new website

Dear all,

Nine years ago I started blogging. I have been quiet the last few years when it comes to blogging. Perhaps I will pick this up again, perhaps not. What changed quite a bit is how I work as a scientist. I am using R now as my default software for analysis, and have also started to use GitHub for version control, as the cool kids nowadays do. Anyways, my website was long due an overhaul. Here it is. Thanks to Thom Volker for helping me out with this. The website should be relatively easy to maintain, so, who knows, I may pick up blogging again. These are interesting times for a survey methods person. We are in the middle of the COVID-19 outbreak - as I write this I have been at home for over 2 weeks now. As much as I agree with people saying that it really is up to the experts - virologists and epidemiologists - I feel there is a role for sampling theory in making sure that we get as accurate statistics on the proportion of the total population get is infected with SarsCov2. The rather far-reaching public measures to battle Covid-19 should be maintained as long as necessary, but not longer than that. The point when to relax or re-instate measures like social distancing depend very much on what proportion of the total population is infected. I believe one can only really estimate this proportion by taking random samples from the general population, and testing everyone in that sample. However, that should not necessarily be a simple random sample, as seems to be current opinion. It will be much better to stratify on geographic area (as Covid-19 spreads locally) and risk factors such as age. More on that later perhaps.

Huge thanks to @ThomVolker for building this site for me!


Peter Lugtig
Associate Professor of Survey Methodology

I am an associate professor at Utrecht University, department of Methodology and Statistics.