A content analysis of six newspaper reports on Ebola in Nigeria in the Telegraph and the Mirror from July 20th 2014 to October 20th 2014; with particular focus on the adjectives used, and sources cited.
In the words of Jawaharlal Nehru (the first Prime Minister of independent India):
A moment comes, which comes but rarely in history, when we step out from the old to the new; when an age ends; and when the soul of a nation longsuppressed finds utterance.
This quote will form the basis of my inquiry into whether the voice of Nigeria was suppressed in the coverage of the Ebola crisis. In addressing my hypothesis I have selected six reports, three articles from each newspaper from the day the virus entered Nigeria (July 20th 2014), to the day the country was declared Ebola free by the World Health Organisation (WHO) — 20th October 2014. Both newspapers were selected by their market segment and political inclination: the Telegraph being a broadsheet right of centre newspaper, while the Daily Mirror is a tabloid newspaper with inclination to the left of the political spectrum.
In addressing Content Analysis I am reminded of a statement by Robson (2002) on making enquiry, a term he borrowed from detectives to describe the process of ascertaining the best research method. In his words,
Put in more usual research language, watching becomes observation; asking becomes interviewing, using questionnaires and administering tests (Robson 2002: p. 223).
- “is a research technique for the objective, systematic, and quantitative description of the manifest content of communication” (Berelson, 1952: 18).
- “the systematic and replicable examination of symbols of communication, which have been assigned numeric values according to valid measurement rules, and the analysis of the relationships involving those values using statistical methods, in order to describe the communication to its context, both of production and consumption” (Riffe et al, 1998: 20).
- “In one sense, all qualitative data analysis is content analysis in that it is the content of interviews, field notes, and documents that is analysed. Although this content can be analysed qualitatively for themes and recurring patterns of meaning, content analysis historically has been very quantitative in nature” (Merriam, 1998: 160)
All three definitions allude to the fact that it is systematic and quantitative. Within the toolkit for researching texts and distinctive in its own right, content analysis has its place with other methods like: textual analysis, discourse analysis, and semiology. It is systematic in the sense that it uses a fixed research guide; and scientific in the sense that it can be replicated by different individuals with certainty that the interpretation of results correspond with data generated from the process. It is also objective and replicable once the parameters are set using a Coding Sheet, on this point it is important to emphasise that content analysis is as valid as the coding scheme is valid to the enquiry at hand. Therefore, the design of the coding scheme is imperative to a successful research project.
According to Robson (2002),
Content Analysis came to prominence in the social sciences at the start of the twentieth century, in a series of quantitative analyses of newspapers, primarily in the United States.’ (p. 351).
It is effective in comparing: representations, contents; and, it can be applied across media and formats e.g. video games, TV programmes, radio programmes, movies, and print. For instance, it could be used to analyse the amount of adverts in a video game, or representation of women in news or a TV drama. Content Analysis can be used to monitor, measure, and count subject matters — as is the case with this research.
However one of its limitation is the fact that resources are finite and our universe is vast, with this in mind the scope and samples size for research needs to be representative in order to allow for generalisation. For example, the first Ebola outbreak in West Africa was reported in Guinea in December 2013, and on 20th July 2014 Mr Sawyer (a Liberian-American diplomat) arrived at the main Nigerian international airport in Lagos with the virus. In terms of news outlet, we have: radio, television, magazines, newspapers, and the internet. In England alone there at least twenty four newspapers, which can be divided into the following segments: broadsheets, tabloid, mid-market broadsheet and freesheets. Clearly the burden on finances, resources and time poses a challenge on a research project of this magnitude. As a result, I am working on a couple of newspapers and six articles.
As part of the quality assurance process, the values need to be clearly thought through and defined; so like a funnel, attributes have to be streamlined to basic units in order to enable computation and analysis. For example, units of analysis (e.g. front page of the Sun), subjects (news about migrants), categories (politics, economics, housing, NHS, Schools), and variables (Africans, EU,).
Robson (2002), listed the following steps in carrying out a content analysis:
- Start with a research question;
- Decide on sampling strategy;
- Define the recording unit;
- Construct categories of analysis;
- Test the coding on samples of text and assess reliability; and
- Carry out the analysis
In comparison to other methods of data collection, Robson (2002) highlighted the following advantages and disadvantages:
- when based on existing documents, it is unobtrusive. You can ‘observe’ without being observed.
- the data are in permanent form and hence can be subject to re-analysis, allowing reliability checks and replication studies.
- it may provide a low cost form of longitudinal analysis when a run of series of documents of a particular type is available.
- the documents available may be limited or partial.
- the documents have been written for some purpose other than for research, and it is difficult or impossible to allow for the biases or distortions that this introduces (note need for triangulation with other accounts/data sources to address this problem).
- it is very difficult to assess casual relationships. Are the documents causes of the social phenomena you are interested in, or reflections of them (e.g.in relation to pornography and/or violence in the mass media)? (p. 358)
Coding Sheet and Sample
According to Merriam (1998),
the name of categories can be derived from three main sources: the researcher, the participants, or sources outside the study such as the literature (p. 182).
