A content analysis of community news sites affiliated with Irish newspapers shows that content selection reflects classic community news values and fulfills roles and functions similar to ones documented in historical research about community press and social organization in the United States. This illustrates the potential for Irish local media sites to be important agents of constructing community identity.
In the United States, one of the deepest parts of the research about community journalism extends from the community ties hypothesis. Starting with Morris Janowitz’s seminal early 1950s research into neighborhood community newspapers and extending through the work of Keith Stamm and others in the 1980s and beyond, this research agenda established a view of the community press as both an indicator of – and an impetus for – social change and community-building through exploration of intimacy of the relationship between local newspapers and their communities.
Little if any work has been done exploring such community/media interactions in the European context, however. This project seeks to address this deficit with a content analysis of online news sites in Ireland to examine whether the classic principles and relationships found in U.S. community media apply among Ireland’s community-news publications today.
LITERATURE REVIEW
A primary theme through the literature of U.S. community journalism is the intimacy that news organizations share with the institutions and individuals they cover, especially as it is articulated through the topics selected for coverage. A classic text, frequently cited to establish principles of community journalism, says that weekly newspapers should report “details of local news not included in stories that are in other papers” (Byerly, 1961, p. 5).
This intimacy between the community media and its audience that develops through news selections was explored deeply by Morris Janowitz (1951, 1967)[1] in a study of weekly newspapers in Chicago neighborhoods around 1950. Janowitz did content analysis of three such papers, and also analyzed neighborhood demographics, surveyed readers, and conducted in-depth interviews with the papers’ managers and residents of the neighborhoods they served. As a result, he concluded that the urban community press was “one of the social mechanisms through which the individual is integrated into the urban social structure” (Janowitz, 1967, p. 9). According to Janowitz, the community press:
- Provides support for and draws support (advertising) from satellite business districts within cities.
- Helps to maintain local consensus through an emphasis on common values rather than on conflicting ones.
- Shapes and reflects the neighborhood social and political structure.
- Provides a forum in which mass communication effects are interrelated with personal communications and social contacts.
In short, local media can help build the communities they cover.
This is accomplished with a high concentration of news coverage on community organizations and institutions, especially voluntary social, cultural, religious, and youth groups. “The community newspaper’s emphasis on community routines, low controversy and social ritual are the very characteristics that account for its readership,” Janowitz noted (1967, p. 130).
Building on Janowitz, Edelstein and Larsen (1960) also determined that content selection could affect community-building. They concluded that coverage of clubs and associations developed community consciousness, news of individual activities and accomplishments contributed to community identity, and crime and accident reports disclosed threats to the community.
Also growing from Janowitz’s work was a body of research in the 1970s and ’80s that explored how media usage, community characteristics, and an individual’s sense of community connectedness related to one another, which came to be known as the community ties research agenda. Many research projects at the time defined community according to geography and examined newspaper usage in light of variables such as home ownership and length of residence in a given geographic market (e.g. Stone 1978). However, Stamm and Fortini-Campbell (1983) introduced the idea that community, which traditionally had been rooted in a physical locale, should be construed on multiple dimensions of not only place (geography) but also structure (community institutions) and process (shared interests and interaction of community members). They further maintained that residents developed ties to each of these independently. This classification became the basis for investigations into the relationship of media usage and development of community attachments (Stamm 1985, 1988).
More recent updates of this work include design and testing of a 22-item index linking the news values reflected in content selection and the process of community-building by Lowrey, Woo & Mackay (2007). Paek, Yoon and Shah (2005) similarly found that news readership increased the likelihood of community participation.
The reasonableness of using these news values documented in American journalism as a framework for investigating Irish news sites rests in theories that explain the export and diffusion of news values from dominant countries such as the United States to the rest of the world. Among the widely used theories of international communication that propose this are Galtung’s structural imperialism (1971), Schiller’s cultural imperialism (1976) and Boyd-Barrett’s media imperialism (1977). In the seminal work in this area, Galtung and Ruge (1965) noted that “consonance” of values such as culture and language would lead to similarities in news presentations. Clearly that sort of cultural and linguistic consonance exists between Ireland and America.
