I wanted to take a break from our regularly scheduled programming to highlight a recent true user success story, which illustrates the importance of collecting all the data – whether you think you need it or not.
This particular academic user- a student researcher – was doing growth studies on a specific model plant organism (Pisum sativum var. saccharatum), testing the effect of various experimental soil amendments on germination and seedling growth. The experiment was conducted indoors under ostensibly climate controlled conditions, using natural light exposure. Germination in non-dormant seeds is triggered by the presence of water – which re-hydrates stored food within the seed and activates hydrolytic enzymes – and oxygen. Under normal conditions, P. sativum is expected to germinate between 7-9 days after planting. The entire experiment was scheduled with a 9-day germination budget as this was a time-critical study and previous trials had yielded germination times in as little as 6 days. Almost as an afterthought, an Elemental Machines Element-A already in the general lab area was placed directly in the experiment pod to monitor light, humidity and temperature. (For reference, Element-As monitor ambient temperature, humidity, air pressure and light levels. )
Ten days after planting, the researcher noticed that not a single sample had germinated. In a mild panic, she went back through her notebook, trying to figure out why things weren’t working.
In retrospect, it’s evident that in this particular situation the issue could have been avoided if temperature had been explicitly defined as a controlled variable. However, it’s not always so obvious which piece of data might turn out to be critical – which is why we firmly believe in the value of collecting ALL the data. With the availability of easy to deploy, highly sophisticated multi-channel sensors, there is simply no excuse for not collecting as much data as possible on environmental variables that may impact your experiment. It’s easier to have the data and not need it then to be left guessing as to what caused an unexpected result.
One final note – I gently ask the reader to not judge this particular student researcher too harshly, especially now that I can reveal her identity as my 7-year-old daughter conducting her experiments for an elementary school science fair at our kitchen table. The very fact that she’s learning the value of collecting and analyzing all the data from a very early age bodes well for her budding career as a serious scientist.