Resources to help you develop your Statistics knowledge & skills
We’ve previously shared blogs, tweets & books to help you develop both Data Visualisation & coding skills, now we turn to Statistics knowledge.
In this post, I share the result of my search for the most helpful resources to develop (or sustain) your statistics knowledge & skills. This focus is a deliberate complement to sharing resources to help analysts code in R, Python or Julia. That’s because we need to remember the importance of robust statistical thinking & design in analytical or data science work.
This purely subjective collection is based solely on my own experience, so I’d welcome your tips too. After reading the recommendations below, feel free to add your own recommendations in the comments box below this post. If there is sufficient interest, I may even extend this to a poll to select the most popular Stats resources to help analysts today.
Given the media I am using to communicate this, let’s start with recommended blogs…
Recommended blogs for your Statistics knowledge
To avoid too long a post, I will limit myself to my top three recommended statistics blogs.
(1) Andrew Gelman’s blog
First, I’ll go with a very popular blog that is also recommended by most Statistics experts (including Prof David Spiegelhalter) that is the blog from Andrew Gelman. In fact, this blog shares posts from both Andrew & his group at Columbia University. A simple, text-heavy, WordPress blog that nevertheless helps you think more clearly about Statistical Modelling, Causal Inference & Social Science:
(2) Annie Pettit’s blog
Second, let me share a resource aimed at the market research/customer insight community. I’m aware that a majority of content on my blog can be skewed towards data, data science or analytics leaders. So, in an effort to remember market research leaders, I recommend Dr Annie‘s blog. She does a great job of both sharing useful stats advice & helping you connect with others in the wider community. It could help you also love stats:
(3) Nate Silver’s blog
For many leaders (of all types) forecasting is a key element of their commercial interest in statistics. To keep on top of best practice in forecasting & to see those principles applied to timely news/sports topics, I couldn’t look past the famous FiveThirtyEight blog. Nate can be a controversial figure, not least in his long-running feud with Nassim Taleb. However, his blog can be both educational & engaging in giving you examples to share that would be of interest to other stakeholders:
Recommended books for your Statistics knowledge
As a blog that regularly features book reviews (and my personal conviction that those who lead read) I couldn’t look past this media. In this section let me first say that I would initially recommend three books that I have reviewed already on this blog:
- “The Art of Statistics” by David Spiegelhalter
- “Causality: Models, Reasoning & Inference” by Judea Pearl
- “How to Make the World Add Up” by Tim Harford (review coming soon)
Beyond those, here are 3 more recommended Stats books. Given the grounding in theory that Spiegelhalter’s book will give you, I’ve focussed on books that include coding examples to apply that theory.
(4) Introductory text with exercises in R
As a start, let me recommend this text from Gareth James. It does a good job of both explaining the key statistical concepts and applying those to key Data Science challenges using R code. Examples & exercises help you apply what you have learned and test your own understanding:
(5) A Bayesian statistics course with exercises in R & Stan
Since reading David Spiegelhalter’s case for Bayesian inference, I have become more interested in this important branch of statistics. One that is growing in influence amongst both statisticians & data scientists. I considered “Bayesian Statistics” by the late Peter M Lee, which is a good text for students. But decided to stick with my challenge of a text with examples in code for readers to practice as they learn. So Richard McElreath won:
(6) Predictive modelling for Marketing, with examples in R, Python or SAS
I’m a little biased with this book recommendation as I feature on the sleeve as a reviewer. But, I know Sameer and both his hands-on modelling experience & marketing understanding shine through in this book. Co-written with experienced data scientist Alun Brain, this book also has the advantage of being available in versions with R, SAS or Python coding. It is very practical and will help modellers avoid many pitfalls:
Recommended podcasts for your Statistical knowledge
Unlike previous collections, as for Data Visualisation, when I have shared recommended blogs, books & twitter experts – in this curation I will focus on podcasts. This is partly influenced by now being a regular podcast host myself. It’s also about giving you a wider range of media for your ongoing CPD (web surfing online, book reading offline or listening anywhere).
(7) More or Less the best podcast available
I can’t walk past the podcast that I listen to most often & have followed for the longest. Similar to the appeal of the FiveThirtyEight blog, but with a UK focus. Plus, this show that investigates the reality behind numbers used in the news benefits from a brilliant host in Tim Harford. If you want to keep up with timely examples of how statistics can help you think more clearly about the news, this podcast can reveal them. It can be a great way to raise statistical topics with others or have an analogy to use with stakeholders:
(8) The official podcast for statisticians
Next, I don’t want to ignore the benefit of professional bodies in advancing both best practice & accreditations. So, I recommend listening to the Royal Statistical Society’s podcast. It is a great way to get to hear from some of the world’s leading statisticians & to find out about major projects happening around the world. Plus you might be inspired to become chartered:
(9) Integrating Stats & Data Science podcast
For my final recommendation, here is a podcast that spans many of the themes our blog covers. Statistics, Data Science, AI and applications all over the world. It has a bias towards students/researchers and new innovations. But it is a great example of how critical statistics is to new developments in Data Science and so a great listen for Data Scientists. Here’s Data Skeptic:
Which resources help you develop your Statistical knowledge & skills?
I hope those recommended resources are helpful for you. If possible, I recommend you select just one to start with and begin to embed a habit of a regular time to read or listen to that one. From that foundation, you can always add more or vary over time. Developing a CPD habit could be the biggest benefit you gain, rather than any one book or podcast.
But, as I mentioned at the beginning of this post, I am also keen to pick your brains. Which blogs, books or podcasts help you think statistically? Please share your favourites in the comment boxes below. It would be great to tap into the wisdom of the crowd of our readers to identify which help most.