702047

research-article2017
PUS0010.1177/0963662517702047Public Understanding of ScienceAlcíbar

P U S
Theoretical/research paper

Public Understanding of Science

2018, Vol. 27(3) 365­–381
Information visualisation as a © The Author(s) 2017
Reprints and permissions:
resource for popularising the sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0963662517702047
https://doi.org/10.1177/0963662517702047
technical-biomedical aspects of the journals.sagepub.com/home/pus

last Ebola virus epidemic: The case

of the Spanish reference press

Miguel Alcíbar
University of Seville, Spain

Abstract
This study explores the role that information visualisation played in the popularisation of the technical-
biomedical aspects of the last Ebola virus epidemic, the most devastating to date. Applying content analysis
methods, the total population of information visualisations (N = 209) was coded and analysed to identify
topics, and to define features and identify patterns in the images. The corpus was based on the record of
articles with graphics appearing in five Spanish reference newspapers from 22 March 2014 to 13 January 2016,
the start and suppression of the epidemic, respectively. The results suggest that information visualisation
was a key factor in the popularisation of the epidemic’s technical-biomedical aspects, as well as contributing
actively to construct, in the words of Myers, a narrative of nature.

Keywords
content analysis, Ebola, infographic, information visualisation, science popularisation

1. Introduction
In 1920, the Swiss-German artist Paul Klee wrote in his Creative Confession: ‘Art does not
reproduce the visible; rather, it makes visible’. To paraphrase him, one could say that informa-
tion visualisation (which is also referred to as ‘graphic’ or ‘visual representation’) allows jour-
nalists not so much to reproduce the visible as to make visible what is not. Indeed, by means of
information visualisation, it is possible to reveal patterns and relationships that are not always
evident in a dataset, to represent entities (viruses, bacteria, organs, the internal structure of arte-
facts, etc.) and to describe molecular processes, biological cycles and the sequential stages of
natural phenomena or technical achievements which would not otherwise be visible, that is,
comprehensible. Visual communication in journalism can be understood as the activity of pro-
ducing visualisations of data and information graphics or infographics. As a matter of fact,

Corresponding author:
Miguel Alcíbar, Department of Journalism I, Faculty of Communication, University of Seville, Americo Vespucio s/n,
41092, Spain.
Email: jalcibar@us.es
366 Public Understanding of Science 27(3)

infographics has become one the best resources for communicating scientific information to
both specialists and laypersons (Frankel and DePace, 2012; Hornmoen, 2010; Silletti, 2015;
Trumbo, 1999). Infographics is a journalistic tool that cannot be reduced to something merely
ornamental, but should rather be conceived as a singular and effective resource for constructing
and popularising certain knowledge among wide audiences. It is important to stress that images
and headlines are the first elements of journalistic texts of which readers take note, as indicated
by eye-tracking studies (see Holsanova et al., 2006).
It is striking, however, that until recent decades, scholars have not paid due attention to science-
based images destined for both peer communications and dissemination and public consumption
(Hüppauf and Weingart, 2008). However, the mass media constitute one of the most characteristic
extra-scientific domains in which scientific information and knowledge is consumed in various
visual formats.
In spite of the fact that communication models recognising that audiences are heterogeneous
and active and need to be heard are postulated in academic circles, the truth is that communication
praxis in the written press is still anchored, in many respects, in the traditional model of dissemina-
tion of science information or science popularisation (see, for example, Schäfer, 2011; Summ and
Volpers, 2016).
This article addresses the analysis of information visualisations with a huge potential for popu-
larising scientific-technical knowledge. Specifically, it explores the role that they played in the
popularisation of specialised knowledge (technical-biomedical aspects) pertaining to the last Ebola
virus epidemic in the Spanish reference press.
The epidemic outbreak of the Ebola virus disease (EVD) in Guinea Conakry in December 2013
has been the most devastating to date. From there, the virus spread to other African countries and,
for the first time since being identified in 1976, reached Europe. On 7 August 2014, the Spanish
government repatriated the missionary Miguel Pajares, who had become infected in Liberia. Teresa
Romero, one of the nursing assistants who cared for Pajares at Carlos III Hospital, Madrid, fell ill
two months later, thus becoming the first person to be infected by the virus in Europe. The arrival
of the missionary from Liberia marked the beginning of a comprehensive coverage of EVD in
Spain’s major national newspapers.

2. Review of the literature

Information visualisation
Visualisation is a blanket term that refers to any visual representation designed for the purpose of
communicating, analysing, exploring and discovering patterns in information (Cairo, 2016). It is
a singular ‘functional art’ that combines words and images with a varying degree of abstraction
(Cairo, 2013). In the specific case of journalism, information visualisation is a powerful and con-
venient tool for constructing events or phenomena and making them understandable by means of
figurative graphics (illustrations, photographs, drawing, diagrams, maps, etc.) and non-figurative
ones (statistical graphs or data visualisations). The former represent physical phenomena and
retain certain similarities between the object and its visual representation, while the latter strike a
conventional balance between the object and its visual representation (Cairo, 2013). Both types
of graphics can form an integral part of infographics. Therefore, information visualisation is par-
ticularly important because it serves as a support for understanding: it not only allows us to
describe and explain objects, complex processes and events, but also helps us to discover, behind
a chaotic dataset, ‘connections, constants, patterns not always obvious to the casual eye which, if
they were not represented visually, would remain hidden from the reader’ (Cairo, 2008: 29; see
Alcíbar 367

