Now that data analysis has taken over my life, I have decided that it is like Buddhism as these two quotes below indicate:
“For every cause there is an effect [in the future, man, this will be represented on a graph].”
“You don’t want to do any of that Bray-Curtis similarity or Multi Dimensional Scaling nonsense for the question you’re asking. It doesn’t show a causal relationship, and that’s what you’re after.” – Professor Byrnes on discussing how I should analyse my data
OK, so I’m obviously embellishing here (Jarrett didn’t use the word ‘nonsense’, and the Buddha didn’t say ‘man’ (although I believe he did use the phrase ‘What up, yo’)) but I feel a bit like a Journo by writing a blog, and not a scientist, so it’s ok. But essentially, it’s all about cause and effect, or karma if you like, something we run into on a daily basis.
Statistics started to invade my normal life when I was deciding whether to take the class or not. Having reached the Holy Grail of 1) fulfilling my course requirements, 2) being on grant money (no more Bio 111 ever), and 3) actually having my project finally working, I was somewhat reluctant to have a routine again. However, not being too good at stats, I really didn’t have a choice. And so, as I have data coming in for both my own project and that of my Research Assistantship, and Jarrett’s class too, I decided to have a semester of nothing but data analysis and R. Having any sort of alternative fun life in parallel would have to involve statistics. My first application of statistics to daily life came when I wondered if I could still do a road trip that I was planning, writing my homework in the car. The first project that came to mind was seeing if there was a correlation between how late I stayed up exploring new cities versus the distance driven the next day drive. Since then, I have caught myself following a similar train of thought on several occasions. Yesterday, for example, I was planning my SCUBA diving for the weekend but also wanted to eat a lot of Chinese food on Friday night. If I just have some chicken with greens, I’ll probably still get up by 8 and be quite motivated. If I add some of those deep fried crabby-cream-cheese-triangle things, we’re looking at 9 a.m. Anything else deep fried and add another hour. This time I realized that, to my horror, I had also thought about the equation for the linear relationship and had visualized the graph in my mind’s eye. Furthermore, if I’m not enough of a loser at this point, I actually tried to write out the R code in my head.
Well, I don’t want to bore you with my ramblings while I have my Pu-Pu Platter for 2 (for 1), I think its time for a graph. The relationship that I would like to explore is that between Jarrett’s enthusiasm (using week number as a approximation) and the amount of homework we get. What do you think, is y equal to x, 2x, is it an exponential relationship???
Turns out that the linear model is y = 3x + 2 or, for those of us that still don’t like to see x and y, the number of chapters that we get to read each week is three times how enthusiastic Jarrett is feeling plus two! Holy homework indeed. Be afraid, be very afraid. So what are our options? If statistics really was like Buddhism then we wouldn’t be able to feign sickness or make up any excuses but in Buddhistics, I’ll let you decide.
I guess my point is that, as much as I hate to admit it, statistics is relevant to real life even if it what effect a diet of MSG will have on any other dependent variable. It is not just an abstract notion of getting through stats as a means to an end to finish your degree. As I find the words ‘function’ and ‘object’ making it into my daily vocabulary in phrases like, “Well, the creaminess of the frosting is a function of how much butter I added” I concede that, like it or not, statistics are here to stay.
I would like to end on a final note of caution from a man some of us know and maybe love:
“Correlation does not imply causality. For example, take gas prices and my age in years. Both have gone up since I was born [when Ron Etter, when were you born?] but they are by no means related.”
And yes, I did make it diving by 8, but p.m counts too right?
Code for graph
enthus <- read.csv("enthusiasm.csv", header = TRUE) ￼ # Plot the data plot ( Chapters ~ Enthusiasm, data = enthus + type = "p", pch = 20 + main ="Holy Homework!" + xlab = "Enthusiam units (week number)", xlim = c (0,3) + ylab = "Number of chapters", ylim = c (0,11)) # Fit regression line abline (lm (Chapters ~ Enthusiasm, data = enthus)) # get equation for limera regression, R-squared and p-value summary(lm(Chapters ~ Enthusiasm, data = enthus))