Number Crunching

This week (and upcoming few days) has been all about number crunching...more formally known as data analysis. With the 18th International Meiofauna Conference starting on Monday, and my presentation occurring on day 1 of the conference, Sean and I have been working to get as much data together as possible for the presentation. Today I finished 35 out of 90 samples, which I am happy with after the multiple setbacks regarding fluorescence and with the hurricanes reducing the time I could spend at the lab.

I think data analysis is an interesting part of the research process. Obviously we analyze data to answer our questions, but the data analysis doesn't provide a complete answer. It is up to the researcher to draft a narrative surrounding or explaining the results, and this is sometimes where bad science can occur. For example, if a researcher finds no difference between their experimental group and control group, they can potentially point to arbitrary factors within the experiment that shaped tis result, like equipment error, a variable the researcher didn't measure/account for, or low sample size. Similarly, if there is a difference between the two groups, a researcher could argue that their treatment was the definitive cause for the difference, even if background variables (environmental or otherwise) may have influenced one group over the other. One of the first things that I learned in my graduate experimental design and statistical analysis classes is that researchers should know how they plan to analyze their data before doing an experiment, and I believe that by envisioning the analysis step, researchers are forced to realize what information they need to account for throughout their experiment to reduce as many confounding variables as possible. For example, in this current work, I have included the row number and column number for each microcosm jar as explanatory variables in case there was an unaccounted for environmental gradient within the experimental chamber that affected nematode behaviors.

While I can't report on the results of this work yet, I am pleased with the work that I get to present internationally this upcoming Monday. My goal is to finish the initial assessments by the end of January (or earlier) so that I can follow-up with the modified fluorescence assessment and get this work done in a timely manner. Next week's blog will likely talk more about my experience at 18IMCO, so stay tuned!