In June 2003 the Federal Communications Commission (FCC) voted 3-2 to relax ownership restrictions of broadcast and print media, allowing a single company or individual to own up to three television stations, eight radio stations, one daily newspaper, and one cable operation in the largest media

markets. Further, the commission raised the ceiling from 35% to 45% of the nationwide audience that a broadcast group owner could reach with its television stations (Munro & Vaida, 2004). Most of those proposed guidelines were blocked by a U.S. Court of Appeals on September 3, 2003, while allowing an interim compromise ceiling of 39% to stand (Plamondon, 2003). Even without the 45% ceiling, U.S. media ownership is more concentrated than at any time over the past 40 years (Turner, 2004). It has been estimated that five media companies control 75% of prime-time programming and 90% of the top 50 channels on cable (Hollings & Dorgan, 2001).

The controversy concerning the proposed revisions to FCC regulations on broadcast-media ownership limits has sparked an intense debate regarding localism. Localism is one of three widely recognized public policy goals [1] used by the FCC to help define the "public interest, convenience and necessity," (McChesney, 1995, p. 18) the standard of performance that broadcasters must meet in order to retain their licenses. Essentially, localism requires a broadcaster to serve the interests of the community covered by its transmission signal; if localism is successfully met, then the public interest is said to be better served. Napoli (2001) defined localism "in terms of a program's geographic point of origin. Thus, any program produced and presented within a local community" (p. 376) can help a broadcaster meet the goal of localism. In the early 1960s as television developed into the dominant medium of news and information, the FCC developed regulations designed to encourage television news decision makers to embrace localism. Further, the Courts have affirmed the importance of localism. In Turner Broadcasting System, Inc. v. F.C.C. (1994), the Supreme Court stressed the importance of local news, local public affairs programming, and matters of local concern (Plamondon, 2003). In the current context of increased media-ownership concentration, one of the important questions that emerge is whether or not broadcasters are meeting--or even attempting to meet--localism.

Proponents of the proposed FCC regulations contend that large media companies have advantages over smaller ones, arguing that the resources of large media groups actually increase the quantity and quality of locally originated news. Then-FCC Chair Michael Powell asserted that network owned-and-operated stations produce 50% more local news than local affiliate stations and have won 200% to 300% more quality awards for their local news content (Crystle, 2003; Online NewsHour, 2003). It has been further observed that large corporate owners of local media have deeper economic resources that can be invested in the qualitative production of local news (Compaine, 2004).

Opponents of the proposed FCC regulations fear that economies of scale will encourage large media chains to homogenize the news they distribute. Critics predict that liberalized ownership rules will cause media channels to become simple "content providers" by using the same news stories across multiple group-held properties (Corrigan, 2003, p. 30), and that group-owned broadcast news departments will rely more on syndicated feeds with no authentic local connection and less on locally produced, locally relevant news. Clement (2003) observed that, in order to cut costs and to increase profits, many larger media chains produce generic news stories that are repackaged by local reporters at each station in order to imprint those stories with a local look and feel (p. 43 A). Such is the practice of NewsProNet, an organization that produces generic news-you-can-use stories and in-depth investigative reports, distributing them nationally to resource-starved local broadcast-news operations. As of 2003, 93 stations had signed up for such canned stories, including major stations in New York City, Atlanta, Chicago, Denver, and Los Angeles (Barrett, 2003).

Questions regarding the quality of local news, regardless of the medium, are of primary importance in a news-rich democratic society. Classical democratic theory suggests citizens must be well informed on local issues via an independent press that provides governmental accountability. We seek to explore the issue of ownership type and the quality of local news by advancing the argument that a television news department's commitment to quality may be assessed using an investment model of news. We predict an inverse relationship between the relative investment of resources made in the production of local news at a given television station and the size of the media group--the larger the media group, the fewer resources will be devoted to locally originated news stories.

Review of Literature

Much of the research cited in the current regulatory debate has been controversial, due to several reasons. One important factor is that many of the proposed regulatory changes are based on studies that were produced outside of the traditional channels of academic peer-reviewed publications, either the by-product of FCC-commissioned research (e.g., Spavins, Denison, Roberts, & Frenette, 2002) or the result of independent think-tank media-watchdog organizations (e.g., PEJ Project, 2003). In addition, several smaller studies and critiques have been sponsored or produced by interested parties (e.g., Baker, 2002; Owen, 2003). The result is a murky, ambiguous, and not-disinterested reading of the research findings.

