Capturing meaning-making in journalism

Performing content analysis is not merely a question of developing and using quantitatively-defined categories in order to investigate a research problem. If you want to go further and look at the meaning of the whole text, your analysis will meet challenges that go beyond numerical classification and coding. The strict numerical procedures of quantifying, classifying and counting have the advantage of objective control and reliability within research communities. However, they tend not to capture the ways meaning is produced, communicated and understood. In this article, we outline a dual procedure for the analysis of meaning-making in journalism. Using the software NVivo 8, we combine a qualitative and quantitative approach to the analysis of news texts. Following Franzosi (2010), we move “from words to numbers” and develop a methodological framework that supports investigation of the ways different cultural context, political reality and journalism culture generate narrative differences and produce alternative meanings. We use the case study of the two-year-long newspaper coverage of the Tasmanian devils sent as a christening gift to Danish Prince Christian to locate the process of meaning-making. Our analysis shows that the reconstruction of an event ‐ a ‘real story’ ‐ generates one or more story frames, which are related to the newsworthiness of the event, shared and hunted down by all journalists regardless of the country of origin or newspaper format. We found that journalists acted differently when selecting frames and refining


Introduction
Meaning-making is a central feature of journalism practice.The minimal success of journalism fundamentally depends on the production of media text units (i.e.printed articles, broadcast programs) which are supposed to be meaningful to readers, listeners and viewers.Units without easily decoded meaning may be found in journalism, but will tend to prevent, complicate or totally block the intended process of public communication.An actual construction of meaning will often be seen to vary within certain limits without getting lost.
In this article, we aim to develop a method for analyzing journalism and construction of meaning in the news text.Acknowledging that meaning is not fixed, but context-specific, contestable and enforced by institutional practice of journalism, we look at the process of meaning-making in a comparative context.Our focus is on two journalism tools used to mediate reality, a frame and an angle.Exploring the coverage of one ongoing event in Australian and Danish newspapers, we want to set up a framework for capturing the variations of meaning-making in a crossnational perspective.Our assumption is that journalism practice creates the logic and shapes the structure of the journalistic field.
Using 'meaning' as a basic concept in scientific investigations is not without difficulties.Disciplines that have language as a primary or secondary research object, such as linguistics, journalism studies, psychology, sociology, philosophy, anthropology, cultural and literary studies, formulate their subject-specific definition of 'meaning' (Ogden andRichards 1985 [1923]) to develop secure frameworks for their investigations (Osgood et al. 1961(Osgood et al. [1957]]; Morris 1946).Despite these conceptual complexities, we chose a method of analysis that takes into consideration the 'meaning' of a story common to journalists and readers.By ana-lyzing the linguistic and narrative devices that make this commonality possible we hope to find out what in the end makes journalism practice meaningful to the public and the ways in which it takes place.Our focus is on the 'narrated story' property of articles.Our supposition is that the story presented in a written text evokes mental pictures of events that can take several forms, but each of the possible forms -in spite of mutual variations -has at least one common element: it has to be true and reliable (i.e.factual) to be trusted.
We are not trying to check out the truth value of the mental pictures or to investigate the relationships between narrated story and political and social reality.Instead, we are focusing on the different forms the story takes in a sample of articles in Australian and Danish quality (broadsheet) and popular (tabloid) newspapers.Our aim is to demonstrate how various journalistic accounts of the same story carry the potential to evoke different meaning in different national contexts.For that analytical purpose, we have developed a model of an invariant structure of a story.This model reflects the story as a whole, and parts and variations of the story can be traced in all articles in our sample.The story model functions as an artificial and hypothetical constant in the analysis and constitutes a framework for developing analytical categories used in the investigation of journalism meaning-making across national boundaries.In this way, we intend to test our supposition: that the story is narrated according to different patterns created by the social context and newspaper format.
Accordingly, we locate the meaning-making process within the process of framing.Using Gamson and Modigliani's (1987: 143) definition of frame as the "central organizing idea or story line that provides meaning to an unfolding strip of events", we looked at "issue specific frame" (De Vreese 2005: 55) to identify how the problem, in this case the gift of the pair of Tasmanian devils, is defined.Taking an 'issue-specific' approach to the study of frame accommodates our objective to develop a method for analyzing journalism and constructions of meaning in the news text.Firstly, the specificity of the story about the gift allows a move towards functional measurement of frame in terms of journalism tools used to produce the text (Entman et al. 2009: 176); secondly, it allows easy comparison of frames in cross-national context; and thirdly, it suits moderation of coding.The aim of the study was to generate empirical evidence for the development of a methodology suitable for capturing the process of meaning-making in journalism in the light of framing seen as a "knowledge structure that is activated by some stimulus and is then employed by a journalist throughout story construction" (Dunwoody 1992: 78).Journalistic frames are professional tools used to organize the story in order to cope with the tide of information, and our identification of four main story telling frames in the 'devils story' articles supports theory building around the tools, rather than the process of framing as such.
We acknowledge both the epistemological and organizational dimensions of frames.The epistemological component (we use the term 'frame' here) links the new information with the existing knowledge about the event, people or phenomenon, while the organizational dimension (we use the term 'angle') determines how this information is structured in the news text (see Grunwald and Rupar 2009).Our objective is to use and test a method that combines a qualitative and a quantitative approach to the analysis of the meaning-producing devices in a sample of news texts.

