A Decade of Immigration in the British Press

Immigration has become one of the most salient topics in the UK public debate. Over the past decade, policymakers and politicians have directed a lot of energy and attention to migration policies, often citing public demand for stronger action to reduce immigration levels or tackle related issues.

Where do the public get their ideas about immigration? One frequently cited source – besides day-to-day contact with immigrants themselves, or what friends and work colleagues might say – is the media. UK media coverage of migration has evolved over the last decade to accommodate an array of profound changes: changing trends in the movement of people; changing governments; changing policies; changing geopolitics; and changing commentators in the debate.

This analysis looks at trends in the language used in newspaper reporting through that period, and considers how these developments relate to the current UK political context. In particular the report identifies six key trends:

• A tendency for journalists themselves to play the role of framing problems in the migration debate, rather than simply reporting on others’ (such as politicians,’ think-tanks,’ or academics’) analysis. This highlights the key role played by journalists and media organisations in shaping the UK migration debate.
• A tendency to blame politicians for the scale of EU migration, while in discourse about ‘illegal’ immigrants, migrants themselves are often blamed. Economic arguments dominated the discussion of problems related to both EU and illegality.
• A sharp increase in the volume of newspaper coverage relating to migration since the election of the Conservative-led coalition government in 2010, particularly after the introduction of measures to reduce net migration in 2011 and 2012.
• An apparent change in how immigration is discussed, with a significant decline in discussion of the legal status of migrants and an increase in the focus on the scale of migration from 2009 onwards. This was accompanied by a rise in the relative importance of discussion relating to ‘limiting’ or ‘controlling’ migration since 2010.
• A sharp increase in the frequency of discussion of migrants from the EU/Europe after 2013, with a particular spike in 2014 when migrants from Romania and Bulgaria achieved full access to the UK labour market.
• A notable change in depictions of refugees between 2006 and 2015, with a sharp increase in references to Syrians coinciding with the escalating Syrian refugee crisis.

The report suggests that press depictions of migrants have focused on concern about high levels of net migration, and particularly EU migration. This numerical focus has eclipsed a waning focus on ‘illegal’ migration and become the leading migration frame in UK national newspapers.

The role of media in shaping public opinion is not clear-cut. It has often been observed that the press is good at setting the agenda – telling readers what to think about – although there is an ongoing debate about the extent to which media coverage either causes or simply reflects the views of its audiences on the topics it discusses.

Immigration, and specifically EU immigration, has clearly emerged as a key factor in the decisions of many people to vote for the UK to leave the European Union. But the significant increase in the profile of EU migration within recent UK media coverage—which has been dominated by a focus on high levels of net migration, and challenges in controlling migration flows – predates the EU referendum debate (the analysis runs until May 2015) and shows that the media was already playing an important role in discussions of the EU and migration in the years leading up to 2016.

In Britain, immigration has become one of the most salient topics in public debate. Over the past decade, policymakers and politicians have directed a lot of energy and attention to migration policies, often citing public demand for stronger action to reduce immigration levels or tackle related issues.

Where does the public get their ideas about immigration? One frequently cited source – besides day-to-day contact with immigrants themselves, or what friends and work colleagues might say – is the media (van Klingeren et al. 2015). Media outputs come from established outlets such as printed newspapers and broadcast as well as online sources including blogs, websites, social media, or apps. These different kinds of media have different ways of operating and producing stories and users can engage with and access online media in more instantaneous and interactive ways (Beckett and Mansell 2008).

This report explicitly focuses on one portion of the whole media landscape – national print editions of newspapers. While there has been much discussion about the demise of print newspapers as key influencers in an increasingly online world it is important to recognise that the traditional press still occupies an important role in public debate and policymaking. This role is necessarily changing due to the rise of other media through which people access and share journalistic content, not all of which is subject to similar standards of editorial scrutiny or transparency (Beckett 2010). But nevertheless, Britain’s national newspapers still have an important position in the engine-room of national policy narratives.

In this context, the role of media in shaping public opinion is not clear-cut. Across many different media and national contexts, it’s often been observed that the press is good at setting the agenda—telling readers what to think about by regularly highlighting certain kinds of information (Chong and Druckman 2007). This may include choices about what to emphasise, or which topics and themes to link together – as well as the choice of narratives used to make sense of a phenomenon. However, there is debate about the extent to which media coverage either causes or simply reflects the views of its audiences in an ‘echo chamber’ (for more discussion, see Boyd 2008).

Motivations for Studying Press Coverage of Migration

This report builds on prior Migration Observatory work examining how the British press covers migration (Allen and Blinder 2013). It covers the period January 2006 to May 2015, and aims to contribute to understanding how the British press has actually talked about this issue.

This report has several aims: (1) to show how the kinds of words used to describe immigration, as well as immigrants, have changed since 2006, based on quantitative analysis of a large database of newspaper articles mentioning migration; and (2) to examine the kinds of problems, justifications, and messengers that have been especially prominent in discussions surrounding two key aspects of the press debate: illegal and EU migration.

Following this introduction, Section 2 focuses on the amounts of coverage mentioning key terms. Section 3 turns attention to patterns of words that the press uses to describe immigration and migrants. Finally, Section 4 focuses on close-reading of a smaller sample of articles on the topics of illegal and EU migration to show how the press uses elements of narratives in articles.

