If there is one thing I am learning throughout my MSc it is that science doesn't always go the way you expect it to. Personally, I think that is half the fun but when I think about some of things I have done over the last few years and some of the results I have generated (sometimes in error) as I learn this process called science I can't help but think that maybe this is why scientists are sometimes viewed as "crazy" by the general public.
Well in the interest of pointing out that science is a process; a long, tedious but fascinating process and to show that those who work in science make mistakes, I have 2 images to share today.
These are the first two exhibitions in a series I'd like to call "hmmm, something seems off here...?".
Making mistakes, doing things the hard way etc is part of the process. It is what makes grad students into future scientists (ahem hopefully filled with integrity and critical thinking skills) and threatens our insanity all at the same time.
Exhibit 1: One of my favourite R outputs
I like stats, I'll say that right now. I dislike learning R and I'm ok with admitting that. But even in this cloud of bitterness I can see its utility and have warmed to it. My "favourite" warning message ever went something along the lines of: "Warning message- results may be meaningless in Bray". Well thank you R for completely dashing my aspirations of statistical signifiance. At least when you fall in R you fall hard and this makes you determined to not make the same mistake again. This means I will never forget to check for colinearity again, and because of this I'm sure that one day R will say "Your data is just perfect, great job!". Ok maybe not...perhaps this will inspire me to one day have the skills to write my own package that will say just that.
But even this warning message was beat this summer when I began to experiment with the commerical version of R. One of the first things I was working on in R was making a simple boxplot for my cosm experiment. On the x-axis I wanted experimental treatments and the Y I wanted emergence success. Here is what I got the first time around:
Intriguing though right? Like I would hang this on my wall as an abstract piece. Would you believe that all I had to do to create this (absolutely useless) masterpiece was to switch the x and y axes? Once I switched them back this transformed into a readable boxplot.
Intriguing though right? Like I would hang this on my wall as an abstract piece. Would you believe that all I had to do to create this (absolutely useless) masterpiece was to switch the x and y axes? Once I switched them back this transformed into a readable boxplot.
Exhibit 2: And then there were worms...
As part of my colleague Anne's (the shore bird researcher from Trent) work this summer she was measuring invertebrate biomass in the mudflats where SEPLs were foraging, to see if peaks in biomass conincided with chick hatching. This may sound simple, but when you are identifying insect larva to family level and then using specific equations based on individual measurements of individual specimens to calculate said biomass you end up with something that looks like this:
Now this is for 5 individuals of a single sample. Tedious? Yes. Accurate? Yes, and probably more so than drying and weighing the samples. I'll look forward to featuring this publication once this work at Trent University is done.