These theories of international communication, which developed around the same time as the evolution of the U.S. community ties hypothesis in the 1970s and ‘80s, generally sought to address the impact of news and other information flows on less-developed regions of the world. Galtung (1971), for example, separated the world into the “core” (developed countries) and “periphery” (developing and underdeveloped ones) and argued that news by and about the core dominates worldwide, even in the periphery. In an example of this, Chang (1998) studied Reuters news service reporting of a World Trade Organization conference in Singapore and concluded that more reporting was done on core countries (especially the United States, Japan, Canada and the European Union) than peripheral ones, and that coverage of peripheral countries was mostly in the context of their relationship with the core.
These theories are not wholly relevant to the current case because Ireland clearly is a part of the core, rather than the periphery. Yet they are relevant for explaining diffusion of news values from dominant players – in this case, the U.S. – to other parts of the world. For example, Boyd-Barrett and Rantanen (1998) said major news agencies (wire services) were agents of the globalization and commodification of news. Rantanen (1998) also described the process by which national news systems developed with values similar to those of larger international news agencies, and McPhail added that wire services “directly and indirectly promote a core-based focus and emphasis in reporting values” (2002, p. 159, emphasis added).
Thussu (2006) notes that many American news organizations such as the Wall Street Journal, Time, Fortune and Forbes produce regional editions for Asian, European and South American markets, which furthers the spread of American news values. The global success of CNN also contributed to proliferation of American news values (McPhail, 2002). A more recent investigation of international diffusion of news practices found partial support for two hypotheses that European news systems over time had developed a “hard news” paradigm with sourcing patterns more similar to U.S. coverage (Esser & Umbricht, 2014). Overall, it is reasonable to assume that the globalization of media described in these theories can predict a homogenization of news values that could include community news.
The specific research question to be explored here is: What degree of consonance exists between historically documented community news values of American papers and contemporary Irish coverage? This will be demonstrated with a comparative analysis of topical themes for community news coverage presented in each country.
This is an especially relevant line of investigation because no direct cross-cultural comparison with American media could be located in literature about Irish news and media. It also is important given the findings that community media can contribute to construction of social identity – in this case, Irish identity. Studies of Irish news content were found that focused on news coverage related to national and regional identity. But mostly these addressed historical events, such as the early 20th century independence struggle (Foley, 2004), the Good Friday peace accords (Baker 2005), Northern Ireland’s annual “orange marches” (Fawcett 2002; Ferman 2013), and the collapse of the Celtic Tiger economy (Cawley 2012a). Studies of media contributions to identity at more local levels in a more contemporary context are lacking.
METHODOLOGY
Answering the question of whether Irish community news sites would present coverage consonant with news values found in American publications was done with a content analysis of material on the websites of Irish newspapers. News items published on the sites were classified by topic (e.g. community events, government, sports, police news, etc.) and the results were evaluated with a factor analysis to discern whether any patterns underlay the topics in ways that related to the purposes and functions described by Janowitz and Stamm. This approach is similar to a content study of print and online U.S. newspapers that used factor analysis to ascertain underlying determinants of coverage in testing a proposed community news index (Lowrey, Woo & Mackay, 2007).
Ireland is a small country; with a population of 4.6 million people on 33,600 square miles it is just slightly larger than the U.S. state that ranks 25th in population and 40th in area: South Carolina
(4.2 million people; 31,117 sq. miles). So like a small U.S. state, Ireland’s universe of media outlets to investigate is relatively small as well. Combining the results of a search in two different listings of online news sites (www.allnewsmedia.com/Europe/Ireland/newspapers.htm and www.abyznewslinks.com) and eliminating duplicates and broken links, a database of 122 Irish newspapers with an online presence was developed. They ranged from large ones based in Dublin with national circulation all the way down to local and regional dailies and weeklies. This small number of outlets made it possible to do a very generous sampling of 50 percent of the available universe (n = 61) for a comprehensive look at the coverage patterns for the country as a whole.
Although comprehensive circulation statistics could not be located, the vast majority of these can be considered community papers. Eight publications in the 122-outlet universe are national papers – but even the largest of them, the Independent, is the size of a mid-market daily by American standards with a circulation of 112,000. Even most of the “nationals” have circulation of around 50,000, which is commonly used by U.S. trade associations as a cutoff for “small” publications (Lauterer, 2006). It is reasonable therefore to consider that the rest of Ireland’s newspapers are smaller still and focused on serving local geographic areas, and therefore that the cross-cultural analysis of examining them for traditional community journalism values as articulated in small U.S. newspapers makes sense. As one journalist from Cork remarked, “Almost every newspaper in Ireland is hyper-local” (Weckler, 2011, para. 11).