Data Sources and Methods

The data for this report come from Nexis UK and Factiva, two archiving services that include full-text versions of many newspapers and other periodicals. The database includes articles containing migration related words in all 19 national UK publications that continuously appeared between 1 January 2006 and 30 May 2015. These are divided into tabloids, midmarkets, and broadsheets as seen in Table 1. In total, there are over 170,000 articles included in the study (for more details, see Table 4 in Appendix A – all appendices are available in the report PDF).

Table 1: National UK tabloid, midmarket, and broadsheet titles included in the study

Tabloids	Midmarkets	Broadsheets
Daily Mirror (Sunday Mirror)	Daily Mail (Mail on Sunday)	The Daily Telegraph (Sunday Telegraph)
Daily Star (Daily Star Sunday)	The Express (Sunday Express)	Financial Times
The People		The Guardian (The Observer)
The Sun		The Independent (Independent on Sunday)
		The Times (Sunday Times)

The report uses three main techniques. The first is frequency analysis, where migration-related words, as well as articles containing them, are totalled up to show how prominent they are. The second is collocational analysis, a method that uses grammatical rules as well as statistical tests to determine how strongly one word is linked with a target word – as opposed to them appearing together by random chance. This helps to identify groups of words that tend to be used together. And the third is manual content analysis, where specific aspects of narratives are identified and counted up for subsequent analysis. The content analysis in this report covers two specific topics – EU/European immigration, and illegal immigration. For full details on the data collection and methods, see Appendix B.

 

The volume of press coverage mentioning ‘immigration’ or ‘migration’ declined from 2006 to 2011 before rising each year from 2011 to mid-2015

Figure 1 shows the average number of articles per month in each year that contained either the words ‘immigration’ or ‘migration’. From now on, when this report presents data about ‘immigration’, results for the term ‘migration’ are also included in the figures unless explicitly stated.

From 2006 to 2011, the British press mentioned immigration in fewer articles – with the exception of 2010. The small increase in 2010 coincides with a General Election which produced a Conservative-led coalition. A key policy focus of the Conservative party in this period was reducing net migration, or the difference between the numbers of people entering and leaving the country, to the ‘tens of thousands’. The government introduced policies to achieve this goal from 2011 onwards. Then, from 2012, the press published increasingly more articles mentioning immigration. Notably, the average month in 2014 – the last year in the dataset with a full year for comparison – contained over twice as many articles mentioning immigration than in 2011.

Figure 1

Over the 2006-2015 period, the prominence of immigration also changed in British public opinion. The Ipsos MORI Issue Index asks people every month to name the most important issue facing the country, as well as any other issues that come to mind. Then, these answers are grouped together to create a percentage of all respondents. Figure 2 shows the percentage of people that named immigration (in 2015) or immigration and race relations (2006-2014) as one of these top issues (actual percentage), compared with four of the other top issues raised by respondents (expressed as six-month rolling averages for clarity).

Figure 2

From 2008 to mid-2014, in the wake of the economic crisis and subsequent recovery, the economy was the top issue raised by respondents. Unemployment also rose in public salience, but both issues declined during 2014-2015 as immigration increased in importance again. The data indicate relatively higher levels of both public concern and press coverage during 2006-2007, lower levels between 2008-2012 (with the exception of the time around the May 2010 General Election), and increasing attention to the issue from 2013 onwards. This trend is similar to that seen in the data on the volume of media coverage in Figure 1. While this does not necessarily show that public concern caused more press coverage, or vice versa, there does appear to be a relationship between the two.

Immigrants get the most coverage in the British press compared to refugees and asylum seekers

Figure 3 shows the average number of articles each month from 2006 to May 2015 that contained at least one of the following terms: ‘immigrant(s)/migrant(s)’, ‘asylum seeker(s)’, or ‘refugee(s)’. Several points are worth mentioning. First is the decline and resurgence of articles mentioning immigrants—a U-shaped pattern also seen in Figure 1 involving mentions of ‘immigration’ over the years. This suggests that, at least in terms of frequency, ‘immigration’ as the phenomenon has similar dynamics to ‘immigrants’ as people. Second is the lack of articles the press as a whole published that mentioned ‘asylum seeker(s)’. These results confirm previous work on the 2010-2012 period that shows how the British press mentioned asylum seekers the least (Blinder and Allen 2016: 21) but for the full 2006-mid 2015 period. But, thirdly, there has been a recent increase in the average number of articles mentioning refugees and asylum seekers: the average month between January and May 2015 contained the highest number of articles mentioning ‘refugees’ over the period, about twice as many as in 2012. Average monthly mentions of ‘asylum seekers’ increased more than three times from 2012 to 2015. Recent events such as coverage of the 2014-2015 migrant crisis likely contributed to these large increases.

Figure 3

Figure 4 compares the amount of coverage mentioning ‘asylum seekers’ with the number of asylum main applications (not the total number of people seeking asylum, which would include children and dependents as part of the same application). Although it is debatable whether the volume of news coverage does or should exactly mirror developments in the real world, the analysis shows that press coverage from 2006 to 2014 roughly corresponded with changes in the number of asylum applications.

