Media sources,credibility,and perceptions of science: Learning about how people learn about science

Abstract

Knowledge about science and technology has become increasingly important in this age of digital information overload. It is also becoming increasingly important to understand what contributes to scientific learning, including information sources and trust in those sources. In this study, we develop and test a multivariate model to explain scientific knowledge based on past theories on learning from the news from the fields of political communication, sociology, and media psychology. We focus on the impact of sources—by platform, such as television and online, and by expertise, such as scientists and the media—in understanding what predicts scientific knowledge. The results show that interest in science not only directly predicts knowledge but also has indirect effects on knowledge through its effects on Internet use, confidence in the press, and perception of scientists. In addition, distrust on the news sources is an important pathway to learning about science.

Keywords

media and science popularization of science public understanding of science scientific literacy

The average citizen is confronted with an array of social, political, and environmental issues involving complex pieces of information he or she might not have learned during formal education. These issues, however, affect the public. Therefore, the average citizen is expected to make sense of, for example, climate change, hydraulic fracturing, and government support on space exploration, among others. Only a few adults have formal education in stem cell research, and “yet we expect citizens to make sense of competing arguments on this question and to render a political judgment when they cast a vote or support an advocacy group” (Miller, 2010: 192).

This is a scientific, technological, and risky world we live in (Beck, 1992; Falk and Needham, 2013; Miller, 2010). This has been illustrated by the numerous natural and man-made crises the world has witnessed over the years: from the deadly, evolving, and transnational bird flu strain that flew into pandemic status in June 2009, to the Fukushima nuclear crisis that shook Japan after an earthquake in March 2011, to the great flood brought by Hurricane Sandy that swept New York in October 2012, and to the powerful winds blown by Typhoon Haiyan when it pummeled the Philippines in November 2013. This is also seen in the new ways that people try to understand these complex issues: by relying on the technology of the Internet (Falk and Needham, 2013; Miller, 2010). The public, in order to make sense of all of these, needs to have an understanding of science.

Knowledge about science, scientific process, and science institutions are considered necessary (although not sufficient) conditions for the development of positive scientific attitudes. Research on the public understanding of science has attempted to explain the differences in people’s perceptions and knowledge about scientific issues with a wide variety of results. Studies in this domain have accounted for some of the differences in people’s understanding of science, from individual-level variables, such as education, political ideology, and gender, to macro-level variables such as wide-spread anti-intellectualism, changes in the political economy of the media, and trust in institutions, among others (Bauer et al., 2000). Research also shows that scientific literacy and knowledge of environmental issues remain relatively low in the United States (Bord et al., 2000; Nisbet and Myers, 2007; Nisbet et al., 2002).

Ungar (2000) argued that ignorance about science and scientific issues is a function of the knowledge-ignorance paradox. The paradox suggests that the over-saturation of the information environment has created specialized knowledge groups and increased the level of overall ignorance about scientific issues. This paradox is explained by factors such as distrust in institutions, the cutback to knowledge-generating institutions (i.e. museums), a wave of anti-intellectualism, as well as a rise in the entertainment media landscape. Ungar (2000) argued that the processes explaining the paradox “create an attention economy that favors celebrity facts and militates against the acquisition of scientific knowledge” (p. 308). This paradox is opposite to some research that suggests that number of sources of information predicts scientific knowledge about specific issues (Kahlor and Rosenthal, 2009).

If knowledge about science and technology has become increasingly important in this age of digital information overload, it is also becoming increasingly important to understand what contributes to scientific learning under this changing information environment, especially among adults who are already out of formal education. We do not attempt to advocate for an information-deficit model approach, but our interest is to further our understanding of the factors that explain scientific knowledge. In this study, we develop and test a multivariate model to explain scientific knowledge, based on previous research examining selective media exposure, deference to scientific authority, and trust in information sources (Brossard and Nisbet, 2007; Eveland, 2001; Eveland et al., 2004; Jang, 2014; Sundar, 2008; Ungar, 2000, 2008), and the literature on what accounts for levels of knowledge about science (Falk and Needham, 2013; Malka et al., 2009; Nisbet et al., 2002), using data from the General Social Survey (GSS). This study borrows perspectives from three main paradigms in public understanding of science: science literacy (i.e. knowledge), public understanding (i.e. attitudes), and science and society (i.e. trust) (Bauer et al., 2007). We focus on the impact of sources—by platform, such as print, television, and online, and by expertise, such as scientists and the media—in understanding what predicts scientific knowledge. In doing so, we contribute to the growing and important discourse on how people learn about science by focusing on sources of information and how people feel about individuals who disseminate this information.

1. Literature review

Science knowledge and perceptions

The public, in general, has a positive appreciation of science. A 2015 survey in the United States found that 79% believed that science had a positive effect on society (Pew, 2015), although this is down from 84% in 2009 (Pew, 2009b). This appreciation of science is matched by a moderate knowledge of science, with about 65% of participants able to correctly answer about 67% of a 12-item science knowledge quiz (Pew, 2009b). Most participants correctly answered questions about things they use in daily life, such as those about aspirin and global positioning system (GPS).The incorrect answers, in contrast, were mostly about more complex science questions, such as those about stem cells, electrons, and lasers (Pew, 2009b). This appears consistent with what Miller (2010) found to be an increase in the last 20 years of American adults’ civic scientific literacy. This refers to knowledge that an individual would need “to make sense of current and emerging science policy issues” (Miller, 2010: 197). From only 10% in 1988, the percentage of American adults who are civic scientific literate increased to 28% in 2008 (Miller, 2010).

However, the politicization of science (Gauchat, 2012) and a recent wave of anti-intellectualism in the United States have been blamed for the lack of support for emerging scientific technologies (e.g. nanotechnology, stem cell research) (Mooney, 2012). Despite recent evidence discussing the deference to scientific authority in the United States (Brossard and Nisbet, 2007), some evidence suggests an increasing criticism to science and scientists. For example, in the case of climate change, recent studies have reported that a small number of the American public started questioning the scientific consensus on climate change (Koteyko et al., 2013; Maibach et al., 2013). The distrust in science, institutions and scientists is immersed in a reciprocal relation with people’s knowledge of science.