Meaning of 'meaning'
Today, most researchers agree on the view that all sorts of sign expressions are representational.Meaning is, accordingly, produced, related to and explainable within the contexts, where signs are supposed to function as tools of human communication.So meaning has to be studied within the interplay between communicating agents in different social situations (Halliday andHasan 1998 [1985]: 89-94;112-118;124-127) or as cognitive processes taking place inside human organisms (Lakoff 1987: 266-268;Lakoff and Johnson 1999;Lakoff 2003Lakoff [1988]: 420-422;]: 420-422;Fauconnier and Turner 2003: 453-514).From these social and cognitive points of view, meaning is supposed not to exist as transcendental ('disembodied') abstractions (as it is, for example, expressed in word lists for content analytical purposes where sin-gle words without context are associated with specific values), but as an integrated ('embodied') part of the human mind reflecting changing characteristics of the human perception of nature and society.From these points of view, meaning associated with texts is described as the results of constructional processes taking place outside the physical appearances of letters or sounds and inside the minds of the language users operating within language communities according to certain grammars, lexica and norms of use.
These principles have an effect upon our study of meaning when we decide below what the story evoked by the text material is all about.Our analytical categories are intended to capture how the narrated story varies in the sample of articles.The study began by analyzing and discussing a small sample of articles (five articles from the Australian sample).We independently identified four dominant frames with different values attached to them.For example, the story about the devils was seen as a story about the gift giving (frame), but sometimes in terms of affection and sometimes as a mistake.We discussed the findings and decided to introduce an 'angle' code to capture the variations of the frame.This 'manual-holistic approach' to finding frames in media content (Entman et al. 2009) was then applied to ten other articles (each coder chose five articles from her/his sample).The analysis confirmed the existence of four frames that could be further diversified into different angles within each frame.We developed a coding sheet by defining all categories and started coding.Each of us dealt with a nation-specific corpus of data (coder EG looked at the Danish papers, and coder VR at the Australian papers).Once it was done, we set up a meeting with a third coder -TB, a bilingual Danish and English journalism studies graduate -who was given the coding sheet and samples of data -ten per cent of the articles -from each corpora.The intercoder reliability was 92 per cent.We quantified the content hoping to get an indication of differences in journalism tools used across national boundaries and newspaper formats; our objective therefore was to get an "indication of something other than what is counted" (Krippendorff 2004: 6) rather than results that lead to statistical hypothesis.
As coders we therefore agreed -before the full coding was performed -on a structure of the meaning-evoking parts of the story which we wanted to investigate.In other words, we discussed and agreed on a commonly accepted, constant version of the story as a basis for the analysis of actual variations within the sample.This constructing procedure, meant for research purposes, uses a close reading of the material with a concept of the story and how it is presented as a background tool.
For the purpose of investigation, we have chosen a narrow and simple approach to the concept of meaning.To address the question of what contributes to the variations of meaning at the level of articles, we start with the concept of frame, which is broadly defined as background or system conditions "structuring our cognitive and social life" (Fauconnier and Turner 2003: 453).Frames function in journalism as observable and describable contextual devices that motivate and give meaning to different ways of seeing.As Gitlin 1980 (p. 7) says, frames are "persistent patterns of cognition, interpretation, and presentation, of selection, emphasis, and exclusion, by which symbol-handlers routinely organize discourse […]".The technological possibilities afforded by different media platforms, different economic and organizational conditions in media houses and news rooms are, at institutional level, frames that regulate professional thinking and action.At practice level, the communicative acts of journalism are framed by the fact that the essential task of journalism is to provide a 'truthful' account of reality; to inform by evoking interest, involvement and comprehension in the audience.
So, meaning is cognitively evoked as a result of the structured, socially-organized teamwork of all involved in the process of mass communication.This we assume to be a central principle which is shared (by the professionals) and persistent over time, which works symbolically and which structures meaningfully the social world of working journalists (cf.Reese 2003: 11).The construction and reconstruction of meaning takes place in different ways within the internationally shared communities of journalism.The patterns of journalism practices in different countries deserve detailed investigation: the choices of journalistic genres, the selection of topics and themes, and the specific linguistic and narrative tools adapted to the varying tasks of meaning-making.But how do we analyze meaning in the different (political, geographical, social, cultural) contexts?And how do we handle the details within a specific journalistic story in a way that shows sufficient consideration for the 'whole' and at the same time pays attention to the smaller elements of a story (e.g.words, phrases, paragraphs, headlines, story lines, themes, places, dates, participants)?
We have addressed the challenges of one method in an analysis of our case study.The story about the gift sent by the Tasmanian government to Denmark as a present for the occasion of the christening of Prince Christian, the first-born child of the Danish Crown Prince Frederik and his Tasmanian-born wife, Crown Princess Mary, ran for nearly two years in Australian and Danish newspapers.In both countries, the papers were covering the spectacular and problematic journey of a pair of Tasmanian devils, wondering if the animals were suffering from a disease which at that time nobody was able to diagnose precisely.
The Tasmanian government's decision to send a pair of devils to Denmark as a christening present elicited heated discussion among politicians, experts and ordinary people both in Tasmania and Denmark.The problem that the living gift might be infected with facial cancer was the central driving force of this long-lived story.Our first analysis of the media coverage of the gift of the devils (Grunwald and Rupar 2009), which focused on journalism curiosity and storytelling frame, provided a unique opportunity to compare journalism practices in a cross-national context and via different newspaper formats (broadsheet and tabloid).In this article, we expand this discussion and turn our gaze towards the methodological questions that arise in the attempt to unpack the ways news frames utilize meaning-making within a news story.