Ireland has consistently been documented as a place with strong newspaper readership (Brennan, 2005; Elvestad & Blekesaune, 2008). Irish newspaper readership tends more strongly toward print than online, although online is growing with a 26 percent increase in 2013-14 compared with 2012-13 in the Joint National Readership Survey (2014). This survey, released in August 2014, said 2.9 million people read an average issue in print with 565,000 online (in a country of 4.2 million population). It reported high crossover between print and online readership, with 3/4 of online average issue readers also reading in print.
The idea of using story topics as the key measurement for this research was drawn from Janowitz, who said that community newspaper content selection could “emphasize values and interests on which there is a high level of consensus in the community” and assist in “building and maintaining local traditions and local identifications” (1967, p. 61). The variables identified for this project generally followed the list he used in analyzing the Chicago papers. However, it combined some categories and also excluded one – trade unions – that made sense for 1950s Chicago but not a contemporary context. Other story topics not used by Janowitz but relevant to modern-day coverage – such as schools, real estate, and transportation/commuting – then were added to create a 17-item codebook. The codebook specified that topical categories applied to local news only; larger-scale stories such as national events were categorized as “Non-local/other.” (See complete list of variables in Table 1.)
A total of 1,425 items that appeared on the home pages of 61 sites selected at random were coded by placing them into one of these categories in a review of the sites during an eight-week period in October and November 2014. A mean of 23.3 items and a median of 22 items per site were coded.
Agreement reliability between the two coders (the principal researcher and a trained student assistant) was determined by having both of them code 10 percent of the sites (n = 6) including approximately 8 percent of the total items (n = 110) used in the final analysis. The rate of agreement in this post-hoc test was 75.5 percent, with a Cohen’s kappa of .714. Scott’s pi also was .714, but Cohen’s kappa is a preferred statistic when a large number of variables (in this case, 17) are coded.
FINDINGS
The amount of coverage Irish online newspapers devoted to different topics varied from 19 percent for sports (n = 275 out of 1,425) to a fraction of a percent for religion coverage (n = 2 for organized religion; n = 2 for individual religion/spirituality). The next-most common topics after sports were arts and cultural coverage (12 percent; n = 171), business and economy coverage (11.2 percent; n = 160), government (10.7 percent; n = 153) and police, court and crime news (8.4 percent; n = 119). (See Table 1 for complete percentage results.)
Because the idea of using news topics as a unit of analysis was drawn from Janowitz and the codebook was a modification of his categories, amounts and types of coverage were compared across his study and the current one. Few similarities emerged. Janowitz’s study, for instance, found 6.5 percent of coverage devoted to religious organizations, compared to a fraction of 1 percent in the Irish sites (4 items out of more than 1,200 analyzed). This difference is especially notable in that Janowitz specifically identified religious groups as one constituent of the coverage of community routines and social rituals (1967, p. 74), but religion was utterly lacking in the contemporary Irish coverage. Likewise, Janowitz discovered that 18.8 percent of news coverage was devoted to social and personal news and 23.6 percent devoted to local volunteer efforts, compared to 4.6 percent and 2 percent in the Irish sample.
In the other direction, Janowitz found only 4.3 percent of coverage devoted to disasters, accidents and police news (combination of two categories) whereas the current study noted 10.2 percent of coverage was about these topics combined. Less than 6 percent of the coverage documented by Janowitz was devoted to sports, compared with more than 19 percent in the Irish coverage.
Some limitations in making these comparisons are worth noting, including the differences in coding categories; Janowitz did not have a separate category for community events as the current study did so an event sponsored by a volunteer group most likely would have appeared as coverage about the group in his study, but in a different category, the “events” one, in this study. Also, the statistics he reported were percentages of space devoted to various types of news while the current study used item-counts – largely because news hole and proportion of space used are impossible to calculate online.
