Occasionally, Facts Can Make a Difference


“Correcting widespread misperceptions about the level of federal science spending may help close the gap between the overwhelming majorities of Americans who support scientific research in the abstract, and the more meager numbers who back increased federal investment in research and development.” 

A longstanding question in science communication research is the extent to which inadequate science literacy explains the persistent divide between public belief and expert consensus on many scientific issues. Beginning several decades ago, a “deficit model” became the favored explanation for this divide. The model supposed
that negative public attitudes toward science were primarily caused by a gap in knowledge. As a corollary, it was argued that the main task of science communicators was simply to educate the public: as members of the public grew more knowledgeable, so the theory went, their beliefs and policy preferences would align with those of the scientific community.

The deficit model has come under scrutiny in recent years, and is now generally considered a poor explanatory model for people’s beliefs about science (e.g., NAP 2016, Box 5.2). A growing body of research suggests that knowledge gaps play an ancillary role in public attitudes toward science, with ideological disposition having a much greater impact. Indeed, it’s often been observed that people will dismiss or minimize evidence when it runs counter to their deeply held ideological beliefs–a cognitive defense mechanism known as motivated reasoning.

A Fresh Look at the Deficit Model

Against this backdrop, a recent study looked at the utility of the deficit model in the context of federal funding for science.

Americans consistently favor public funding for scientific research—support has hovered at around 80% for decades. However, when it comes to increasing the federal investment in science, support drops to under 40%. This suggests a disconnect with the scientific community, whose most prominent representatives consistently call for more federal resources devoted to research.

The study looked to see whether the deficit model was at play: do Americans have an accurate understanding of the current level of funding for science? If not, does simply supplying information alter their support? Or, have questions of federal spending become so politically charged that ideologically-driven motivated reasoning will impede efforts to build a political consensus for more funding?

To test this hypothesis, the researchers asked 1,000 participants in an online survey to estimate the level of federal spending on science research. In 2015, when the study was conducted, $61.5 billion was spent on non-defense research, from a budget of $3.8 trillion—around 1.6% of total federal spending. When probed, however, respondents reported a highly variable and generally inaccurate perception of the science budget. The median estimate was 10%, a more than five-fold overestimation of actual spending. Only one-third of respondents gave a reasonably close estimate (0–3%), and a quarter believed that spending on research consumed over 20% of the federal budget.

Given the disparate perceptions of federal largesse with respect to science, how did support for funding change when these perceptions were challenged? Did ideology influence their response to new evidence?

After providing an initial estimate, respondents were randomized to two different treatment groups. An experimental group was informed of the actual level of spending on research, while a control group received no additional information. Both groups were then asked whether they supported increased spending on research.

The results were striking. A strong correlation between perceived current spending and support for more spending was found in the control group: for every 10% increase in a respondent’s budget estimate, the median likelihood that they would support more spending dropped by 6 percentage points. At the extreme end—those who guessed that 50% of the budget went toward research—support fell to 20% or less.

Party affiliation and ideological makeup both influenced support for spending in the control group: Democrats were more likely to support increased spending, and ideological conservatives less likely. However, these partisan differences disappeared when respondents were provided actual budgetary numbers. Across the entire experimental group, support for increased spending consistently hovered around 60–65%, regardless of respondent’s initial estimate of the federal science budget. Among respondents who overestimated actual spending the most, providing accurate information increased their likelihood to support more spending dramatically—from 20% to nearly 60%.

Simple Knowledge Dissemination Works, in a Limited Sphere

The study suggests that certain issues are more amenable to a deficit model approach to knowledge dissemination. This appears to be particularly true when the opinion gap between the public and scientific community doesn’t revolve around a clash of values, but rather an incongruence of facts that can be readily remedied by injecting empirical evidence into a discussion.

Interestingly, it was recently noted at Making Science Public that the deficit model does not have much of an academic paper trail: it appears to be more of an intuitive assumption than a true model founded on empirical support. Ironically, then, the findings in this study represent one of the rare instances where the deficit model has been shown to work in practice.

Study: ­

Goldfarb JL and Kriner DL. Building Public Support for Science Spending: Misinformation, Motivated Reasoning, and the Power of Corrections. Science Communication. 2017;39:77–100.

Leave a Reply

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 )

Google+ photo

You are commenting using your Google+ 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 )

Connecting to %s