Meaning units
In a frequently-cited article by Robert Entman, the framing process at the textual level is described by three steps.Entman emphasizes his points by using italics: "To frame is to select some aspects of a perceived reality and make them more salient in a communicating text […]" (Entman 1993: 52).Three steps with relevance for our analysis are expressed in these few words.They are: (1) the perceived reality, which is presupposed by the next step; (2) the selection of some aspects, which again is presupposed by the third step; (3) the salience (of the selected aspects) in a communicating text.
We as researchers did not have direct access to the events (perceived reality) covered by the papers in Denmark and Tasmania.But by reading the 66 article items from the four newspapers, Berlingske Tidende, B.T., The Australian and The Mercury, which communicated and made salient different aspects of that perceived reality, we have been able to reconstruct the whole 'real story', its participants, places, driving forces and developments over time.Our reconstruction reveals differences in the performances of steps 2 and 3 above and indirectly shows different ways of handling, structuring and interpreting different perceptions of step 1, the perceived reality.
We claim that this reconstruction of an event -a 'real story'generates one or more frames (i.e. the story frames), which are related to the newsworthiness of the event, shared and hunted down by all journalists regardless of the country of origin or newspaper format.Our hypothesis is that journalists act differently when refining frames into angles, when they select and make salient aspects of a story frame at the level of the perceived reality.Our analysis generates support for this hypothesis, indicating that the choice of angles relates to a specific social context (country) and institutional setting (media format).
Our central point concerning sampling the text was that each article collected from the four papers should contribute to the story in question as a whole, i.e. sampling in this case should be controlled by qualitative criteria.Central keywords in searching the databases in both countries were 'Tasmanian devils', 'Denmark' and 'gift'.Assuming that each single article in our text corpus communicates a smaller part of a greater story, we let a close reading of all 66 articles motivate the development of our analytical categories.So sampling and categorizing were intended to follow a qualitative design and avoid the sort of quantitative top-down-procedure which characterizes much of today's con-tent analysis.Our procedure resembles what Roberto Franzosi in a book calls QNA, i.e.Quantitative Narrative Analysis (Franzosi 2010).According to Franzosi, "the categories of QNA are based on linguistic properties of texts […] Content analysis, on the other hand, bases its categories on the investigator's theoretical and/or substantive interests" (Franzosi 2010: 3).

Categories and measures
So our analytical procedure ends up quantitatively with numbers that reflect the coverage in the whole sample of the appearance of categories measured in number of words used to tell and expose (parts of ) the story to the reader.This combination of methods allows us to go "from words to numbers" (Franzosi 2010: 1), but we are following our own route to fulfill the task.The steps between the words and numbers are supported by the use of NVivo, a software application designed for combining a qualitative with a quantitative approach.
Drawing on Halliday and Hasan (Halliday andHasan 1998 [1985]), we presuppose that meaning is a result of contextual influence upon language use.Language per se is a complex, and in principle endless, system of meaning producing potentialities, which are selected and actualized only by their use.You cannot observe or grasp meaning without relating language to the contexts where it functions as a communication tool.With this background, our categories are connected not only to the occurrences of words or single expressions, but to the structure, structure parts and motivating devices of the whole journalistic story, which is followed and gradually exposed by means of articles written by Danish and Tasmanian journalists.For that purpose, we intend to show how NVivo software can be used to enrich the investigation of journalism's meaning-making activity.
We use the term story, a textual and cognitive concept, to make it distinct from events, which are primarily understood as socially organized actions and motivated announcements (Franzosi 2010: 17-18).The content of a story is organized and expressed in narrative text sequences (patterns of narrative discourse) which in journalism are described by a row of well-known models of narration.They all encompass possibilities of plotting a story of journalism into a text under the pattern names as 'the Inverted Pyramid', 'the Third Way of Narration', 'the Wall Street Journal Guide', 'the Drama Model', 'the Story Model', etc. (Clark 2008: 135-139;Blundell 1986;Franklin 1994Franklin [1986]]).

