[Publication] The Visibility and Communication of Cohesion Policy in Online Media

Download the Study

The visibility of cohesion policy in online news media stories is relatively low, with many references to the policy being little more than an acknowledgement of their contribution to projects. Cohesion policy is more visible in national media, but regional and local media are more likely to provide some depth to their coverage at the project level.

The tone of news media coverage is generally positive, particularly so in regional/local media. News stories tend to be framed around the socio-economic consequences of the funds in terms of economic development, jobs, infrastructure and social inclusion. Institutional bargaining is a very dominant theme, with stories giving prominence to the reform of the Multiannual Financial Framework.

There are significant differences across countries in visibility, tone, framing and bias. Member States with higher levels of cohesion policy funding tend to have greater visibility and more positive tone. The proportion of bias (or myths) in cohesion policy news stories is less variable and relatively low (7 percent of stories). Most of the myths propagated in online news are about fraud, mismanagement and the lack of added value. News stories with myths are most prevalent in some (though not all) of the net payer countries.

Turning to social media, the visibility of facebook posts on cohesion policy by EU institutions, European political parties and interests groups is relatively low, constituting 6 percent of their total facebook activity. Of these, almost 60 percent are by DG REGIO followed (a long way behind) by the CPMR and Commissioner for Regional Policy. Most FB posts are neutral in tone, providing factual information, although DG REGIO posts are generally more positive than posts by other actors.

The visibility of cohesion policy in Twitter is low among the 2019 European Parliament election candidates relative to their overall Twitter activity and in terms of frequency. Around 15 percent of the candidates with a Twitter account tweeted about cohesion policy during the period January-June 2019, on average 3 times, and generally in positive terms. DG REGIO produced the largest number of tweets followed by ‘EU influencers’ (typically journalists, EU officials and academics) and interest groups. The tone was overwhelmingly positive, significantly more so than for the EP candidates. Public discussion about cohesion policy are relatively low key but often lively in terms of politicisation and sentiment.

Cohesion policy is well represented in European Commission press releases relative to other EU policy domains, but less so in the press activity of other EU institutions and political parties. Overall, press releases from EU institutions tend to focus on socio-economic issues, whereas the political parties are more likely to frame their press releases in terms of power issues (institutional bargaining, empowerment, conditionality etc.).

The main conclusion is that the communication of cohesion policy in online media has been weak in recent years, despite the increased political priority placed on communication by EU institutions. Media coverage is often shallow, with limited depth of understanding or analysis and there is little appreciation of the role of the EU or wider impact of the policy.

A key precondition for greater visibility and better communication is greater citizen engagement in the policy, particularly at the programming stage. The main recommendation of this study is the need to promote a citizen-centred approach to programming through democratic innovations, including the piloting of participatory budgeting. The European Parliament should strive to ensure that the public have a real say on what is funded in their local area by one of the largest and most visible areas of EU expenditure impacting on their daily lives.

Link to the full Study: http://bit.ly/629-196

Please give us your feedback on this publication

Selection of maps and figures:

This slideshow requires JavaScript.

Figure 1 lists the results of the news search for ten EU member states using a set of harmonised keywords for cohesion policy, the funds (ESIF, Structural Funds) and individual funds (ERDF, CF, ESF). This amounted to just over 60,000 news stories over the period 2014-2019, after the removal of duplicates.

Figure 2 shows the evolution of online media stories over time revealing a general increase in most cases, as depicted by the regression line in red. At the same time, the peaks and troughs in activity suggest a clustering around specific events.

The visibility of key cohesion policy terms in online news media across countries is illustrated in Figure 3. For visualisation purposes, the acronym are used for the ERDF, ESF and ESIF (rather than the full terms) and in English (rather than the national language).

. Figure 4 shows the distribution of visibility as measured on two different scales, a five-point and a three-point scale. The key finding is that the visibility of cohesion policy in terms of its prominence within stories is low. On the five-point scale, the most frequent value is very low.

Figure 5 shows the chi-square residuals for testing the association between the territorial level of a news source and a story’s visibility. The plots shows whether an observed count for a category is above, below or within an expected range. Blue bars indicate that observed count is significantly higher than expected, while red bars indicate the inverse i.e. that the count is significantly lower than expected. Grey bars indicate that the observed count for a particular category is within an expected range. Figure 5 shows that there is a clear and significant association between the territorial level of a news source (regional or national) and the visibility of cohesion policy in news stories. In national media cohesion policy is more likely to have a high visibility in a story. The opposite is the case for regional media, where the visibility of cohesion policy in a news story is significantly lower than expected.

In Figure 10 the negative score is subtracted from the positive score to give an overall view of positive tone across the ten Member States. Only in Austria and Germany are the negative tone scores higher than positive scores leading to an overall negative score.

Figure 16 shows the proportion of myths detected across the ten Member States. The percentages are on the whole very small for the individual myths. Very few cohesion policy news stories contain myths as defined above. Nonetheless, the number of myths do add up just over 6.6 percent of stories.

The evolution of each topic over time is shown in Figure 19. Some topics have lots of overlapping error bars, such as the topics ‘Economic development’, ‘Employment, training and education’ and ‘Research and Innovation’ (with the exception of the year 2019), which indicates a more equally distributed discussion of topics and themes over time. Other topics appear to be punctuated by emphasis during particular periods. This is quite obviously the case with the Brexit topic or the Greek crisis and conditionality topic. The latter is highly prevalent between 2014 through to 2016 then almost disappears and only reappears in 2019.

Figure 24 plots the average rate of engagement of all FB posts on cohesion policy by actor. It is clear that the posts with the highest engagement are by DG REGIO (‘EU in my region’), which accounts for 13 of the top 15 posts – represented by purple dots with engagement values greater than 2.5. However, the box plots also show that other actors score higher engagement ratios on average across all of their posts. In particular, the interest groups CALRE, CPMR and AEBR have particularly high engagement rates. While the European Parliament has the highest engagement rate, it is important to note that it only has two cohesion policy posts under its corporate account and that the high score is accounted for by a single post.

The evolution of Twitter activity does not reveal a pronounced change over time, as shown by the rather flat regression line in the first panel of Figure 30. However, there are some noticeable peaks in Twitter activity.

Figure 33 shows the engagement ratio by follower count with some examples of outlier cases from each of the three groups. Amongst the EP candidates, the MEP Marine Le Pen has by far the largest number of followers. Further inspection reveals that this MEP posted one original cohesion policy tweet during the period under study that generated a relatively high number of retweets (26), which constitute a very small proportion of her follower count. Not surprisingly, among the EU institutional groups, it is the European Commission and the European Parliament with the highest following by some degree.

Sentiment analysis was performed on the tweets to assess the tone of cohesion policy discussions. As can be seen in Figure 35, the average sentiment polarity is in the positive range. This is particularly the case for the EU institutional actor category, and to a slightly lesser degree for the EP candidate tweets. The ‘other’ category, on the other hand, has more polarization in terms of positive and negative tone.

The distribution of press releases over time are shown in Figure 36 distinguishing those on cohesion policy from other all other EU policy domains. There is clearly an upward trend in the number of press releases. The drop in the number for the year 2019 is due to the fact that the data collection was for the first half of the year. The proportion of cohesion policy relevant press releases is very small. At 207 relevant press releases, the number of cohesion policy related press releases appears to be rather low, accounting for just over 1.5 percent of total press releases.