This proportional comparison was augmented with a principal components factor analysis used to assess the underlying structure of the set of variables to discern whether any patterns underlay the topics in ways that related to the purposes and functions described by Janowitz and Stamm. Factor analysis is a data-reduction tool that creates derived variables (called factors) representing the degree to which variables in the larger initial set may be representing related characteristics by clustering them together in more homogeneous groupings. The factor analysis (with Varimax rotation) used a variable set that included 14 of the original content categories. Three were discarded because of the small number of items found in them; they were commuting/transportation (n = 7) and both of the religion categories (n = 2 for each). Results suggested five factors that explained 64.7 percent of the variance among the variables, grouped as follows:
- Factor 1 (accounting for 16.6 percent of variance): Government, Business/Economy, Real Estate/Land Development, and Community History. The four categories accounted for 25.7 percent of the total items coded (n = 367). Factor loadings ranged from .830 to .460.
- Factor 2 (14.3 percent of variance): Sports and Culture/Arts. The two categories accounted for 31.3 percent of the total items coded (n = 446). Factor loadings ranged from .875 to .825.
- Factor 3 (15.6 percent of variance): Community Events, Education, Volunteer Activities, Social/Personal News, and General Local News. This category accounted for 22.1 percent of the total items coded (n = 316). Factor loadings ranged from .705 to .558.
- Factor 4 (9.5 percent of variance): Police/Courts/Crime and Accidents/Disasters. This category accounted for 10.23 percent of the total items coded (n = 144). Factor loadings ranged from .710 to .456.
- Factor 5 (8.7 percent of variance): Other (non-local) news. This category accounted for 9.9 percent of the total items coded (n = 141). Its factor loading was .841.
(Complete results in Table 2.)
Unlike the simple proportional comparison, the underlying coverage determinants as suggested by the factor analysis indicate Irish news sites are fulfilling community orientation functions as described by Janowitz and Stamm.
One of the four key roles in Janowitz’s description of the community press was that it would shape and reflect the neighborhood social and political structure. Factor 1 shows that reporting on local “power structures” – government, business, and community development/real estate – has a common determinant. The collection of “structural” items loading on this factor indicates that the sites are fulfilling the function Janowitz identified regarding reflection of and support for community structure, including its economic players.
Factor 2, meanwhile, finds a common determinant to coverage of sports and arts/cultural coverage, which might be taken together as “diversions” or entertainment. Factor 3 includes personal/social news (job promotions, civic awards, obituaries, and the like) along with community events, education and volunteer activities. These items represent many types of coverage of which “social ritual” is made and thus this factor reflects Janowitz’s finding that community reporting focuses on “community routines, low controversy and social ritual” (Janowitz 1967, p. 130).
Factor 4, which groups accident/disaster coverage and police/crime news, represents reporting on threats to the community, as described by Edelstein and Larsen (1960) in their follow-up to Janowitz’s work (which also was based on his methodology). It groups two coverage variables that also relate to another social value, that of community safety.
Coverage of non-local news loaded on a factor of its own, indicating it has different coverage determinants than the community news topics. This illustrates that community coverage is separate and discrete from non-community news in the Irish publications.
Thus, the general trend with coverage variables as they were grouped by the factor analysis support the hypothesis that Irish news sites are fulfilling the functions ascribed to the U.S. community press in Janowitz’s classic work.
The factor solution also offers evidence of the Irish sites’ coverage patterns associating with community ties in the ways postulated by Stamm. He described the development of community ties as not only a matter of place (geography) but also of structure (institutions) and process (shared values/common activities). While this taxonomy does not completely overlap with Janowitz’s, the two approaches do intersect. Both scholars describe news about local institutions as coverage of “community structure.” Further, Janowitz’s description of coverage “emphasizing values and interests on which there is a high level of consensus in the community” (1967, p. 61) closely parallels what Stamm calls “process” coverage, or news that helps build community identity by illustrating “common endeavor and shared interest” (Stamm, 1985, p. 18).
Using Stamm’s taxonomy, the separation of non-local coverage (Factor 5) from everything else indicates a geographic determinant to news decisions by the Irish journalists. Factor 1 could be labeled “structure” for its collection of coverage variables about community institutions; Factors 2, 3 and 4 could be labeled “process” for their items that illustrate shared values, including community safety (Factor 4).
