M-c Shanahan

This summer project was inspired by a panel that I sat on at LogiCon this spring. The moderator, Desiree Schell asked us whether we would describe ourselves as evangelists for science and for scientific thinking. I answered that in my everyday dealings with friends and family I try to be a stealth evangelist, sharing my own enthusiasm as a gentle approach to encourage others to do the same. After the panel though, I felt that I’d cheated a little bit with my answer and not thought about my experience as a science education researcher. That’s what these posts are meant to do, take a look at the research literature in science education and ask what it might have to offer to communicating about science and scientific ideas more generally.

I find online science communication fascinating. I am enthusiastic about its possibilities and intrigued by its challenges. With an interest in online communication, comes an interest in text. While videos, animations and images are powerful too, the written word is often the simplest and the default mode of online communication–-think blog posts, tweets, status updates, and comment sections, mostly all written or at least including written elements. In the world of online science communication, these are all texts but what makes a text good for communicating about science and, in particular, what makes a text good for helping readers understand and accept scientific ideas about the world?

Science education has been kind of text (especially textbook) obsessed for a long time. In the late 19^th century textbooks acted as de facto curricula for schools that aimed for some cohesion as they spread out across a North American landscape that was still being settled. And we’ve never quite been able to let that go. Questions about what makes texts good and what makes them convincing, have been a recurring theme.

Last week, Christie Wilcox began a series on her blog Science Sushi, part of the Scientific American blog network. She started with this introduction:

“People believe a lot of things that we have little to no evidence for, like that vikings wore horned helmets or that you can see the Great Wall of China from space. One of the things I like to do on my blogs is bust commonly held myths that I think matter. For example, I get really annoyed when I hear someone say sharks don’t get cancer (I’ll save that rant for another day). From now onward, posts that attack conventionally believed untruths will fall under a series I’m going to call ‘Mythbusting 101.’”

I read it and thought, “A-Ha, Inspiration!” (not like the A-Ha! tuna I was once offered at a restaurant, that’s another story)[i]. What guidance can the science education literature offer for doing this kind of blogging well? Are there ways to more effectively change readers’ minds about common misconceptions, myths and everyday notions that are less than scientific?

As I wrote in Part 1 of this series, changing peoples’ conceptions is hard, very hard. The way we understand the world is shaped by all of our interactions with it and with all of the people in our lives. We don’t just have a set of ideas that sit on a shelf like books and can easily be replaced one for another. Ideas about the world are more like tangled webs of connected information, experiences, and beliefs. A complex ecosystem is a better analogy than a bookshelf. This means that writing to bust myths, convince people about scientific evidence or change their minds takes more than just communicating clearly. If that were all it took, science teaching would be easy and there would be few public controversies about accepted scientific ideas.

Explainers, like Chris Rowan’s post about the Japanese earthquake, are excellent when the issue is missing information. For example, I think I have reasonably scientific views about earthquakes (I did fairly well in undergrad geology and have taught some very rudimentary earth science in schools) but my views are patchy in places. It’s not so much that I have serious misconceptions but instead holes in my understanding. Good explainers fill in these gaps with clear descriptions and new information. They aren’t usually narratives and they aren’t usually arguments. They are typically purely expository, and they are excellent for filling in patchy places in a reader’s understanding.  This process is sometimes described as assimilation – the new ideas are like a new species introduced into the ecosystem. If there’s a niche for them and they fit into the existing structure, they are assimilated with little conflict and change. (This analogy kind of breaks if you try to take it as far as invasive species). When the problem is misunderstanding though, explainers aren’t as helpful.

There have been two major reviews of research done on the ways that written texts can support conceptual change, the kind of conceptual change that causes the whole conceptual ecosystem to be altered. In 1993, Barbara Guzzetti and her colleagues published a statistical meta-analysis of studies up to that point, comparing the different approaches that had been used and the effect that they had on students from elementary school up to undergraduate classes.[ii] Christine Tippett updated their review last year with an overview and a thematic analysis.[iii] Both reviews show consistent evidence that explainers are not the best type of writing for conceptual change. The most effective texts were those that directly addressed and refuted common misconceptions.

Refutation texts always include at least two parts: a) a statement illustrating a common or likely misconception and b) direct statements that contradict the misconception and emphasize more scientific views. Usually there is some sort of refutation cue as well, such as labelling something as a myth or saying directly “but this is not true.” Tippett gives this example written for young children (the misconception is in red, the cue in blue and underlined and the refutation in green):

“Some people believe that a camel stores water in its hump. They think that the hump gets smaller as the camel uses up water. But this is not true. The hump stores fat and grows smaller only if the camel has not eaten for a long time. A camel can also live for days without water because water is produced as the fat in its hump is used up.” (p. 952)

In her Mythbusting 101 post, Wilcox does something very similar. She lays out four myths and common beliefs and then carefully explains why each is not true or at least isn’t as simple as it first sounds. Her post has the structure of a refutation text, pointing out to the reader something that many believe to be true and then explicitly saying that it isn’t (colours are the same as the example from Tippett).

“Myth #1: Organic Farms Don’t Use Pesticides

When the Soil Association, a major organic accreditation body in the UK, asked consumers why they buy organic food, 95% of them said their top reason was to avoid pesticides. They, like many people, believe that organic farming involves little to no pesticide use. I hate to burst the bubble, but that’s simply not true. Organic farming, just like other forms of agriculture, still uses pesticides and fungicides to prevent critters from destroying their crops.”

Wilcox’s text also illustrates another element of effective conceptual change writing – straight and direct expository refutation. Sometimes education authors will try to explain science concepts through stories. The misconception is brought up as part of the narrative on the assumption that narratives are more comfortable, more interesting, and easier to understand. In Guzzetti’s analysis, though, only young children benefited from having narrative included as part of the refutation. High school and undergrad students reponded better to the straight expository texts. Tippett also points out that older students seem to prefer to read in this style.