From words to numbers
Our route from words to numbers is based on investigation of those parts of the whole story which are identified in specific places in the single articles of our sample.We intend to reveal and formalize the story without making further references to the encompassing discussions over several decades of the relations between fabula (story) and sujet (narrative discourse) as narratological concepts introduced by the Russian formalists in the late twenties.
Instead, we use the seven well-known journalistic tools as key devices in search of the story and its constituents: who (person), what (event), where (place), when (time), why (cause), in which way (mode), with which consequences (effects).The first five of these questions are basic and make a framework for our categories.Our working hypothesis is that journalistic articles principally break up their selected stories into narrative salient-making sequences of text.The choice of a communication model for the 'real story' differs according to the individual characteristics of a journalist and newspaper formats.Different sets of news value patterns, plotting devices and genre models stand in a framing background.In order to achieve our objective of comparing actual ways of journalistic writing performances, it is necessary to use the reconstruction of the common story as an artificial constant that lies behind the narrative sequences brought in the newspapers.We decided on the story of the Tasmanian devils because it was a story where we could take a transnational approach to our research problem.
The research questions in the process are: (1) how the story content is distributed in the material; and (2) where the constituents are found in a model of the journalistic narrative discourse; a discourse constantly influenced by a set of criteria called news values.In short: by means of NVivo we have marked the variations of our qualitative categories and their limits in the material in order to be able to express them in number of words used within the categories.Each category is integrated as part of the narrated story.Our analysis shows variations of narration expressed in numbers of different papers and countries.

The story and its drive
In the construction of texts, the drive is an essential selection parameter when journalists seek, select and combine the elements of events into a story.The drive is evoked by narrative turning-point-values such as complications, conflicts, unsolved, essential and newly-emerged problems of the events in question.A drive creates the uncertainty which makes the elements of a story play together, evoking suspense.Any judgment about newsworthy reality is based on several characteristics of the event: consequences, proximity, prominence, human interest, conflict and oddity.These criteria, along with timeliness, which Masterton 1998 acknowledges as a pre-condition for any interest, are used in this study as a basis for judging how the events were selected and represented in the news.Judgment about the newsworthiness of reality is integrated in the process of choosing a story-telling frame.In journalism, the combination of news values, which we call drive, turns out to be an essential selection parameter for the elements of a story.
In our case, the persistent conflict between a ceremonious gift giving between nations and a severe risk that the solemnly received gift might be infected with an incurable disease constituted an ongoing story drive for nearly two years.There were other news values (cf.Grunwald and Rupar 2009) which evoked public interest in our story, but none of these seemed to account for the relatively long duration of the coverage in the media.The uncertainty whether the facial cancer disease of the Tasmanian devils will lead to an outbreak in Copenhagen Zoo still remains.
The following recurring story frames (issue frames) were identified: (1) the conflict creating drive and suspense evoking uncertainty; (2) the Tasmanian devils, their natural history and their situation as threatened animals; (3) the gift giving -the values and attitudes expressed by such an action; (4) the royals as infallible celebrities and symbols of national identity.These frames are mutually dependent and count as integrated parts of what we call below the whole story.Following Kenneth Burke's description of the constituents and motives of a story (Burke 1945), we formalized the frames and categories of the investigated story into the following network model (cf. Figure 1):

Analytical categories
Figure 1 conceptualizes the common story of the papers in both countries.
To obtain an answer to the question of how these basic elements are managed across the sample and what the differences are, we coded the material with the following categories as background: The three lists of categories were chosen to describe the variation of construction forms within each article treated as an analytical case.The article structure describes the plot principle or sequential structure followed in the article.Knowledge of this makes it possible to describe relations between story frames, angles and the plot principles known from most of the teaching literature of journalism.The following research questions can be 60 // answered by means of this coding: Do the article combine different story frames and in which way?Where in the sequential discourse structure is the angle exposed?Are there differences on these points between paper types and countries?In this way, the salient-making strategy followed by papers and journalists becomes visible.Coding for the use of story frames in each article makes it possible to observe differences between papers and countries with regard to selections and saliencies of parts of the whole story.Coding for the use of angles follows the same patterns of answer: Which angles are used in the articles?Is there more than one angle in the article?What is the relationship between story frames and angles in the different papers and countries?The basis for the construction of the lists of story frames and angles is our common and negotiated close reading of the articles in our sample.
As explained earlier, the coding was conducted by the authors of this article.To test the intercoder reliability, we involved the third coder and undertook a manual reliability check, but we also chose five coded articles from the English sample and conducted a NVivo test.The NVivo check showed a relatively high agreement rate (cf. the Kappa coefficient and the agreement percentage in Table 1).This high agreement rate is the result of continuous discussion of the way we interpreted the text fragments in relation to the categories we had developed.Some cells in the first row can be explained as follows: The five test articles con- sisted of 10,768 characters.As coders we fully agreed that 1.65 per cent (= 178) of these characters are integrated in the functions as headlines in the articles, and we agreed that the rest (98.35 per cent) of the characters do not serve this purpose.There is no disagreement on these observations.Following on from that there is no text which coder A had coded and B not coded, nor any text which coder B had coded and A not.The test result of the category coding of the entrances of the articles show the lowest Kappa coefficient (0.8645) and should be explained in the same way.The disagreement number here is explained by the fact that coder B coded something which coder A did not code.The program enabled the coders to find this text fragment in the data base of NVivo and discuss the reason for disagreement.

