Thursday, 2 February 2012

Putting the D'oh in Dr...

Throughout my studies I've noticed that us PhD students have a habit of being incredibly inefficient. The usual culprit is doing something by hand that could be done 100x faster on some form of computer software. When we are shown the error of our ways, shown how much time and effort we are wasting, we all have that moment where you see the light and generally a 'D'oh' moment is born. There's nothing technically wrong with the way you may be doing something, in fact, a few decades ago it would have been your only choice and you would have been commended on your attention to detail and dedication to get those time-gobbling tasks done. Referencing is one of the main culprits, but for us psychologists, statistical analysis is an equally guilty party.

I had such a 'D'oh' moment today...

I do not claim to be an expert in statistics, or even proficient. I definitely don't have any expertise in the use of SPSS (a particularly unhelpful statistical software package that the psychology world keeps in business despite it's complete and utter lack of a logical user interface). It would be safe to say that my four years of Undergraduate stats training, a year of Masters level stats training and a year and a half of working at PhD level have left me with a 'workable' level of SPSS knowledge, but only if I have a large cup of tea in one hand and my trusty Andy Field SPSS book in the other. Ok...I admit it...I can do one specific statistical test on it and that's about it....

Today, I was trying to put a lot of categorical data into percentages (basically what percentage of people said one thing compared to another). I thought to myself, this seems like something SPSS would be able to do for me, but I didn't have the slightest idea of where to even look for that sort of command. Nobody was around to ask so I decided to calculated the percentages the old fashioned way - well, with my iPhone, not with pen and paper! Funnily enough this was taking a looooooong time so I decided to revisit old SPSS...

Unsurprisingly, SPSS is quite capable of calculating categorical percentages and in fact I had been but two button clicks away from having the information right in front of me.

D'OH!!!

These little moments could, by a positive thinking person, be considered to be a moment of learning to be celebrated but it's very difficult to be that positive person when you realise just how much time you've been wasted. These moments happen a little too frequently and are completely soul destroying. My only solace is the fact that they happen to everyone else in the office too!!

Ciao for now!

2 comments:

  1. May I remind you of the time I completely forgot to even account for the within-subjects part of my design in my analysis...

    D'oH!!! xx

    ReplyDelete
  2. Nifty! I tend to avoid doing anything extra in SPSS if I can. Excel is a much better friend to me! Though testing in SPSS is my bitch, or at least for the tests that I need to know (not random crap like Cluster Analysis. What a waste of a Friday morning that lab class was)...

    ReplyDelete