At least two of the commonly cited think-tank studies were "reformatted" studies in the sense that the characteristics of ownership and local news content were not the primary variables of interest in those studies. Further, the original datasets that provide the foundation for these studies have not always been made available to other scholars for external review. In addition, much of the conclusions provided rely on subjectively based assessments, e.g., evaluative "reports cards" or a quantification of subjectively based journalistic awards. The FCC commissioned several studies that supported the loosening of media-ownership restrictions, but they suffered from a perceived lack of objectivity and have been criticized within the FCC itself. Commissioner Jonathan Adelstein pointed out that the methodology employed in one study resulted in the faulty conclusion that there were more television stations licensed in Sioux Falls, SD than Detroit, MI.

We now turn our attention to examining a set of studies not only for their findings but for each study's method as well. The issues are material. Only one of the FCC-commissioned studies dealt directly with the issue of localism and the quantity and quality of local news based on ownership characteristics. Comparing network owned-and-operated (O&O) stations--local broadcast stations owned by the major networks such as Viacom, Disney, etc.--with network-affiliated stations, Spavins, Denison, Roberts, and Frenette (2002) found the O&Os "appear to produce, on average, a greater quantity of local news and public affairs programming than affiliates in markets where the two station types compete directly" (p. 3), citing 22.8 hours of local news per week by the O&Os compared to 18.5 hours by network-affiliated stations. However, the authors did not analyze the data based on the size of group ownership. Further, most of the affiliated stations were also owned by large media chains that made the comparison less meaningful. Baker (2002) also criticized the FCC study on the grounds that it did not control for factors "such as the age of the station which may impact the findings and did not examine trends in news quality over time" (pp. 8-9). However, it should be noted that Baker's methodological critique was funded by the AFL-CIO Department for Professional Employees Union, which has publicly opposed the loosening of FCC ownership restrictions.

It is important to note how Spavins et al. (2002) measured news program quality. Three secondary measures were used--ratings during a November-sweeps period, the number of awards won in competition sponsored by the Radio and Television News Directors Association (RTNDA), and the number of duPont Silver Baton awards won--raising questions regarding the integrity of the conceptual link to the construct of news quality. The study found no meaningful differences in the sweeps ratings of the local evening newscasts, and the O&Os won significantly more RTNDA and duPont awards than the affiliates. Plamondon (2003) criticized the index because those awards require entry fees, more likely to be paid by stations owned by a broadcast network.

A longitudinal study by the Pew Project for Excellence in Journalism (PEI) analyzed 172 news programs from 50 different media markets nationwide over a 5-year period (1998-2002) providing a database of approximately 23,000 stories (PE] Project, 2003). Newscast quality was assessed by how many topics were covered (termed "covering the community"), how many sources and points of view were presented in each story ("balance and accuracy"), the "authoritativeness" of story sources, how much effort was demonstrated in reporting the story ("enterprise"), the degree to which stories were made locally relevant ("localism"), and the degree to which stories touched on underlying themes, issues, or trends ("significance and informativeness") (pp. 2-3). Points were tallied for each station on each variable, story by story. The data were sorted to reflect different ownership characteristics, points were tabulated by variable, and grades "A" through "F" were assigned based on comparative point totals. The PEJ study found that smaller station groups "tended to produce higher quality newscasts than stations owned by larger companies" (p. 1) by a significant margin, while locally owned stations' news "tended to be below average in quality" (p. 13). Despite the overall lower grade, both small-station companies and locally owned stations "were about as likely" (p. 13) to display "enterprise" and demonstrate "localism" in their coverage of local stories.

The PEJ methodology has been criticized on several points. The data used in the study were secondary, originally collected for the Local TV Project's annual report card on television news at stations around the country. Several journalists who helped compile the grading criteria used in the original study said they were not comfortable with the way data were collected and used to make one set of conclusions, then recategorized to make another set of conclusions (Greppi, 2003):

   Many of the news professionals whose names were listed as part of a
   design team say they feel blindsided. They say this was not part of
   the project they agreed to participate in, and they were never told
   it was coming out or that their names would be on it. They also say
   they were not given a chance to comment on the final draft before
   it was released. (p. 4)

Ultimately two members of the original research design team resigned after the report was issued (Trigoboff, 2003).

