Hi there,

Today, I cover Hans Rosling’s classing talk on statistical visualisation and the free app used to create it, the R code development guide, pins and some resources for students.

Let’s dive in!

Have you watched Hans Rosling explain statistics? A Swedish physician and academic, Prof Rosling, explains numbers and how we feel them. We believe we know the world, but do we? This talk shows how our beliefs don’t align with data anymore.

(I’m not going to hold it against you if you leave now and go watch all his TED talks. Seriously.)

Consider COVID-19 vaccination. In your head, rank these five countries based on their coverage percentage, i.e. percentage of people who got at least one dose. China, Cuba, India, Iran and United States. Scroll down to find the correct ranking as of today.

Once you recover from the awe-attack by Prof Rosling’s explanation, you’d be curious about how to develop such beautiful time-series charts. Gapminder Tools Offline provides a time-series visualisation tool, the one we saw in the previous talk, for free. It has a drag-and-drop interface and comes with many inbuilt datasets.

Looking for ways to make fundamental R contributions to core R? Here’s the best way to get started. This guide covers essential details such as what counts as a “bug” in R, how to track and review it, what are patches and how they work, how to find the source code of functions and many other questions.

What are pins? Pins are a versatile way to publish R objects on a virtual corkboard so you can share them across projects and people. You can pin just about any object: data, models, JSON files — any object. A frequent use case is pinning small data sets that don’t belong in a database but are useful for analyses.

Pins can be stored in Dropbox, Microsoft OneDrive, Azure, AWS and many others. Check it out!

Advices, advices, advices. We don’t take it seriously when we get them, and it’s usually late when we seek them.

Here are some bits from Dr Scherer’s blog:

pins package publishes data, models, and other R objects, making it easy to share them across projects and with your colleagues. Read the vignette here.

fusen provides methods to build R packages from a single R Markdown document. One file, that’s it! Read the vignette here.

terrainr provides functions for the 3D geospatial data visualisation, working with ‘Unity’ rendering engine. Read the vignette here.

Diffify provides you with a comparison between different versions of any R package stored on CRAN. Say you were using a particular version of a package in a project and now a new version of that package is available. With Diffify you are easily able to check what has been changed in the new release. Read the vignette here.

Here is a translator between core visual concepts, statistical jargon and machine learning!

Here is the correct ranking as of today: Cuba (95%), US (78%), Iran (78%), India (74%), Egypt (46%) and South Africa (37%). This is from the NYTimes tracker but I believe it’s not been updated for some time as India’s coverage (that I know of) is significantly higher (BBC).

Hope you learnt something new today. Like always, your feedback and suggestions are always welcome. See you next week!

Harsh