The use of Internet in newsgathering among European science journalists
61
Invitations to participate
The first invitation to participate in this survey was sent by mail on 17 November
2003 to all 208 journalists, in a cover letter with the logo of the University of
Leeds16. The letter included a unique link to the survey website, so journalists
could go there and answer all the questions17. Nineteen journalists answered the
survey before being sent the first e-mail.
The first reminder was sent by e-mail on 2 December 2003. At that particular
time, 23 e-mails bounced back, which means they were not received by the
respondents. The main reason for the rejection of the messages was the
inexistence of the specific e-mail address, either because it was wrong or the
journalists had already left the newspaper/magazine/news agency. A copy of this
message can be seen on appendix IVa.
Subsequent reminders were sent on 7 January 2004 and 27 January 2004, with 21
and 20 rejections, respectively. Copies of these messages can be read on
appendixes IVb and IVc, respectively.
The survey was closed on 9 February 2004, with 97 respondents (46.6 percent).18
16
Dillman (2000) says that researchers should use both e-mail and postal mail for notification, as it
can help the contact with respondents.
17
A copy of this letter can be seen on appendix III.
18
The percentage of respondents takes into account every journalist who was invited to participate
in the survey. If we were to eliminate the 23 journalists who only received the letter, because the e-
mails bounced back, and the 3 journalists whose e-mail was impossible to get, the percentage of
respondents would have risen to 53.3 percent (N=97/182).
62
3.2. The interviews
The second part of this research project consisted in conducting interviews with
some of the European science journalists who answered the survey. The idea
behind these interviews was to collect more information about the way journalists
relate with the Internet and how they feel about it. In this case, as Wimmer and
Dominick (2000) put it, we thought that the interview could be of an enormous
help to explore these questions more deeply:
The most important advantage of the in-depth interview is the wealth
of detail that it provides. Futhermore, when compared to more
traditional survey methods, intensive interviewing provides more
accurate responses on sensitive issues. (Wimmer and Dominick
2000, p.122)
To guide our work, and because the survey had already given us some information
on the use of Internet by European science journalists, we set two main goals for
these interviews: to understand how these journalists recognize the Internet as a
reporting tool; and to gather information on how Internet is changing the routines
of these professionals.
Drawing on these goals, on the literature review, and on our own results from the
survey, we planned the questions for the interview. Because we were dealing with
journalists with very tighten schedules, and because we would have to use the
telephone to conduct most of these interviews (the respondents were too scattered
to allow face-to-face contact), we decided to limit them to a maximum duration of
30 minutes.
This decision had an important effect on the quantity of information we could
collect from the interviews. Longer interviews would have allowed us to probe
even deeper on the aspects we wanted to study, but would also certainly
discourage some of the interviewees, too busy to spend one hour or more being
63
interviewed by phone. As we already had some good quantitative information
about the use these professionals made of the Internet on a daily basis, it seemed
wise to shorten the interviews, so we could get a good level of acceptance.
Nevertheless, as we will see in table 1, we still had 11 rejections.
For the preparation of the interviews, we followed mainly Bryman (2001),
specially the chapter on “Interviewing in Qualitative Research”. We chose to
conduct a semi-structured interview that would allow us to cover some specific
topics and probe deeper if and when we thought necessary. The complete set of
questions for the interviews can be seen on appendix V.
Sampling
In order to get a diverse set of interviewees, we separated the journalists who
answered the survey into three different groups, according to the number of hours
spent on the Internet per day while at work19. From each of these groups, we
randomly chose four people to be interviewed, for a total of 12 interviews.
After the sampling, we usually contacted the respondents by e-mail, in order to
ask them for the interview. If they refused, or didn’t answer after three e-mail
messages, we would randomly choose another person from that same group and
did the same procedure, until completing 4 interviews for each group. To achieve
the final number of 12 interviews, a total of 23 journalists were contacted.
To randomize the sample, we gathered the subjects in three different lists and
chose to contact every third person. Every list was perceived as a continuous loop,
so we would not run out of possible interviewees everytime the end of the list was
reached.