also Kirk, 2012). The combination of iconic and verbal elements offers a greater explanatory
power than mere words and images on their own. Thus, information visualisation, in addition to
providing a certain degree of aesthetic pleasure, facilitates data analysis.
Data visualisation can be defined as the representation of numerical values (abstract informa-
tion) by means of charts, graphs and tables, in such a way that raw data are transformed for their
graphic presentation. It can be a very efficient way of showing a large set of numbers in a limited
space (Krum, 2014). For its part, an infographic, acronym stemming from the expression ‘informa-
tion graphic’, displays simultaneously a large quantity of data and information about the history of
an event, phenomenon or issue (Dur, 2014). It is usually made up of different parts (figurative and
non-figurative graphics, as well as written text) united by thematic juxtaposition to convey one or
several specific messages and to provide an explanation and context. According to Krum (2014: 6),
‘data visualizations by themselves are no longer considered to be complete infographics but are a
powerful tool that designers often use to help tell their story visually in an infographics’.
The term ‘data visualisation’ is useful for referring to those visual representations that are
produced in an algorithmic fashion (they allow for some customisation, but their basic design is
completely computerised), can be reused easily with different datasets and are aesthetically mini-
malist and rich in data (they can handle large volumes). On the contrary, infographics is a useful
term for referring to manually produced illustrations that, as a result of this process, present a
personalised treatment of information, are specific to a dataset (they cannot be reused with others),
are aesthetically rich (with a strong visual content keyed to engaging readers and sustaining their
interest) and are relatively poor in data (since each piece of information must be coded manually)
(Iliinsky and Steele, 2011). Both are multimodal texts, to wit, coherent entities of semiotic interac-
tion in which each mode, the visual and the verbal, plays a specific and balanced role (Kress and
van Leeuwen, 2006). In other words, they not only exhibit additional information, but also
allow readers to explore different levels of reading and establish less obvious data connections
which, otherwise, would not be so clear.

Classification of information visualisations

Given the wide range of existing graphics, hierarchical classifications and taxonomies have prolif-
erated (Desnoyers, 2011; Lester, 2006; Lohse, et al., 1994; Racine, 2002; Valero et al., 2014). It has
been assumed here that data visualisation and infographics are two different ways of visualising
information, although the former can be included in the latter. Moreover, the typology proposed by
Lester (2006) has been adopted and adapted. His basic idea is to differentiate between ‘statistical
infographic elements’ and ‘non-statistical infographic elements’. The first group includes graphics
displaying a scale which contains concise and comprehensible numerical information in order to
highlight data trends, which if presented in a purely verbal form could escape readers. The most
important are numerical data tables, line charts, pie charts, bar charts, area charts, bubble charts,
radar charts and scatter plots. Quantitative or statistical maps (‘data maps’, in the terminology of
Lester), namely, maps that combine geographical information with quantitative data (choropleth
maps, dot maps, proportional symbol maps, isoline maps and cartograms), also belong to this first
group.
The second group encompasses qualitative graphics: drawings, 2D and 3D illustrations, dia-
grams, flowcharts and floor plans. Diagrams are widely used in journalistic information. Indeed,
the most eye-catching infographics usually involve their use (Lester, 2006). There are two classes:
(1) structure diagrams, which statically describe a physical object and (2) process diagrams,
which describe the interrelations and processes associated with physical objects (Lohse et al.,
1994). These diagrams are normally accompanied by diverse visual devices (e.g. the enlargement
368 Public Understanding of Science 27(3)

of some detail of the representation, presentation of different perspectives, use of colours, arrows
or pictograms) and verbal ones (e.g. labels, callouts, explainers, etc.), which help readers to
understand the information (López-Manjón and Postigo, 2014).
This second group also includes, on the one hand, text-based graphics: fact boxes which
‘contain a series of statements that summarize the key points of a story’ (Lester, 2006: 188); time-
lines which allow for the visualisation of the most significant events of a phenomenon or history
along a horizontal or vertical line on which relevant dates are indicated (Lester, 2006) and sche-
mata which arrange information by means of arrows or opening curly brackets, and, on the other,
qualitative maps (‘non-statistical maps’), namely, maps that are not based on numerical data
(locator maps, explanatory maps and chorochromatic maps). Nonetheless, locator maps (with or
without a scale) are considered here as mixed (qualitative and quantitative), since on occasion,
they contain numerical data.
Furthermore, there is a third group of hybrid maps that simultaneously provides qualitative and
quantitative data (chorochromatic-proportional symbol maps and chorochromatic-dot maps).
And, lastly, there are flow maps that, by means of arrows (proportional or not to a numerical
quantity), represent the movement of objects, animals or people from one place to another.

Epidemic diseases in the mass media

Since the 1980s, medicine and health-related issues have dominated the public debate on science
(Bauer, 1998). In recent decades, scholars have shown a growing concern for the coverage that the
media dedicate to emerging infectious diseases that, on occasion, lead to epidemics. In addition to
EVD, media coverage has focused above all on the HIV/AIDS epidemic (see Mayer and Pizer,
2005), H5N1 avian influenza (see Hellsten and Nerlich, 2010; Koteyko et al., 2008), the H1N1
influenza pandemic (see Lee and Basnyat, 2013; Wagner-Egger et al., 2011) and the SARS epi-
demic (see Ding, 2009; Lewison, 2008; Wallis and Nerlich, 2005).
EVD, as with other types of epidemics, is a technical-biomedical problem (e.g. epidemiol-
ogy, vaccine development), as well as a socio-economic (e.g. prevention and research costs,
negative impacts on the affected populations) and political one (e.g. resource allocation, spend-
ing on vaccines and treatments) (Hellsten and Nerlich, 2010). The public health crisis arising
from the most recent Ebola epidemic claimed 11,000 lives and had dramatic socio-economic
consequences for the African countries concerned. In the Spanish case, the attribution of
responsibilities to the different administrations involved in the political management of the
crisis and the socio-economic repercussions of the health alert were, at the time, hot topics in
the press. While the political and socio-economic aspects of the epidemic tended to be better
understood by wide audiences, its scientific-technical explanation might have seemed more
difficult to grasp. Information visualisation had apparently a lot to do with the popularisation
of the epidemic’s technical-biomedical aspects.
A number of previous works have focused on the coverage of the Ebola outbreak of 1995 in
Zaire (Ungar, 1998), the representation of EVD in the British media as a remote disease (Joffe and
Haarhoff, 2002), the nature and scope of the Ebola epidemic in the United States (Basch et al.,
2014; Kane, 2015) and Spanish (Barberá and Cuesta, 2015; Revuelta et al., 2015) written press, the
institutional communication of the crisis in Spain (Micaletto and Gallardo, 2015), the metaphoric
discourse of the disease and its victims in the American liberal press (Trčková, 2015) and the rheto-
ric employed by the World Health Organisation (WHO) and its impact on the propagation of the
virus (Condit, 2015).
Despite the capacity of graphic representations to communicate the different technical-biomedical
aspects of EVD, there is a dearth of literature in this respect. Welhausen (2015) is the only author
Alcíbar 369

who has addressed the analysis of visual strategies in risk perception with regard to EVD. Hence,
this work intends to fill that void.