Figure 4

This section shows how the words associated with the phenomenon of ‘immigration’ as well as migrant groups themselves have changed over time. Instead of looking at key terms themselves, this section looks at the other kinds of words that are closely associated with these terms. In a similar way to ‘immigration’, when this report talks about ‘immigrants’, it includes mentions of ‘migrants’ unless explicitly mentioned.

‘Mass’ was the single most common way of describing immigration from 2006 to 2015. The press mentioned aspects of immigration’s scale and pace more often than legality

Looking at modifiers of the terms ‘immigration’ and ‘migration’ shows those aspects the press tends to highlight over others. Modifiers are words that describe, characterise, or intensify something. In the Migration Observatory dataset, the press mentioned either ‘immigration’ or ‘migration’ a total of 121,026 times. In about 20% of these cases (23,853 instances), there was a modifier present.

Figure 5 shows the top ten modifiers of the terms ‘immigration’ or ‘migration’ ranked by their share of all modifiers.

Figure 5

When British newspapers have chosen to describe immigration in some additional way over the 2006-2015 period, about 15% of the time they explicitly use the word ‘mass’. This is closely followed by ‘net’ and ‘illegal’. However, Figure 5 is an aggregated snapshot of the whole period: as explained below, the apparent predominance of these latter two modifiers has changed over time. Some examples illustrate how these terms appear in the press:

Yet, astonishingly, the government has never made a serious attempt to justify mass immigration. (Broadsheet 2008)

With net migration reaching 298,000 it is apparent that immigration from the EU is out of control. (Midmarket 2015)

Violent crime is up, illegal immigration is out of control, the NHS is failing and hospital-acquired diseases are needlessly slaughtering patients. (Tabloid 2006)

Looking at the top modifiers as a whole, words related to the scale or pace of migration can be grouped together to show their prominence over time. Figure 6 shows the relative shares of ‘scale and pace’ modifiers compared to ‘legality’ modifiers in each year of press coverage.

Figure 6

A key finding is the decline of illegality from comprising nearly a quarter (22%) of all modifiers in 2006 to less than 10% of instances in 2015. Meanwhile, the share of scale-related modifiers rose over the same period, peaking at 63% of instances in 2015. What this means is that when the news articles explicitly used a word to describe ‘immigration’ in the first five months of 2015, about 6 out of 10 times it was with a word related to its scale or pace (see footnote 8 for the complete list of these words).

Another way of looking at scale or pace is to examine the kinds of actions done to ‘immigration’ or ‘migration’. This involves analysing the frequency of verbs associated with either term. For example, in the sentences ‘the government is reducing immigration’ and ‘we should encourage migration’, the verbs ‘reduce’ and ‘encourage’ signal some kind of action that is done to either immigration or migration. Looking at the most frequent actions associated with both terms over the whole period revealed a category of verbs expressing efforts to limit or control movement—what might be called ‘limit verbs’. Figure 7 shows how the visibility of these ‘limit verbs’, both in terms of raw frequency and as a share of all actions done to ‘immigration’ or ‘migration’, has changed from 2006 to 2015.

Figure 7

The chart shows that the overall frequency of these words, indicated by the solid line, increased by about four times between 2006 and the high point of 2014—from 283 instances to 1,414. Also, since 2010 these verbs have consistently made up about 30-40% of all verbs used in connection with ‘immigration’ or ‘migration’. These findings suggest that, although the visibility of this category became prominent in absolute terms beginning 2013, press coverage was already moving towards using these kinds of actions from 2010.

‘Net migration’, or the difference between the numbers of people entering and leaving the UK, became more prominent in public debate once the Conservative Party pledged to bring this figure down to the ‘tens of thousands’ as part of its 2010 General Election manifesto. Figure 8 shows how the press used this phrase more often as the Coalition government went on.

Figure 8

Since the idea of net migration has been closely linked with policy developments and rationales, it provides an opportunity to see how press coverage matches actual migration data. Figure 9 places the 2006-2014 annual net migration figures from the Office for National Statistics (ONS) Long Term Immigration (LTIM) data against the number of articles that each subset of the press published mentioning ‘net migration’ or some variation. These are expressed per 1,000 articles within each group, to allow comparison between subsets of the press.

Figure 9

Comparing net migration levels with media coverage of net migration, two patterns emerge. First is the clear increase in the proportion of migration-related articles mentioning net migration in 2010: this is likely due to coverage and subsequent discussion of the Conservative manifesto announcement. Second is the increase of tabloids and broadsheet articles mentioning net migration from 2011 to 2014, and of midmarket articles from 2012. These increases took place alongside increases in actual net migration. However, relatively high levels of net migration from 2006-2008—before the introduction of the net migration target—were not associated with the same level of coverage.

