When attempting to communicate effectively with the public about a science-related debate, which is more important, framing the message or conveying science-based facts about the topic?  A forthcoming study (Word) at the Journal of Communication by Northwestern University researchers James Druckman and Toby Bolsen sheds new light on this long standing question.

As I will be highlighting at this blog, previous research consistently finds that the public typically form opinions in the absence of factual information, instead relying on mental short-cuts based on personal experience, values, and the selective presentation—or “framing”--of an issue. 

Frames influence perceptions and decisions because they focus on just one dimension of a complex topic over another, in the process communicating why an issue matters, why it might be personally relevant, and why a related action might lead to specific benefits or risks. 

Understanding and applying research on framing is particularly relevant to engaging the public on emerging issues such as nanotechnology and genetic engineering, but it also applies to communicating about entrenched policy debates such as climate change.  For example, in a recent study I conducted with Ed Maibach and colleagues, we find that when climate change is re-framed as a health problem rather than an environmental one, this re-interpretation is evaluated favorably and positively by a broad cross-section of Americans.  A frame in essence switches the train of thought for an audience, leading to a different set of attributions and conclusions.

Framing is an unavoidable aspect of human communication.  There is no such thing as unframed information.  On science-related issues, this idea is difficult to grasp for some advocates and scientists who still view communication through the lens of what scholars call the “deficit model” which assumes that opinion formation is a direct consequence of knowledge (or alternatively ignorance).  If the public only better understood the facts of a scientific topic they would more likely view the issue as scientists do and controversy would go away.

There is no question that deliberate decisions to selectively frame an issue can be used to deceive, but they can also be used to more effectively explain and engage audiences, boosting interest, attention, and learning.  As an example, in a recently published book chapter, I discussed the audience research approach that the National Academies used in framing the structure of a report on the teaching of evolution in schools (PDF). 

A second example is our recent research on the potential to re-frame climate as a health problem.  Not only does this new focus likely increase personal significance and relevance among Americans but it also communicates about objectively real and scientifically well-documented health risks that the public should know about.   It also starts to promote greater attention to adaptation policies and strategies--such as evacuation procedures, water and agricultural sanitation policies, improved housing, cooling stations during heat waves, and new transportation infrastructures-- that are needed to protect people and communities and that also result in healthier and higher quality lives.

Frames vs. Facts on Nano and GMOs

The forthcoming study by James Druckman and Toby Bolsen provides new understanding and data on how frames influence public perceptions and decisions about emerging areas of science.  Later today, I will post an interview with Druckman about his research (read it here).  In the rest of this post, I discuss and provide background on the study.

Druckman and Bolsen were interested in understanding whether the inclusion of facts--specifically reference to the findings of a scientific study---added additional power and influence to the framing of nanotechnology and genetically modified food. 

These issues--like many science debates--are often debated and talked about in terms of benefits and risks, typically in very general terms without much reference to actual scientific research.  Does providing more precise information about scientific findings relative to benefits and risks matter to public judgments or are more general assertions what really drive perceptions?

Importantly, Druckman and Bolsen also wanted to know whether after a frame was set on an issue, how did subjects then interpret information about scientific studies on the topic?  Were subjects open to re-considering their views or did they interpret the studies as fitting with their pre-conceived opinion?

On election day in 2008, Druckman and Bolsen assembled 20 teams of students to conduct exit polls of 621 voters in the Chicago region, querying voters on their perceptions of carbon nanotubes (CNTs) and genetically-modified foods (GMOs).  For the interviews, voters were randomly assigned to separate frame and issue conditions. 

For different groups of voters, CNTs or GMOS were defined using either a “fact” free frame or fact-based frame, with an emphasis on either benefits or risks.  In the case of CNTs, respondents were read the following introduction followed by one of the following frames, depending on their assigned experimental condition.  A similar method was used on GMOs (see the paper for more details):

One of the most pressing issues facing the nation—as has been clear from the election—concerns the limitations to our energy supply (e.g., with regard to coal, oil and natural gas).  One approach to addressing this issue is to rely more on carbon nanotubes or CNTs. CNTs are tiny graphite with distinct chemical properties. They efficiently convert sunlight into electricity, and thus, serve as an alternative to coal, oil, and natural gas. The uncertain long-term effects of CNTs are the subject of continued study and debate.

Fact Free Benefits of Nanotechnology

Most agree that the most important implication of CNTs concerns how they will affect energy cost and availability. A recent study on cost and availability showed that CNTs will double the efficiency of solar cells in the coming years.

Fact Free Risks of Nanotechnology

Most agree that the most important implication of CNTs concerns their unknown long-run implications for human health. A recent study on health showed that mice injected with large quantities of CNTs reacted in the same way as they do when injected with asbestos.

Fact-based Benefits of Nanotechnology

A recent study on cost and availability showed that CNTs will double the efficiency of solar cells in the coming years.

Fact-based Risks of Nanotechnology

A recent study on health showed that mice injected with large quantities of CNTs reacted in the same way as they do when injected with asbestos.

Subjects were asked on a 7 point scale to express their level of support or opposition for either CNTs or GMOs depending on the frame condition.  Other questions measuring pre-existing levels of scientific knowledge, demographic background, and generalized views about science were asked and used as controls in their analysis.