The use of visual aids to communicate other technical-scientific topics

Only a few studies have paid due attention to the role played by visual aids, besides photographs,
in the news coverage of other technical-scientific topics. Some authors have analysed the contribu-
tion of visual aids to the popularisation of technical information on climate change in the British
(Smith and Joffe, 2009), Canadian (DiFrancesco and Young, 2010) and US press (Rebich-Hespanha
et al., 2015). By the same token, it has been observed that for communicating environmental infor-
mation, the messages included in infographics are more dissuasive than others without visual aids
(Lazard and Atkinson, 2015).
There are also several studies that focus on the effects of graphic representations on audiences
in the field of health risk communication, in general (e.g. Ancker et al., 2006; García-Retamero and
Cokely, 2013; Lipkus, 2007; Stone et al., 2015), and as regards breast cancer risk, in particular
(Occa and Suggs, 2016; Schapira et al., 2006). The literature on visual aids for displaying epidemic
information is practically non-existent: Wong et al. (2013) have studied the influence of interactive
maps on the perception of the H1N1 flu pandemic; for her part, Shin (2016) explores the strategic
and visual characteristics of infographics used by health organisations and news media for dissemi-
nating epidemic information.

3. Research questions
Since there are no previous studies of the topic, there are currently no data on core issues, such as
how frequently information with technical-biomedical content is visualised in the written press.
Thus, the first research question was as follows:

RQ1. What is the proportion of technical-biomedical-related information visualisations with

respect to the total number of visualisations of Ebola-related information published during the
study period?

The suitableness of information visualisation for disseminating scientific information and

popularising specialised knowledge begged the following two questions:

RQ2. What technical-biomedical content associated with the Ebola epidemic is visually rep-
resented? How is this content distributed depending on the newspaper?
RQ3. Is there any relationship between information visualisation and technical-biomedical
content variables?

Expert information sources play a relevant role in disseminating scientific information, framing
specific interpretations and giving meaning to messages related to public health (Nelkin, 1995;
Rock, 2005). It was thus necessary to identify the sources employed by the media to cover the EVD
epidemic:

RQ4. What sources have provided journalists with information on EVD for their visual
representations?
370 Public Understanding of Science 27(3)

Moreover, it is interesting to determine the relationship between the graphics and the journal-
istic texts in which they were inserted. In addition, the presence or absence of references in the
written texts and/or in the graphics themselves to their representative nature could provide insights
into the type of narrative structure underlying the discourse on EVD (see Myers, 1990). To this
end, the following questions were posed:

RQ5. Is there any link between the infographics and the texts in which they are inserted? If so,
are there any significant differences between the link that an individual graphic establishes with
the text in which it is inserted and that established by a graphic visual representation in an
infographic?
RQ6. Are there any references to their representative character in the written text or in the
information visualisation itself?

Based on the syntactic structure of the headlines of the visual representations, the communica-
tion purpose of Ebola-related information visualisations was explored.

RQ7. What syntactic structures do the headlines of the graphics adopt?

4. Materials and methods

In order to answer the research questions, a content analysis was performed on the total population
of information visualisations relating to technical-biomedical issues arising from the last Ebola
epidemic in the Spanish reference press (El País, El Mundo, La Vanguardia, ABC, and El Periódico).
The study period lasted from 22 March 2014 (when Guinea Conakry officially declared that
there had been outbreak in the country) to 13 January 2016 (when the WHO officially declared an
end to the epidemic).
Using the databases of the aforementioned newspapers, all the journalistic texts containing the
keyword ‘Ebola’ were reviewed, extracting those that included graphics. After excluding those
texts with graphics that were not directly related to the epidemic’s technical-biomedical aspects (29
articles), 93 were obtained. Since 43% (n = 40) of the articles contained just one visual representa-
tion (e.g. a line chart) and 57% (n = 53) more than one (e.g. a bar chart, a structure diagram and a
fact box), individual information visualisations were taken as a unit of analysis. Consequently, the
total population of information visualisations analysed was N = 209 (40 individual visualisations
and 169 incorporated into infographics).
A coding sheet with the pertinent research categories was developed using an inductive and
deductive integral approach. After a review of the existing literature, a pilot study of a sample of
information visualisations (n = 50) was conducted. This hybrid approach facilitated the definition
of two large sets of categories: (1) basic identification data (date, newspaper, section, genre and
source) and (2) the individual features of each information visualisation (syntactic structure of the
headline of the graphic, reference in the body of the text to graphic content, representative charac-
ter of the image, type of visual representation and nature of the technical-biomedical content).
Two independent coders, including the author, coded all of the categories, though the intercoder
index was only applied to the non-descriptive ones (second set of categories). A post-graduate
assistant was trained as a coder. Regular meetings were held over a period of 3 months to unify
criteria and allow the assistant to familiarise herself with the code book and the research objec-
tives and questions. Krippendorff’s alpha, a widely employed index in communication research
Alcíbar 371

(Krippendorff, 2011), was used to determine intercoder reliability. On the whole, a coefficient of
α ⩾ 0.80 was considered reliable, whereas coefficients of 0.80 > α ⩾ 0.67 indicated that the con-
clusions should be regarded as tentative (Krippendorff, 2013). The following alpha values were
obtained: syntactic structure of the headline (α = 0.87), textual reference to graphic content
(α = 0.74), reference to the representative character of the image (α = 1), type of visual representa-
tion (α = 0.83) and nature of the technical-biomedical content (α = 1).