One possible interpretation is that as the concept of net migration gained greater traction during the Conservative-led coalition government from 2010, the press paid more attention to the actual quarterly ONS reports, with rising numbers providing opportunities for coverage:

The net migration figures have, in fact, barely changed since 2010 in spite of the Coalition ‘s cap on migration for people outside the European Union. (Broadsheet, 2014)

ONS figures show that net migration rose from 154,000 in 2012 to 212,000 in 2013, which suggests we are a long way off David Cameron ‘s target of 100,000 in 2015. (Midmarket, 2014)

No one asked the British population if we wanted an increase in net migration to this level, which is twice that of the previous decade. (Midmarket, 2013)

When the press explicitly described immigrants and migrants during 2006-2015, 3 out of 10 times (30.4%) it was with the word ‘illegal’

When it comes to describing immigrants as people, all subsets of the press used the term ‘illegal’ most often. As shown in Table 2, when newspapers characterised immigrants from 2006 to mid-2015, 3 out of 10 times (30.4%) it was with the word ‘illegal’. This is in contrast to the words ‘EU’ or ‘European’ which were the second-most frequent modifiers: they were only used 7% of the time. And, geographic terms like ‘African’, ‘Polish’, and ‘Irish’ were also used to describe immigrants’ origins.

Table 2 – Top ten modifiers of ‘immigrants’ or ‘migrants’, all publications, 2006-May 2015

Rank	Modifier	Frequency	Share of All Modifiers
1	illegal	14,202	30.4%
2	EU/European	3,271	7.0%
3	many	1,935	4.1%
4	new	1,395	3.0%
5	more	1,043	2.2%
6	African	921	2.0%
7	Jewish	782	1.7%
8	Polish	777	1.7%
9	Irish	750	1.6%
10	recent	551	1.2%

Figure 10 shows how the different press subsets used ‘illegal’ to describe immigrants and migrants since 2006. As a proportion of all the modifiers they use to describe immigrants, broadsheets consistently use ‘illegal’ the least compared to tabloids and midmarkets. This rate has dropped by nearly half to about 14% of all modifiers of immigrants in broadsheet coverage through May 2015. Meanwhile, tabloids and midmarkets tended to use ‘illegal’ between 30-50% of the times they describe immigrants. But all subsets show a downward trend: January 2013 to May 2015 has lower rates of using ‘illegal’ compared to 2006.

Figure 10

Returning to Table 2, the second most frequent way that the press described immigrants is with the terms ‘EU’ or ‘European’. This is actually a relatively new phenomenon: as seen in Figure 11, the use of these modifiers increased dramatically from 2012 onwards. At its peak in 2014, when newspapers described immigrants and migrants, nearly 1 out of 5 times (19%) it was with ‘EU’ or ‘European’. There is little evidence of an increase in references to EU migrants in 2007, the year that Romania and Bulgaria actually joined the EU.

Figure 11

Refugees tend to be described in terms of their geographic origins more than immigrants: since 2012, the press mentioned Syrian refugees the most. Prior to that point, ‘Palestinian’ was the most frequent modifier of ‘refugees’.

As seen in Table 3, 1 in 10 (10.3%) instances of the press using a word to describe refugees from 2006-2015 involved ‘Syrian’. Other terms related to either geographic origins or site of conflict also appear in the top ten most frequent modifiers: ‘Iraqi’ (in 2.6% of instances) and ‘Afghan’ (2.1%).

Table 3 – Top ten modifiers of ‘refugee(s)’, all publications, 2006-May 2015

Rank	Modifier	Frequency	Share of All Modifiers
1	Syrian	1,336	10.3%
2	Palestinian	907	7.0%
3	Jewish	766	5.9%
4	many	338	2.6%
5	political	339	2.6%
6	Iraqi	340	2.6%
7	other	308	2.4%
8	more	277	2.1%
9	Afghan	278	2.1%
10	young	205	1.6%

Figure 12 shows that before 2011 the press as a whole rarely if ever mentioned ‘Syrian refugees’. Rather, as the conflict grew in importance and scale, average monthly mentions also increased nearly three-fold between 2012 and the first part of 2015. Prior to 2012, the word most often modifying ‘refugees’ in the press as a whole was ‘Palestinian’. Meanwhile, the presence of ‘Jewish’ in the top ten modifiers comes mainly from references to people or places related to the Second World War.

Figure 12

Summary

On the issue of immigration, a complicated and wide-ranging topic, newspapers are important sources of information for the UK public (Blinder and Allen 2016). The analysis Section 3 shows that immigration has risen in media visibility since 2012, to levels higher than in 2006 (Figure 1). Immigrants and migrants received more explicit mentions than refugees and asylum seekers over the whole period (Figure 3). Over the entire 2006-2015 period, ‘mass’ was the single most frequent modifier of ‘immigration’, followed by ‘net’ and ‘illegal’ (Figure 5). Meanwhile, the press increasingly described the phenomenon of immigration in terms of its scale, pace, and direction rather than legality (Figure 6). Although immigrants are most often described as ‘illegal’, (Table 2), this trend has been declining across all subsections of the press since 2010 (Figure 10).

This section looks in more detail at two aspects of press coverage on immigration and immigrants: legal status and the role of the European Union.

Manual content analysis methods were used to identify specific aspects of news articles, opinion pieces, and letters from the public that corpus linguistic methods used in the previous section are less immediately suited for: do articles highlight mainly problems or achievements, which kinds of problems or successes are mentioned more often, what kinds of justifications are particularly frequent, who is claimed to be responsible for these occurrences? Drawing upon a subsample of the larger corpus, the research analysed 412 articles that dealt with either EU immigration or illegal immigration in the UK context. In the following section, where a difference between the two topics is mentioned, it is statistically significant. More details about the specific procedures used, as well as full statistical tables reporting all findings, appear in Appendix B.

