Becoming a scientist is a process that reminds me of the saying, “you can’t see the wood for the trees.” If that’s not immediately obvious, stick with me, you may find that this applies to you too…
When you’re studying or conducting research in the Sciences you’re so busy staring at your data (the trees), that you overlook your development as a person and a scientist (the wood).
When I started out studying triple sciences at A-Level and secretly hating science, I couldn’t even make the simplest chemistry experiment work. I was in fact so bad that I made myself a comical ‘dunce’ hat to wear in class. And don’t get me started on Physics! My favourite subject was Psychology, which I later pursued at degree level; and although my theoretical knowledge was good, my technical ability, logical reasoning, and practical skills were average at best. The weight of these limitations was a constant burden throughout my career, until mounting evidence suggested what I could not believe to be true… I had somehow unwittingly shed my shackles of ineptitude and become what can only be known as… A Scientist.
There’s no absolute test to see if you have unwittingly become a Scientist. I rather like to imagine it as “Scientist Syndrome” – diagnosed by the observation of a cluster of symptoms. And it’s not an easy syndrome to live with. You can use the following symptom checker to see if you too, have become a scientist.
1) Data Rage: Any reporting of data annoys you.
Do you find that you have to question any basic reporting of data in the media? Are you left with remaining questions regarding the validity of said data; plagued with intrusive thoughts after such an ordeal? Then you may have Data Rage…
Whilst watching BBC News one morning, I learned about the crisis UK milk producers are suffering regarding the price of milk. It sounded quite the dire situation – the price that supermarkets pay for milk has fallen year on year, meaning that some farms can no longer afford to carry on. Terrible news! The reporter went on to present a bar graph of the price of milk by year to make his point. “Great idea”, I thought. But no, this was an epic fail that, for me, completely undermined the story. At a glance I could already see that at least one of the bars was not smaller than its predecessor, suggesting that the price of milk did not fall that year. What’s more disturbing – I had grave concerns that the price of milk from year to year was not significantly different – i.e. it didn’t look to me as if the difference in price was large enough to say for sure that the price was really falling year by year rather than just fluctuating in the normal way prices tend to. To know this I wanted to see the standard error of milk prices for each year. But, to my horror, no standard error was presented. How could the BBC make such an oversight?! Breakfast ruined.
2) Matlabitis: inflammation of the matlab gland.
If you are regularly caught extolling the benefits of MATLAB to your poor uneducated Excel-using friends, then you may just be suffering from a bad case of ‘Matlabitis’. For those (un?)lucky enough not to know what Matlab is – it’s a life-changing ‘high level’ programing language, which is great for management and analysis of large data sets and, which includes a number of useful toolboxes for specialist analysis (like SPM (Statistical Parametric Mapping) for neuroimaging research). With matlab the world is your oyster! And, it’s exactly this kind of thinking that is symptomatic of Matlabitis.
When I started my PhD I did a lot of my data management in Microsoft Excel – nothing wrong with that, but it wasn’t easy. For example, in a complex data set you may have many columns of data (let’s say relating some demographic information and questionnaire responses). So far, so good. But what if you want to look at a subset of these data, like only data from males? “Use the Sort function” I hear you cry? Indeed! But what if you want to look at a subsection within a subsection; or what if there are more than two conditions that specify the data of interest (males, over 30 years old, living in South Manchester, who have a disability)? I found this tedious and difficult in Excel, however, in matlab I can write a simple function that loops through each row and selects only the data that satisfy my conditions. If I want to, I can then save it as a new variable (organised like a spreadsheet), and manipulate (organise) it to display however I wish. It’s like a dream, I’m telling you! It’s when you start using matlab outside of work that you should worry…
3) Scientist Syndrome Sleep Disturbances.
Do you wake up in a cold sweat, wondering whether the analysis you left running overnight has finished? Have you had dreams about your research? Are your night-times plagues by nightmares of mislabelled graphs, insignificant t-tests and negative reviewer’s comments? If so, you may be suffering from Scientist Syndrome sleep disturbances.
You spend so many hours of your life at work that when you leave for home you need a peaceful, work-free environment. However, if you haven’t properly decompressed from the day, you can inadvertently bring you work home with you. This can lead to troubles falling asleep, early waking, and night terrors. One time when I was deep in programming (writing a code – in matlab of course – that would present my experiment on a computer screen) I seriously had a nightmare that I could only talk in ‘for’ / ‘if’ loops and logical statements. I know it’s a common joke that scientists might as well talk in binary code, but this was no joke, it was terrifying!
4) Science-related sight difficulties: You see ANOVAs everywhere.
Like many syndromes, Science Syndrome can adversely affect our senses and our cognitions – the way we think. If you’ve found yourself looking at simple objects of beauty in a new and slightly odd way, or you’ve started interpreting art as science, this may be a sign that you have Science-related sight difficulties.
During my Masters studies, I think I over-indulged in statistics a little, until one day I had a temporary breakdown. I was at band practice with my housemate (we called ourselves The Gamma Band, which should have been an early warning sign of the syndrome) guitar in hand, vocal chords warmed. And then it hit me. The guitar was like a very large ANOVA (a statistical analysis of variance). ANOVAs test the statistical relationship between a number of factors, which can have many levels. In this case the factors were: Strings (with 6 levels, one per string); and Frets (with 19 levels, one per fret). The combination of these levels and factors create distinctly different sounds, and therefore I reasoned that this demonstrated a significant ANOVA. When I explained this to my housemate, she was not enthused. We never spoke of the guitar ANOVA again…
5) SNR hypersensitivity: You explain everything as Signal to Noise Ratio (SNR).
Have you started seeing your environment differently? Maybe your perception of the environment (what you actually see) is the same, but the way you interpret and navigate it is different? If this sounds like you, you may be suffering from SNR hyperactivity.
I noticed this myself last week when driving through drizzly Manchester. Although well-known for its downpours, on this particular day the rain was happily rather light. I was driving home from the office, listening to XFMs daily feature “that’s good innit” when I had my own “that’s good innit” moment. I found the optimal windscreen wiper setting for the weather conditions. I’m ashamed to say that not only did this realisation accompany a rather long inner monologue which hinged upon scientific concepts but, that I also felt utterly delighted. I reasoned to myself that the size and frequency of the rain drops, along with the velocity of my car had created a deficit in the usual signal to noise ratio of driving. The proportion of signal – in this case, the visual information my eyes could detect about the road, the position of other cars etc. – was lower than the proportion of noise – in this case, the disruption to my visual perception caused by rain on the windscreen making things look blurry. Fortunately this disaster was averted by choosing the correct windscreen wiper speed, which weighed the SNR (signal to noise ratio) in favour of the signal by eradicating enough noise (rain) to drive safely. As Clint Boon of XFM would say – “that’s good innit!”
If you have been affected by any of the issues raised in this article, you should “get help now”; or to use my (newly) native tongue: “01100111 01100101 01110100 00100000 01101000 01100101 01101100 01110000 00100000 01101110 01101111 01110111”.(binary code taken from http://www.unit-conversion.info/texttools/convert-text-to-binary/).
Post by: Gemma Barnacle