The measurement of science knowledge has been the focus of considerable research and discussion, from both a conceptual and a measurement perspective (Bauer et al., 2000). Studies that have focused on the factors that explain science knowledge vary on their results, but they nevertheless provide important insights on some common predictors. Bauer (1996) argued that in the context of the knowledge-ignorance paradox, ignorance is distributed based on socio-demographic variables.Most studies report important and significant differences in science knowledge based on such demographic variables, specifically education and gender (Kahlor and Rosenthal, 2009; Zhao, 2009).For example, men and those with more years of formal education report higher levels of perceptions and knowledge in science and technology (Falk and Needham, 2013; Nisbet et al., 2002), a finding that seemed to have persisted across time, as an early study on scientific knowledge had also found that men with better education and higher income were more likely to seek science information from more than one source (Wade and Schramm, 1969).

Scientific knowledge is important because it can partly explain environmental concern, behaviors, and trust in institutions and individuals (Malka et al., 2009). Similarly, scientific knowledge is related to information-seeking behaviors. For example, knowledge about nanotechnology mediated the influence of web science use and newspaper science use in the support of nanotechnology (Lee and Scheufele, 2006). A study found that scientific knowledge played a minor role in shaping people’s attitudes toward stem cell research, with its effects on attitude moderated by people’s value predispositions (Ho et al., 2008). Indeed, scientific knowledge can be a strong predictor of attitudes toward science, but this knowledge–attitude link is mediated by contextual factors, such as demographic variables (Sturgis and Allum, 2004). Nevertheless, studies have found that knowledge is an important factor in explaining behavioral intentions, such as taking voluntary actions and supporting policies related to climate change (Bord et al., 2000), as well as future information-seeking actions (Zhao, 2009).

Science knowledge and media use

Contrary to the stereotype of scientists being alone in laboratories without contact with the outside world, scientists themselves are also concerned about media coverage of their work not only out of altruism and believing in their work but also because the legitimacy conferred by the news media matters to scholars and academic institutions (Tsfati et al., 2011). Despite their critical view of the media, scientists’ own personal experience with journalists has been mostly positive (Besley and Nisbet, 2013). A majority of scientists also believe that a lack of interest in science news contributes to limited public knowledge (Pew, 2015).

In general, the news media do not devote the same amount of attention they devote to politics and crime to coverage of science and technology (see Baker et al., 2012, for a recent example). This likely corresponds with the lack of newsworthiness of the topics, challenges faced by journalists in reporting scientific issues, and journalists’ perception of the news audience’s literacy level that discourages journalists from writing about science topics that they believe will not increase readership (Baker et al., 2012). This last explanation is ironic, considering the amount of work that found the news media as an important agent of learning about complex issues such as climate change (Kahlor and Rosenthal, 2009; O’Neill and Nicholson-Cole, 2009).

In the field of political communication, the availability of media options has been found to increase political knowledge and involvement among the politically engaged through selective exposure, although the same availability increases entertainment media consumption for those already disengaged, increasing the chances of completely tuning out of politics (Prior, 2005). Jerit et al. (2006) argued that differences in knowledge related to education became more pronounced “in environments in which information is plentiful” (p. 278). More specifically, studies found that newspaper use is more efficient for educated people, while TV use is more efficient for the less educated (Kleinnijenhuis, 1991; Wade and Schramm, 1969).

Selective exposure refers to the “behavior that is deliberatively performed to attain and sustain perceptual control of particular stimulus events” (Zillmann and Bryant, 2008, p. 2). In the context of media use, selective exposure refers to the selection of media outlets that match people’s beliefs and predispositions (Stroud, 2008, 2010). Recent scholarship in this area points toward the effect of increasing political polarization and ideological argumentation on selective exposure and avoidance of counter-attitudinal information, something considered as a threat to deliberative democracy (Garrett, 2009; Garrett et al., 2013; Garrett and Stroud, 2014). This phenomenon has become more salient with the increased availability of information via the Internet, but more research is needed to determine inconsistencies found in different cultural contexts (Trilling and Schoenbach, 2014). What the concept of selective exposure highlighted is the idea that audiences can be active in their use of media. Indeed, Rubin and Perse (1987) argued that media use can be either ritualized or instrumental. Ritualized media use refers to a less intentional and non-selective orientation focused mostly on the medium rather than the content. Instrumental media use refers to intentional and selective exposure to specific content. These media uses have also been reported in Internet use (Papacharissi and Rubin, 2000).

But people interact with science information differently from how they interact with political information (Jang, 2014). In a study examining selective exposure related to various scientific issues, Jang (2014) reported that individuals who perceived that their science knowledge was sufficient were more likely to display confirmation bias. That is, people tended to seek information that challenged rather than supported their positions. This was present in genetically modified organisms (GMOs) and stem cells, but not in evolution and global warming. The study suggested that selective exposure to certain media (i.e. television) about science issues such as global warming results as a by-product of channel selection, not as a tendency to seek skeptical views on the issue. On the other hand, when examining online media, individuals have more freedom to seek science information and might not be attracted to congruent news. It is plausible that selective exposure to media sources of scientific information then influences factual scientific knowledge.

Indeed, some studies found significant and positive effects of news attention on knowledge about controversial science issues (Ho et al., 2008; Lee and Scheufele, 2006) and on more accurate science-based beliefs (Zhao et al., 2011). However, attention specifically to political news is associated with less accurate science-based beliefs (Zhao et al., 2011). In terms of learning about climate change, the use of individual media was not a significant predictor of knowledge, but the number of news media sources and prior information-seeking efforts were among the strongest predictors (Kahlor and Rosenthal, 2009). These competing findings suggest a need to further explore the relation between media consumption and scientific knowledge by examining possible mediators such as trust in the sources of information.

