11 JULY 2008 VOL 321 SCIENCE www.sciencemag.org 204 P revious research, as well as anecdotal evidence among researchers and jour- nalists, often leads to perceptions of “barriers” to a more active involvement of scientists in public communication [e.g. (1–3)] or of a “gap” between science and journalism (4) or to areas of potential conflict between the two professions (5, 6). Recently, researchers have begun to recognize the sym- biotic character of many scientist-journalist interactions (7). Nonetheless, negative expe- riences with the media still dominate peer communication about science-media rela- tions. On the basis of extensive survey data, we now challenge several of the negative impressions of science-media interactions that are still all too common. Although surveys of scientists’ interac- tions with the media have been conducted in several countries [e.g. (1, 4, 6)], little empiri- cal research has compared scientists’ public communication attitudes and activities across countries in a rigorous way. One might expect cross-cultural differences in the science- media interface for several reasons, among them possible differential benefits of public visibility because of variance in competitive research funding environments, differences in the nature of professionalism in science jour- nalism or science public relations, or cultural differences in public acceptance of science and technology across countries. Our analysis was based on a mail survey in 2005–06 of 1354 researchers in the United States (n = 358), Japan (n = 239), Germany (n = 283), United Kingdom (n = 281), and France (n = 193), the top countries for research and development (R&D) at the time of the study. Averaged across countries, the response rate was 43%; the sampling bias is unlikely to invalidate our main findings [for details, see (8)]. We used two research fields—epidemiology and stem cell re- search—as case studies. The sample com- prises 648 epidemiologists and 706 stem cell researchers who had published during 2002–04 in peer-reviewed journals. With few exceptions, the results for the two research fields studied were quite similar. We, there- fore, present aggregated research field results here and focus on cross-country comparisons. The supporting material (8) will provide the reader with breakdowns by field. Across the countries under study, scientist- journalist interactions were not the province of a small set of scientists but, rather, were more common than anticipated (fig. S1A). Of the respondents, 30% said they had been engaged in more than five media contacts in the past 3 years, and another 39% reported one to five contacts. In all countries, epidemiolo- gists had more contact with journalists than stem cell researchers, but there were no signif- icant differences across countries (table S2). The primary type of media contact was the media interview; nearly two-thirds of the respondents (64%) said that they have been interviewed by journalists at least once in the past 3 years (fig. S1B). Frequency of contact with journalists was clearly associated with leadership functions and research productiv- ity; a scientist’s personal attitude was also a factor but was relatively less important (fig. S2). The amount of contact, as well as its assoc- iation with leadership role, re- search productivity, and per- sonal attitude, formed a pattern that is quite similar across coun- tries (tables S2 and S5). We posed 16 motives, both positive and negative, that could influence scientists’ willingness to interact with journalists and asked respondents to rate the importance of each factor to themselves personally. Despite some differences regarding the perception of risks and benefits across countries, three main findings emerged (table S8). Increasing the public’s ap- preciation of science was the most important benefit men- tioned by scientists as an incen- tive to interact with the media. More than 9 in 10 respondents (93%) indicated that achiev- ing “a more positive public attitude toward research” was an important motivator; about as many (92%) similarly identified “a better- educated general public.” In interactions with the media, many scien- tists indicated they felt uncertain and perceived a lack of control. Nine in 10 respondents iden- tified the “risk of incorrect quotation” in stories as an important disincentive, and 8 in 10 felt that the “unpredictability of journalists” was also a problem. Norms of the scientific community com- mitting researchers to strong peer orientation and highly precise information (delivered in a formal, impersonal style) have historically been regarded as major deterrents to scientists’ interactions with journalists (3). However, those norms seem to be playing a more nuanced role today as only 34% of our sample identified “incompatibility with the scientific culture” as an important concern. Furthermore, the impact of scientific norms seemed to be perceived inconsistently (table S10). Although “possible critical reactions from peers” were considered important concerns for 42% of the respondents, a similar proportion (39%) found “enhanced personal reputation among peers” to be an important outcome of media contacts. POLICY FORUM Interactions with the Mass Media Hans Peter Peters, 1 * Dominique Brossard, 2 Suzanne de Cheveigné, 3 Sharon Dunwoody, 2 Monika Kallfass, 1 Steve Miller, 4 Shoji Tsuchida 5 SCIENCE COMMUNICATION 1 Program Group Humans-Environment-Technology, Forschungszentrum Jülich, 52425 Jülich, Germany. 2 School of Journalism & Mass Communication, University of Wisconsin–Madison, Madison, WI 53706, USA. 3 Shadyc— Sociologie, histoire, anthropologie des dynamiques cul- turelles, l’École des Hautes Études en Sciences Sociales (EHESS)–CNRS, 13236 Marseille, France. 4 Science and Technology Studies, University College London, London WC1E 6BT, UK. 5 Department of Psychology, Kansai University, Osaka 564-8680, Japan. *Author for correspondence. E-mail: h.p.peters@fz-juelich. de 100 90 80 70 60 50 40 30 20 10 0 France (n = 120) Germany (n = 187) Japan (n = 165) UK (n = 183) USA (n = 241) Mostly positive Mostly negative Relatively balanced No impact at all Responses (%) Perceived impact of media contacts on career by country. Distribution of answers to the question: “Consider the totality of your media contacts over your career. How great has their positive or neg- ative impact been on you professionally?” Only respondents report- ing media contact(s) in the past 3 years are included in the graph. See table S4 for a breakdown of responses by country and research field. A survey reveals that media contacts of scientists in top R&D countries are more frequent and smooth than was previously thought. Published by AAAS on March 31, 2009 www.sciencemag.org Downloaded from