When papers mention either EU or illegal immigration, they tend to focus on perceived problems rather than achievements. Coverage of EU immigration tends to mention the numbers of migrants, while articles mentioning illegal immigration highlight how existing laws are ineffective as well as increased criminality

About 7 in 10 articles (69%) mentioning EU immigration, and about three-quarters (76%) of articles in the illegal immigration sample, contained only mentions of problems. But, the types of problems mentioned in either topic differed. The analysis identified eight main types of problems in each sample (more details about each problem type can be found in Appendix B). Figure 13 shows the percentage of articles that mentioned one or more of these problems, divided by topic.

In the EU immigration topic, 39% of articles mentioned the number of migrants (either actual or potential) as a problem, compared to 20% in the illegality topic. Meanwhile, 41% of articles in the illegality sample argued that existing laws and rules were weak, abused, or ineffective, whereas only 20% of articles in the EU sample did so. In addition, 41% of the articles in the illegality sample (compared to 10% in the EU sample) mentioned crime and insecurity in the UK as a problem.

Under this Government £30 million less is being spent on protecting our borders than five years ago and for seven years the Home Office has not checked how many illegal immigrants are in Britain. These figures suggest that the Government is simply not serious about cracking down on what it is a flagrant and widespread act of criminality. When illegal immigrants come here, many of them take on cash-in-hand work which robs the Treasury of tax income but these immigrants also frequently become involved in organised crime. (Midmarket, 2014)

Figure 13

(Note: percentages within topics do not total 100% because articles could be assigned up to two problem types).

Meanwhile, much of the debate about EU migration has centred on its scale and anticipated impacts.

Tougher controls on migration could include curbs on freedom of movement to delay an exodus from new EU countries, David Cameron said yesterday…He acknowledged fears about an influx of non-skilled workers from Romania and Bulgaria when restrictions are lifted on January 1 and condemned Labour for giving Poles and other eastern Europeans instant access to British jobs in 2004. (Midmarket, 2013)

Whether regarding EU or illegal immigration, articles place the blame for problems—as well as the credit for successes—with the government in power. But, migrants themselves are also portrayed as the causes of problems with illegal immigration rather than EU immigration

Articles often attribute responsibility to specific groups or people when they report a problem or success. In both topics, articles usually put the spotlight on the government in power or one of its agencies. This happens in nearly half (49%) of articles mentioning a problem related to EU migration, and a similar proportion (46%) in the illegality sample. Meanwhile, when an article mentioned a success or achievement, about a third of the time (34%) in the EU sample it was the government that got credit. This happened a quarter (25%) of the time in the illegality sample.
But a distinction emerges in who or what gets blamed for problems specifically associated with each topic, as seen in Figure 14. Migrants tend to be cast as the cause of problems related to illegal immigration (28% of articles), much more than EU migration (9% of articles). Although there also is some evidence suggesting that the press blames smugglers and traffickers in the illegality sample, blame is attached less often to this group than to migrants themselves. Conversely, EU bodies and other European states are blamed for problems relating to EU migration (10% of articles) rather than illegal immigration (fewer than 1% of articles).

Four illegal immigrants sneaked into Britain in the back of a lorry carrying Tony Blair’s new armoured car. The asylum seekers slipped into the truck shortly before it was shipped across the Channel in a ferry…A police spokesman confirmed four men were arrested for alleged immigration offences on Tuesday. The asylum seekers have now been handed over to the immigration service. Sources said they will soon be released into the community while their asylum applications are processed. (Midmarket, 2007)

Non-EU migrants have swamped France. Instead of deporting them, France aims to offload them on us. Most are delighted to come, since our economy and benefits are better. And France is using our EU contributions to ease their passage. It is truly insane. (Tabloid, 2015)

Figure 14

(Note: percentages within topics do not total 100% because articles could have multiple actors blamed for problems or credited for successes).

Meanwhile, when articles report on a perceived success or achievement, the only significant difference between the two topics appears in regard to the police or security services. In 23% of successes mentioned in the illegality sample, compared to none in the EU sample, the police receive credit for achieving something. Closer inspection revealed that articles attributing success to this group are generally mentioning the successful arrest of illegal immigrants.

Two Brazilian illegal immigrants who set up Britain’s biggest fake passport factory were jailed for five years yesterday…When police raided it last November after a tip-off from a Central London stationers, officers found ‘wall-to-wall machinery’ along with 12,000 passport covers, hundreds of passport photographs, driving licences and utility bills. Detective Sergeant Tony Lynes said the passports could easily have been used in benefit, credit card, loan and mortgage fraud. (Midmarket, 2006)

Whether dealing with a problem or success, the people identified as the source of the main message are most often the article authors themselves

The analysis identified who says that something matters, or the messenger of an issue. In nearly half of both samples (about half of articles mentioning either EU or illegal immigration), the author of the article is the person who is communicating the main issue—asserting whether it is problem or success. This includes instances of opinion pieces, commentaries, and letters. (In articles where the author was clearly writing as a representative of a particular group, such as a business, it was coded as part of that group). The second-most frequent group that articles use to transmit a message consists of MPs or other politicians: for example, 18% of EU and illegality articles containing a problem cite politicians as the messenger (Figure 15) .