Distribution and combination of story frames and angles
Assuming that story frames are ways of structuring information (Gitlin 1980) into textual themes in order to provide knowledge about a topic, and given that this study has identified four distinct story frames which organize the news material in the 66 investigated articles, queries by means of NVivo make it possible to describe how the structuring is actually performed by each paper.The four frames are distributed and combined in differ-

Table 2
Distribution of story frames within the four investigated papers. 1     ent ways (cf.Table 2), and the analysis shows consecutively that each of them, whenever activated, takes a specific angle.
The conflict frame is activated when a text or a part of it focuses on different opinions and disagreements, especially between or among politicians and experts.The conflict frame gets a relatively high priority in Tasmanian papers (23.26 per cent and 41.43 per cent) where the heaviest controversies concerning the ethic and political implications of the gift giving decision take place, whereas the coverage of this conflict in the Danish papers is remarkably low (6.95 per cent and 9.32 per cent).
The devils frame is activated when a news text or a part of it focuses on the animals as the main theme.The priority of this frame is relatively high as it is reflected in the coverage of all four papers.Notable is the Danish paper Berlingske Tidende, where more than half of the coverage (56.25 per cent) in the paper is dedicated to the devils, especially by comparison with the relatively low number in B.T. (22.08 per cent).The numbers of the coded story frames show how the different editorial priorities of the papers investigated are reflected in their coverage selection.
The gift giving frame is activated when a news text or a part of it focuses on gift giving as a theme involving symbolic acts of friendship and affection.The gift giving as such has a relatively small presence, but a later analysis of the angles taken by the gift giving shows how this part of the coverage can contribute to the general story drive in the material.
The royals frame is activated when a text or a part of it focuses on the members of the Royal Family, their privileged and protected lives and the events in which they participate.This frame facilitates the reported events within the actual combination of story frames with pure positive values, i.e. national identity themes and a stimulating touch of fame and celebrity.This story frame has the highest priority (51.84 per cent coverage) in B.T. and -surprisingly -the lowest priority in The Mercury.
Each of the treated story frames takes, when it is used, a specific angle (cf. Figure 2, group '03.Angles').An investigation of the combinations between story frames and angles shows differences between countries and paper styles, which will be reported below.

The devils story frame
Our analysis shows that the approach to the Tasmanian devils story can take positive and negative values in both countries.But the distribution of the attitudes towards the destiny of the animals varies remarkably within each country.The positive attitude is expressed in taking an angle that describes devils as unique, attractive and exciting creatures.This angle has a relatively higher priority in the Danish paper coverage (61.33 per cent) compared to the distribution of the same angle in the Tasmanian coverage (38.17 per cent) where the negative attitude prevails with the devils described as threatened (83.67 per cent) and exposed to the risk of disease and extinction.
One may say that the numbers reflect differences in perspective on the topic explained by contextual differences -different political and cultural agendas in each country.Seeing the devils as a threatened species became an argument in the political controversy in Tasmania, while their uniqueness became the dominant angle in B.T.
Berlingske Tidende gives a higher priority (46.59 per cent) to the disease threat angle, as opposed to that of the devils as unique creatures, than B.T. does (27.38 per cent).The same seems to be the case when The Australian (100 per cent) and The Mercury (68.86 per cent) are compared.The parallel choices of angle values in the papers may be explained by the differences of paper style: Berlingske Tidende and The Australian are both quality newspapers, while B.T. and The Mercury belong to a category of popular papers which reflect and give priority to topics that appeal to the mass market in their respective countries.
The negative angle value (disease) in the Australian papers and the relatively positive marking in the Danish papers (unique creatures) is supported by the results of a word frequency investigation, where the three most frequent nouns in the Australian papers are 'devils', 'devil' and 'disease'.In the Danish papers, the most frequently used nouns are more neutral with 'djaevle' [devils] or 'pungdjaevle' [marsupial devils] as numbers one and two and [Crown Prince] 'Frederik' as number three.