Many people do not seek information across all possible sources, however. For example, most of today’s youth get their news from online sources (Beaujon, 2012). The consumption of hard news, in general, also increases political knowledge (McLeod et al., 1996). But the type of medium can also explain knowledge and engagement in politics. Indeed, use of specific media, such as newspapers (Chaffee and Kanihan, 1997; Eveland et al., 1998; Kim, 2008), television news (Hively and Eveland, 2009), and the Internet (Kim, 2008; Xenos and Moy, 2007) were found to increase political knowledge across different ages. For scientific knowledge, newspapers were found to be the dominant source in a media environment composed of only traditional media outlets (Wade and Schramm, 1969). Falk and Needham (2013) found that more than formal schooling, “reading books and magazines about science and technology, using the Internet, and watching science related documentaries and videos were collectively” the strongest predictors of self-reported knowledge about science and technology (p. 446). In terms of civic science literacy, Miller (2010) found that Internet use and print media use are significant and positive predictors of knowledge. In contrast, television use had a negative impact on science literacy (Miller, 2010), consistent with the finding of Nisbet et al. (2002) that TV obstructed appreciation for science. In the context of science communication, Maier et al. (2014) suggested that the role of the Internet in allowing individuals to choose their media consumption at a very specific level makes selective exposure more relevant than ever.

Interest in science

Our proposed model of learning about science is informed by previous research that examines media use based on motivated reasoning. Eveland (2001) argued that

news attention and elaboration of news content are under the motivational control of the individual, and a motivation to be aware of the happenings in one’s environment and to have information available for decision making will lead individuals to engage in these effortful behaviors to obtain the desired results. (pp. 576–577)

This is particularly salient in understanding media use for science information. Since the news media do not regularly report on science news (e.g. Baker et al., 2012), those interested in science seek particular media sources for the information they are interested in.

Nisbet et al. (2002) found that people who attained higher levels of education not only tend to know more about science than those with lower educational attainment, but they also read newspapers more often and view television less often. Thus, demographic background, such as gender and educational attainment, tend to account for media use patterns and knowledge about science. Guided by the evidence presented in previous works on ritualized and instrumental media use (Rubin and Perse, 1987), and selective exposure (Stroud, 2008, 2010) discussed above, the first part of our proposed model includes interest in science as a possible predictor of using particular media sources and knowledge about science. Exposure to certain mediums should be related specifically to the content and therefore serve an instrumental function. This is consistent with the finding of Bauer et al. (1994) that public interest in science is linked to knowledge of science and technology. Thus, we ask the following question:

RQ1. What is the relationship between interest in science and seeking information from particular media sources?

Credibility of information sources

A way to navigate an oversaturated information environment is to rely on shortcuts to determine which pieces of information to believe in and which to dismiss. This is especially true among young people who depend on the Internet for information (Sundar, 2008). A group of cues that young people use to evaluate information is agency, which refers to the source of the information (Sundar, 2008). Information about science can be considered as coming from two main sources: scientists and the media who report about them. This is the second part of our theoretical framework, drawing from theories on the effects of source perceptions (Petty et al., 2009; Petty and Cacioppo, 1986; Sundar, 2008; Ungar, 2000).

For this study, an assumption of interest is the role of source trustworthiness. Priester and Petty (2003) found that information from sources with low trustworthiness “engender greater elaboration” than information from sources with high trustworthiness (p. 413). This is because individuals want to hold what they perceive to be correct attitudes, and when a source is positively perceived, individuals become confident of the information from this source (Petty et al., 2009). This is consistent with the so-called authority heuristic used to evaluate the credibility of a source. “The authority heuristic is likely to be operational whenever a topic expert or official authority is identified as the source of content” (Sundar, 2008: 84).

It is logical to assume that positive perceptions of the source will increase learning because people proceed to trust the information from this source. For example, Ungar (2000) argued that an explanation for the knowledge-ignorance paradox, among other factors, was the distrust in institutions. A Gallup poll showed that 57% of Americans have little or no trust in the media—a record high since the group started polling in 1973 (Morales, 2010). In a 2009 survey, the Pew Research also found that 63% of Americans agreed that news stories had been inaccurate and 60% said news organizations were politically biased—the worst marks since 1985 (Pew, 2009a). Other public institutions too—the executive, the legislative, and even the judiciary—are distrusted by Americans (Gronke and Cook, 2007; Pew, 2009a). The economic design of the news media has made them vulnerable to economic pressures difficult to disentangle from political influence (Barendt, 1994; Davis and Craft, 2000). It is not surprising, therefore, whether citizens see the news media no longer as a watchdog as what it was ideally perceived but as an institution “enmeshed with the other national institutions” it was supposed to monitor (Gronke and Cook, 2007). Thus, the Pew (2009a) survey also showed that only 20% believed news organizations are independent of powerful people and organizations.

The mushrooming of blogs and other related sites brought about by the new media provides the public with alternatives. This leads to a declining reliance on the mainstream news media and an exposure to diverse perspectives and information online (Tsfati, 2010). It has become easy for the public to verify what the mainstream news media say. It has also become easier for partisans to expose themselves only to information sources that are consistent with their views, that a collection of contradictory messages in a supposedly objective account by the mainstream news media becomes alienating, if not repulsive (Stroud, 2008, 2010). Therefore, particular forms of media use can influence how users perceive journalists. In this respect, we ask the following questions:

RQ2. How do particular media uses influence how people perceive journalists?

RQ3. What is the relationship between interest in science and audiences’ perception of journalists?