5. Results and discussion

RQ1. It was observed that 87.8% (N = 209) of the total number of Ebola-related information visu-
alisations registered during the study period corresponded to those with technical-biomedical content
(basically, epidemiological data, viral life cycle, pernicious effects of the infection, or control
measures for treating patients to minimise the risk of infection). The other 12.2% (not considered
in this study) concerned non-medical information on the Spanish patients (especially Miguel
Pajares and Teresa Romero) and the epidemic’s collateral aspects (e.g. locator maps of a purely
illustrative nature, graphics showing diverse social parameters of the African countries affected,
versus the same parameters for Spain, or graphics establishing a parallel between EVD and the fall
in the stock market of several companies in the air transport sector).
Other studies have obtained similar results. In her analysis of infographics relating to epidemic
risks, Shin (2016) compared the main sources that produced them with their communication goals.
She observed that, in media infographics, nearly 85% of the messages were of an informative
nature, namely, they displayed technical-biomedical information about the disease. On the con-
trary, in the infographics of health organisations, informative messages only accounted for 46% of
the total.
RQ2. Most of the content was published in August and October 2014 (34.7% and 35.8%,
respectively), coinciding with the repatriation of the Spanish missionary Miguel Pajares from
Liberia (6 August) and the assistant nurse Teresa Romero’s infection in Madrid (6 October) (see
Figure 1).
Eight technical-biomedical aspects were identified in the media coverage of this issue: ‘epide-
miological data’ (56%), ‘prevention and control measures’ (16.3%), ‘injurious effects’ (12.4%),

Figure 1. Distribution of frequencies of articles with information visualisations during the epidemic.
The sole article published in 2016 (January) is not shown.
372 Public Understanding of Science 27(3)

Table 1. Distribution of frequencies of the three information scenarios used to contextualise

epidemiological data.

Frequency Percentage
Scenario 1 91 77.8
Scenario 2 22 18.8
Scenario 3 4 3.4
Total 117 100.0

‘therapies and treatments’ (6.7%), ‘causes of infection’ (3.8%), ‘viral life cycle’ (3.3%), ‘viral load
in body fluids’ (1%) and ‘molecular structure of the virus’ (0.5%).
The first (‘epidemiological data’) refers basically to the distribution of the geographical areas of
West Africa affected by the disease; the time evolution of the number of people infected, deceased
or cured; and mortality rates. During a public health crisis, covering epidemiological information
as it is produced is a priority for the mass media, hence its high frequency here. This epidemiologi-
cal information was published in all of the months preceding and following the peaks occurring in
August and October 2014 (69% of information visualisations), which indicates that it was a recur-
rent and continual phenomenon throughout the media coverage of the issue. In April 2014, several
days after the official announcement of the epidemic outbreak, the media began to publish epide-
miological data in qualitative map format (reliable numerical data were still unavailable). It was
not until June that quantitative data of an epidemiological nature were first published.
Table 1 shows the three scenarios that the written press created to contextualise the epidemio-
logical information. Scenario 1 was the most frequent (77.8%) and pertained exclusively to
information on the ongoing Ebola epidemic, Scenario 2 compared data on the current epidemic
with those of other previous Ebola outbreaks (18.8%) and Scenario 3 compared data on the
ongoing Ebola epidemic with those pertaining to different viral epidemics occurring beforehand
(e.g. SARS, bird flu, Middle East Respiratory Syndrome [MERS]) (3.4%).
From the data shown in Figure 2, several interesting conclusions can be drawn about the dis-
tribution of technical-biomedical content by newspaper. ABC (a conservative, monarchical and
Catholic daily) and El Periódico (a progressive Catalan newspaper) were those that gave most
coverage to this kind of content (25.4% and 24.9%, respectively); the newspaper that paid the
least attention to technical-biomedical content was La Vanguardia, a centre-right Catalan daily
with a national circulation (13.9%), followed by the liberal newspaper El Mundo (16.7%).
However, El País (19.1%), a progressive daily with the largest circulation in Spain, was the news-
paper that offered a greater diversity of technical-biomedical content; in effect, it was the daily
that provided information on the molecular structure of the virus and the target cells that it infects
(endothelial cells and hepatocytes, among others) (El País, 7 October 2014) and a unique diagram
of the process which showed the windows of time during which the viral load (quantity of the
virus) is positive in different corporal fluids (e.g. blood, faeces, saliva, semen) (El País,
12 October 2014, reused in another article published on 7 November 2014).1
Likewise, it is noteworthy that the two Catalan newspapers (La Vanguardia and El Periódico)
paid less attention to information on control and prevention measures (23.5%) than the dailies
published in Madrid, namely, ABC, El País and El Mundo (76.5%). A plausible explanation for this
could be that the Madrid newspapers paid greater attention to the technical aspects of safety (how
to put on and take off protective suits and the action protocols keyed to minimising the risk of
health workers contracting the disease) to calm public opinion, since the infected missionaries
repatriated from Africa were admitted to Carlos III Hospital, Madrid, where the assistant nurse was
Alcíbar 373

Figure 2. Distribution of technical-biomedical content by newspaper.

also infected. Nonetheless, linked to the journalistic rule of geographical proximity, the influence
of popular culture on news coverage should not be overlooked, inasmuch as both the safety meas-
ures and the protective gear are easily associated with stereotypes belonging to the realm of science
fiction and, therefore, attractive topics for the media.2
In order to answer RQ3, the possible association between the nominal variables of ‘visual
representation’ (qualitative, quantitative, mixed and hybrid) and ‘technical-biomedical content’
was studied by applying the chi-square test. The alternative hypothesis (the existence of associa-
tion) was substantiated, χ2(21, N = 209) = 121.88, p < 0.001.
The close relationship between the content and its forms of representation is described below.
All the quantitative graphics and quantitative maps (n = 66) were used exclusively to represent
epidemiological data (see Table 2). This result is consistent with the fact that people tend to place
more trust in numerical information on health- and risk-related issues than in other formats, but
usually have difficulty in understanding it (Bell et al., 2006). Visual aids allow for a better and
easier interpretation of quantitative information (see, for example, Stone et al., 2015; Tufte, 1983).
The findings here concur with the results obtained by other authors (DiFrancesco and Young, 2010;
Rebich-Hespanha et al., 2015; Smith and Joffe, 2009). For instance, to represent statistic data on
climate change (e.g. the volume of greenhouse effect gases emitted by country or temporary
changes in atmospheric CO2 levels), the British press relied on a large number of bar charts and line
graphs (Smith and Joffe, 2009).
For Welhausen (2015), the use of maps and line charts highlighted the sensation of risk during
the last EVD epidemic: the former because they transcended national and international borders, and
the latter due to the fact that they showed a continuous increase in the number of people infected by
the virus and the rising death toll in West Africa. In the Spanish case, this feeling was even more
pronounced, since the first case of infection in Europe occurred in Madrid on 6 October 2014.
Likewise, the qualitative maps (chorochromatic, n = 21, and explanatory, n = 1), flowcharts
(qualitative, n = 3), locator maps (qualitative, n = 5, and quantitative, n = 9) and hybrid maps
374 Public Understanding of Science 27(3)

Table 2. Types of quantitative graphics and quantitative maps used to represent epidemiological data.