The conflict story frame
Disagreements between protagonists in relation to the transfer of the devils from Tasmania to the Copenhagen Zoo in Denmark were expected to take place within the 'conflict' story frame.Following the close reading of the text, we refined the conflict frame into several angles: conflict taking place between politicians (32.01 per cent) and between politicians and experts (67.99 per cent).The NVivo query showed that these angles on the conflict are not exclusively activated within a single story frame, but may as well be included and activated within other story frames too.So the gift giving story can take both the conflict between politicians (45.34 per cent) and between politicians and experts (54.66 per cent) as angles.This is probably because these conflict angles give the most logical explanations of the complications that may spoil the rituals within the gift giving story frame.When the conflict angles are used within the devils story frame, it is exclusively as a conflict between politicians and experts (100 per cent).
The two conflict angles, political conflict and conflict between politicians and experts, prevail and keep the stories running because a precise diagnosis of the disease problem is not available, nor is a cure for the disease.The high number of the politicians' and experts' conflict angle within the devils story frame (100 per cent) may be interpreted as the experts, as participants in the controversy, having personal interests involved: it is often owners of wilderness parks who express the view of the devils as threatened species.An example from The Mercury (November 19 2006) shows this involvement of private interests: In April, devils from Mr Kelly's park were exported -against some expert advice -to Denmark as a christening gift for a Danish prince.
Environmental consultant and former Tasmanian national park ranger Steve Cronin is critical of the decision to send devils overseas and interstate.
These investigations show further that the well-known principle of journalism -an article should contain one story and the jour-nalist should take a single angle on a story -falls short in practice.The coverage of the devils story shows that a story may be influenced and structured by more angles and a single article may contain more than one story.For example, there are columns that sum up several interesting events in the previous week, one of them being the devils story.The investigation also shows (cf.Table 3) that the conflict in the Tasmanian papers is primarily presented as a conflict between politicians and experts (70.94 per cent), while in the Danish B.T., for instance, it is angled and understood in a one-dimensional way as a straightforward (and far away) controversy between politicians (100 per cent).On this point, Berlingske Tidende gives a more substantial and varied picture where the conflict is shown as taking place both between politicians and experts (60.96 per cent) and politicians alone (39.04 per cent).The explanation of this difference in choosing an angle seems to be a result of the difference in the newspaper format (popular press and quality press).The gift giving story frame The analysis reveals a continuous interplay between angles and story frames, which has to be kept in focus: selecting and structuring a flow of events needs multiple rhetorical devices to underline salient aspects of the stories.For example, the devils play a role as a social hub of the gift giving, which is described as a ritual accomplished within systems (Maus 2000(Maus [1923(Maus -24, 1950]: 35).Gift giving takes place on several occasions and with different scope but as a social act it expresses and confirms positive relationships and values between participants.As such, the gift giving is vulnerable to external factors which might negatively affect what it sought to accomplish.Therefore, the angles on gift giving (as an act of affection, a mistake or an act of cheating) to some degree serve as a continuously motivating power or driving force of nearly all story frames in the material (cf.Table 4 on the distribution of angle values on gift giving across story frames in all investigated papers).Every time the gift giving takes a negative angle, it functions as a story drive towards renewed balance, where the negative values are neutralized or replaced with positive values.There are two recurrent angles which structure the devils story frame.One angle presents the devils as a threatened species requiring careful and empathic protection in their natural environment.The other presents them as a unique attraction of high entertainment value.We have already indicated (see the earlier section 'The devils story frame') that Australian papers look at the devils as threatened, whereas the Danish papers present them as an attraction worth watching and enjoying simply for amusement.This is confirmed by the numbers in Table 4 where both angles show a high percentage of appearances (90.45 per cent and 80.50 per cent).These numbers show how the angles are used within the devils story frame, and how two of the three angles on gift giving are used mainly within the gift giving story frame (75.89 per cent and 100 per cent).The exception is the angle on the gift giving as a mistake (79.10 per cent).This clearly negative angle serves as an argument in the referred controversies within the conflict story frame.
The numbers in Table 4 show further that the angles of seeing devils either as threatened or unique creatures are primarily activated within the devils story frame.This could have been expected.What had not been expected was that these angles show effects on other story frames too.
Within the gift giving story, the act of affection angle is far more frequently used (75.89 per cent), than the angle of devils seen as unique creatures (15.85 per cent).The following examples from B.T. (May 31 2005), where the theme of threat is not mentioned at all, show the way it is executed: Hidsig tasmansk djaevel til prinsen [Hot-tempered Tasmanian devil for the prince] De fleste nyfødte bliver overdaenget med tøjdyr -den lille prins har nu fået et aegte plysdyr.Den tasmanske djaevel er dog for hidsig til at putte ned i vuggen.
[Most newborns are showered with soft toys -the little prince now has got a genuine plush animal.However, the Tasmanian devil is too hot-tempered to tuck into the cradle.] In Berlingske Tidende, the uniqueness and entertainment value can be observed in the following examples concerning the gift giving story from, respectively, March 30 and April 11 2006: Når fire tasmanske pungdjaevle, to af hvert køn, i naeste uge saetter poterne på dansk jord, kan de se frem til at rykke ind i mondaene omgivelser, der prismaessigt kan sammenlignes med en ganske nydelig menneskebolig.
[When four Tasmanian marsupials, two of each gender, put their paws on Danish soil within the next week, they may look forward to moving into fashionable surroundings, which in relation to what you get for the price you pay, may be compared to a rather posh human residence.]"Vi glaeder os til at se lille Christian med vidt opspaerrede øjne betragte disse dyr fra den anden side af jorden," sagde kronprinsesse Mary i sin takketale, hvor hun også omtalte Zoos nye royale pensionaerer som "langtfra nuttede".
["We look forward to seeing little Prince Christian wide eyed, observing these animals from the other side of the world," Crown Princess Mary said in her speech of thanks, in which she also referred to the Zoo's new royal lodgers as "far from cute"] Seeing the devils as a positive and amusing gift object differs significantly from seeing them as threatened species -e.g. in this piece from The Mercury on October 18 2005: Of critical concern in the decision to give the devils to Copenhagen Zoo is the need to ensure they are free of the deadly and debilitating Devil Facial Tumour Disease.
The disease, which has killed between 30 and 50 per cent of the devil population over the past four years, is notoriously hard to detect before its appearance as unsightly mouth lesions and gross facial deformities.