19
The data were collected from question 25 of the survey (see table 28, page 108).
64
Invitations for the interviews
Journalists were contacted to take part in the interviews by e-mail20. The first
group of e-mails was sent on 19th April 2005 to 15 different journalists, five from
each group.
In order to get all the 12 interviewees, we contacted seven people from the first
group (respondents who use the Internet for less than 3 hours a day), seven from
the second group (respondents who use the Internet from 3 to 4 hours a day) and
nine from the third group (journalists who use the Internet for more than 4 hours a
day). A summary of these contacts is provided in Table 1. (see next page)
The 12 interviews for this project were conducted by telephone, from 22
nd April
2005 to 9th June 2005, in English. Only one of these interviews (with a Portuguese
journalist) was done in person. For consistency reasons, this interview was also
conducted in English.
20
The content of the e-mail can be read in appendix VI.
65
Subject
number
GROUP I
<3hrs/day
GROUP II
3-4hrs/day
GROUP III
>4hrs/day
1st 2
e-mail
nd 3
e-mail
rd 1
e-mail
st 2
e-mail
nd 3
e-mail
rd 1
e-mail
st 2
e-mail
nd
e-mail
1 x x x - - - - - -
2 x x - - - - - - -
3 V - - - - - - - -
4 x V - - - - - - -
5 V - - - - - - - -
6 - - - x x x - - -
7 - - - V - - - - -
8 - - - x x V - - -
9 - - - x x x - - -
10 - - - x V - - - -
11 - - - - - - x x x
12 - - - - - - x V -
13 - - - - - - x x -
3rd
e-mail
14 - - - - - - x x V
15 - - - - - - x x x
16 x x x - - - - - -
17 V - - - - - - - -
18 - - - x x x - - -
19 - - - x V - - - -
20 - - - - - - x - -
21 - - - - - - x x x
22 - - - - - - V - -
23 - - - - - - x V -
x – Didn’t answer the e-mail or refused to be interviewed
V – accepted the interview
Table 1 - Contacts for the interviews in the three groups of respondents
66
4. Survey results
This chapter will present the results of a survey of science journalists working on
staff for general print media and news agencies in 14 different countries of the
European Union (Austria, Belgium, Denmark, Finland, France, Germany, Greece,
Ireland, Italy, Netherlands, Portugal, Spain, Sweden and the United Kingdom).
The survey was conducted through the Internet, between 17 November 2003 and
9 February 2004. Details on its methodology are elucidated in chapter 3 of this
thesis.
The results will be presented according to several sub-chapters that will follow the
main journalistic constraints recognized in the literature review and probed in the
questionnaire: general characterisation of the respondents, their relations and
thoughts about sources, competition, internal pressures, audiences, and their
feelings about the current situation of their job. Finally, we will present the results
about the use European science journalists make of the Internet and the influence
this new resource is having on their daily routines.
As we have stated in the previous chapter, invitations to participate in the survey
were sent to 208 journalists from 102 different media. Ninety-seven (97) valid
answers were obtained, which means this survey had a response rate of 46.6
percent.
67
4.1. Characterisation of the respondents
The first questions of the survey were aimed at characterizing the European
science journalists who work for general print media and news agencies
throughout the European Union. In the next few pages, you will contact a general
overview of the respondents to this survey, according to country, sex, age, type of
media, size of science section they work in, number of years as a journalist and as
a science journalist, and, finally, level and area of education.
The science journalists who responded to this questionnaire came from 14
different countries of the European Union. Table 2 shows the number of
respondents per country. As stated earlier, it was not possible to identify a single
science journalist working on staff for a general print media or news agency in
Luxembourg. At the same time, only one Irish science journalist could fill the preset
conditions for this survey.
