This semester I’m teaching my “how to teach” class: The Teaching of Statistics at the University Level. (Stat 6600, or those of you here at Columbia.) I’ll post more on that in a bit. Here I want to talk about an idea I had as I was falling asleep last night, of a new course I’d like to teach sometime.
The new course will be called Statistical Communication and it will cover the following topics:
– Graphical presentation (not just of raw data, also visualization of inferences)
– Writing research reports
– Writing computer code that can be used by others
– Working with colleagues (including “consulting” but also research collaborations)
– Email, blogging, hallway conversations, and other informal interactions
I think there was some other aspect of statistical communication that I wanted to include that I can’t remember right now. The big idea is that maybe something is to be learned by thinking about all these activities as modes of communicating statistical ideas.
This sounds like a great professional development course!
One thing that I try and emphasise in my stat computing course is code as a medium of communication: to the computer, to yourself in the future and to others. Learning how to express your ideas cleanly and succinctly in code is becoming increasingly important (a related skill is learning how to read other peoples code, despite differences in vocabulary and dialect). It's also useful to draw parallels with the different types of writing that you do – jotting down personal notes (scraps of code) involves a total different process to writing a paper that communicates your ideas to others (a software package).
Dealing with email is another important professional skill that no one is teaching. On average, each month I receive around 10,000 emails, read about 4,000 and write around 300. I've done a lot of reading around this area and I think I have an efficient system, but it's clear that many others in academia are struggling.
Other important communication challenges are delivering presentations and presenting posters. Was either of those the other aspect you were thinking of?
Sounds like an interesting idea. I would be glad to see the syllabus if you ever decide to carry it out. Thanks!
Great idea, I would love to take a course like this. As another potential topic, how about some examples of common pitfalls in statistical communication that are either done unintentionally or with the intent to mislead.
Great idea. Please teach this class online, so those of us at other universities can attend…as lots of us statistics grad students are acutely aware of the lack of this type of training.
Reviewing and editing reports/papers?
Believe Tukey once reported "editing" as his main/major activity
Neat idea – perhaps some ideas can be mined from the reproducible research literature.
Keith
Sounds interesting except for "Writing computer code that can be used by others". Surely there is a reason it is called "code" … :)
Great idea. Please show us this class online, so those of us at other universities can learn some this.
Again, great idea, I'd like to see the material if you implement. It might also be interesting to include a component on visually communicating the methods through process/flow diagrams. Something that provides a conceptual overview what the analyst(s) did from data collection, query/manipulation,exploration, and analysis.
Great idea!
Another topic could be – "communicating complex statistics to the lay person". I think a lot of us statisticians working professionally have come across this issue – whether it be communicating to the policy maker, clients or even your boss!
Although not ideal, sometimes it's unavoidable to have to try to explain complex analysis in a very summarised form.
Also – I agree with the others, it would be very much appreciated if you make an online component or share with us some of your materials – for those of us who are not able to attend the course =)
I will just point out that the term "Statistical Communications Theory" has long been used in electrical engineering to refer to modern communications engineering. See, for example, the course EE 630 offered Spring 2008 at Columbia University in NYC.
Perhaps a name less likely to result in the confusion of two worlds would be "Communicating Statistics".
Chuck Jackson
See also
http://www.amazon.com/Introduction-Statistical-Co…
I second Linda's point of view. I'd say if there is sufficient interest among students in going into industry, there ought to be a Communications for Statisticians II that focuses on how to present findings to non-statistical audiences. One of the key learning objectives is precisely what Linda referred to: how to summarize a project in 5 sentences or less. Much as many of use hate Powerpoint, being able to produce effective presentations is also crucial in an industry setting.
Chuck – good observation. I have in fact taken such classes in electrical engineering and it is in my mind an exciting area of applications for statistics.
I agree with your comment. It would be great to learn about ways to avoid making mistakes when communicating statistics. Very often the media just jumps on 1 line of your paper or statement and when that 1 lines is reported it can be very misleading.
I think the type of class you are propsing would be great. The topics you mention are at times some of the most important things we do and yet are never correctly taught how to do them.