The report was also criticized by Economists Incorporated (El), an ostensibly independent economic consulting firm based in Washington, DC. EI argued that the PEJ, which issued its ownership-quality report on the eve of a FCC hearing on deregulatory issues, had not made the supporting data fully available for external review. A rebuttal article issued by EI stated, "statistical tests, run on the limited data provided, found that none of the principal findings is statistically significant" (as cited by Phipps, 2003, p. 6). Bruce M. Owen, the primary author of the rebuttal, specifically criticized the subjectivity and arbitrary aspects of the "grading system" (Plamondon, 2003, p. 83; Sloan, 2003). In a formal report submitted to the FCC, EI argued:

   There is an unacceptably high risk that the PEJ findings are
   attributable to random noise in the data. The PEJ Study reports no
   differences in percentages of newscasts that received a particular
   grade, but it fails to provide any statistical testing on these
   results. (Owen, 2003, p. 570)

EI has in turn been criticized as being nothing more than a "hired gun" of the corporate media chains. Tom Rosenstiel, PEJ director, said, "This is a rebuttal from a group of consultants hired by the networks to help them lobby for deregulation, and they were hired to discredit our study, not to do an independent analysis" (News Briefs, 2003, p. 6).

A study cosponsored by the Lear Center Local News Archive and the Wisconsin Newslab project at the University of Wisconsin-Madison found local news content disparities based on ownership characteristics in the coverage of the 2002 midterm elections (Lear Center Local News Archive, 2002). The 122-station sample contained 45 stations owned by large media groups (with nationwide audience reach of more than 20%), 50 by midsized groups, and by 23 small groups. The stations in small and midsized groups offered more coverage of local races than the national average, while stations in large groups provided less. The same pattern appears in individual media market comparisons: in 16 of the 22 markets sampled where comparisons could be made, stations owned by small or midsized media groups aired more local campaign coverage than their large-group counterparts in the same market. The findings were based on analysis of the highest rated half-hour news program aired during the early evening (4:00-7:30 p.m.) and the highest rated half-hour of late local news (9:00-11:30 p.m.) every night of the week on 122 randomly selected local television stations in the top 50 media markets in the United States. The broadcasts analyzed in the report aired from September 18, 2002 through November 4, 2002 (Rabinowitz & Derris, 2004).

While the Lear NewsLab study did not generate the level of controversy associated with the FCC and PEI studies, it was also flawed in some respects. First, the study was not designed to correlate ownership to local news content; only two paragraphs in the 26-page report discussed ownership content issues. Further, the data provided only one snapshot of campaign stories during the 2002 general election. In addition, its focus on election campaign stories limits the ability to generalize the data to all story types. The Local News Archive (at the time of this writing) only contained campaign-related news. Thus, the overall quality and character of the local newscasts were not assessed.

However, the Lear study did have several strengths. First, the simplicity of the method minimized potential controversy regarding interpretation of the findings. The data were generated on a straightforward quantification of story type, absent of any additional evaluative criterion. This approach suggests a methodological direction for conventional statistically based hypothesis-testing research. The intense politicization of the issue clearly demonstrates the need for a clear method, for objectively derived data, for open and rigorous review, and for a perceived lack of bias on the part of authors.

Toward an Investment Model of News Quality

In this article we argue for an investment-driven model to assess news quality. Since local television news is expensive to produce, we believe that the resources devoted to the production of news are a revealing indicator of the organizational commitment to news quality. Although respective philosophies of newscast quality may vary, the cost of producing local news is fairly constant (depending on regional costs of living differences) and is a highly salient and thoroughly scrutinized element in every domestic broadcast television news operation. A fiscal perspective provides an appropriate basis to evaluate the qualitative commitment a given news organization makes to local news--the higher the investment, the higher the qualitative commitment. Earlier research had partially incorporated this approach; the PEJ (2003) study defined the "enterprise" index to include how "much effort went into creating the story" (p. 19), including such factors as whether the station's assignment editor sent a full team (composed of at least a reporter and a videographer, and perhaps a producer) to cover a story or if only a videographer was sent. These elements (along with others in this category, such as the more subjectively assessed aspects, e.g., the "gutsy" nature of the story) would result in a higher or lower "grade" for that particular PEJ-category construct.

Production costs are increasingly relevant in the assessment of local news. Managers in local television news departments are pressured by falling revenues caused by continually fragmented news audiences. For example, a 2002 Nexstar acquisition of two Montana television stations was accompanied by the announcement that the recently acquired stations' local newscasts would be replaced with network news and/or chain-based syndicated programming because the "elimination of the expense of producing local news will mean the difference between profit and loss for these television stations" (Eggerton, 2003, p. 16). And in 2002 the Sinclair Broadcast Group eliminated locally originated newscasts at its ABC affiliates in Winston-Salem, NC, the nation's 44th market, and in St. Louis, the 22nd (Bergen, 2002). In the postderegulation media environment, television news departments must be revenue-generating units.