Scientists, in contrast to journalists, seem to enjoy favorable perceptions by the public. A 2009 survey found that 70% of Americans believed that scientists contribute a lot to the well-being of society (Pew, 2009b). In the same survey, only 38% had a favorable image of journalists (Pew, 2009b). In this respect, several studies have attempted to explain the variance in the levels of trust in scientific sources. For instance, Brewer and Ley (2013) examined the factors influencing levels of trust in scientists, environmental organizations, and media, finding that the media have the lowest levels of trust. Males, those more educated, those with higher income, and those who trust scientists, tend to have lower levels of trust in the media. They also suggest that trust in scientists can increase trust in sources of science information. On the other hand, deference to scientific authority and the use of public affairs media predict trust in governmental institutions and scientists (Anderson et al., 2012). Science media use and knowledge about nanotechnology predicted trust in scientists (Anderson et al., 2012). Attitudes, more than perceived knowledge, explain most of trust in science and technology (Roberts et al., 2011). In addition, trust in scientists mediates the relationship between media use and perceptions of scientific issues, such as climate change (Hmielowski et al., 2013; Leiserowitz et al., 2013). It is therefore important to understand predictors of the public’s trust in scientists. However, the relationship between instrumental use of media sources related to science and people’s perception of scientists has received little attention in the past. Therefore,

RQ4. How do particular media uses influence how people perceive scientists?

H1. Interest in science will be associated with a positive perception of scientists.

Theoretical framework

This review described the different factors that have been previously examined in different models that explain knowledge about science. This discussion and the first four questions we had posed provide the building blocks for the theoretical model we are proposing in the study and which we will explain below that seeks to expand previous efforts in this area of research.

In this study, we attempt to improve our understanding of the predictors of scientific knowledge, with a special emphasis on the role of media sources, interest in science, credibility of media sources, and perceptions of scientists. Previous research has documented the relationship between knowledge gaps and media use and education (Eveland and Scheufele, 2000). Although theories that incorporate the effects of source perceptions (e.g. elaboration likelihood model) focused on effects on attitude and credibility evaluations (Petty et al., 2009; Petty and Cacioppo, 1986; Sundar, 2008), this study seeks to test an extension of the effects of source perceptions on learning about science. Specifically, we argue that the effects of media use on knowledge about science are also affected by how the sources of information are perceived by individuals. We also bring into the model the effect of interest in science, which has been reported to be positively influenced by exposure to science shows (Retzbach et al., 2013). In this study, we examine the same relationship, but under the assumption that existing interest in science (explained by other factors such as education) heightens seeking and exposure to science information. Therefore, to complete our proposed theoretical model, we also test the following hypotheses:

H2. Media use will be positively associated with scientific knowledge.

H3. Interest in science will predict scientific knowledge.

H4. Having a positive perception of scientists will predict scientific knowledge.

H5. Having a positive perception of journalists will predict scientific knowledge.

These research questions and hypotheses are represented in our proposed theoretical framework (see Figure 1).

Figure 1.

Theoretical framework.

2. Method

In order to test the proposed theoretical framework, we used the 2012 GSS conducted by the National Opinion Research Center at the University of Chicago. It is a nationally representative survey of non-institutionalized adults in the United States conducted almost annually since 1972 and asks questions on a range of topics on social characteristics and attitudes. It also uses face-to-face interviews to gather data from its full-probability sample.

The 2012 survey has 1974 respondents, but for the purposes of this study, we used data from a sub-sample of 949 respondents, excluding respondents who were not asked about the variables we are analyzing. Since the GSS asks questions in rotation (for a full explanation of the GSS methodology, please visit www3.norc.org/GSS), almost half of the respondents in the original sample were not asked to report their main source of information about science, aside from other questions. However, we made sure that our sub-sample remained comparable to the original sample in terms of demographics. The average age for our sub-sample is 48.19 years (standard deviation (SD) = 17.7). About 49% finished high school as their highest degree. Some 75% were White Americans, while 15% were Black Americans. Some 45% were males (see Table 1). These characteristics of our sub-sample (n = 949) are very similar to the demographic profile of the original sample (N = 1974).

Table 1.

Demographics.

	Original sample (N=1974)	Study’s sub-sample (N=949)
Age, years	49.39 (SD=17.7)	48.19 (SD=17.7)
Educational level	48.6% high school	49.4% high school
Race	73.7% White Americans	74.8% White Americans
Sex	46.2% males	44.8% males
Political identification	18.2% strong Democrats	19.1% strong Democrats
	9.8% strong Republicans	9.1% strong Republicans
	19% independent	18.3% independent

The characteristics of our sub-sample (n=949) are very similar to the demographic profile of the original 2012 sample (N=1974).

Variables

Interest in science

This scale is based on three items. Respondents were asked to rate how interested they were in space exploration, new scientific discoveries, and technologies. They had to rate each item in a 3-point scale: from very interested, to moderately interested, to not at all interested. The scale is reliable, Cronbach’s alpha = .77.

Source of science information

The survey also asked respondents to indicate their main source of information about science and technology. There were nine possible categories: newspapers, magazines, the Internet, books, television, radio, government agencies, family, and friends. Since in the original sample only 27 respondents chose radio and only one respondent chose the government, we excluded them from the analysis. We reduced the remaining seven items into the following: traditional print news media (magazines and newspapers), books, television, interpersonal communication (family and friends), and the Internet. We dummy coded each category for purposes of regression analysis. The Internet was the main source for 41% of the respondents. TV was a close second with 37.6%.

Confidence in the press

The respondents were asked how much confidence they have on the press in general. They had to rate the item in a 3-point scale, from a great deal, to only some, to hardly any. Some 10% of the respondents said they have a great deal of confidence in the press.

Perception of scientists

This scale is based on three items rated on a 4-point Likert scale from strongly agree to strongly disagree. The respondents rated their level of agreement to the following: Scientists want to make life better for the average person, scientists work for the good of humanity, and scientists help solve problems. The scale is reliable, Cronbach’s alpha = .72.

Factual science knowledge

This scale is based on 10 items that tested the factual knowledge of the respondents about science. These questions, answerable by true or false, have been validated in earlier GSS surveys and even in other surveys as a measure of knowledge about science. The scale is based on the mean score across 10 questions to be able to adjust for missing values. The respondents were asked to indicate whether each of the following statements was right or wrong: Lasers work by focusing on sound waves, antibiotics kill viruses as well as bacteria, all radioactivity is man-made, it is the father’s gene that decides whether the baby is a boy or girl, electrons are smaller than atoms, the earth goes around the sun, the center of the earth is very hot, human beings are developed from earlier species of animals, the universe began with a huge explosion, and the continents have been moving. The first three items were wrong, while the rest were correct.