Type of graphic Frequency Percentage

Line chart 15 22.7
Bar chart 12 18.2
Bubble chart 10 15.2
Area chart 8 12.1
Numerical data table 5 7.5
Proportional symbol map 10 15.2
Choropleth map 6 9.1
Total 66 100.0

(chorochromatic-proportional symbol, n = 5, and chorochromatic-dot, n = 2) registered were also

used merely for visualising ‘epidemiological data’. This technical-biomedical information was
also represented by timelines (n = 7), fact boxes (n = 1) and illustrations (n = 1).
‘Prevention and control measures’ appeared in practically any qualitative format (n = 34), except
timelines. These featured above all structure diagrams (n = 12), thanks to which the elements and
primary functions of the protective suits worn by health workers in contact with patients were
described. Process diagrams (n = 7) and floor plans (n = 6) were relevant as well. The former were
basically diagrams with illustrations of health workers in the process of putting on or taking off
their protective suits, while the latter were diagrams of the floor of Carlos III Hospital, Madrid,
where the Ebola patients were isolated. The floor plans contained additional written information to
describe the different elements of rooms (e.g. ‘class II biosafety laboratory’, ‘closed-circuit venti-
lation system’, ‘medical hazardous waste container’) and the actions performed by health workers
(e.g. ‘health workers should wash their hands before entering and leaving’).3
‘Injurious effects’ were generally represented by process diagrams (n = 23), with two different
designs that, nevertheless, had the same function: (1) diagrams in which human pictograms were
used (indicating the damaged areas) and, sometimes, arrows (indicating the direction of temporal
movement),4 and (2) the graphic designers found an ingenious solution for saving space and visu-
ally engaging readers in the use of a transparent plan of the human body and different colours to
represent the anatomical areas affected and their associated symptoms (with written text), depend-
ing on the stage of the disease (see Figure 3).
To represent ‘therapies and treatments’, fact boxes (n = 8) were used above all: this is an ideal
tool for synthesising, in an orderly way, the different preventive strategies, therapies and treat-
ments. By the same token, suitable process diagrams (n = 4) were employed to describe the sequen-
tial stages in the process of obtaining effective drugs, as well as timelines (n = 2).
‘Causes of infection’ were only represented by text-based graphics (fact boxes, n = 4, and sche-
mata, n = 3), which allowed for the organisation and hierarchisation of the different aetiological
aspects of EVD.
‘Viral life cycle’ was only displayed with process diagrams (n = 7), the best way of representing
this attractively with illustrations, pictograms and arrows indicating the direction of movement.
Lastly, the two cases of ‘viral load in body fluids’ were represented with process diagrams, and
the sole example of ‘virus structure’, by a structure diagram (see explanation above).
Answering RQ4 is no easy matter. In infographics, it is not always possible to relate a specific
visualisation to a particular source, since its source lines usually include multiple sources. However,
several thought-provoking conclusions can be drawn from the data.
Alcíbar 375

Figure 3. Process diagram representing the symptoms associated with the different stages of infection.
Source: reproduced with permission from La Vanguardia, 2016, La Vanguardia, 11 October 2014.

The WHO was the most frequently cited information source (71.8%). On the whole, expert
sources (institutions or research centres like the WHO or the Centre for Disease Control and
376 Public Understanding of Science 27(3)

Prevention (CDC), or specialised journals such as Science or The New England Journal of Medicine)
were the most regularly cited (49.2% as a sole source; 79.4% in conjunction with other types of
sources, such as media companies or press agencies). These findings suggest that expert sources
played an exceedingly relevant role in the production of information visualisations. Whereas the
public discourse on the government’s handling of the crisis and its damaging effects on the economy
involved contradictory postures, acrimonious debates and bitter controversies, the visual discourse
of the technical-biomedical aspects was constructed apodictically on the basis of supposedly sound
information supplied by sources with cognitive authority and social legitimisation.
In order to answer RQ5, it was understood that there was a textual reference to graphic content
only if there was a certain (partial or complete) discursive development of the content of the visu-
alisation in the body of the text. The chi-square test was employed for the individual visualisa-
tions and those included in infographics, alike. The analysis produced a p value = 5.78 10—9 < .001
(significance level of 99%), χ2(2, N = 209) = 35.97, p < .001. This result suggests that there was
indeed an association between the visualisations and the journalistic texts in which they were
inserted. A more in-depth analysis revealed that this association was highly dependent on the
individual or integrated character of the information visualisations. The individual ones did not
appear to function independently of the text in which they were inserted and, therefore, compre-
hension depended on the written information that the article in question provided. Nonetheless,
the visualisations incorporated into infographics were interconnected by thematic juxtaposition to
narrate a complex story. Thus, it could be said that the infographics were capable of functioning
independently, behaving as a unit of meaning.
RQ6 examined whether or not there was any reference to the representative character of the
image in the written text or the information visualisation itself. There was no such reference in any
of the cases. This finding is consistent with studies such as those of Myers (1990) who underscores
the fact that scientific discourses are based on two opposing visions. On the one hand, professional
discourse (‘research articles’) constitutes a narrative of science that sustains and emphasises the
arguments of the scientist and the conceptual structure of the discipline. On the other hand, public
discourse (‘science popularisation’) represents a narrative of nature in which entities and natural
phenomena – instead of scientific practices – are the focal point. By presenting scientific facts as
irrefutable and conclusive, science popularisation contributes decisively to naturalising the repre-
sentation itself, charging its scientific referent with ontology (Roqueplo, 1983). Indeed, by ‘charg-
ing itself with ontology’ science popularisation discourse generates the sensation that scientific
facts speak for themselves and are thus self-evident. In this way, popularisation contributes to
minimising, when not concealing, the role of the person who makes or interprets a scientific
discovery.
The lack of references to the representative nature of the information visualisations reinforced
the narrative of nature model that Myers (1990) observes in products keyed to science popularisa-
tion. Information visualisation would then be a direct tool at the service of journalists to narrate
the unquestionable reality of the Ebola epidemic, rather than a graphic resource that contributed
to the social construction of the outbreak.
RQ7: Nominalisations had the greatest frequency of occurrence (88%; e.g. ‘Advance of Ebola
in Africa’). Lagging far behind these were declarative sentences (4.8%; e.g. ‘All of the outbreaks
have occurred in Equatorial Africa’), demonstrative sentences with the adverb ‘like’ (1%; e.g.
‘This is what the drug against Ebola is like’), interrogative sentences (1%), indirect interrogative
sentences with ‘what … like’ (1%; e.g. ‘What protective gear against biological and chemical
agents should look like’) and sentences expressing probability with a conditional verb (0.5%; e.g.
‘Ebola might have already been responsible for 70 deaths in Guinea’). In all, 3.8% of the visuali-
sations did not have headlines.
Alcíbar 377