A tightly constructed investment model of local news quality can avoid the subjective methodology that has plagued earlier research. The organizational commitment toward local news can be more objectively assessed by examining factors associated with the production costs. By eliminating a subjective "grading" process across diverse (and sometimes undifferentiated) category constructs, our investment model avoids some of the perceived distortions that have limited earlier studies. More specifically, production elements--whether a story originated locally or from a distant location, whether a story utilized video produced by local news-department personnel as opposed to another so-called cost-effective source, and whether an on-air field-based reporter was utilized or if the reporter appears only by voiceover--can produce an empirical, evidential foundation to assess how much capital a news department invests in a given story, a relative measurement of the extent to which a news operation commits its limited resources to the production of local news. A quantitative approach to the measurement of these variables will provide insight into local news practices across differing ownership types (small groups versus large groups, e.g.).

An investment model of news quality should also attempt to quantify the extent to which promotions are used in a broadcast. In recent years the line between reporting the news and marketing the news--such as previewing provocative upcoming stories within a newscast--has become blurred. A study of 36 local affiliate newscasts found that about 9% of the newscast was devoted to such news promotion (Buchman, 2000). Thus, increased use of promotions is an inexpensive way to fill a news hole and to serve the concurrent function of "free advertising" for upcoming station programming.

This article attempts to document a method to measure investment patterns in the production of local broadcast television news across ownership types in the Tulsa, OK, television market. The underlying premise is that some news stories require a higher level of investment than others. Overall, it is hoped that an analysis of the Tulsa market can provide an adequate pilot test of our investment model of news quality across ownership types.

Based on economies of scale, it would be expected that larger chain-based media outlets would tend to use nonlocal "generic" news stories that can be shared with sister stations in other markets within the chain. Therefore, we propose the following hypothesis:

H1: A television news department operated by a small chain will air more local news (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations.

For larger chain-based media outlets, the production of locally identifiable video would limit its utility to be aired in multiple markets and would thus represent an additional expense. Consequently, we propose the following hypothesis:

H2: A small-chain television news department will air more locally produced video (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations.

The use of local reporters in the field represents an additional logistical expense for the news organization. In addition, featuring local reporters in news video limits the ability of news organizations to broadcast the story across multiple markets. Therefore, it would be expected that larger media chains, seeking to maximize profits, would make less use of local on-air reporters. Thus, we propose the following hypothesis:

H3: A small-chain television news department will air more stories featuring local reporters (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations.

News promotions are a cost-effective way to fill a news hole and may be more symptomatic of larger group ownership that makes extensive use of syndicated feeds. Accordingly, we propose the following hypothesis:

H4: A small-chain television news department will air fewer news promotions (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations.

Method

We now will describe briefly the nature of ownership types of the three major broadcast network affiliates in the Tulsa, OK, market. [2] KJRH, an NBC affiliate, is a part of the E. W. Scripps Company, a top-25, publicly traded, horizontally diversified media group headquartered in Cincinnati, OH. Scripps owns ten television stations, all affiliated with a major broadcast network, but it is perhaps better known as a newspaper company with over 20 daily publications around the country. Scripps also operates four cable and satellite television programming networks and a television retailing network. We consider Scripps to be a large media group owner. KTUL is owned by Mlbritton Communications, headquartered in Washington, DC. Allbritton operates television stations in eight markets, all affiliated with the ABC television network. We consider KTUL to be owned by a midsize media group--it operates only those eight TV stations and has no nationwide newspaper operation like Scripps. KOTV is owned by Griffin Communications, a small media group that owns only two television stations--KOTV and KWTV in Oklahoma City, [3] both affiliated with CBS. Griffin claims to focus their "resources in the state of Oklahoma--Oklahoma businesses owned and operated by Oklahomans" (Griffin Communications, n.d.). Griffin's corporate headquarters is in Oklahoma City, a 90-minute drive. Thus, KOTV comes the closest of all three stations to having local ownership. KOTV has historically won the late evening news ratings battle in the Tulsa market, with KTUL consistently coming in second (Sherrow, 2003, p. D6). Cumulatively, the three stations provide a representative cross-section of the relevant ownership types. For the purposes of our study both KJRH and KTUL are representative of larger multi-state media chains, whereas KOTV represents a rare example of a small intrastate-owned media television outlet.