To this end I arrived at the following headers for my coding sheets (attached as appendices): Article ID, Date, Newspaper, Word count, Adjectives (two subcategories: positive or negative), Sources cited (three subcategories: reliable, vague or none), and Other (for anything that is relevant, but does not fit into any of the headers). I hasten to clarify that reliability of sources is based on either the individual’s proximity to the event or victim, expertise on the subject, or position of power. For example, a relative of the victim, a knowledge expert on the subject, a health professional based in Nigeria, a staff from Medicine San Frontiers (MSF), a WHO official, or a member of the Nigerian government.
My search was carried out on LexisNexis using query for ‘Ebola’ and ‘Nigeria’in ‘headlines’. With the start and end dates known, I worked out the total days of the event (93 days) in arriving at a median date. Three articles from each newspaper was selected based on reports on the start-date (July 20th 2014), median-date (4th September 2014), and, end-date (20th October 2014). Where no reports were available on any of these days, an article closest to the date was selected. And if more than one article was available on the same day the one with the most words was used.
Stereotype and News Values: Is the African voice being suppressed?
Using the analogy of the mechanical press to explain the impact of a negative image and the stereotype it leaves in the minds of people, Downing and Husband (2005), mentioned that:
a figurative stereotypes becomes a social and psychological definition of an ethnic or other social group, as something produced as a result of enormous, irresistible pressure that in consequence is completely fixed, ‘carved in stone’ so to speak, totally resistant to change or adaptation. (p. 32).
These images and stereotypes are likely to be evident in the types of words (especially adjectives) used and sources cited in the coverage of this news. By way of background the Nigerian government was able to track victims and contain the Ebola virus; using local expertise, innovative ideas and a fit-for-purpose strategy. Sadly, the country, its volunteers and health professionals did not get the praise and coverage they deserved — especially after all the negative publicity the nation and nationals have suffered from the English press over the years.
In preparing for this research, I read through a broad range of online materials, sources include: the BBC, Aljazeera, Bloomberg, CNN, BBC, British Medical Council, Vanguard (Nigeria), Guardian, The Independent, CBS, Huffington Post, The Atlantic, Walls Street Journal, The Times, Mail online, Daily Express; with a view to spot trends and observe themes.
During the course of my reading exercise, I was surprised to know that following the success by the Nigerian authorities, a contingent was sent to lead the African Union task force against Ebola in Liberia, Sierra Leone and Guinea. This team comprised of medical epidemiologist, field epidemiologist, physicians, nurses, laboratory technologists, laboratory scientist, data managers, and psycho-social professionals. Armed with this information, I turned to Galtung and Ruge’s (1965) seminal work on News Values, for answers. They defined news values
as specific standards adhered to by media professionals in the structuring, selection and production of news stories worthy of reporting to the audience by news reporter, journalist, and broadcasting media professionals also as a general guidelines and criteria that determines the worth of news and how much prominence is given by newspapers, radio, or television reporters.
They listed twelve factors responsible for the construction and reporting decisions in newspapers and broadcast news — a term they coined‘newsworthiness’. Therefore, an event is as worthy as the amount of criteria it addresses among these twelve factors. Through their research they worked through three basic hypotheses:
(1.) additivity hypothesis: the more factors an event satisfies, the higher the probability that it becomes news; (2.) complementarity hypothesis: the factors will tend to exclude each other; (3.) exclusion hypothesis: events that satisfy none or very few factors will not become news (pp. 64–91).
Application of Galtung and Ruge’s (1965) twelve factors:
- Frequency: the events occurred unexpectedly, and with Lagos State being densely populated, it was feared to be a disaster. However as the authorities stepped in to bring a speedy end to the virus, the news suddenly died out;
- Threshold: the threshold value witnessed a diminishing return once no new incidents were reported by the World Health Organisation (WHO);
- Negativity: the negativity factor quickly dried out as the virus was curbed through the efforts of the government of Lagos State and the joint task force delegated to deal with the outbreak of the virus;
- Unexpectedness: It was an extra-ordinary event when it was known that Mr Sawyer travelled via plane with seventy two other passengers to Nigeria, infected with the virus;
- Unambiguity: because of the negative perception of Nigeria, and the shock factor, the press were hesitant to report about the turn of events in Nigeria even after the WHO had given the country the all clear;
- Meaningfulness: other than the fact that Nigeria is an ex-colony and a member of the common wealth, the cultural proximity is quite wide in distance; as a result once the course of the story became positive, its news value dropped significantly;
- Personalisation: the outbreak could not be personalised beyond the death of the doctor Dr Stella Adadevoh who diagnosed and alerted the authorities in Nigeria of the first victim. After her death the personalisation factor dropped moved to her family, then it faded;
- Reference to elite nations: although an oil rich elite nation in an African context, Nigeria is not a global power. As a result there wasn’t much attention once the infection was curbed;
- Reference to elite persons: similar to the point raised with regards to personalisation, the virus did not affect any rich, powerful, famous and infamous get more coverage;
- Consonance: the initial outbreak fitted with the media’s expectations so it received a lot of coverage, but as time passed the media’s readiness to report on the Nigeria dwindled;
- Continuity: with the population, and perceived hygiene and educational level of the masses, the story gathered momentum. However, this changed very quickly as the outbreak was contained through effective public information and mass mobilisation of volunteers and government officials;
- Composition: stories from Liberia, Sierra Leone and Guinea took centre stage as Nigeria contained the virus.