3. Results

RQ1 asked about the relationship between interest in science, a measure of motivation, and seeking information from particular media sources. A bivariate correlations test found two significant relationships. First, those interested in science tend to seek information about science from the Internet, r(945) = .127, p < .001. Second, interest in science is negatively correlated with seeking information from television, r(945) = −.153, p < .001.

Having established the media sources that have significant relationships with interest in science, we then proceeded with an operational framework that specifies media sources from the theoretical framework we had earlier proposed (see Figure 2). This operational framework also allows us to test an entire model using path analysis, a variant of structural equation modeling (SEM), in analyzing the data.

Figure 2.

Operational framework.

Path analysis

Path analysis allows visual modeling and testing of the relationships between variables of interest. It provides a graphical way to represent an assumed theory as well as an empirical estimate about the relationships within a proposed framework (Byrne, 2010). Thus, rather than arguing for causality, Meehl and Waller (2002) proposed a focus on the notion of “verisimilitude” or closeness to the truth when dealing with the results of path analysis. Path analysis tests a proposed structural model by simultaneously testing direct and indirect relationships specified among variables within the proposed model.

Guided by the assumptions of science learning, source selection, and trust discussed above, we proposed a theoretical framework that accounts for knowledge in science among adults (see Figure 2). Since the literature found that age, gender, and educational level affect knowledge levels, we also controlled for these in our path analysis model, which we ran using Mplus Version 6.11.

The initial operational model fits the data well, χ²(15) = 18.11, p > .05; root mean square error of approximation (RMSEA) = .015, PCLOSE = .99; comparative fit index (CFI) = .99; Tucker–Lewis index (TLI) = .99, and standardized root mean square residual = .019 (see Figure 2). For model parsimony, non-significant pathways were excluded to come up with the final model (see Figure 3 and Table 2).

Figure 3.

Final structural model.

Table 2.

Parameter estimates and factor loadings.

From	To	β	p
Educational degree	Interest in science	.181	<.01
Sex		−.211	<.01
Age	TV as main source of science information	.195	<.01
Educational degree		−.335	<.01
Interest in science		−.084	<.01
Age	Internet as main source of science information	−.223	<.01
Educational degree		.069	<.01
TV as source		−.579	<.01
Internet as source	Perception of scientists	−.163	<.01
Interest in science		.097	<.01
Internet as source	Confidence in the press	−.063	.051
Interest in science		.068	<.05
Interest in science	Factual knowledge about science	.170	<.01
Internet as source		.104	<.01
Educational degree		.324	<.01
Sex		−.060	<.05
Perception of scientists		−.156	<.01
Confidence in the press		−.073	<.05

RMSEA: root mean square error of approximation; CFI: comparative fit index; TLI: Tucker–Lewis index; SRMR: standardized root mean square residual.

Final model that emerged after non-significant pathways was excluded; χ²(15)=18.11, p>.05; RMSEA=.015, PCLOSE=.99; CFI=.99; TLI=.99; SRMR=.019.

RQ2 asked how particular media sources influence perception of journalists. Based on the final model, Internet use was found to negatively predict confidence in the press (β = –.06, p = .05). RQ3 asked about the effect of interest in science on perceptions of journalists. The analysis showed that interest in science positively predicted confidence in the press (β = .07, p < .05).

RQ4 asked how particular media sources influence perception of scientists. The final model showed that Internet use was also negatively related to trusting scientists (β = –.16, p < .01). H1 predicted that interest in science will lead to positive perceptions of scientists. This is supported (β = .10, p < .01).

H2 predicted the direct effects of media sources on scientific knowledge. Internet use was found to be a positive predictor of scientific knowledge (β = .10, p < .001). In terms of the control variables, males had higher scores on scientific knowledge than females (β = −.06, p < .05) while those who have higher educational degrees also knew more (β = .32, p < .01).

H3 predicted that interest in science will lead to scientific knowledge. This is supported (β = .17, p < .001). Finally, H4 and H5 predicted that having positive perceptions of information sources about science—scientists and journalists—will lead to learning more about science. Both hypotheses are rejected. There is a negative relationship between confidence in the press and scientific knowledge (β = −.07, p < .05). Similarly, those who trust scientists scored lower on the scientific knowledge scale (β = −.16, p < .01).

4. Discussion

The goal of this study was to advance our understanding of the complex pathway to learning about science, especially as we live in an increasingly science and technology–driven world. Studies in this area examined above have explored the relationship between media use and science knowledge, but have not fully examined the role of media sources and the impact of people’s trust in such sources. By combining various theoretical and conceptual perspectives on learning from the media, we proposed a theoretical framework that looks at the pathways from demographic variables, interest in science, sources of science information, perception of news sources, to knowledge about science. Our analyses revealed a complex relationship between the variables we tested. We found that males and those with higher educational attainment tend to know more about science, consistent with previous studies (Falk and Needham, 2013; Wade and Schramm, 1969). They are also more interested in science, which leads them to use the Internet to search for science information. This Internet use leads to lower confidence in the mainstream press and a less favorable perception of scientists. Being critical of these information sources finally leads to higher knowledge about science.

The final model uncovers three interesting findings. First, we found that interest in science not only directly predicts knowledge but also has indirect effects on knowledge through its effects on Internet use, confidence in the press, and perception of scientists. This is consistent with the argument that motivation is an integral initial step in the complex pathway to learning (Eveland, 2001).