Using content analytic techniques, we analyzed broadcast television newscast segments from September 1, 2003 to October 9, 2003, of the 10:00 p.m. newscasts of the three major network-affiliated stations in the Tulsa, OK, market. All segments of each broadcast--news, weather, and sports--were coded. The selection of the timeframe makes the sample a purposive sample, intended to fall outside of ratings weeks to avoid potential distortions resulting from attempts at hyping the ratings. The sample resulted in a total of 87 broadcasts and 2,527 news stories and segments. The three coauthors of the article defined the content variables and served as the primary coders. Each coder coded approximately one third of the total sample. The category constructs used for this study are:

(1) Local news: stories that originate within the broadcast reach of the station or originate within Oklahoma and are given local relevance.

(2) Locally produced video: stories that utilize video produced by the local station news personnel.

(3) On-air field reporters: stories that picture a local station reporter in the field or on the scene of a story, not including stories that feature reporters delivering a report from different parts of the newsroom or conducting stand-ups near the newsroom.

(4) Newscast promotions: newscast segments used to preview upcoming news segments or other network shows, including segments dealing with station promotional activities (a station staff member winning an award, e.g.) and transitions during the newscast that promote news personalities or that promote the high quality of the newscast.

Each of the coding categories is independent and dichotomous. The alternative (mutually exclusive) category is implicit (e.g., nonlocal news, no use of an on-air field reporter, no local video used in the story, nonpromotional segment).

An intercoder reliability test was conducted of 271 coding decisions (10.71% of the total sample). Since each content category was independent, four separate reliability coefficients were computed. A widely used coefficient of reliability can be assessed as the percentage of observed agreement between the coders (see Holsti, 1969). Table 1 reports the average percentage level of agreement for all three coders. Based on this measure, all categories exceeded 90% intercoder agreement. However, this statistic provides an inflated view of intercoder agreement since it does not take into account the level of agreement due to chance. This is particularly a problem when using dichotomous coding categories; by chance alone, agreement will increase as the number of categories decreases (p. 140). Based on this issue, intercoder reliability was recalculated using Scott's pi, which takes into account the extent of intercoder agreement that may result from chance (Scott, 1955). Scott's pi differentiates between the percentage-observed agreement and the percentage-expected agreement. Computation of Scott's pi was performed using the PRAM software (version 0.4.4) for reliability assessment with multiple coders. The minimal level of reliability is approximately 0.75 when using Scott's pi (Wimmer & Dominick, 2000, p. 154). Table 1 shows the levels of composite reliability for all content categories surpassing minimal levels of reliability.

The original nominal-level data were transformed to a ratio-level measure by using percentages based on story time length. Each story (the coding unit) was timed in seconds and as a result each broadcast had a cumulative total length in seconds. This allowed for a sum total (in seconds) of each variable category per broadcast (e.g., local versus nonlocal news, etc.). To produce a percentage average, this number was divided by the total number of seconds in the newscast. Several considerations led to our decision to convert our data from the nominal-level to the ratio-level. Since the news hole for each broadcast slightly fluctuated on a daily basis, it was felt that a percentage-based comparison would allow for more accurate comparisons between stations per broadcast (as opposed to a frequency count). A time measure would also allow us to assess the proportion of newscasts based on the relevant variables in "real" terms. The actual length of a story is a key point of relative emphasis within a broadcast. Finally, the creation of ratio-level data allowed for a more robust statistical procedure. Holsti's (1969) discussion of "systems of enumeration" discusses the possibility of using space/time measures that can serve as a measure in descriptive studies comparing mass media content (p. 121). To test the research hypotheses, the data from the two larger media-chain stations were pooled and group t tests were computed to compare stations by ownership type (large versus small).

Results

The first hypothesis predicted that a small-chain television news department would air more local news (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations. This hypothesis was supported. The mean percentage of local news was significantly higher (p < .0001, one-tailed) for the small-chain news department (M= 81.03, SD = 7.15) when compared to the two larger media chain stations (M = 70.8, SD = 9.37), t(85) = 5.17.

The second hypothesis predicted that a small-chain television news department would air more locally produced video (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations. Once again this hypothesis was supported. The small-chain news station had a significantly higher (p < .001, one-tailed) mean percentage of locally produced video (M= 56.06, SD = 8.38) when compared to the two larger media chain stations (M = 47.69, SD = 11.55), t(85) = 3.47.

The third hypothesis posited that a small-chain television news department would air more stories featuring local reporters (measured as an averaged proportion of a newscast) than news departments operated by larger chain-based stations. This hypothesis was partially supported. The small-chain news station had a significantly higher (p = .001, one-tailed) mean percentage of stories featuring local reporters (M= 34.01, SD= 9.61) when compared to the two larger media-chain stations (M= 26.34, SD = 10.73), t(85)= 3.25. However, the small media-group station KOTV, when directly paired with the larger chain-based station KJRH (M = 30.37), failed to produce a significantly different finding (p = .087, one-tailed) t(56) = 1.39. [4]

The fourth hypothesis predicted that the small-chain television news department would air fewer news promotions (measured as an averaged proportion of a newscast) than stations operated by larger chain-based news departments. This hypothesis was strongly supported. The small-chain news station had a significantly lower (p < .0001, one-tailed) mean percentage of promotions during the newscast (M= 9.69, SD= 2.12) when compared to the two larger media-chain stations (M = 13.08, SD = 2.83), t(85) = 5.67.