After completing the interviews on Election Day, participants were re-contacted 10 days later.  At this time, depending on their frame condition, they were presented with reminder information about CNTs and GM foods.  Then, for each technology, respondents evaluated the “effectiveness” of three distinct factually based scientific studies “in providing information or making an argument” (on 7-point scales with higher scores indicating increased effectiveness). Respondents also rated the extent to which each study opposed or supported the technology (on 7-point scales with higher scores indicating increased effectiveness), and re-reported their overall support for each technology.

In their analysis Druckman and Bolsen find that on both issues, frames that emphasize general statements about either the benefits or the risks of the technology influenced perceptions, with slightly stronger effects for the risk conditions.  Yet importantly, when frames included “facts” that reference specific findings from scientific studies, the inclusion did not significantly enhance the power of the frames. 

As they conclude, contrary to the strong emphasis of the deficit model on promoting scientific literacy as the key to communication, referencing facts adds little to how framing influences individuals’ opinions about new technologies.

Finally, when they analyzed the results of the follow-up interviews, they found that how subjects evaluated information presented about research on the technologies was strongly influenced by their pre-existing views.  In other words, much like a strong Democrat or Republican tends to view even favorable news coverage as reinforcing their pre-existing opinion of an opposing party's candidate, when subjects were presented with information about scientific studies on GMOs and CNTs,  they tended to re-interpret these studies in line with their pre-existing judgments.

Here is how Druckman and Bolsen describe the significance of their study:

We find at every stage of the decision-making process, the processing of factual information is fraught with imperfections. First, facts have limited impact on initial opinions—no greater than alternative considerations including values and perceptions about science credibility (also see, e.g., Scheufele & Lewenstein 2005). Second, we find that when provided with frames that lack factual information and frames that include facts, individuals do not privilege the facts (also see, e.g., Nisbet & Mooney, 2007). Facts do not enhance frame strength (although facts do have effects equivalent to that of frames without facts). Third, once they form initial opinions, individuals process new factual information in a biased manner (also see, e.g., Kahan et al., 2008). Specifically, they view information consistent with their prior opinions as relatively stronger and they view neutral facts as consistent with their existing dispositions.

Of course ours is just one study on two particular technologies, and as a result, caution needs to be taken in generalizing. It does seem clear, however, that factual information is not always as it appears (to a neutral observer). Our results suggest that the best route to facilitate reasonable opinion formation may be to provide alternative ways of thinking about new technologies—that is, different frames—and then to encourage individuals to weigh these frames against one another. Under distinct circumstances, facts may play a more salient and less biased role. Indeed, there undoubtedly are situations where facts matter and this could depend on a range of factors including contextual elements, individual motivations, and precise presentation.[i] For example, in his study of framing and nanotechnology, Cobb (2005) reports that adding explicit statements about health risks and/or benefits to relatively weak philosophical frames (that discuss the role of science) significantly increase the impact of the frames.

It may be that different facts—such as evidence based on an accumulation of scientific studies (e.g., meta-analyses)—may have more of an added impact. Moreover, while we have incorporated competition and time into the study of opinion formation, we have not done so explicitly with regard to facts. Facts themselves are contestable (e.g., Shapiro & Block-Elkon, 2008; Katz & Strange, 1998), as is made clear by the growing literature on how perceptions of scientific studies depend on economic and political pressures (Pielke, 2004; Jotterand, 2006). Additionally, individuals often receive related facts over time. The impact of “fact competition” and over-time exposure to facts may matter; for example, repeated exposure to facts might moderate motivated reasoning.[ii] Only once we explore these types of variations can general statements about factual information be made. We also encourage future work to further probe the factors that enhance frame strength and explore the relationship between competing frames and motivated reasoning. Most important is to continue expanding studies of opinion formation to account for the realities of competition over time.


Takeaway: Whether it is politics or science, audiences rely heavily on information short cuts to reach judgments. As part of this process, research finds that how a particular message "frames" or selectively defines an issue strongly shapes public opinion. On science topics, frames of reference that make general assertions about benefits and risks often dominate debate.

A forthcoming experimental study using as subjects exit voters in the 2008 election looked at whether specific reference to the findings of scientific studies made a difference to the influence of risk/benefit frames about nanotechnology and GMOS.

More generalized frames--absent direct reference to scientific findings--were found to be more impactful than frames that included such detailed emphasis. Scientists, their organizations, and many advocates dedicate significant resources to emphasizing scientific findings with the goal of engaging the public or quelling controversies. For the wider public--who are generally unmotivated to pay attention to this type of detailed scientific emphasis--a more effective strategy is to rely on generally framed messages about benefits or values that while scientifically accurate do not feature technical details.


Druckman, J. & Bolsen, T. (in press). Framing, Motivated Reasoning, and Opinions about Emergent Technologies. Journal of Communication.

Maibach, E., Nisbet, M., Baldwin, P., Akerlof, K., & Diao, G. (2010). Reframing climate change as a public health issue: an exploratory study of public reactions BMC Public Health, 10 (1) DOI: 10.1186/1471-2458-10-299

See also:

Study: Reframing Climate Change as a Public Health Issue.