Comments for Sam Clifford

Comment on A proud moment by Sam Clifford

Sam Clifford — Tue, 07 May 2013 13:48:09 +0000

The collaborator’s not as fluent in R as they’d like to be but they have definitely come to the realisation that reusable code in R is a better approach than a bunch of cell references and filled formulae in Excel.

One of my SEB113 students asked me today if there was a better way to write equations than the way Word does it. My eyes must have lit up. They’re a physics student, so they may as well learn about LaTeX sooner rather than later.

Comment on A proud moment by JegarP

JegarP — Mon, 06 May 2013 01:55:42 +0000

Classic. I like your comparison to the stages of grief, as the researcher slowly realises that the Office Suite is ‘dead’ and experiences that separation.

‘Sorry for the inconvenience’ is my favourite bit. Like taking your car in for a service a apologising for it not being a steam engine.

Comment on An argument for why careful coding in R is better than an Excel spreadsheet by blenz3

blenz3 — Fri, 19 Apr 2013 03:03:51 +0000

I’d be interested to see how that turns out! Most of my past and present coworkers have been barely able to use the f(x) bar in Excel let alone understand the basics of an R program or interactive session. Is the class more focused on the programming aspects or will the emphasis be on working with the R shell to run the analyses?

Comment on An argument for why careful coding in R is better than an Excel spreadsheet by Sam Clifford

Sam Clifford — Fri, 19 Apr 2013 02:28:25 +0000

I’m tutoring in a new first year quantitative methods class. We’re teaching a bunch of 17-18 year olds, most of whom have zero background in programming, how to use R. I don’t expect that at the end of the semester they’ll all be proficient at using R to do all of their data analysis but by showing them that they can operate on entire columns with functions rather than by click-dragging a formula from a cell I’d hope that we can convince them that R makes certain things easier.

Comment on An argument for why careful coding in R is better than an Excel spreadsheet by blenz3

blenz3 — Thu, 18 Apr 2013 18:30:52 +0000

The problem is that few professionals outside of IT/software dev really understand programming well enough to use R. I have been encouraging coworkers and friends to learn VBA so that they can stay within the comfort of their Excel environment and generalize their analyses, but even this has been impractical. I like R, but I just think the complexity is too great for the majority of non-programmers.

Comment on Something I’ve learned for BUGS/JAGS by Matt

Matt — Sat, 23 Mar 2013 10:02:58 +0000

The BUGS manual suggests 3 different ways of handling unbalanced or “ragged” datasets – I think I’ve tried all 3 at one time or another.

Comment on New things in Science and Engineering at QUT by Sam Clifford

Sam Clifford — Mon, 18 Feb 2013 10:51:50 +0000

Oh. I should add that the “poster session” section of the day was held using the CubIT system, so people could interact with the posters by zooming in on sections with their fingers. Hopefully we get some really neat software to allow people to build posters with proper interactive/multimedia functionality such as animation, pull-out graphs, highlightable text, etc.

Comment on Presenting data and mathematical ideas by Sam Clifford

Sam Clifford — Sun, 16 Dec 2012 04:25:29 +0000

Ah okay, yeah. I’ve been playing around with violin plots but I think the boxplot’s IQR-based summaries are a bit rubbish. I think you only need half a violin plot (the symmetry is unnecessary) with the mean and median marked.

Comment on Presenting data and mathematical ideas by Sam Clifford

Sam Clifford — Sun, 16 Dec 2012 00:30:18 +0000

What’s a violin plot? Is it like a bean plot? Density and some other meaningful summaries?

Comment on Presenting data and mathematical ideas by Matt

Matt — Sun, 16 Dec 2012 00:28:45 +0000

I like the stripped-down boxplot, and thanks for posting the code – that would’ve been my first question :-)

What do you think of violin plots, which do convey meaningful information apart from the median and IQR? Or do you think that’s an instance where Andrew Gelman would recommend that multiple plots would do a better job, rather than overloading a single plot?