The nominal sentence is a rhetorical device with ‘important ideological functions such as
deleting agency and reifying processes’ (Billig, 2008). By concealing the subject, lending inex-
pressiveness to the sentence and reifying the referents, the nominal style highlights the facts and
gives them a patina of ‘objectivity’.
In this analysis, it must also be stressed that demonstrative sentences with the adverb ‘like’
were used to headline information on ‘therapies and treatments’. Specifically, ABC used the same
diagram at different times to describe the manufacturing process of drugs for combating the action
of the virus (‘This is what the drug against Ebola is like’, 11 August 2014, and ‘The drugs against
Ebola are manufactured like this’, 13 August 2014). The diagram displayed a logical sequence of
steps taken by researchers to obtain the end product (the drug). Each step in the chain of actions
was numbered and included an explainer (an additional text that started with an impersonal pas-
sive sentence using ‘it’ in the present tense). The impersonal passive voice with which these steps
were related allowed the journalist in question to focus the narrative more on natural entities
(mice, monoclonal antibodies, viruses, tobacco plants and patients), than on the procedural and
experimental decisions that were being adopted by researchers in an attempt to obtain a viable and
effective drug.5

6. Conclusion and key findings

Given the lack of existing literature on the subject, this study could contribute to a better under-
standing of the role that information visualisation plays in the popularisation of scientific content
in general, and the technical-biomedical aspects of EVD in particular. Based on a relevant theo-
retical frame, a content analysis was performed to explore issues relating to the content, structure
and function of information visualisations in the Spanish written press.
The results suggest that visual representations were very important for popularising technical-
biomedical content of a certain complexity to wide audiences. They also seem to imply that, firstly,
there was a significant association between the content and its graphic formats of representation.
Secondly, the most relevant information sources were expert ones (above all health organisations
and research centres), which enjoy a greater level of cognitive authority and social legitimacy.
Thirdly, that the understanding of individual representations (nearly always data visualisations)
depended to a certain extent on the written text in which they were inserted. On the contrary, there
is evidence that the information visualisations incorporated into infographics helped to convert
these into units of meaning that functioned independently of the text in which they were inserted.
Fourthly, the absence of any written indication of the representative character of the graphics
appears to indicate that the representation was really regarded as the object, rather than its discur-
sive construction. And lastly, in consonance with the previous finding, the analysis of the syntactic
structure of the headlines of the graphics revealed the predominance of sentences that tended to
suppress the subject and create an impression of objectivity.
These last two findings appear to suggest that the information visualisations followed the narra-
tive of nature model that Myers (1990) observed in products destined for science popularisation.
Nearly 90% of the visual representations published during the study period popularised technical-
biomedical content of some complexity. Thus, the epidemiological information and the damaging
effects provoked by the virus in humans in each stage of the infection cycle were the predominant
topics, a finding consistent with the basic feature of all epidemics: their dynamics.
It has been suggested that information visualisation is one of the most efficient tools for display-
ing large epidemiological/statistical datasets, since the public apparently tends to place greater
trust in numerical information than in other types of data (e.g. Bell et al., 2006; Stone et al., 2015).
This study has shown that all the quantitative graphics and maps were used by the journalists to
378 Public Understanding of Science 27(3)

represent epidemiological data alone. Moreover, it was observed that each newspaper employed its
own graphic design templates (primarily, line and bar charts) which were periodically updated with
new epidemiological information provided by expert sources. It has been established that the type
of graphic format influences risk perception (Schapira et al., 2006) and that bar and line charts are
ideal for making comparisons and showing trends over time, respectively (Lipkus, 2007).
As noted by Jordan et al. (2009), ‘content analysis research is limited in terms of what content
patterns can tell us about effects of health messages on the audience’. Therefore, the need arises to
develop future lines of research in order to explore how visual representations affect public aware-
ness and understanding of science. Empirical studies of science literacy have neglected, to a great
extent, the visual aspect of the public communication of science. In an attempt to tackle this prob-
lem, Bucchi and Saracino (2016) have performed perception analyses based on different ‘empirical
indicators of visual science literacy’ (p. 812). The preliminary results seem to indicate that the
respondents are more familiar with visual than written information on science, which suggests that
images can be an efficient way of engaging the public with scientific research results. Accordingly,
information visualisations could be an effective tool for facilitating technical-scientific knowledge
and encouraging certain prophylactic behaviours (Occa and Suggs, 2016).
Additional work may be required to explore the reactions of readers to online information with
or without graphic representations. The interactivity of online formats allows us to gain insights
into these reactions and to ascertain to what extent information visualisations can influence public
understanding of technical-scientific content.
Likewise, it would be interesting to perform a comparative analysis of graphic representations
and their communication purposes (e.g. to inform, persuade, appeal to) in both Western media and
those of the African countries affected by EVD.
Lastly, a key ethical implication can be derived from this study: graphic journalists who work
with information that is sensitive for the public should strike, as recommended by Cairo (2013), a
balance between rigorous reporting and an aesthetic format.