The royals story frame
The royal family, as presented in the newspaper material, live a privileged and protected life and endow those events in which they participate with positive values.They take on a role as national identity objects for both the present and the future and give an aura of fame and celebrity to people and events with which they connect and participate in.An investigation of possible angles on the royals story frame shows a distribution pattern almost exclusively in favour of this frame: the different uses of the celebrity angle and the national symbol angle fall inside the royals story frame (100 per cent).
An exception is the gift giving story frame, where the meaning of the gift giving story, to a limited degree (11.27 per cent), is exposed through an angle on royal imitative behaviour functioning as a symbol of friendship and open affection between people from the two nations.An example of this kind of exposure comes from The Mercury of April 11 2006: CROWN Princess Mary and Crown Prince Frederik giggled and tried to copy the sounds of four Tasmanian devils when they [: the devils] arrived at Copenhagen Zoo as a christening gift to the couple's baby son yesterday.
The royal couple tried to emulate the loud growls of the small carnivorous animals, which were given to the zoo by the State Government.
The four devils -two males and two females -will be kept in a specially built facility in a section of the zoo that has other animals from Australia, including kangaroos and emus.
Frederik, Denmark's future king, kneeled as he tried to get the animals' attention through a thick window.
Princess Mary was more familiar with the animals of her homeland.
Tourism Minister Paula Wriedt, in Denmark for the presentation, said the group of captive-bred devils arrived safely at their new home at Copenhagen Zoo last week and were formally handed over yesterday."'This is an exciting time for both Tasmania and Copenhagen and this unique gift further strengthens the ties between us,'" Ms Wriedt said.But how are the three possible angles distributed in each paper within the royals story frame, and how can these results be interpreted?
In Denmark, Berlingske Tidende is (100 per cent) a newspaper that follows the activities of the royal family closely -the implications of social values connected to the royal institution regularly get very high priority and extensive coverage.But we have to be slightly guarded in making such a strong generalization because of the limited items in the material.The best way to document this priority is to look at some statements from a text.Here is a reportage (Berlingske Tidende, 4 August 2006) where personal attitudes of joy and admiration around Mary as a key character are communicated in conjunction with factual observations from a historical event.It is an example of the use of the royal story frame in a text, where the royals are seen (angled) as a significant and a positively-validated link between the nations involved: Mary mellem to nationer Royalt portraet: Så er kronprinsesse Mary blevet malet officielt.I går afslørede hun maleriet, som skal haenge på Frederiksborg Slot.
[The paintings in the orangery have been replaced with mirrors, which reflect her from behind with a view of scenery including the Hobart waterfront.The Dannebrog can just be seen in the ceiling decoration.The connection between Tasmania and Denmark has been established.] These value implications may also exist in the editorial policy of B.T., but they are not expressed in such a direct way.Instead, B.T. angles its version of the royal story frame as a story about admired celebrities walking around in an exotic universe (50.38 per cent) or else they make the royals symbols of national identity (49.62 per cent).All three angles in the Danish papers can be traced and documented in the Australian material too, but here they take different numerical values (cf.