Ok, so that’s two tips so far – direct refutation is important and it’s most effective when it’s straight expository refutation (except when it’s for young children). What about the context in which texts are read and the thinking processes of the reader?

Both Tippett and Guzzetti were able to look at several comparisons in how refutation texts were used: texts on their own, texts used with classroom discussions, texts read before and after classroom demonstrations, and texts used with writing activities. Given how powerful direct experiences can be, I was surprised that both of the reviews showed that the most effective strategies were always combinations that included text and that text on its own was more powerful that any of the other methods on their own (e.g., discussions and demos). This says a lot about the power of what we read.

Of course there are several possible explanations for this, not the least of which is that you can return to a text and read it several times to remind yourself of its content, something you can’t do with a discussion. The strength of text alone shouldn’t be taken as absolute as neither Tippett nor Guzzetti were able to make comparisons to videos and interactive animations which would presumably have some of those same benefits.

Given that texts are important, what made particular texts more effective than others? Across all of those combinations, the texts worked better when students had a chance to think about their own conceptions first (sometimes called activating or priming their prior conceptions) and then had their own ideas directly challenged. This makes sense from a conceptual change perspective, where the difficult task of rearranging and changing conceptions is thought to happen as a result of cognitive conflict or disequilibrium – creating an internal discrepancy.  The discussion around cognitive dissonance in relation climate change and evolution, for example, also views this conflict as potentially negative, where placing ideas side by side leads people to want to resolve the conflict, often by relying on their prior views and warping the new information to suit. At the same time real conceptual change is unlikely to happen unless this same conflict occurs.

Just asking people to think about or priming their prior knowledge without explicitly challenging it was not enough. The most effective texts (and text-activity combinations) asked students to think about and apply their own conceptions and then challenged them directly. In writing and blogging then, activating or priming misconceptions would mean more than just stating common misconceptions. Sometimes people don’t think they hold a particular misconception until you ask them to make a prediction, explain a particular situation or make a hypothetical decision. And it’s easier to gloss over or ignore mythbusting when you don’t think you hold the myth or that it doesn’t apply to you. Good activation asks the reader to recognize how they view the world, so then the writer can go on to refute it. The chance for a meaningful discrepancy between ideas (the myth and the scientific conception) is higher when the conflicting ideas are recognized as your own. In Part 1 I wrote about one of my favourite teaching techniques (the POE: Predict, Observe, Explain).[iv] It serves the same basic function. When presented with a situation, asking students to predict what will happen activates their prior knowledge and brings it forward to be challenged. It’s even better when you have them explain the reasoning behind their predictions. With a POE demo, the refutation comes when it doesn’t happen like they expected it would. In text, in comes from the refutation cue (“I hate to burst the bubble, but that’s simply not true”) and the scientific conception  presented by the author. The effect of activating the reader to think about their own prior conceptions can add to the chance that these refutations will work.

So let’s go back to Wilcox’s post for a moment. After the brief description of her mythbusting series that I copied above, there are two opening paragraphs that discuss organic foods generally and introduce the idea that there are a lot of myths out there about them. The one thing that might be missing, though, is a challenge to the reader to actually think about their own views, in other words a chance to activate their prior conceptions. I’ll admit it here: I was once (in what now seems like a past life) a vegan and committed to only eating natural foods. It’s taken a long time (and a lot of bacon) for me to sort through my conceptions of food and agriculture and to make sense of which ideas are supported by evidence and which are everyday notions that I still cling to. Wilcox’s mythbusting is directed exactly at someone like me and might be even more effective if those readers had an opportunity to bring their own ideas to the front of their minds to be recognized. From my own perspective, on the surface I don’t think that I subscribe to these myths anymore but I know deep down that there are pieces of them still there in the ways that I think. Good conceptual change activation would start by digging into these deeper patterns and challenging me to recognize where I too subscribe to some elements of these myths. One way might be to present a hypothetical decision making problem, for example asking the reader to examine fictitious statements from farmers at a farmers market on the topic of organic foods and decide which they would choose to buy from. This would ask the reader to commit, at least to a hypothetical degree, to their conceptions making them more likely to be challenged. When students have these opportunities in classrooms, they are more likely to change their minds towards more scientific conceptions.

So what hints are there in the conceptual change literature about writing to change people’s mind?

1. When challenging difficult myths and misconceptions, direct refutation seems to work best.
2. Refutations that are written in expository rather than narrative language seem to be both preferred and most effective.
3. Refutations are especially useful when they not only state common misconceptions but activate the reader to think about and commit to their own views before having them challenged.

Of course people and their ideas are very complex. None of these strategies will guarantee that any reader will change their mind. There are many other factors involved, including motivation, interests,  and social relationships that are built on shared beliefs and ideological commitments. One of the studies in Tippett’s review that surprised me the most, though, asked if the students who were more committed to their conceptions experienced less conceptual change. To my surprise, the researchers didn’t find any relationship. Students who were strongly and weakly committed to their ideas we just as likely to change their minds. Much more important was students’ understanding of science processes and scientific evidence. Those with sophisticated views of science were, not surprisingly, more convinced by scientific evidence[v] – adding weight to ongoing efforts to emphasize the processes of science both in schools and public science outreach. This relationship is important to remember as a public communicator. No matter how well written and clear your explanation, no matter how direct your refutation, readers struggling to understanding scientific evidence will more likely struggle to be convinced by it. Communicating about scientific ideas is difficult, doing it with the intent of changing people’s minds even harder, but I hope that some of the lessons learned in science education might offer a few strategies for making that road a little bit easier to travel.

This is also cross-posted at the Scientific American Guest Blog.