Aim 1 is so very nearly done!
Week 39 as a photo collage L to R, top to bottom: chopping cocoons all the way through to freezing the final solution.
Repeatability is everything in science.
An effect may be observed – great! – but it means nothing unless measured in separate trials and those measurements are NOT found to be significantly different when compared by statistical analysis. Usually, the bare minimum to be able to compare is three repeats, or n=3. The more repeats, the more robust your statistics, the stronger the evidence for your claim of the effect you observed.
I am writing my first journal article to come out of my PhD research. In it, I will propose a new method for the preparation of certain silk components for the benefit and use of others in my field (more details once I’m published 😉 ). This week, I set out to prepare the materials for the final bit of data I need: the yield – how much of my target compounds does my new method isolate? (And how long should the method run to maximise yield?)
As this paper will describe a newly devised method, I need to be extra sure that my claims of advantage over previous methods are correct, so I prepared enough samples to achieve n=5. I had to perform two additional repeats of the one of the test conditions due to timing errors allowing the sample to cool too much. Those samples had started to gel (not what I was aiming for) and did not filter very well. Also, by cooling more than the others they had been treated differently, so it was not a true repeat. I had to exclude those two points of data and re-do them, thus my tally came to 22 samples over four test conditions, but 20 points of usable data (each trialled five times, thus n=5 x 4 gives 20 measurements).
Yield, in this case, is calculated gravimetrically by weighing the dried remnants of a known volume of solution. A vessel is weighed while empty, then again once the known volume has dried. The increase in weight is attributed to the dissolved solids that did not evaporate, and the concentration of the original solution as %weight/volume (%w/v) is worked out thus:
The rest of the sample was frozen in readiness for freeze-drying (the bottom right image in the earlier collage) – but that is a task for Week 38. Once I had all the data points for yield, I entered them into the statistics software package, checked that my repeats of each test condition were similar enough (i.e. NOT statistically significantly different) with an analysis of variance (“ANOVA” for short) and turned them into a neat little graph ready to add to my paper.
I must say, it feels really good to be ticking things off the giant Gantt chart and getting closer to my first publication in this field*, little by little. I finished weighing the last mini Petri dish and noticed a drying artefact that had formed a ring a few millimetres wide near the edge of the dish and tried to catch the light on it and take a picture. It took a few goes to get the angle of the light just right and this photo was one of the “failed” attempts. Instead of the ring standing out, the beautiful blue sky was framed by the rim of the dish and I couldn’t get the jazz standard “Blue Skies” by Irving Berlin out of my head for the rest of the day. I know there will be darker days in the weeks to come, when stress boils up and threatens to overcome me, but for now I’ll take the chance to fully immerse myself in this happiness stemming from completion and competence.
The tune was so entrenched by the end of the day that I asked my wonderfully obliging partner to please learn the chords and accompany me on guitar so that I could try to get it out my head 🙂
Nothin’ but Blue Skies from now on!
P.S. The plan didn’t work. Now we both have Blue Skies stuck in our heads! 😀