2011-06-22 09:46:41MADE TO STICK Part 2: Unexpected
John Cook


As communicators, we need to achieve two things - first grab the reader's attention, then keep it.

One simple method to get attention (not the only way but a useful tool) - defeat expectation, surprise reader. This achieves several things:

  • Gets their attention
  • Helps them remember your message. The more unexpected, the stickier (surprise sears things into memory).
  • Arouses curiosity to know more. People want to resolve why they were surprised, opening the opportunity for you to explain the science

So how do SkSers use surprise? One possible process would be to identify your core message then determine if there's anything counterintuitive about your message. If yes, then communicate your message in a way that builds up reader expectation then defeats it.

Probably more useful to SkSers is techniques to maintain interest. To sustain interest, unfold your explanation like a mystery. What makes people interested is when you expose a knowledge gap (pose a question, stimulate curiosity) then close the gap (provide the answer). Scientific investigation lends itself to this approach so there's plenty of opportunity for SkS blog posts to be structured this way.

The book characterises this approach quite nicely - tease and flirt.

2011-06-22 11:55:42How to make even scientific data....exciting!
Tom Smerling


Speaking of surprind suspence...

We all know that nothing induces audience slumber faster a complicated, confusing and barely-readable scientific chart.

But Andy Jorgenson of the University of Toledo, and a driving force behind CAMEL (Climate Adaptation and Mitigation E-Learning) recently demonstrated a simple but extremely effective technique to make graphs actually exciting....using the principle John describes above.

He animates his bars, pie slices or lines so he can reveal them unfold bit by bit.    With a little narration to add to the suspense, the viewer is kept guessing, what's coming next?    It's particularly effective where there's first a buildup of boring, predictable findings then at the very end.....bam!   A startling (usually alarming) revelation.     The hocky stick graph of temperture begs for this treatment, but there are many more places it works.

In a variation, Andy will project a slide with a question and 5 possible answers, like a little quiz:    "Which do you think is that actual [number, amount, percentage, etc]?   A...B...C... D...or E?"

The effect is to keep the audience guessing, actively engaged, and often -- when the correct answer is revealed -- gasping!

The kicker is that he said that he almost always hands out clickers to the audience, so they can "vote" and he can report aloud, in real time, how many audience members picked each of the 5 possible answers.

Hard to imagine anybody nodding off while this is going on!







2011-06-22 12:08:33


Let's not make everything like that. Personally, I hate these type of games.

2011-06-22 12:43:21Depends on the audience
Glenn Tamblyn


Neal, this is where many of us coming from science or technical backgrounds can introduce a bias in how we communicate to a more general audience. When our working livces are about assimilating and assessing technical information we have usually become adept at it and in fact probably wouldn't actually be in our professions if we weren't. Unfortunately, we simply aren't typical of the human race.


What you find most annoying may actually be exactly the right approach for the average punter.


Looking at the following is quite interesting http://en.wikipedia.org/wiki/Myers-Briggs_Type_Indicator This system classifies personality types by 4 different categories and it is interesting to try and locate yourself in the system. I recently spent some time with a psychologist for some personal issues and we discussed MBTI. I identified myself as probably INTJ or INTP types. And his response was that 'you are probably only about 1% of the population, and the world isn't built for people like you'

Often we have to ask the question 'How do I write for people who think very differently from me?' Even if the best approach for them does give me the s#!^s.

2011-06-22 12:54:58Don't get lost in the details
John Cook

The principle is opening gaps in understanding, then filling it. Or put another way, tease and flirt. So audience voting is one technique. Personally, I'm more interested in the slow reveal with graphs and will try to incorporate that in my talks.
2011-06-23 10:15:07Start with the audience, not oneself
Tom Smerling


I think the main point is that it all depends on your audience.   If you focus on the audience first -- not on yourself and your message -- you'll find clues about what will work and what won't.

Some techniques may seem way to gimmicky for a certain audience, but will be fantastic with others.   And of course, we all have to understand our limits and what we're comfortable with.

Marine-scientist-turned-filmaker Randy Olsen uses scriptwriter jargon to describe the "opening gaps, then filling it."   he calls it "arouse and satisfy."

There are lots of ways to do it, and still stay within one's own comfort zone.   For example, some of the most spellbinding  scientific presentations, even at academic conferences, start with the researcher's own "journey of discovery."    How they became interested, what puzzled them, what hunches they had, their mistakes and dead-ends, etc.

But frankly, this is the exception.   Most of what I observed in graduate school about effective scientific presentations -- how to be credible and convincing to other scientists --  is counterproductive with general audiences.

2011-06-23 10:28:14I think examples will help
John Cook


It doesn't require we necessarily go weird and wacky with our blog posts. It can be as simple as a slight tweak in the wording of a blog post so you restructure the way you unfold the information - pose a question, get the reader intrigued (hopefully), then answer it. Eg - Dana is working on comparing denier predictions to subsequent observations. One way we discussed possibly structuring the post was to first show the original prediction, ask "well, how accurate did it pan out?" then show the final result. That approach pulls the reader along, instead of just hitting them up front with the prediction and observations in the first graph together.

So just keep in mind the technique - "open gaps then fill them" or "arouse and satisfy" or "tease and flirt". There are many ways of approaching the same idea.