DISCUSSION
Many theories of international communication developed from the 1960s through the 1980s to predict and explain the globalization and homogenization of news values did so with a quite jaundiced eye toward developed nations, and the United States in particular. The general thread of this thinking was that core nations and their media organizations exercised hegemony over the periphery by exporting media content and associated cultural values. This even can be seen in the names of the theories. “Imperialism” is seldom seen in a positive light, yet that is exactly the term used by Galtung (1965) (“structural imperialism”), Schiller (1976) (“cultural imperialism”) and Boyd-Barrett (1977) (“ media imperialism”) to describe global flow of news and other media forms. The influence of the international news system, especially the large wire services or news agencies, in spreading common news values around the world has been clearly documented. This homogenization of news and news values around the world may be one explanation for the findings observed in this project.
But it is not possible to say this for certain because it also is plausible that the types of news coverage that exemplify community are similar without regard to national borders. In documenting the adoption of online news in Ireland, Cawley observed that,
The news content of their websites originated with the print newspapers and were aligned to established routines of information gathering from regional and local institutional sources and traditional journalistic judgments of what constituted local news: principally, local council and court reports, local commerce and sports. (Cawley 2012b, p. 228).
In that short list, he identifies three of the top four areas of coverage found in this content analysis that account for 42 percent of the items found. Those same three items account for a third of the coverage documented by Janowitz, too.
The largest single category of news coverage in the current analysis was sports, particularly of local teams in the Gaelic Athletic Association (GAA). Sports and team attachment play a large role in Irish cultural identity (Hassan, 2002; Fulton & Bairner, 2007) and therefore in devoting so much coverage to sport the community news sites are reflecting an important part of the social structure and local identity. Artistic and cultural endeavors also have long been a part of the construction of Irish identity (e.g. Foley, 2011; McLoone, 1994). Those topics comprised the second-largest category of coverage.
The lack of coverage about religion also can be explained by the adherence to community news values identified by Janowitz. For the setting he investigated (mid-20th-century Chicago), religious institutions were a key part of the community via neighborhood churches serving ethnic immigrants, especially Italians, Poles and – ironically – Irish. Coverage of churches therefore was a part of documenting the “social ritual” and “consensus values” aspects of living in those communities (Janowitz, 1967, p. 74), and constituted 6.5 percent of the coverage items he discovered in his content analysis. In historical and contemporary Ireland, on the other hand, religion is a point of major contention and conflict (e.g, Fahey, Hayes, & Sinnott, 2005). Avoiding coverage of religion, rather than putting news resources toward it, would serve the news value of emphasizing consensus values in the community and avoiding larger controversies.
So, rather than illustrating news imperialism of any sort, the findings that community news values documented in America more than a half century ago persist in Ireland today may be saying more about the enduring value of community coverage that “[satisfies] a basic human craving … the affirmation of the sense of community, a positive and intimate reflection of the sense of place” (Lauterer 2006, p. 33).
Janowitz’s work, and later Stamm’s, were noteworthy in explaining how community media could influence the way individuals connected with their communities and also could serve as an agent for community building. Ireland always has been known as a place where culture, identity and geography are tightly intertwined, especially in the North (Hayward, 2006). This project’s finding that Irish websites exhibit some of the same characteristics in news coverage as found in U.S. community papers that served as community builders therefore is a significant one in light of that long-standing struggle for articulating community identity in the Emerald Isle.
WORKS CITED
- Baker, S. (2005). The alternative press in Northern Ireland and the political process. Journalism Studies, 6(3), 375-386
- Boyd-Barrett, O. & Rantanen, T. (1998). The Globalization of News. In Boyd-Barrett, O. I Rantanen, T. (Eds), Globalization of News (pp. 1-13). London: SAGE Publications Ltd.
- Boyd-Barrett, O. (1977). Media imperialism: towards and international framework for the analysis of media systems. In J. Curran, M. Gurevitch and J. Woollacott (eds), Mass communication and society. London: Edward Arnold.
- Brennan, E. (2005). Irish journalists’ attitudes towards, and use of, internet technology. Seventh Conference of the European Sociological Association, Torun, Poland, 2005. Retrieved from http://arrow.dit.ie/aaschmedcon/22/
- Byerly, K.R. (1961). Community journalism. Philadelphia: Chilton.
