Based on the observation of twelve PhD students, the likelihood for the spontaneous appearance of beads of sweat while coming across a capital sigma (Σ) followed by square roots and indices is actually rather high – biologists are not terribly famous for their affinity to maths and statistics.
But wait – was the sample size of twelve highly motivated ImmunoSentation IITB PhD students attending the statistical literacy course conducted by Dr Rick Scavetta and Dr Irina Czogiel this March large enough? Which prerequisites have to be fulfilled to estimate the mean statistical knowledge of young scientists in general? One thing is for sure, all workshop participants had a fantastic chance to refresh their rusty basics of statistics and learn that statistics is not only necessary because their boss or reviewer requests it, but because it helps to really understand the results.
Accompanied by a population of Martians (those amiable green friends from our neighbouring planet) with certain heights, nose colours and so forth, we started with descriptive statistics. How can you correctly visualise the data distribution as well as measures of location and spread? The second important part was inference, i.e. the process of deriving general conclusions from the available data. This ultimately led to hypothesis testing and we talked about some common tests like the t-test, the Χ2-test, linear regression and finally ANOVA.
Compared to some other statistics courses, this course really focused on understanding the general concepts of statistics and then see how common tests work in particular. Definitively, not a “My data look like that – which test do I have to use” course with the risk of perfectly learning which test to use but not increasing any understanding at all. In a nutshell, my impression was that all attendees learned how large the contribution of statistics to science is, and could revive their knowledge of fundamental statistical concepts and how they are applied in statistical tests. The course is especially useful for students at the beginning of their PhD thesis because it not only helps to understand data but also teaches where statistics are needed while planning a set of experiments.