Countries
Number
of answers Percent
Austria 6 6,2
Belgium 5 5,2
Denmark 5 5,2
Finland 6 6,2
France 8 8,2
Germany 8 8,2
Greece 4 4,1
Ireland
Italy
1
14
1,0
14,4
Netherlands 12 12,4
Portugal 9 9,3
Spain 8 8,2
Sweden 6 6,2
UK 5 5,2
Total 97 100,0
Table 2 -Origin of the science journalists who responded to this survey
68
Because the number of journalists who answered the questionnaire is too low to
allow a country per country analysis, we decided to divide these journalists into
three different and bigger groups, according to the Gross National Income (GNI)
per capita of their respective countries21.
Gross National Income (GNI) is the key indicator for classifying regions in
the research works conducted by international organisations, such as the World
Bank and the United Nations (World Bank 2003). GNI per capita is considered a
‘summary indicator’, because it is directly related with a series of other indicators
as health, education, access to communication technologies, and so on.
Using GNI per capita as an indicator to separate the members of the European
Union can allow us to better understand the subtle differences among a series of
countries who share the same economic space. Once the analysis country per
country is statistically impracticable, given the sample size, the grouping of
countries around GNI per capita is a way of guaranteeing that the clustering is as
fair as possible.
Other studies on the public understanding of science, for instance, have clustered
EU countries around industrialization (Durant, Bauer et al. 2000). However, the
weight of the industrial sector in developed countries has dropped in the last few
years, which means that the levels of industrialization are no longer closely linked
with GNI per capita. This fact and the realization that GNI per capita is
increasingly used as a ‘summary indicator’ to measure the relative development of
countries lead us to choose it as a variable in this study.
21
World Bank (2005) World Development Indicators 2004, World Bank, Washington D.C. See
table 2.
69
Countries GNI per capita (USD)
Denmark 30,600
Ireland 29,570
Austria 28,910
Netherlands 28,350
Belgium 28,130
France 27,040
Germany 26,980
United Kingdom
Italy
26,580
26,170
Finland 26,160
Sweden 25,820
Spain 21,210
Greece 18,770
Portugal 17,820
Table 3 - GNI per capita of the countries of the respondents - World Bank (2005)
In the first group, we joined countries with a GNI per capita lower than 22,000
USD (Spain, Greece and Portugal); in a second group, countries with a GNI per
capita higher than 22,000 USD but lower than 28,000 USD (France, Germany,
United Kingdom, Italy, Finland and Sweden). Finally, in the third group, countries
with a GNI per capita higher than 28,000 USD (Denmark, Ireland, Austria,
Netherlands and Belgium). Table 4 shows the groups formed by this division.
GNI per capita Number of cases Percent
<22,000 USD 21 21,6
22,000-28,000 USD 47 48,5
>28,000 USD 29 29,9
Total 97 100,0
Table 4 - Number of respondents according to GNI per capita of their country
70
Sex
The first conclusion that can be drawn from the results of this survey is that the
majority of European science journalists22 is male, but not for long. Although 63.9
percent (N=62)23 of the respondents to this survey are male, we can notice that the
number of women becomes higher as we move towards the younger journalists.
See table 5.
Sex Total
Male Female
25-34 years 14 13 27
35-44 years 17 10 27
45-54 years 18 9 27
55-64 years 12 3 15
61 35 96
χ² (3) = 3.932, p > 0.05
Table 5 - Sex of science journalists according to age
At the same time, the observed majority of males is influenced by countries with a
higher GNI per capita. When put into three different groups – countries with a
GNI per capita lower than 22,000 USD, from 22,000 to 28,000 USD, and higher
than 28,000 USD –, we can notice that, in the first group, the number of women is
the double of the number of men.
Sex Total
Male Female
GNI <22,000 USD 7 14 21
22,000-28,000 USD 30 17 47
>28,000 USD 25 4 29
Total 62 35 97
χ² (2) = 14.764, p = 0.001
Table 6 - Sex of science journalists according to GNI of the country per capita
22
From now on, the expression “European science journalists” will be used to characterise the
surveyed population. It is not absolutely correct – it should be “European science journalists
working on staff for general print media and news agencies in 14 countries of the European
Union” –, but it is certainly shorter and avoids very long and awkward sentences.
23
Sixty-two (62) respondents were male, but one of them decided not to state his age. That is the
reason why, in table 4, the number of males is only 61.