Second, it appears that people interested in science turn to the Internet for science information, instead of using traditional news media platforms such as newspapers, television, or even books. In contrast, interest in science is negatively related to using television as source for science information. This is consistent with earlier work about the relationship between television use and science literacy (Miller, 2010; Nisbet et al., 2002). Indeed, recent research also noted how the rise of the Internet is displacing traditional sources of science information, such as books and even science museums (Segev and Baram-Tsabari, 2012). Future studies should explore the implications of this apparent switch to the Internet for people who seek information about science. It is possible that the speed, convenience, and the amount of information available online are important incentives for people curious about scientific developments. However, this could also be a symptom of a general distrust on traditional media as sources of information (Anderson et al., 2012; Brewer and Ley, 2013). Traditional news organizations have also created an online media presence. However, the Internet offers a wealth of information from both traditional and non-traditional sources of information. Scientists themselves can now directly interact with the public through blogs and social media. In addition, future studies should consider issue attributes, as previous research in science and health communication suggests that the role of the Internet can differ based on the issue and other contextual factors (Jones and Biddlecom, 2011; Koch-Weser et al., 2010). For example, in the case of climate change, individual media use did not predict scientific knowledge (Kahlor and Rosenthal, 2009).

Third, a contribution of our study to our current understanding of how people learn about science from the media is our adoption of the influence of source perceptions. Drawing from contrasting theories on the effects of source perceptions, we found that distrust on the news sources was an important pathway to learning about science. Those who have lower confidence levels in the press and who have less favorable attitudes toward scientists tend to score higher in the factual science quiz administered in the survey. A plausible explanation comes from conceptualizing source perceptions as a heuristic (Sundar, 2008). When people trust the press and the scientists being quoted in the press, they use that trust as a heuristic to believe the information they are presented. This leads them to skip a deeper processing of that information or what others would call critical thinking. It is also possible, however, that learning about science is a cycle. Those who know more about science tend to be critical of the information they encounter, and hence critical of the sources of this information. This leads them to have less favorable attitudes toward the press and scientists in general. Being critical of these sources leads them to a careful processing of the information they get, prompting them to access other sources of information, which leads to more learning about science. Although this alternative relationship is currently not supported by the data, it is an interesting proposition that future studies can test. An important concept is the process of elaboration or when people think about the news (Eveland, 2001). It is possible that elaboration of news can also explain some variance in how much learning about science an individual gets. Our current model only accounts for some 22% of the variance, and future studies can build on what we have started in this exploratory study to investigate more factors that contribute to learning about science.

This study is not without limitations. First, we relied on GSS data. Therefore, some of the questions presented limitations regarding the variables included in our model. For instance, we mentioned that the questions about scientific knowledge were relatively basic. Despite the normal distribution of the responses, questions with various degrees of difficulty might present a better picture of respondents’ scientific knowledge. Future studies could test and expand our proposed model using measures of knowledge of specific science issues. Second, guided by the theories we discussed, we proposed a framework that focused on interest in science and motivated media consumption and perception of sources. Interest in science is just one of the plausible motivations for information seeking about science. Incidental, unmotivated exposure to information about science might also have effects on learning about science, and this was not fully taken into account in the current design (Rubin and Perse, 1987). Still, we believe our exploratory study in proposing and testing a framework on scientific learning by focusing on sources—by platform, such as print, television, and online, and by expertise, such as scientists and the press—demonstrated important relationships that will help our understanding and provoke more questions about how people learn about science.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biographies

Bruno Takahashi (PhD, SUNY ESF) is an Assistant Professor at Michigan State University with a joint appointment with the School of Journalism and the Department of Communication. He is also the research director of the Knight Center for Environmental Journalism. His research focuses on environmental and science communication, environmental journalism practices, and the links between media and policymaking.

Edson C. Tandoc Jr (PhD, University of Missouri) is an Assistant Professor at the Wee Kim Wee School of Communication and Information at the Nanyang Technological University in Singapore. His research focuses on the sociology of message construction and the impact of journalistic roles and audience feedback on the various stages of the news gatekeeping process.

References

Anderson

Scheufele

Brossard

. (2012) The role of media and deference to scientific authority in cultivating trust in sources of information about emerging technologies. International Journal of Public Opinion Research 24(2): 225–237. DOI: 10.1093/ijpor/edr032.

Baker

Williams

Lybbert

. (2012) How ecological science is portrayed in mass media. Ecosphere 3(1): 9. DOI: 10.1890/es11-00238.1.

Barendt

(1994) Media ownership and freedom of expression. Index on Censorship 23(4–5): 224–227.

Bauer

Allum

Miller

(2007) What can we learn from 25 years of PUS survey research? Liberating and expanding the agenda. Public Understanding of Science 16(1): 79–95. DOI: 10.1177/0963662506071287.

Bauer

Durant

Evans

(1994) European public perceptions of science. International Journal of Public Opinion Research 6(2): 163–186. DOI: 10.1093/ijpor/6.2.163.

Bauer

(1996) Socio-demographic correlates of DK-responses in knowledge surveys: Self-attributed ignorance of science. Social Science Information 35(1): 39–68. DOI: 10.1177/053901896035001003.

Bauer

Petkova

Boyadjieva

(2000) Public knowledge of and attitudes to science: Alternative measures that may end the “science war.” Science, Technology & Human Values 25(1): 30–51. DOI: 10.1177/016224390002500102.

Beaujon

(2012) Pew: Half of Americans get news digitally, topping newspapers, radio. Available at: http://www.poynter.org/latest-news/mediawire/189819/pew-tv-viewing-habit-grays-as-digital-news-consumption-tops-print-radio/

Beck

(1992) Risk Society: Towards a New Modernity. London: SAGE.

10.

Besley

Nisbet

(2013) How scientists view the public, the media and the political process. Public Understanding of Science 22(6): 644–659. DOI: 10.1177/0963662511418743.

11.

Bord

O’Connor

Fisher

(2000) In what sense does the public need to understand global climate change? Public Understanding of Science 9(3): 205–218. DOI: 10.1088/0963-6625/9/3/301.

12.

Brewer

Ley

(2013) Whose science do you believe? Explaining trust in sources of scientific information about the environment. Science Communication 35(1): 115–137. DOI: 10.1177/1075547012441691.

13.

Brossard

Nisbet

(2007) Deference to scientific authority among a low information public: Understanding U.S. opinion on agricultural biotechnology. International Journal of Public Opinion Research 19(1): 24–52. DOI: 10.1093/ijpor/edl003.

14.

Byrne

(2010) Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Programming. New York, NY: Routledge.

15.