- Cawley, A. (2012a). Sharing the pain or shouldering the burden: News-media framing of the public sector and the private sector in Ireland during the economic crisis, 2008-2010. Journalism Studies 13(4), 600-615.
- Cawley, A. (2012b). Towards a historical perspective on locating online news in the news ecology: The case of Irish news websites, 1994-2010. Media History, 18(2), 219-237.
- Chang, T-K. (1998). All countries are not created equal to be news: World system and international communication. Communication Research, 25(5), 528-564.
- Edelstein, A.S. & Larsen, O.N. (1960). The weekly press’ contribution to a sense of urban community. Journalism Quarterly, 37, 489-498.
- Elvestad, E. & A. Blekesaune, A (2008). Newspaper readers in Europe: A multilevel study of individual and national differences. European Journal of Communication. 23(4), 425-447.
- Esser, F. & Umbricht, A. (2014). The evolution of objective and interpretive journalism in the Western press: Comparing six news systems since the 1960s. Journalism and Mass Communication Quarterly, 91(2), 229-249.
- Fahey, T., Hayes, B. C., & Sinnott, R. (2005). Conflict and consensus: A study of values and attitudes in the Republic of Ireland and Northern Ireland. Institute of public administration.
- Fawcett, L. (2002). Why peace journalism isn’t news. Journalism Studies, 3(2), 213-223.
- Ferman, D. (2013) A parade or a riot: A discourse analysis of two ethnic newspapers on the 2011 marching season in Northern Ireland. Journal of Media and Religion, 12 (2), 55-70.
- Foley, M. (2004). Colonialism and journalism in Ireland. Journalism Studies 5 (3), 373-385.
- Fulton, G., & Bairner, A. (2007). Sport, space and national identity in Ireland: The GAA, Croke Park and Rule 42. Space & Polity, 11(1), 55-74.
- Galtung, J. (1971). A structural theory of imperialism. Journal of Peace Research, 8 (2), 81-117.
- Galtung, J. & Ruge, M.H. (1965). The structure of foreign news. Journal of Peace Research, 2(1), 64-91.
- Hassan, D. (2002). A people apart: Soccer, identity and Irish nationalists in Northern Ireland. Soccer & Society, 3(3), 65-83.
- Hayward, K. (2006). Reiterating national identities: The European Union conception of conflict resolution in Northern Ireland. Cooperation & Conflict, 41(3), 261-284.
- Janowitz, M. (1951). The imagery of the urban community press. Public Opinion Quarterly, 15(3), 519-531.
- Janowitz, M. (1967). The community press in an urban setting: The social elements of urbanism (2nd ed.). Chicago: University of Chicago Press.
- Joint National Readership Survey. (28 August 2014). Joint national readership survey topline data 2013-2014. Retrieved from www.jnrs.ie/pdf/Intro2013-2014.pdf.
- Lauterer, J. (2006). Community journalism: Relentlessly local (3rd ed.). Chapel Hill: The University of North Carolina Press.
- Lowrey, W., Woo, C.W. & Mackay, J. (2007). A test of a measure of community journalism. Paper presented to the annual convention of the Association for Education in Journalism and Mass Communication, Washington, D.C., August 2007.
- McLoone, M. (1994). National Cinema and Cultural Identity: Ireland in Europe. In Hill, J, McLoone, M. & Hainsworth, P. (Eds). Border Crossing: Film in Ireland, Britain and Europe (pp. 146-173). Belfast, NI: QUB Institute of Irish Studies.
- McPhail, T. (2002). Global communication: Theories, stakeholders and trends. Boston: Allyn & Bacon.
- Paek, H-J., Yoon, S-H. & Shah, D.V. (2005). Local news, social integration and community participation: Hierarchical linear modeling of contextual and cross-level effects. Journalism and Mass Communication Quarterly, 82(3), 587-606.
- Rantanen, T. (1998). The struggle for control of domestic news markets. In O. Boyd-Barrett & Rantanen, T. (Eds.). Globalization of News (pp. 35-48). London, UK: SAGE Publications Ltd.
- Schiller, H. (1976) Communication and Cultural Domination. New York: International Arts and Sciences Press.
- Stamm, K.R. (1985). Newspaper use and community ties. Norwood, NJ: Ablex Publishing.