Acknowledgements
The author would like to thank Beatriz Morales for her suggestions and her valuable assistance.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article

Notes
1. These graphics can be consulted at: http://elpais.com/elpais/2014/08/09/media/1407595020_722065.
html and http://elpais.com/elpais/2014/08/11/media/1407779252_015727.html
2. A number of significant examples can be seen in the following links of the digital versions of ABC, El
País and El Mundo, respectively: http://www.abc.es/sociedad/20141016/abci-ebola-trajes-ventilacion-
carlosiii-201410152044.html; http://elpais.com/elpais/2014/10/08/media/1412795366_182631.html;
http://hemeroteca.lavanguardia.com/preview/2014/10/12/pagina-44/94534796/pdf.html
3. For illustrative examples, see: http://hemeroteca.abc.es/nav/Navigate.exe/hemeroteca/madrid/abc/2014/
10/16/040.html; http://elpais.com/elpais/2014/10/08/media/1412795366_182631.html; http://www.
elmundo.es/salud/2014/10/09/5435a1ea22601d9b468b4592.html
4. For illustrative examples, see: http://hemeroteca.abc.es/nav/Navigate.exe/hemeroteca/madrid/abc/2014/
10/10/018.html; http://hemeroteca.abc.es/nav/Navigate.exe/hemeroteca/madrid/abc/2014/10/10/019.html
5. See example at: http://hemeroteca.abc.es/nav/Navigate.exe/hemeroteca/madrid/abc/2014/08/11/029.
html
Alcíbar 379

References
Ancker JS, Senathirajah Y, Kukafka R and Starren JB (2006) Design features of graphs in health risk com-
munication: A systematic review. Journal of the American Medical Informatics Association 13(6):
608–618.
Barberá R and Cuesta U (2015) El virus del ébola: Análisis de su comunicación de crisis en España [Ebola:
Analysis of crisis communication in Spain]. Opción 31(4): 67–86.
Basch CH, Basch CE and Redlener I (2014) Coverage of the Ebola virus disease epidemic in three widely
circulated United States newspapers: Implications for preparedness and prevention. Health Promotion
Perspectives 4(2): 247–251.
Bauer MW (1998) The medicalization of science news: From the ‘rocket-scalpel’ to the ‘gene-meteorite’
complex. Social Science Information 37(4): 731–751.
Bell BS, Hoskins RE, Pickle LW and Wartenberg D (2006) Current practices in spatial analysis of cancer
data: Mapping health statistics to inform policymakers and the public. International Journal of Health
Geographics 5: 49.
Billig M (2008) The language of critical discourse analysis: The case of nominalization. Discourse & Society
19(6): 783–800.
Bucchi M and Saracino B (2016) ‘Visual science literacy’: Images and public understanding of science in the
digital age. Science Communication 38(6): 812–819.
Cairo A (2008) Infografía 2.0: Visualización interactiva de información en prensa [Infographics 2.0: Interactive
visualisation of information in the press]. Madrid: Alamut.
Cairo A (2013) The Functional Art: An Introduction to Information Graphics and Visualization. Berkeley,
CA: New Riders.
Cairo A (2016) The Truthful Art: Data, Charts, and Maps for Communication. Berkeley, CA: New Riders.
Condit C (2015) The spread of Ebola: How the world health organization’s rhetoric contributed to virus trans-
mission. Mètode Science Studies Journal (6): 121–127.
Desnoyers L (2011) Toward a taxonomy of visuals in science communication. Technical Communication
58(2): 119–134.
DiFrancesco DA and Young N (2010) Seeing climate change: The visual construction of global warming in
Canadian national print media. Cultural Geographies 18(4): 517–536.
Ding H (2009) Rhetorics of alternative media in an emerging epidemic: SARS, censorship, and extra-institutional
risk communication. Technical Communication Quarterly 18(4): 327–350.
Dur BIU (2014) Data visualization and infographics in visual communication design education at the age of
information. Journal of Arts and Humanities 3(5): 39–50.
Frankel F and DePace AH (2012) Visual Strategies: A Practical Guide to Graphics for Scientists and
Engineers. New Haven, CT; London: Yale University Press.
García-Retamero R and Cokely ET (2013) Communicating health risks with visual aids. Current Directions
in Psychological Science 22(5): 392–399.
Hellsten I and Nerlich B (2010) Bird flu hype: The spread of a disease outbreak through the media and
Internet discussion groups. Journal of Language and Politics 9(3): 393–408.
Holsanova J, Rahm H and Holmqvist K (2006) Entry points and reading paths on newspaper spreads:
Comparing a semiotic analysis with eye-tracking measurements. Visual Communication 5(1): 65–93.
Hornmoen H (2010) ‘Making us see science’: Visual images in popular science articles and science journal-
ism. Journalistica 2: 79–99.
Hüppauf B and Weingart P (2008) Science Images and Popular Images of the Sciences. New York, NY;
London: Routledge.
Iliinsky N and Steele J (2011) Designing Data Visualizations. Sebastopol, CA: O’Reilly.
Joffe H and Haarhoff G (2002) Representations of far-flung illnesses: The case of Ebola in Britain. Social
Science & Medicine 54(6): 955–969.
Jordan AB, Kunkel D, Manganello J and Fishbein M (2009) Advancing the science of content analysis.
In: Jordan AB, Kunkel D, Manganello J and Fishbein M (eds) Media Messages and Public Health.
New York, NY; London: Routledge, pp. 246–257.
380 Public Understanding of Science 27(3)

