Help crowd source better science communication methods

Recently, studies aimed at better quantifying and communicating the “consensus” on climate change have become more popular. To take advantage of the increasing monetary flow in this direction, and to advance the science even further, our institute, meaning me, have been designing a new research protocol. In the spirit of the “open science” movement, we/I thought it would be good to get some public feedback on potential flaws and possible improvements to this protocol. There are several advantages of this “crowd sourcing” approach to blog science, not the least of which is avoidance of placing a rather tacky “Tip Jar” icon on one’s home page.

What we want to know is what sort of message will really stick with people, make them think. New research has shown that communication methods are very important in this regard. For example, van der Linden et al. (2014) showed that “simple text” and pie charts are slightly superior to mixed metaphors involving doctors’ opinions regarding bridge failures. This is an important advancement that we want to build upon, leveraging the moment to effectively effect an optimal messaging paradigm that cannot be falsified.

One improvement that can be made involves how the experimental units are chosen. van der Linden et al. (2014) queried about 1000 volunteers, but these were chosen from among a “nationwide panel of people who are willing to participate in online surveys”. Those people want to be asked random questions by unknown people having unknown motives, which being abnormal, is not representative of the entire population. A better approach is to just target everybody, and a good way to do that is to confront them on the street while they are minding their own business, before they have any idea what you’re up to really.

Another issue is the treatments themselves; van der Linden et al. used a set of treatments involving pie charts, metaphors and numbers. This is nice but c’mon we put a man on the moon; we believe we can achieve more here. Our design applies less pedestrian treatments to these pedestrian experimental units, each chosen after careful thought. Our procedure is similar however. That is, we first ask what each unit believes that scientists believe about the climate, record the response, then apply the randomly chosen treatment, repeat the original question, and record the second response. Pretty simple really; the whole thing hinges on the treatments, which are:

Treatment 1:
The unit is shown a pie chart with the AAAS logo below it, indicating that 97 percent of scientists believe that climate change is real.

Treatment 2:
The unit is shown a pie chart with images of kittens and Jesus below, with statement as above.

Treatment 3:
The unit is shown a rerun of an old Sesame Street episode featuring the numbers 9 and 7, in sequence, over a backdrop picture of a hurricane.

Treatment 4:
The unit is informed that only Australian Aborigines and Death Row inmates are unaware that 97 of scientists believe that climate change is real.

Treatment 5:
Free dinner and beer at a nice local pub is promised to the unit for all answers over 95 regarding what percentage of scientists believe in climate change.

Treatment 6:
“97% consensus” and “Mother” are tatooed prominently on the unit’s right inner forearm.

Treatment 7:
The unit’s face is situated ~ 0.3 meters proximate to the front end of an OSHA-certified megaphone and unit is informed three times, at the “riot control” setting, that 97 percent of scientists believe in climate change.

Treatment 8:
Justin Verlander is placed approximately 60.5 feet from unit, facing, and delivers a ~97 mph fastball to unit’s upper left rib cage quadrant, while yelling “Get a real-time feeling for what 97 is all about partner”

Many more treatments than this are possible of course. For example we can certainly improve upon the “Indirectness Factor” (IF) one or more steps by asking people what they think other people think scientists believe about the climate, what they think they would think if exposed to a particular treatment, and so forth. There is a rich garden for potential studies following this path.

Thank you in advance for any contributions to the science that you may have, the world will be a better place for it. If you would like to donate $1000 or more that would be fine as well.

Advertisements

4 thoughts on “Help crowd source better science communication methods

  1. “Get a real-time feeling for what 97 is all about partner”

    Ah yes, the persuasion technique that Randy Johnson called “Mr. Snappy.” If the authors had just listed the Lewandowsky and Oreskes citations (I’m imagining/hoping that Kahan is frowning) and left out all the other drivel, they would have achieved their goal and I could have absorbed the information much more efficiently.

  2. Now Jim… paragraph 2 broke my buzzwords filter. And I’m not in the mood to be coding that rascal at the moment…

    This is a brilliant experimental design, and if I thought surveys of random Homo sapiens meant anything I might be persuaded to get on board. Unfortunately I suffer from a severely biased interest in other members of the species. And I’m guessing most of the 97% fall into a category of limited interest.

    Oh, and personally – I like treatment #3 the best – am an ol’ Sesame Street fan. Oscar the Grouch and Cookie Monster may not be members of the species, but I can relate to them and will usually spare a minute to see what they think.

    • It’s Verlander’s fault Clem, he hit your buzzword filter with a rising four-seamer coming up and in.

      I’m sure you’re aware that the Cookie Monster was very fond of “Cookie Charts”.

Have at it

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s