- Stamm, K.R. (1988). Community ties and media use. Critical Studies in Mass Communication, 5(4), 357-361.
- Stamm, K.R. & Fortini-Campbell, L. (1983). The relationship of community ties to newspaper use. Journalism Monographs, 84.
- Stone, Gerald C. (1978). Using community characteristics to predict newspaper circulation. ANPA News Research Report 14. Washington, DC: American Newspaper Publishers Association.
- Thussu, D. (2006). International Communication (2nd ed.). London: Hodder Arnold.
- Weckler, A. (2011, Dec. 1). Startup of the day: iMedia Revenue. Daily Business Post. Retrieved from www.businesspost.ie/#!story/Home/News/Startup+of+the+day%3A+iMedia+Revenue/id/ 89127228-1884-ed74-a076-ac6d85864387.
APPENDIX
TABLE 1. CONTENT CATEGORY PERCENTAGE COMPARISON
| Current study | Janowitz[1] | ||
| Sports | 19.3% | Sports | 5.9% |
| Arts/Cultural News | 12.0% | Entertainment | 5.1% |
| Business and Economy | 11.2% | Business | 8.7% |
| Government and Politics | 10.7% | Mun. Svcs/Pub Affairs/Politics | 18.8% |
| Police, Courts and Crime | 8.4% | Accidents/disasters, police (as combined by Janowitz) | 4.3% |
| Accidents/ Disasters | 1.8% | — | |
| Community Events | 6.7% | — | |
| General News | 5.4% | Other local news | 7.0% |
| Personal/Social | 4.6% | Personal/Social | 18.8% |
| Education (combined cats) | 3.2% | — | |
| Real Estate/Land Dev. | 2.4% | — | |
| Volunteer grps (non-event) | 2.2% | Volunteer assocs (2 cats) | 23.6% |
| Community History | 1.4% | Community History | 1.2% |
| Commuting/Transportation | 0.5% | — | |
| Religion (combined cats) | 0.4% | Religion | 6.5% |
| Non-local (other) | 9.9% | — | |
| Total | 99.9% | Total | 99.9% |
TABLE 2: FACTOR ANALYSIS OF NEWS ITEM TOPICS
| Factor 1 “Power Institutions” (Structure) | |||||
| Government and Politics | 0.83 | 0.006 | 0 | 0.117 | -0.071 |
| Real Estate/Land Dev. | 0.754 | -0.038 | 0.251 | 0.087 | -0.003 |
| Business and Economy | 0.734 | 0.102 | -0.034 | -0.146 | 0.19 |
| Community History | 0.46 | 0.119 | 0.193 | -0.442 | -0.332 |
| Factor 2 “Diversions” (Process) | |||||
| Sports | -0.085 | 0.875 | 0.107 | 0.09 | 0.038 |
| Arts/Cultural News | 0.098 | 0.825 | 0 | 0.016 | -0.098 |
| Factor 3 “Social Ritual” (Process) | |||||
| Community Events | 0.143 | 0.486 | 0.595 | -0.112 | -0.298 |
| Education | 0.285 | 0.161 | 0.643 | -0.26 | 0.041 |
| Volunteer grps (non-event) | 0.233 | 0.187 | 0.687 | -0.086 | 0.189 |
| Personal/Social | 0.123 | -0.111 | 0.558 | 0.536 | -0.256 |
| General local news | 0.206 | -0.118 | 0.705 | 0.065 | -0.022 |
| Factor 4 “Community Safety” (Process) | |||||
| Accidents/ Disasters | -0.004 | 0.111 | -0.133 | 0.71 | 0.038 |
| Police, Courts and Crime | 0.298 | 0.419 | 0.093 | 0.456 | 0.39 |
| Factor 5 “External news” (Geography) | |||||
| Non-local (other) | 0.038 | -0.081 | 0.027 | 0.025 | 0.841 |
[1] Janowitz’s original work on the topic appeared in a 1951 article in Public Opinion Quarterly, and was elaborated upon in a book published in 1952. A second edition of that book, with a new preface and epilogue but otherwise still focused on the same early-1950s project and data, was published in 1967. That later work is the one cited in this article.
About the Author
Jack Rosenberry is an associate professor at St. John Fisher College in Rochester, NY. He would like to acknowledge and thank student Katie Weidman for her assistance on this project.