As reflected in all content categories (amount of local news, locally produced video, use of promotions, and use of reporters in the field), the small-chain television news department would appear to have a greater commitment to news quality than the larger chain-based stations (see Table 2).

Discussion

We take the findings to support two claims we wish to advance. First, the use of investment-based content categories is a promising indicator of ownership influence on news quality. Application of the model documented in this study demonstrates the model shows potential in illustrating news characteristics based on station ownership type, indicating clear differences between two large media groups' news operations and a locally based small-group station. We conclude the investment model of commitment to news quality offers substantial methodological advantages over other methods. A strong feature of the investment model is that it avoids the subjective debate over what constitutes "quality." By using an inferred monetary standard, the investment model gauges organizational commitment to the production of quality news in an era of newsroom-as-profit-center. Since many newsroom decisions are guided by how limited resources may best be directed and the subsequent impact on the newsroom's ability to maximize its profit-making potential, a monetary standard is clearly warranted.

The second claim we wish to advance resides in the arena of public debate regarding the effects of increased concentration of the domestic media. The data in this study contradict the assertions of former FCC Chair Michael Powell regarding the quantity and quality of local news provided by larger media chains. The results of this research offer support to critics who voice concern over the effects of increased concentration of ownership. In this case, the effects of large-corporate ownership on local news detrimentally impact the FCC's public policy goal of localism, news that is part of the vital function of the press. We have documented statistically significant differences in local news content (again, with the qualified exception of the use of on-air reporters) among three television news departments that have characteristically heterogeneous ownership types. Ironically the findings are consistent with a recently leaked internal FCC draft report that analyzed 4,078 individual news stories broadcast in 1998. The internal report found local ownership of television stations added almost 51/2 minutes of total news to broadcasts and more than 3 minutes of "on-location" news. However, this report has never been officially released and only came to light during a confirmation hearing of current FCC Chairman Kevin Martin (Associated Press, 2006).

The findings partially contradict the PEJ findings that "grade" locally owned stations as having lower quality newscasts. Ultimately, this disparity was based more on differences on how to measure news quality. However, when comparing the findings to the PEJ category of "story enterprise" (the category that comes closest to an investment measure of news quality) both studies produce similar results. In sum, our criterion supports the assertion that the small (locally owned) media group's television station made a greater investment in local news during the sample period.

There are two limitations of the present research that require acknowledgement and assessment. First, it could be argued that the small sample size used--three television news departments in one media market--would limit the generalizability of the results. One of the realities of the contemporary U.S. consolidated media landscape is that there simply aren't many small-group, locally owned television stations left in operation, suggesting that there is no population to which to generalize the results. The Tulsa television marketplace--one that engages a small-, a medium-, and a large-group station into direct competition with each other--is becoming increasingly rare and represents an increasingly unique environment to apply our metric. A 2005 review of the major network-affiliated stations in the top 50 markets indicated that only 1.5% are small-group, locally owned television stations, [5] If this trend continues, the findings of this research may provide insight into a passing historical phase in local television journalism. Nonetheless, an examination of other U.S. media markets that feature similar ownership dynamics represents one direction for future study, but finding such markets would be problematic. As such, the best use of the results of this study may be not to generalize to other markets but instead to describe the practices of a small-group-owned station in direct competition with larger group-owned stations within a single television market. This perspective has value on two levels: it can describe the effects of media consolidation on news practices in one particular market, and it can articulate how an investment model of commitment to local news might operate.

Second, Griffin Communications' sole two television stations, KOTV and KWTV in Oklahoma City, are arguably better situated to generate and to share statewide (read "local") news stories. However, this advantage may not be much different from a media company that operates a duopoly (a single company that operates both a newspaper and a television station within a single market). The PEJ study rated chains that operated duopolies as producing the highest quality news. In this sense, this could be an argument in favor of more finely tuned FCC regulations that would allow consolidation on a regional level or across media types. This regulatory approach could create strong local- or regional-based ownership groups that would not dominate a single market but could pool newsgathering abilities on a local or regional scale (as opposed to a general syndication of generic news stories being disseminated through local affiliates).