Chaffee

Kanihan

(1997) Learning about politics from the mass media. Political Communication 14(4): 421–430. DOI: 10.1080/105846097199218.

16.

Davis

Craft

(2000) New media synergy: Emergence of institutional conflicts of interest. Journal of Mass Media Ethics 15(4): 219–231.

17.

Eveland

McLeod

Horowitz

(1998) Communication and age in childhood political socialization: An interactive model of political development. Journalism & Mass Communication Quarterly 75(4): 699–718. DOI: 10.1177/107769909807500406.

18.

Eveland

(2001) The cognitive mediation model of learning from the news: Evidence from nonelection, off-year election, and presidential election contexts. Communication Research 28(5): 571–601. DOI: 10.1177/009365001028005001.

19.

Eveland

Scheufele

(2000) Connecting news media use with gaps in knowledge and participation. Political Communication 17(3): 215–237. DOI: 10.1080/105846000414250.

20.

Eveland

Marton

Seo

(2004) Moving beyond “just the facts”: The influence of online news on the content and structure of public affairs knowledge. Communication Research 31(1): 82–108. DOI: 10.1177/0093650203260203.

21.

Falk

Needham

(2013) Factors contributing to adult knowledge of science and technology. Journal of Research in Science Teaching 50(4): 431–452. DOI: 10.1002/tea.21080.

22.

Garrett

(2009) Politically motivated reinforcement seeking: Reframing the selective exposure debate. Journal of Communication 59(4): 676–699. DOI: 10.1111/j.1460-2466.2009.01452.x.

23.

Garrett

Stroud

(2014) Partisan paths to exposure diversity: Differences in pro- and counterattitudinal news consumption. Journal of Communication 64(4): 680–701. DOI: 10.1111/jcom.12105.

24.

Garrett

Carnahan

Lynch

(2013) A turn toward avoidance? Selective exposure to online political information, 2004–2008. Political Behavior 35(1): 113–134. DOI: 10.1007/s11109-011-9185-6.

25.

Gauchat

(2012) Politicization of science in the public sphere: A study of public trust in the United States, 1974 to 2010. American Sociological Review 77(2): 167–187. DOI: 10.1177/0003122412438225.

26.

Gronke

Cook

(2007) Disdaining the media: The American public’s changing attitudes toward the news. Political Communication 24(3): 259–281. DOI: 10.1080/10584600701471591.

27.

Hively

Eveland

(2009) Contextual antecedents and political consequences of adolescent political discussion, discussion elaboration, and network diversity. Political Communication 26(1): 30–47. DOI: 10.1080/10584600802622837.

28.

Hmielowski

Feldman

Myers

. (2013) An attack on science? Media use, trust in scientists, and perceptions of global warming. Public Understanding of Science. Epub ahead of print 3 April. DOI: 10.1177/0963662513480091.

29.

Brossard

Scheufele

(2008) Effects of value predispositions, mass media use, and knowledge on public attitudes toward embryonic stem cell research. International Journal of Public Opinion Research 20(2): 171–192. DOI: 10.1093/ijpor/edn017.

30.

Jang

(2014) Seeking congruency or incongruency online? Examining selective exposure to four controversial science issues. Science Communication 36(2): 143–167. DOI: 10.1177/1075547013502733.

31.

Jerit

Barabas

Bolsen

(2006) Citizens, knowledge, and the information environment. American Journal of Political Science 50(2): 266–282.

32.

Jones

Biddlecom

(2011) Is the internet filling the sexual health information gap for teens? An exploratory study. Journal of Health Communication 16(2): 112–123. DOI: 10.1080/10810730.2010.535112.

33.

Kahlor

Rosenthal

(2009) If we seek, do we learn? Predicting knowledge of global warming. Science Communication 30(3): 380–414. DOI: 10.1177/1075547008328798.

34.

Kim

(2008) Testing the knowledge gap hypothesis in South Korea: Traditional news media, the internet, and political learning. International Journal of Public Opinion Research 20(2): 193–210. DOI: 10.1093/ijpor/edn019.

35.

Kleinnijenhuis

(1991) Newspaper complexity and the knowledge gap. European Journal of Communication 6(4): 499–522. DOI: 10.1177/0267323191006004006.

36.

Koch-Weser

Bradshaw

Gualtieri

. (2010) The internet as a health information source: Findings from the 2007 health information national trends survey and implications for health communication. Journal of Health Communication 15(Suppl. 3): 279–293. DOI: 10.1080/10810730.2010.522700.

37.

Koteyko

Jaspal

Nerlich

(2012) Climate change and ‘climategate’ in online reader comments: A mixed methods study. The Geographical Journal 179(1): 74–86. DOI: 10.1111/j.1475-4959.2012.00479.x.

38.

Lee

Scheufele

(2006) The influence of knowledge and deference toward scientific authority: Amedia effects model for public attitudes toward nanotechnology. Journalism & Mass Communication Quarterly 83(4): 819–834.

39.

Leiserowitz

Maibach

Roser-Renouf

Smith

Dawson

(2013) Climategate, public opinion, and the loss of trust. American Behavioral Scientist 57(6): 818–837. DOI: 10.1177/0002764212458272.

40.

Maibach

Leiserowitz

Cobb

. (2012) The legacy of climategate: Undermining or revitalizing climate science and policy? Wiley Interdisciplinary Reviews: Climate Change 3(3): 289–295.

41.

Maier

Rothmund

Retzbach

. (2014) Informal learning through science media usage. Educational Psychologist 49(2): 86–103. DOI: 10.1080/00461520.2014.916215.

42.

Malka

Krosnick

Langer

(2009) The association of knowledge with concern about global warming: Trusted information sources shape public thinking. Risk Analysis 29(5): 633–647.

43.

McLeod

Daily

Guo

. (1996) Community integration, local media use, and democratic processes. Communication Research 23(2): 179–209. DOI: 10.1177/009365096023002002.

44.

Meehl

Waller

(2002) The path analysis controversy: A new statistical approach to strong appraisal of verisimilitude. Psychological Methods 7(3): 283–300. DOI: 10.1037/1082-989x.7.3.283.