Figure 15

By way of example, the first of the following two quotes shows a case in which the author of the article is the ‘messenger’ of the problem, while the second shows a case in which the messenger is a politician.

Violent extremism, illegal immigration, climate change and rising food prices are not problems we can address by acting only within our borders. Aid focused on tackling poverty and nurturing civil society can play an important part in tackling the root causes behind such challenges. (Broadsheet, 2014)

Tony Abbott, the Australian prime minister, has been quick to recommend his approach to Europe after hundreds of migrants drowned in the Mediterranean. “The only way you can stop the deaths is to stop the people smuggling trade. The only way you can stop the deaths is in fact to stop the boats, “he said. “That’s why it is so urgent that the countries of Europe adopt very strong policies that will end the people smuggling trade across the Mediterranean. (Broadsheet, 2015)

Other actors outside of government agencies or Parliament do not appear to be prominent as messengers in discussions of problems or achievements related to EU or illegal migration.

Assertions in media coverage also tend not to be accompanied by other sources—either in support or contrast to the main points.

Authors have the option to bring in other actors, whether through passing attribution or explicit quotation. But in the cases of both topics, articles tend to lack references to additional sources. This happens more often in the context of illegal immigration (56%) rather than EU immigration (41%). When sources do appear, they tend to be politicians (mentioned in 23% of articles in the EU sample; 15% of articles in the illegality sample), or civil servants (19% of articles in the EU sample, and 12% of articles in the illegality sample). Although the samples include some genres of newspaper content that might be reasonably assumed to naturally feature the author’s own voice—one-off opinion pieces or regular columns, for example—the fact that other sources tend not to be brought in to complement or strengthen the author’s arguments highlights the significance of journalists’ roles in framing migration narratives.

In both topics, problems and successes were most often justified on the basis that the UK’s own prosperity and well-being should come before others. But, justifications related to public order and security were very strongly associated with illegal immigration rather than EU migration. Meanwhile, arguments based on sovereignty were common in the EU sample

Another key part of press coverage involves how issues are ethically framed, or how authors draw upon different kinds of values to make their arguments. Some of these values might place more emphasis on the host country’s own welfare (‘domestic prosperity’), while others might highlight the importance of states being able to govern themselves (‘sovereignty’). In nearly half (48%) of articles mentioning EU immigration, the authors justified their argument on the basis of domestic prosperity, or the idea that ‘immigration should be controlled to deliver the best possible economic, social and welfare conditions for citizens’ (Balabanova and Balch 2010: 384). A similar proportion of articles in the illegality sample (44%) also contained justifications based on domestic prosperity.

Figure 16

(Note: percentages within topics do not total 100% because articles could be assigned up to two justification types).

The following two examples illustrate how ‘domestic prosperity’ is communicated by emphasising differences between interests in the UK’s own economy and society and the interests of those outside Britain.

There is no doubt that Britain needs to address its skills deficiencies. Improved education and training programmes are crucial. There is, of course, another way of tackling the problem, and that is by importing skilled labour. This has happened to a considerable degree over the past two years as a large number of people from Eastern Europe have migrated here. There is no doubt their presence has been a fillip for employers and helped to hold down inflation by restraining wage increases, if not undercutting wages. On some estimates, they contributed about 1/2%, possibly more, to GDP growth in 2005 and a similar contribution can be expected for 2006, thus bolstering the Chancellor’s GDP data. But large-scale immigration cannot be the long-term solution to our skills problems. There are too many downsides in a crowded island. (Broadsheet, 2006)

The Commons home affairs select committee’s damning report on immigration controls…says that the Home Office is clearly ‘not in a position’ to think through the wider implications of immigration on labour markets or society and that a cross-government committee should take charge of this. (Broadsheet, 2006)

Previous research by Balabanova and Balch (2010) measured the quantities and types of ethical framings that appeared in British and Bulgarian coverage about EU migration in 2006. They found that in nearly half (48%) of all framings identified in their UK sample (52 out of 108), the main justification was ‘domestic social justice’, which corresponds with this report’s code of ‘domestic prosperity’ (see Appendix B for full details of each code). Subsequent research that compared coverage in 2006 and 2013 (Balch and Balabanova 2016) found that the emphasis towards domestic concerns had intensified over the period.

As seen in Figure 15, the kinds of justifications made between the two topics differed, too. Arguments citing the value of public order—that is, the claim that any action on immigration should preserve safety and the rule of law as well as combat anti-social behaviours—appeared most in the context of illegal immigration (40% of articles containing a problem) rather than EU migration (12% of articles containing a problem).

More than 50,000 illegal immigrants who were told to leave the country have been granted a ‘de facto amnesty’ after officials admitted they had no idea where they were. [The director-general of immigration enforcement at the Home Office] added that because the department had a ‘limited amount of enforcement resource’, it was used to pursue those immigrants whose location was known and who were considered to be likely to cause harm. ‘If there is any evidence that they are causing harms, i.e., criminal offending, then of course we would pursue them’, she said. (Broadsheet, 2014)

Meanwhile, arguments based on the justification of sovereignty, or the belief that actions should be based on Britain’s right to act according to its own rules and laws, frequently appeared in connection with discussion about the problem of EU migration: 18% of articles in the EU sample contained this justification, compared to 8% in the illegality sample.

