any website, and use it to present your own data – this is seriously cool. Check out his blog-post for the details.

An interesting little comment on “Why Google isn’t good enough for academic search”. Google scholar tends to be my first port of call these days, but the points made in this discussion are pretty much bang-on.

A fantastic PNAS paper by Kosinski et al. (2013; PDF) that demonstrates that personal attributes such as sexual orientation, ethnicity, religious and political views, some aspects of personality, intelligence and many others, can be automatically and accurately (to a fairly startling degree, actually) predicted merely from analysis of Facebook ‘Likes’. A fantastic result, that really demonstrates the value of doing research using online data.

Next up is Google Refine – an interesting little idea from Google intended to assist with cleaning up and re-formatting messy data. Looks like it could be promisingly useful.

A really seriously great website on the stats language R, designed to make the transition for SPSS and SAS users as easy as possible – very clear, very nicely explained. Beautiful stuff.

Another cool website called citethisforme.com; you fill in fields (author, title, etc.) for sources you wish to cite, and it creates a perfectly formatted bibliography for you in the style (APA, Harvard etc.) you choose. A cool idea, but in practice, filling out the fields would be incredibly tedious for anything more than a few sources. Good place to learn about how to format things for different types of reference though.

I’ve previously written about the use of U-HID boards for building USB response devices; I’ve just been made aware of a similar product called Labjack, which looks even more powerful and flexible. A Labjack package is included in the standard distribution of PsychoPy too, which is cool. I’m becoming more and more a fan of PsychoPy by the way – I’m now using it on a couple of projects, and it’s working very well indeed for me.

Now a trio of mobile apps to check out. Reference ME is available for both iOS and Android, and creates a citation in a specific style (Harvard, APA, etc.) when you scan the barcode of a book – very handy! The citations can then be emailed to you for pasting into essays or whatever.

The Great Brain Experiment is a free app from the Wellcome Trust (download links for both iOS and Android here) created in collaboration with UCL. The aim is to crowdsource a massive database on memory, impulsivity, risk-taking and other things. Give it a whirl – it’s free!

Lastly Codea is a very cool-looking iPad-only app that uses the Lua programming language to enable the (relatively) easy development and deployment of ‘proper’ code, entirely on the iPad. Very cool – Wired called it ‘the Garage Band of coding’, and while it’s probably not quite that easy to use, it’s definitely worth checking out if you want to use your iPad as a serious development tool.

If you’re still hungry for more internet goodies, I encourage you most heartily to check out my Links page, which is currently in an ongoing phase of rolling development (meaning, whenever I find something cool, I put it up there).

TTFN.

 

My favourite graphing and plotting software

Preparing graphics for experimental write-ups is always a bit of a minefield. Everyone has their favourite software for preparing histograms, plots and charts and if you’re happy with a program and have a good handle on how it works, you’re probably best off sticking with what you know. For me though, the important aspects when choosing a bit of software to use in this area are primarily aesthetic – I like very clean-looking, uncluttered charts which maximise clarity and readability. Before talking about my favourite bits of software I’ll talk about one that you definitely should not use – Microsoft Excel.

A ‘classic’ MS excel chart. Fugly. Avoid.

Excel is great for a lot of things, but making plots is not one of them. The older versions of Excel in particular just look awful, yet I still see these kinds of plots in so many published papers – every time I see one I do a massive internal facepalm; it just looks totally amateur-ish. The newer versions look a bit better, but still – to be avoided if you want to be taken seriously. Also, for the love of God, never use anything like this:

Excel 3D chart. This way doth madness lie.

3D-effect histograms do not make your data look cool and professional. They are for shiny-suited advertising executives and madmen. End of discussion.

So, what should you use? For years, I was a big fan of SigmaPlot; a very powerful program with a whole host of awesome features that produces some really beautiful results. Unfortunately, I never found the user interface very friendly or intuitive – for most users there’s a fairly steep learning curve involved in using it, but it does produce really nice results, so it’s worth persevering. The level of customisation and formatting options available for your plots are fantastic, and well beyond anything you can achieve in Excel. One other great feature is that SigmaPlot will export plots as very high-quality bitmaps or tiffs (up to 600dpi) for incorporation into figures for papers.

SigmaPlot screenshot with a very pretty plot.

I stuck with SigmaPlot for years because I liked the results, and I (eventually) got comfortable with the interface. However, since switching to (mostly) using a Mac a couple of years ago, I’ve been searching for a good replacement (SigmaPlot is unfortunately PC-only). I think I’ve finally found one – GraphPad Prism. Prism produces really nice, high-quality, simple-looking plots, plus the interface is mercifully friendly, with a lot of built-in demos using sample data which you can modify with your own data very easily.

A sample Prism plot – simple, clear, clean… nice.

The unfortunate downside of both SigmaPlot and Graphpad Prism is that they are both commercial pieces of software which cost real money (although Prism does have a 30-day trial period in which you can try it out). I normally like to recommend free software on this site, but unfortunately I’ve never found anything which compares to these two recommendations in the freeware/shareware world. People have told me that R can produce some very nice plots, and I’m sure they’re right, but because of the command-line interface it’s something that I’d hesitate to recommend for beginners/students. I’ve also had a good poke-around online and haven’t found any decent online tools for making nice-looking basic plots. Well, there’s this one, which (bizarrely) seems to be aimed at kids, and this one which seems OK-ish (but requires sign-up, so y’know… fail) but they’re nothing I’d heartily recommend.

If anyone has any other suggestions for their favourite (preferably free!) software for this kind of thing, whether online or offline, please let me know in the comments! Happy plotting!

TTFN.