45.

Miller

(2010) Adult science learning in the Internet era. Curator: The Museum Journal 53(2): 191–208. DOI: 10.1111/j.2151-6952.2010.00019.x.

46.

Mooney

(2012) The Republican Brain: The Science of Why They Deny Science – And Reality. Hoboken, NJ: John Wiley & Sons.

47.

Morales

(2010) Distrust in U.S. media edges up to record high. Available at: http://www.gallup.com/poll/143267/distrust-media-edges-record-high.aspx

48.

Nisbet

Myers

(2007) The Polls—Trends. Public Opinion Quarterly 71(3): 444–470.

49.

Nisbet

Scheufele

Shanahan

. (2002) Knowledge, reservations, or promise? A media effects model for public perceptions of science and technology. Communication Research 29(5): 584–608. DOI: 10.1177/009365002236196.

50.

O’Neill

Nicholson-Cole

(2009) “Fear won’t do it”: Promoting positive engagement with climate change through visual and iconic representations. Science Communication 30(3): 355–379. DOI: 10.1177/1075547008329201.

51.

Papacharissi

Rubin

(2000) Predictors of internet use. Journal of Broadcasting & Electronic Media 44(2): 175–196. DOI: 10.1207/s15506878jobem4402_2.

52.

Petty

Cacioppo

(1986) Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York, NY: Springer-Verlag.

53.

Petty

Brinol

Priester

(2009) Mass media attitude change: Implications of the elaboration likelihood model of persuasion.In: Bryant

Oliver

(eds) Media Effects: Advances in Theory and Research, 3rd ed. New York, NY: Routledge, pp. 125–164.

54.

Pew (2009a) Press accuracy rating hits two decade low. Available at: http://www.people-press.org/2009/09/13/press-accuracy-rating-hits-two-decade-low/

55.

Pew (2009b) Public Praises Science; Scientists Fault Public, Media. The Pew Research Center for the People and the Press. Available at: http://www.people-press.org/files/legacy-pdf/528.pdf

56.

Pew (2015) Public and Scientists’ Views on Science and Society. The Pew Research Center for the People and the Press. Available at: http://www.pewinternet.org/files/2015/01/PI_ScienceandSociety_Report_012915.pdf

57.

Priester

Petty

(2003) The influence of spokesperson trustworthiness on message elaboration, attitude strength, and advertising effectiveness. Journal of Consumer Psychology 13(4): 408–421. DOI: 10.1207/S15327663JCP1304_08.

58.

Prior

(2005) News vs. entertainment: How increasing media choice widens gaps in political knowledge and turnout. American Journal of Political Science 49(3): 577–592.

59.

Retzbach

Maier

. (2013) Effects of repeated exposure to science TV shows on beliefs about scientific evidence and interest in science. Journal of Media Psychology: Theories, Methods, and Applications 25(1): 3–13. DOI: 10.1027/1864-1105/a000073.

60.

Roberts

Reid

Schroeder

. (2011) Causal or spurious? The relationship of knowledge and attitudes to trust in science and technology. Public Understanding of Science. Epub ahead of print 18 October. DOI: 10.1177/0963662511420511.

61.

Rubin

Perse

(1987) Audience activity and television news gratifications. Communication Research 14(1): 58–84. DOI: 10.1177/009365087014001004.

62.

Segev

Baram-Tsabari

(2012) Seeking science information online: Data mining Google to better understand the roles of the media and the education system. Public Understanding of Science 21(7): 813–829. DOI: 10.1177/0963662510387560.

63.

Stroud

(2008) Media use and political predispositions: Revisiting the concept of selective exposure. Political Behavior 30(3): 341–366. DOI: 10.1007/s11109-007-9050-9.

64.

Stroud

(2010) Polarization and partisan selective exposure. Journal of Communication 60(3): 556–576. DOI: 10.1111/j.1460-2466.2010.01497.x.

65.

Sturgis

Allum

(2004) Science in society: Re-evaluating the deficit model of public attitudes. Public Understanding of Science 13(1): 55–74. DOI: 10.1177/0963662504042690.

66.

Sundar

(2008) The MAIN model: A heuristic approach to understanding technology effects on credibility. In: Metzger

Flanagin

(eds) Digital Media, Youth, and Credibility. Cambridge, MA: The MIT Press pp. 72–100.

67.

Trilling

Schoenbach

(2014) Challenging selective exposure. Digital Journalism. Epub ahead of print 24 April. DOI: 10.1080/21670811.2014.899749.

68.

Tsfati

(2010) Online news exposure and trust in the mainstream media: Exploring possible associations. American Behavioral Scientist 54(1): 22–42. DOI: 10.1177/0002764210376309.

69.

Tsfati

Cohen

Gunther

(2011) The influence of presumed media influence on news about science and scientists. Science Communication 33(2): 143–166. DOI: 10.1177/1075547010380385.

70.

Ungar

(2000) Knowledge, ignorance and the popular culture: Climate change versus the ozone hole. Public Understanding of Science 9(3): 297–312.

71.

Ungar

(2008) Ignorance as an under-identified social problem. The British Journal of Sociology 59(2): 301–326. DOI: 10.1111/j.1468-4446.2008.00195.x.

72.

Wade

Schramm

(1969) The mass media as sources of public affairs, science, and health knowledge. Public Opinion Quarterly 33(2): 197–209. DOI: 10.1086/267691.

73.

Zhao

(2009) Media use and global warming perceptions: A snapshot of the reinforcing spirals. Communication Research 36(5): 698–723. DOI: 10.1177/0093650209338911.

74.

Zhao

Leiserowitz

Maibach

Roser-Renouf

(2011) Attention to science/environment news positively predicts and attention to political news negatively predicts global warming risk perceptions and policy support. Journal of Communication 61(4): 713–731. DOI: 10.1111/j.1460-2466.2011.01563.x.

75.

Zillmann

Bryant

(2008) Selective Exposure to Communication. New York, NY: Routledge.