What is a confounding factor? According to the Oxford English Dictionary, confounding is “destroying, confusing, perplexing, amazing, etc”. In statistics, a confounding factor or variable worms its way between your carefully chosen cause and effect, making a nonsense of your research results. A survey of students on a course (say, Statistics for Social Scientists 101) reveals that attending lectures is correlated with certain physiological effects (say, sleepiness). However, it is possible that social scientists with an interest in quantitative methods are also party animals and that their sleepiness in lectures is caused by late nights, drinking and dancing. Their status as party animals confounds your initial result.
Confounding is a handy concept for a blog like this one. I’m still learning my trade, so I am frequently confounded (or destroyed, confused, perplexed, amazed etc.) by the science of statistics. I want this to be an academic blog. I’m starting it in the first year of my PhD studies and I would like to use it as an informal record of my research process and findings, as well as to explore some of the ideas that come up in lectures, seminars and around the corridors of the university. I’d also like to able to use it to poke a little fun at bad stats stories in the mainstream media, but I don’t know whether I’ll have time. So here it is.