Non-EU relatives of EU citizens currently need visas before they can come to the UK even if they are living within the community. But the European Court of Justice has ruled that Britain has had no right to enforce travel controls on non-Europeans who have been granted a residency permit by another member state. The EU court’s ruling, expected to be upheld in the New Year by Britain’s own High Court, is a fresh blow to the Government’s border powers. (Midmarket, 2014)

These results about the kinds of values and justifications used in news coverage about EU and illegal immigration suggest reasons why these topics matter as either problems or successes. Illegality is strongly linked with ideas of criminality and the need for keeping the public safe, while EU immigration is coupled with concerns about how the UK’s ability to govern itself.

Summary

The analysis in Section 4 shows that discussion of EU immigration tends to focus on the scale of migration, while coverage mentioning illegal immigration highlights how existing laws are ineffective (Figure 13). Across both topics, the journalist themself is most often the “messenger”, rather than simply reporting a third party’s concerns. The government receives most of the blame for problems—although migrants themselves also receive blame in reference to illegal immigration, while EU states and organisations are also implicated in problems relating to European immigration (Figure 14). And, in line with previous research (Balch and Balabanova 2010; 2016), many claims about achievements or problems relating to these topics were based on arguments that emphasised the UK’s own domestic prosperity, whether in economic, social, or security terms (Figure 15).

Media coverage of migration issues divides opinion. To some, British newspapers are barely-regulated juggernauts of opinion, pushing political agendas and shaping the UK’s policy landscape. Others see the UK press as champions of common sense, reflecting the voices of the people and dealing with complex subjects in a forthright manner. Regardless of one’s position on these issues, it is clear that there is an important relationship between the nature of media coverage, policymaking and the public debate.

The data in this report suggest that media coverage should be seen in the context of both the changing dynamics of migration and unfolding policy debates. Looking at trends in media coverage from 2006 to 2015, there are several clear examples of media coverage responding to events on the ground, such as the growing number of displaced people following the conflict in Syria. In other cases, media coverage appears to have been driven by policy positions, such as the government’s net migration target from 2010 onwards.

Over the last decade, words related to the scale and pace of migration to the UK have increasingly dominated discourse, pushing those relating to the legal status of migrants into a distant second-place. In fact, the prevalence of ‘illegal’ as a way of describing immigration has declined. This has occurred alongside a sharp increase in the number of stories appearing in British national newspapers mentioning European and EU migrants. A striking feature of media coverage around EU migration over the period of this study is that it was less evident in the earlier period of this study, from 2006-2008. This was despite the fact this was a period when migration to the UK from new EU member states became very high in a very short period of time. But, newspaper stories about EU migration increased in frequency rapidly from 2012 onwards—continuing at elevated levels into 2015—which coincided with the end of transitional labour market controls on Romanian and Bulgarian workers.

The growing emphasis on scale and numbers rather than illegality also has implications for the nature of coverage, since—as the qualitative analysis shows—the two are covered in different ways. For example, coverage of EU migration, which tended to focus on the scale of migration, is less likely to emphasise links with criminality or attribute responsibility for the problem to migrants themselves.

Recent changes in the focus of the debate are pertinent for understanding the context of the recent debate about Brexit, in which the scale of EU migration and the UK’s ability to restrict access of low-skilled EU migrants to the UK labour market was a key issue. Although further research is needed to understand what role the media played in shaping or reflecting public attitudes to EU migration and EU membership, the evidence in this report leaves the possibility open that it has played some part.

The quantitative design and analysis (seen in Appendices A and B) has some important limitations. First, it relies upon the quality of archiving provided by Nexis UK and Factiva. Although checks were conducted throughout the data collection process to ensure no systematic errors or gaps existed, it is possible that both services may miss some data due to human error. Given the large-scale nature of the analysis, any random errors should not affect the findings meaningfully. Second, counting up the frequencies of key terms—as well as the words associated with them—reveals only some aspects of language. It is not as well-suited as closer reading or content analysis for identifying either how arguments are formed or the overall tone of an article.

Meanwhile, the manual content analysis design (explained further in Appendices B and D) also has limitations. First, determining whether a claim is a ‘success’ or ‘problem’ is subjective: it may look different depending on one’s perspective. So, decisions about which categories the articles fit into were taken by at least two people who separately read each article. This helped minimise potential bias introduced by any single person. Second, the design specifically considered whether people or organisations were the ones to blame or credit for developments about EU and illegal immigration. Although much research seems to reflect this approach (see for example Statham and Geddes 2006), it is possible that wider factors that aren’t related to any single actor also contribute to events covered in the news. The tendency to attribute problems and successes to more immediately visible actors is consistent with the possibility that structural drivers of the scale and nature of migration—such as economic growth, demographic shifts, and the globalisation of labour and product markets—are underappreciated in the public debate. However, the content analysis for this report did not seek to separately identify cases where responsibility was attributed to structural factors.