We believe the investment model is an important methodological departure from previous studies in media consolidation research, bringing into sharper relief an issue associated with recent trends in mass media research--the proliferation and subsequent use of independent and commissioned media research instead of the traditional channel of academic blind peer-review for the dissemination of research findings. Over the long term, the willingness of independent academics to address larger media issue areas may decline. In this sense, too few big research eggs are currently being placed into far too few baskets. Ironically, the field of media studies may he experiencing its own form of consolidation.

It could be argued that the circumvention of the lengthier blind peer-review process may result in questionable methodological practices. Further, the production of faculty-driven, small-scale research generated within the academic, blind peer-reviewed context can perform a valuable form of cross-validation. The collective findings produced by small-scale academic-based research could provide a better sense of general trends in media phenomena. While independent groups may produce potentially valuable findings with their ability to perform studies on a national scope, they also tend to preempt traditional academic research in media studies. If general trends emerge from the small-scale approach, it would be difficult to dispute the collective findings as suffering from source bias. In order for small-scale studies to be comparable to larger research efforts, greater methodological continuity is needed. Our metric can make a contribution to that end, leading to measurable levels of validity, perceived objectivity, and generalizability.

References

Associated Press. (2006, September 15). Report allegedly found that ownership rule changes could hurt local TV news. Akron Beacon Journal, p. A3. Retrieved February 18, 2007, from Newsbank database, http://www.newsbank.com.

Baker, D. (2002). Democracy unhinged: More media concentration means less public discourse is a critique of the Federal Communications Commission (FCC) studies on media ownership. Retrieved August 16, 2005, from http://www.dpeaflcio.org/pdf/FCC_Critique.pdf

Barrett, R. (2003, November/December). Canned news: What does it mean when local TV news isn't local? Columbia Journalism Review, 42(4), 9.

Bergen, K. (2002, February 9). Television networks begin to trim local news shows as costs rise. Chicago Tribune, p. C47.

Besen, S. M., Krattenmaker, T. G., Woodbury, J. R., & Metzger, A. R. (1984). Misregulating television: Network dominance and the FCC. Chicago: University of Chicago Press.

Broadcasting & cable year book. (2005). New Providence, NJ: R. R. Bowker.

Buchman, J. G. (2000). Television newscast promotion and marketing. In S. T. Eastman (Ed.), Research in media promotion (pp. 265-296). Mahwah, NJ: Lawrence Erlbaum Associates.

Clement, J. (2003, May 30). A mega media mess is on the horizon. Long Island Business News, p. 43A.

Compaine, B. (2004, January). Domination fantasies. Reason, 35(8), 26-34.

Corrigan, D. (2003, June). FCC ignores public trust; Aligns with big boys. St. Louis Journalism Review, 33(257), 30.

Crystle, L. M. (Executive Producer). (2003, June 4). PBS NewsHour [Television broadcast]. New York and Washington, DC: Public Broadcasting Service.

Eggerton, J. (2003, October 6). Nexstar nixes newscasts. Broadcasting & Cable, 133(40), 16. Retrieved January 6, 2008, from Communication & Mass Media Complete database. http://web.ebscohost.com.

Greppi, M. (2003, March 3). Study raises nackles. Electronic Media, 22(9), 4-6.

Griffin Communications, L.L.C. (n.d.). Retrieved May 19, 2006, from http://www.griffincommunications.net/

Hollings, E. F. & Dorgan, B. (2001, June 20). Your local station, signing off. The Washington Post, p. A27.

Holsti, O. R. (1969). Content analysis for the social sciences and humanities. Reading, MA: Addison-Wesley.

Lear Center Local News Archive. (2002). Local TV news coverage of the 2002 general election. Retrieved September 13, 2004, from http://www.localnewsarchive.org

Levin, H. J. (1980). Fact and fancy in television regulation: An economic study of policy alternatives. New York: Russell Sage Foundation.

McChesney, R. W. (1995). Telecommunications, mass media, and democracy: The battle for control of U.S. broadcasting, 1928-1935. Cary, NC: Oxford University Press.

Munro, N. & Vaida, B. (2004). Free market cheerleader. National Journal, 36(27), 2092.

Napoli, P. M. (2001). The localism principle in communications policymaking and policy analysis: Ambiguity, inconsistency, and empirical neglect. Policy Studies Journal, 29(3), 372-386.

News Briefs. (2003, March 17) Television Week, 22(11), 6.

