For someone who’s meant to be finishing his PhD I sure do have a lot of other peoples’ papers on my plate. Today and Friday I had a chat with another PhD student from the UPTECH project about looking for spatial variation within the UPTECH schools. We’ve got some divergent ideas about how to go about it but we sat down this afternoon and spent some time going over regression modelling versus exploratory/summary statistics and how we can move from using Spearman’s rank correlation coefficient to doing non-parametric function estimation. I wrote some code and commented it as we went, so it should be fairly straightforward to write the accompanying methodology subsection for the paper.
I’m going to spend the week focussing on my final thesis paper (a spatio-temporal model for data from a split panel design). I got some comments back from my supervisors last week and there’s a lot to be done checking certain sources of variation. I spent a fair bit of time today looking over a former students’ papers, checking old email threads, discussing a few things with the co-authors of his papers who are still around and have even been tracking down which particular instruments were used to perform the measurements.
It’ll be important to check that my spatial analysis of the within-school variation matches up with the other UPTECH student’s analysis. If we come to opposite conclusions then at least one of us is wrong. I’m now tossing up whether I should check my results from the supercomputer or wait until tomorrow.