The Observatory would like to thank the staff at Sketch Engine for their help with the data collection: in alphabetical order, Miloš Jakubíček, Vojtěch Kovář, Jan Michelfeit, and Akshay Minocha. We are also grateful to Simón Escoffier and Shireen Walton who completed the manual coding; Andrew Cutts for statistical support; Alex Balch for feedback on the manual content analysis design and results; Robert Ford, Will Somerville and Bridget Anderson for feedback on a previous draft of the report; and Michael Rundell for feedback on the corpus linguistic methods and techniques. In memory of Adam Kilgarriff (1960-2015), co-author of the Sketch Engine whose involvement in the ‘Migration in the Media’ project since 2012 had been invaluable.

• Allen, William L, and Scott Blinder. 2013. ‘Migration in the News: Portrayals of Immigrants, Migrants, Asylum Seekers and Refugees in National British Newspapers, 2010 to 2012’. Migration Observatory Report. University of Oxford: COMPAS.
• Associated Press. 2013. ‘Illegal Immigrant’ No More. https://blog.ap.org/announcements/illegal-immigrant-no-more.
• Baker, Paul. 2006. Using Corpora in Discourse Analysis. London: Bloomsbury Academic.
• Balabanova, Ekaterina, and Alex Balch. 2010. ‘Sending and Receiving: The Ethical Framing of Intra-EU Migration in the European Press’. European Journal of Communication 25 (4): 382–97.
• Balch, Alex, and Ekaterina Balabanova. 2016. ‘Ethics, Politics and Migration: Public Debates on the Free Movement of Romanians and Bulgarians in the UK, 2006–2013’. Politics 36: 19–35.
• Beckett, Charlie. 2010. ‘The Value of Networked Journalism’. Value of Networked Journalism Report. London School of Economics: POLIS. http://www.lse.ac.uk/media@lse/POLIS/Files/networkedjournalism.pdf.
• Beckett, Charlie, and Robin Mansell. 2008. ‘Crossing Boundaries: New Media and Networked Journalism’. Communication, Culture & Critique 1 (1): 92–104.
• Blinder, Scott, and William L Allen. 2016. ‘Constructing Immigrants: Portrayals of Migrant Groups in British National Newspapers, 2010–2012.’ International Migration Review 50 (1): 3-40.
• Boyd, Danah. 2008. ‘Can Social Network Sites Enable Political Action?’. International Journal of Media and Cultural Politics 4 (2): 241–44.
• Chong, Dennis, and James N Druckman. 2007. ‘Framing Theory’. Annual Review of Political Science 10: 103–26.
• Gabrielatos, Costas and Paul Baker. 2008. ‘Fleeing, Sneaking, Flooding A Corpus Analysis of Discursive Constructions of Refugees and Asylum Seekers in the UK Press, 1996-2005’. Journal of English Linguistics 36 (1): 5–38.
• Helfer, Luzia. 2016. ‘Media Effects on Politicians: An Individual-Level Political Agenda-Setting Experiment’. The International Journal of Press/Politics 21 (2): 233–52.
• John, Peter, Anthony Bertelli, Will Jennings, and Shaun Bevan. 2013. Policy Agendas in British Politics. New York: Palgrave Macmillan.
• Kilgarriff, Adam, Vít Baisa, Jan Bušta, Miloš Jakubíček, Vojtěch Kovář, Jan Michelfeit, Pavel Rychlý, and Vít Suchomel. 2014. ‘The Sketch Engine: Ten Years on’. Lexicography 1 (1): 7–36.
• Krippendorff, Klaus. 2012. Content Analysis: An Introduction to Its Methodology. SAGE.
• Marcus, Mitchell P., Mary Ann Marcinkiewicz, and Beatrice Santorini. 1993. ‘Building a Large Annotated Corpus of English: The Penn Treebank’. Computational Linguistics 19 (2): 313–30.
• McEnery, Tony, and Andrew Hardie. 2011. Corpus Linguistics: Method, Theory and Practice. Cambridge University Press.
• Miller, Joanne M, and Jon A Krosnick. 1996. ‘The Anatomy of News Media Priming’. In Do the Media Govern? Politicians, Voters, and Reporters in America, edited by Shanto Iyengar and Richard Reeves. Thousand Oaks, CA: SAGE.
• Pomikálek, Jan. 2011. ‘Removing Boilerplace and Duplicate Content from Web Corpora’. PhD Thesis. Brno: Masaryk University.
• Soroka, Stuart. 2002. ‘Issue Attributes and Agenda-Setting by Media, the Public, and Policymakers in Canada’. International Journal of Public Opinion Research 14 (3): 264–85.
• Statham, Paul and Andrew Geddes. 2006. ‘Elites and the “Organised Public”: Who Drives British Immigration Politics and in Which Direction?’ West European Politics 29 (2): 248–69.
• van Klingeren, Marijn, Hajo G. Boomgaarden, Rens Vliegenthart, and Claes H. de Vreese. 2015. ‘Real World Is Not Enough: The Media as an Additional Source of Negative Attitudes Toward Immigration, Comparing Denmark and the Netherlands’. European Sociological Review 31 (3): 268–83.