Online NewsHour. (2003, June 2). Michael Powell transcript. Retrieved June 15, 2003, from http:// www.pbs.org/newshour/bb/media/jan-june03/powell_6-2.html

Owen, B. M. (with Baumann, M. & Mickkelsen, K.) (2003). The project for excellence in journalism's

PEJ study of ownership and quality of newscasts: A critique. Submitted to the FCC Biennial Review. Washington, DC: Federal Communications Commission.

PEJ Project. (2003). Does ownership matter in local television news: A five year study of ownership and quality. Washington DC: Project for Excellence in Journalism.

Phipps, J. L. (2003, March 17). News briefs. Television Week, 22(11), 6-9.

Plamondon, A. L. (2003, August). Proposed changes in media ownership rules: A study in ventriloquism? Communications & the Law, 25(2) 47-94.

Rabinowitz, S. & Derris, J. F. (2004, May 10). Ads trump news in broadcasting. Oklahoma Observer, p. 16.

Scott, W. A. (1955). Reliability of content analysis: The case of nominal scale coding. Public Opinion Quarterly, 19, 321-325.

Sherrow, R. (2003, June 21). KOTV tops May TV ratings. Tulsa World, D6.

Sloan, R. (2003, March 25). PEJ vs. networks on study significance. Retrieved on August 23, 2004, from http://poynteronline.org/

Spavins, T. C., Denison, L., Roberts, S., & Frenette, J. (2002). The measurement of local television news and public affairs programs. Washington, DC: Federal Communications Commission.

Trigoboff, D. (2003, March 17). Network O&O newsies slam report slamming them. Broadcasting & Cable, 133(11), 6.

Turner Broadcasting System, Inc. v. FCC 512 U.S. 622 (1994).

Turner, T. (2004, July/August). Break up this band! Washington Monthly, 36(7/8), 30-37.

Wimmer, R. D. & Dominick, J. R. (2000). Mass media research. Belmont, CA: Wadsworth.

Notes

[1] Localism, competition, and diversity are three of four policy goals currently used by the FCC to define "the public interest." The fourth, service to all geographical areas of the United States, was an artifact of the emerging television area in the late 1940s and early 1950s and is no longer regarded as a relevant policy goal (see Besen, Krattenmaker, Woodbury, & Metzger, 1984; Levin, 1980).

[2] Currently the Tulsa market is ranked 60th based on the number of television households.

[3] In October 2005, Griffin Communications announced their intent to purchase KWBT in Tulsa, which would bring their total to three television stations, all within the state of Oklahoma.

[4] Although this was not a hypothesis of interest, the two larger chain-based stations, KJRH and KTUL, demonstrated striking similarities. With the exception of "on-air reporters" (p < .005; two-tailed), three out of the four content categories did not reveal any significant differences between the two stations.

[5] This review was based on data provided in Broadcasting & Cable Yearbook (2005) and individual station and corporate Web sites. A small television chain was defined as three stations or less. Local ownership was defined as the corporate or ownership headquarters as being within an approximate 100-mile radius of the television station.

David K. Scott (Ph.D., University of Oklahoma, 1994) is a Professor of Communication Studies at Northeastern State University (Tahlequah, OK); Robert H. Gobetz (Ph.D., University of Oklahoma, 1992) is an Associate Professor of Communication at the University of Indianapolis (Indianapolis, IN); Mike Chanslor (Ph.D., University of Oklahoma, 1995) is an Associate Professor of Communication Studies at Northeastern State University (Tahlequah, OK). The authors would like to thank Renee Ridge and Amy Aldridge Sanford for their assistance in the preparation of the manuscript. An earlier version of this article was presented to the Mass Communication Division of the 2005 National Communication Association Convention, Boston, MA. Correspondence to: David K. Scott, Department of Communication Studies, Northeastern State University, 609 North Grand Avenue, Tahlequah, OK, U.S.A. E-mail: scottd@nsuok.edu

Table 1 Intercoder Reliability

Content Category   % of Agreement   Scott's Pi

Local News             0.961          0.900
Local Video            0.925          0.849
On-Air Reporter        0.926          0.751
Promotions             0.975          0.962

Table 2 Percentage (Per Broadcast) Mean of Content Categories by
Station

                   KOTV *            KJRH            KTUL

                    M      SD      M      SD       M      SD

Local News        81.03   7.15   70.96    9.49   70.64    9.42
Local Video       56.06   8.38   48.34   12.68   47.05   10.49
On-Air Reporter   34.01   9.06   30.37   10.38   22.31    9.65
Promotions         9.70   2.12   12.88    2.74   13.27    2.96

* KOTV was the locally owned television station; KJRH and KTUL are
considered large chain-owned stations for this study.