Summary of findings
In all material, the four story frames are continuously activated, and each of them generates one of the ten possible angles.It was assumed that each group of angles is preferentially associated with only one story frame, but an investigation modifies this assumption in so far as frames and angles are combined in patterns which cannot be predicted precisely.The conflict story frame is preferably angled as a story about a conflict between politicians and experts (55.91 per cent), and the devils story is preferably written as a story about the devils as a threatened species (61.84 per cent).When it comes to the act of gift giving, the angle of affection is chosen most frequently with the highest coverage (62.85 per cent).The story about the royals is most often told as a story about celebrities (39.36 per cent), but the angle of the royals as symbols of national identity is not far behind (31.23 per cent).Deviations from the average numbers within the story frames can be observed in each paper.However, the basis for drawing sharp conclusions again seems to be relatively weak because of limitations in the material.But tendencies can be seen, and to a certain degree interpreted, and so can the possibilities of using the method on greater and more secure empirical material.
In each paper, fewer angles than the possible ten are activated.Furthermore, the spread between numbers of each of the chosen angles are greater, making the differences of the angling procedure and content focus clearer.
Where B.T. sees the conflict as a threat and uses the mistake angle on the gift giving as a way of exposing the conflict, Berlingske Tidende points to its roots by mentioning the protagonists of the conflict among politicians and experts.Where B.T. draws on emotions, Berlingske Tidende looks to those involved for attitudes and reasons which might explain the grounds for conflict.
The picture in The Australian is, similarly, relatively clear cut.But we have to guard against generalizations because of the limited number of article items (5) from this paper.
In The Mercury, the patterns of angling the conflict and the devils stories (Table 6) are similar to the patterns of The Australian.The slight differences between the papers are probably caused by a greater number of article items (36) in The Mercury.The gift giving story is angled as in B.T. with a preference for the affection angle (76.06 per cent) and, within the royals story, preferences for the royals as celebrities (31.10 per cent) and as symbols of national identity (36.04 per cent).This may be because of differences of paper style, but that would need to be confirmed by investigations of a sample which comprised more items from quality papers.

Conclusions
The objective of this paper was to develop a model for analyzing journalism and its production of meaning using NVivo as an analysis tool.We aimed to capture the variations of meaningmaking by looking at the news texts about the gift of the devils published in two different countries (Australia and Denmark) and in two different newspaper formats (popular and quality).Following Franzosi's (2010) methods of analyzing text "from words to numbers", we developed a methodological framework that consists of eight steps: (1) close reading of the text to identify the central organizing ideas of the story; (2) defining, naming and classifying these ideas as dominant story frames; (3) identifying variations within each frame; (4) defining, naming and classifying variations of frames as dominant story angles; (5) coding of news texts to identify frames and angles; (6) using NVivo to identify the frequency and coverage of the frames and angles appearance; (7) using NVivo to identify patterns of use in comparative perspective; (8) interpreting data in the light of social and institutional context.Our analysis shows that the reconstruction of an event -a 'real story' -generates one or more story frames, which are related to the newsworthiness of the event, shared and hunted down by all journalists regardless of the country of origin or newspaper format.We found that journalists acted differently when refining frames into angles, where the choice of angles relates to a specific national and media format setting.
The main advantage of the applied method is the precision in identifying journalistic tools used to produce a specific meaning.The variation of frames across the countries, Australia and Denmark in this case, is usually attributed to the influence of a social context.Indeed our analysis confirms this finding, but it would be impossible to reconstruct the specific, journalism-produced meaning of the events surrounding the gift of the devils (the 'real story'), if frames alone are examined.It is the comparison of the angles -influenced both by the national context and the model of journalism exercised in quality and popular papers -that explains why the Danish public got the story about the gift of the devils as a story about unique creatures, while the Australians focused on the devils seen as a threatened species.This type of analysis provides empirical material that generates comprehensive description and interpretation of differences in the newspaper coverage of events.However, for the purpose of drawing significant theoretical conclusions about the relationships implied in a study, such as the tension between frame and angle as journalistic tools, further investigation is needed.

Figure 1
Figure 1Model of the whole story, framing the varying presentations in papers and countries.

Table 1
Categories with highest and lowest kappa coefficient in an inter coder reliability test.

Table 3
Distribution of angles on the conflict story frame within countries and papers.

Table 4
Distribution of angles within the devils and the gift giving story frames in all papers.
Table5).The celebrity angle gets greater coverage (70.27 per cent) in The Australian than in The Mercury where the highest percentage (42.32 per cent) is taken by the royals seen as national identity symbols.Two aspects have to be investigated further in order to explain the number variations: (1) editorial values and priorities, which might affect the choice between angles; (2) influences from the sample, which in our case contains relatively few article items.

Table 5
Distribution of angles on the royals within all four papers.

Table 6
Connections between angles and story frames in The Mercury.