Applying Content Analysis
From Table 1 (see appendices), we can deduce that all the articles had at least one source cited, and they were all experts. However nine (64%) were likely to be close to the event in the sense that they were either in Nigeria, West Africa or directly responsible for staff on the ground; while (36%) were unlikely.
The total word count for both sets of articles was close; the telegraph was 2,177 and the Mirror 2,263; this was a surprise as I was expecting the coverage from a broadsheet newspaper to be bigger than that of a tabloid newspaper. I didn’t find any report on 24th September 2014 (the median date).
The Telegraph reported on 20th October (the day Nigeria was declared Ebola free), however the report was very condescending. The success was down to luck and the competence of a British trained doctor — directing the success to the UK. Within the same article, an unrelated news about the Boko Haram insurgency also stole this rare opportunity for the country to shine. The people involved were detached from the country in the report, they were praised almost as if they were not Nigerians — which I found bizarre.
For a nation that is a former colony and member of the common wealth, the reports lacked depth and width, as a result, I suspect that most Nigerians and others concerned about this epidemic would have had to look to else for in-depth coverage.
Very few adjectives were used as reporters seemed very distant from the event. A couple of the Mirror’s reports came across as alarmist, and they did not reflect the content of their headlines. In one case, a report with a headline on Nigeria suddenly switched to Sierra Leone half-way through with clearly no direct link within the story.
This exercise has been an eye opener for me, as I grappled with different theories around race and gender reporting.
With Nigeria’s recent history, a statement by Downing and Husband (2005), comes to mind,
It may be hostile or it may be held supportive — White people over the past millennium have often held supportive stereotypes of themselves and negative stereotypes of those they colonized — but it is rigid. (p. 32).
This line of thought was also highlighted recently in Blurs’ Damon Albarn’s criticism of the image of Africa portrayed in the Band Aid 30 song. In an interview with Cathy Newman (Channel 4) he said:
“Having been to many countries and gotten to know many people, it always seems that we have only one view of it,”
In reflection content analysis is a very effective tool, standing to its name as a method of analysing the contents of a piece of information.
There was evidence of little or no competition from newspapers to report on the subject, in fact most of the newspaper coverage were hardly dissimilar.
There were likely to have been logistical challenges especially to do with health and safety of reporters and their crew, however the risk factors are not different from the risks of war — perhaps this accounts for the lack of depth and width of coverage.
Perhaps advertisers were not likely to be associated with poor disease ridden Africans. This leads us to another salient point: where and how people in diaspora access news. In recent times we have seen a big drift to internet sources; TV channels like CNN and AL Jazeera; and radio programmes like BBC’s World Service. So, may be news from Africa is no longer financially viable for newspapers. May be we are not interested in news from the continent, may be we have exhausted our emotional capital, may be the newspapers have left that share of the business to bigger news companies…
Finally, for an event that ran for ninety three days, a seven percent sample is insufficient for thorough analysis, but sufficient enough for testing the hypothesis, and providing scope for further work on some of the questions raised in this report.
Balnaves, M., Shoesmith, B., and Donald, S. (2009) Media Theories and Approaches: A Global Perspective. Palgrave Macmillan.
Berelson, B. (1952). Content Analysis in Communication Research. Glencoe, Ill: Free Press.
Husband, P. C, with Professor John D. H. Downing (2005) Representing race.Thousand Oaks, Calif: London: SAGE, 2005.
Galtung, J and Ruge, M. (1965): “The Structure of Foreign News. The Presentation of the Congo, Cuba and Cyprus Crises in Four Norwegian Newspapers.” Journal of Peace Research, vol. 2, pp. 64–91.
Kelly, M., Mazzoleni, G. and McQuail, D. (2003) Media in Europe: The Euromedia Research group; Ed. by Mary Kelly. London: Sage Publications Ltd.
Merriam, S. (1997) Qualitative research and case study applications in education. 2nd edn. San Francisco: Jossey-Bass Publishers.
Robson, C. and Robson, P. (2002) Real World Research: A Resource for Social Scientists and Practitioner-Researchers. 22nd edn. Wiley, John & Sons, Incorporated.
Ryan, C. (1991). What’s newsworthy? In Prime time activism: Media strategies for grassroots organizing. Boston: South End.
Table 1: Data from Coding Sheet
6 completed coding sheets
3 Newspaper articles from The Telegraph
3 Newspaper articles from The Mirror