Kane ML (2015) Covering an epidemic. The Ebola virus and AIDS in the news. Electronic Thesis, The
University of Arizona, Tucson, AZ.
Kirk A (2012) Data visualization: A successful design process. Birmingham: Packt Publishing.
Koteyko N, Brown B and Crawford P (2008) The dead parrot and the dying swan: The role of metaphor sce-
narios in UK press coverage of avian flu in the UK in 2005–2006. Metaphor and Symbol 23(4): 242–261.
Kress GR and van Leeuwen T (2006) Reading Images: The Grammar of Visual Design. London: Routledge.
Krippendorff K (2011) Agreement and information in the reliability of coding. Communication Methods and
Measures 5(2): 93–112.
Krippendorff K (2013) Content Analysis: An Introduction to Its Methodology. Thousand Oaks, CA: SAGE.
Krum R (2014) Cool Infographics: Effective Communication with Data Visualization and Design.
Indianapolis, IN: John Wiley & Sons.
Lazard A and Atkinson L (2015) Putting environmental infographics center stage: The role of visuals at the
elaboration likelihood model’s critical point of persuasion. Science Communication 37(1): 6–33.
Lee ST and Basnyat I (2013) From press release to news: Mapping the framing of the 2009 H1N1 A influenza
pandemic. Health Communication 28(2): 119–132.
Lester PM (2006) Visual Communication: Images with Messages. Belmont, CA: Thomson Wadsworth.
Lewison G (2008) The reporting of the risks from severe acute respiratory syndrome (SARS) in the news
media, 2003–2004. Health, Risk & Society 10(3): 241–262.
Lipkus IM (2007) Numeric, verbal, and visual formats of conveying health risks: Suggested best practices and
future recommendations. Medical Decision Making 27: 696–713.
Lohse GL, Biolsi K, Walker N and Rueter HH (1994) A classification of visual representations. Communications
of the ACM 37(12): 36–49.
López-Manjón A and Postigo Y (2014) Análisis de las imágenes del cuerpo humano en libros de texto espa-
ñoles de primaria [Analysis of human body images in Spanish primary school textbooks]. Enseñanza de
las Ciencias 32(3): 551–570.
Mayer KH and Pizer H (eds) (2005) The AIDS Pandemic: Impact on Science and Society. Amsterdam;
London: Academic Press.
Micaletto JP and Gallardo L (2015) La comunicación institucional en la crisis del ébola en Europa: El caso de
la crisis española de 2014 en sus inicios [The institutional communication on Ebola crisis in Europe: the
first moments of the 2014 Spanish crisis]. Revista Internacional de Relaciones Públicas 5(9): 89–110.
Myers G (1990) Writing Biology: Texts in the Social Construction of Scientific Knowledge (Science and
Literature series). Madison, WI: University of Wisconsin Press.
Nelkin D (1995) Selling Science: How the Press Covers Science and Technology. New York, NY:
W. H. Freeman Publishers.
Occa A and Suggs LS (2016) Communicating breast cancer screening with young women: An experimental
test of didactic and narrative messages using video and infographics. Journal of Health Communication
21(1): 1–11.
Racine N (2002) Visual Communication: Understanding Maps, Charts, Diagrams, and Schematics.
New York, NY: Learning Express.
Rebich-Hespanha S, Rice RE, Montello DR, Rezloff S, Tien S and Hespanha JP (2015) Image themes and
frames in US print news stories about climate change. Environmental Communication 9(4): 491–519.
Revuelta G, de Semir V, Armengou C, Cots E, Gonzalo C, Saladié N, et al. (2015) La comunicación pública
sobre la enfermedad del Ébola [The public communication on the Ebola virus disease] (Informe Quiral
2014). Barcelona: Universitat Pompeu Fabra.
Rock M (2005) Diabetes portrayals in North American print media: A qualitative and quantitative analysis.
American Journal of Public Health 95(10): 1832–1838.
Roqueplo P (1983) El reparto del saber: Ciencia, cultura, divulgación [Sharing knowledge: Science, culture,
popularisation]. Buenos Aires: Editorial Gedisa.
Schäfer MS (2011) Sources, characteristics and effects of mass media communication on science: A review of
the literature, current trends and areas for future research. Sociology Compass 5–6, 399–412.
Schapira MM, Nattinger AB and McAuliffe TL (2006) The influence of graphic format on breast cancer risk
communication. Journal of Health Communication 11(6): 569–582.
Alcíbar 381

Shin H (2016) Epidemic and risk communication: An analysis of strategic and graphic characteristics of
infographics. Electronic Thesis, Iowa State University, Ames, IA.
Silletti AM (2015) The role of illustrations in popularizing medical discourse. Linguæ & - Rivista di lingue e
culture moderne 14(2): 65–81.
Smith NW and Joffe H (2009) Climate change in the British press: The role of the visual. Journal of Risk
Research 12(5): 647–663.
Stone ER, Gabard AR, Groves AE and Lipkus IM (2015) Effects of numerical versus foreground-only icon
displays on understanding of risk magnitudes. Journal of Health Communication 20(10): 1230–1241.
Summ A and Volpers A-M (2016) What’s science? Where’s science? Science journalism in German print
media. Public Understanding of Science 25(7): 775–790.
Trčková D (2015) Representations of Ebola and its victims in liberal American newspapers. Topics in
Linguistics 16(1): 29–41.
Trumbo J (1999) Visual literacy and science communication. Science Communication 20(4): 409–425.
Tufte ER (1983) The Visual Display of Quantitative Information. Cheshire, CT: Graphics Press.
Ungar S (1998) Hot crises and media reassurance: A comparison of emerging diseases and Ebola Zaire.
British Journal of Sociology 49(1): 36–56.
Valero JL, Catalá J and Marín BE (2014) An approach to the taxonomy of data visualisation. Revista Latina
de Comunicación Social 69: 486–507.
Wagner-Egger P, Bangerter A, Gilles I, Green E, Rigaud D, Krings F, et al. (2011) Lay perceptions of col-
lectives at the outbreak of the H1N1 epidemic: Heroes, villains and victims. Public Understanding of
Science 20(4): 461–476.
Wallis P and Nerlich B (2005) Disease metaphors in new epidemics: The UK media framing of the 2003
SARS epidemic. Social Science & Medicine 60(11): 2629–2639.
Welhausen CA (2015) Visualizing a non-pandemic: Considerations for communicating public health risks in
intercultural contexts. Technical Communication 62(4): 244–257.
Wong NCH, Harvell LA and Anaz N (2013) Comparing two modes of presentation on perceptions of flu
threat and attributions for its global outbreak. European Journal of Geography 4(2): 17–33.

Author biography
Miguel Alcíbar is currently a professor at the University of Sevilla, in Spain. He specialises in science com-
munication and journalistic discourse analysis. He holds a degree in biology and a PhD in communication
studies. His research interests include science communication in new media, rhetoric of science and media
representation of scientific controversies, especially those related to biomedical research.