Measuring Salience in EU Legislative Politics

Measuring Salience in EU Legislative Politics

The article “Measuring Salience in EU Legislative Politics” evaluates democratic policy of the EU trough the measurement of its political importance. Since the same legislative act can be categorized differently, due to the usage of different salience estimations, it is very important to measure salience empirically. This article compares salience measures based on three different types of data sources—expert interviews, text analysis and media coverage—using legislative proposals that cover different policy fields and types of proposals.

In order to address the issue, it is necessary to first define salience. The word “salience” is very close in meaning with the word “importance”. Its concept basically defines the importance one attaches to the issue. This importance can be based on the policy impact the issue has, its political sensitivity, or attention it gets from its electorate. Salience has two components: actor-specific and issue-specific. The issue-specific component means that some policies are more important than the others, actor-specific component implies that different actors might disagree on the relative importance of a particular policy. In addition, the estimation can be made at different levels, thus one can look at the policy as a whole, or at the individual legislative proposal, or even issues within a proposal. Estimation can also be made retrospectively or simultaneously. For the study of legislative politics the first perception of the proposal, while directly adopting or implementing it, is considered the most relevant.

Empirically, salience is an important feature when it comes to the level at which the decision is made while it is already reached. This includes whether the decision is made at the administrative or ministerial level in the Council. The transposition of the EU Directives into the national law also depends on the salience. Thereafter, the article distinguishes such sources of information as: expert interviews, secondary sources, text analysis, public opinion surveys, media coverage and procedural information.

The most flexible data-gathering technique is expert interview, because it shows the importance of the issue at different levels, and includes the actor-specific component of the salience measure. However, the biggest disadvantage of this data gathering process is its cost. Using expert interviews it is very hard to create a time-series data for a large numbers of data, because the persons can be unavailable over the time. Also, depending on the expert’s organization, different experts can disagree on the salience level of a particular issue.

Secondary sources data makes the data-gathering process a lot easier, since it avoids some practical issues concerned with direct interviews. An example of such data can be a data set created by International Labor Organization used to study legislative productivity in a comparative manner. However, secondary sources have several disadvantages, which include giving just a crude measure of salience, reflecting retrospective rather than simultaneous salience, also the selections might be made on the criteria other than political importance.

Text analyses are used to calibrate the relative importance of the issue by comparing the space devoted to the issue in party manifestoes and by analyzing the speeches. However, these measures are restricted to the level of policy fields, they do not compare different legislative proposals. Thus text analysis show positions rather than salience of the proposals.

Legislative proposals themselves are often introduced with the number of reasons why they have to be adopted. The number of these reasons can be used as a measure of the importance of the act. However, the number of declamations can exhibit the scope of the proposal, whether or not it is a part of a larger legislative agenda, or even indicate the level of political controversy.

Salience is also measured by the amount of media coverage the topic receives. This can include both examination of the topic or its criticism, because either one determines the demand for the information on the given topic. However, media coverage does not always measure salience. Thus it can reflect the complexity of the topic, or the controversy it has raised. Moreover, the measures are based on the viewpoints of the given target’s audience, and hence is not representative of legislative actors or the population itself.

Public opinion surveys are used to evaluate relative importance of the policy fields. This type of data-gathering includes the actor-specific component as well as expert interviews. Actor-specific indicator can be constructed by differentiating the views of different parties supporters or by different countries citizens. However, whether or not public opinion surveys measure salience depends on the responsiveness of the political system.

Procedural aspects, such as the time spent debating a proposal, number of hearings or legislative sessions devoted to it, or the amount and type of legislative activity the topic generates can be used to measure salience. Since the measures differ according to the legislative setting, they capture the actor-specific aspect. However, the amount of attention devoted to the topic may depend on its controversy rather than importance, thus procedural aspects do not always reflect salience. Moreover, the procedural aspects are often explained as a consequence of the salience, therefore it cannot be used as a measure of salience or as an independent variable.

The author of the article uses the data set from DEU which is a sample of legislative proposals covered in the Council in the period from January 1999 to December 2000. The data was collected at the issue level. To ensure at least some degree of salience of the proposals, only those ones that were covered in over five lines were collected. Typically, a proposal addresses a number of policy concerns, expert interviews were used to identify the most important issues within the proposals. To be included in the data set, at least some experts had to disagree on the issue. The number of the issues per proposal ranges from one to six, with the mean value of 3.05. There are 174 issues across 70 proposals. Being based on the experts interviews, most of the data reflects observations of other actor rather than views of the actor concerned. The author claims that relying on the expert respondents rather than using self-reporting is a potential source of measurement error. Also, most of the issues addressed are from the field of Agriculture and Internal Market.

Experts provided their salience values, ranking from 0 to 100, for all 15 member states of the EU, European Parliament and European Commission. The data uses tree aggregate values: mean, weighted mean and the maximum. The mean value assumes that all the states determine the Council’s behavior equally. The weighted mean is taken because one can argue that some member states are more powerful than the others. The maximum value is taken because the Council’s behavior as a whole might be driven by concerns of a single member state. Which of these values is the most appropriate in the empirical analysis depends on a particular issue.

The tree legislative bodies value the salience of the issues differently. Thus, the state that is interested in particular issue will consider it more important than the Commission or the Parliament. This is the case in 65 percent of the issues. Similarly, the mean of the Parliament typically lower than the mean for the Council.

In 6.9 percent of the issues, the maximum of the salience is 0. This indicates that all Council members have agreed on the salience of the issue. If at least one member state considers the issue to be utmost important and the other values it as not important at all, the maximum value of the salience would be 100. This is the case in 10.3 percent of the cases. The range has an average value of 50.7. The average standard deviation lies at 17. Thus, member states have quite divergent views on the importance of issues according to the expert interviews.

Besides the expert interview data, the article also provides the data from the recitals and media coverage.

Recitals provide the reasons for the legislative proposal. However, some of the recitals are devoted to the previous legislative activities, rather than giving the reasons for adopting a proposal. The article uses two measures of recitals: The overall number, and the number of recitals excluding those referring to the current EU law (pure recitals). The two types are strongly correlated.

The Brussels-based newsletter “Agence Europe” was used to code the media coverage of the proposals. Only articles that gave a substantial coverage of the issue were taken in the data, excluding those that were just covering the procedural issues, and those that covered more than four issues with no clear focus on the relevant proposal. The measure of the salience was based on the word count, assuming that the more extensive the coverage of the issue is, the more important the issue is.

In general, there is no statistically significant relationship between media coverage and interview based values. However, there is a weak, but statistically significant relationship between the number of issues addressed in the media and the number of recitals.

I agree with the author that the expert interviews are the most flexible data-gathering technique, however, I think that using this technique the interviewer must pay close attention to the structure of the questions. The questions must be addressed in such a way that the expert specifically defines the level of the importance of the issue. I believe this technique to be the least ambiguous, since its direct purpose is the measure of salience.

Measuring recitals number and amount of media coverage is a very controversial aspect. I think that the measure of the salience has to be addressed to the public, and reflect the opinion of the public. Therefore I do not believe recitals to be particularly valid measure of importance. Recitals are created by those who address the issue, and are designed to convince the Council to adopt the proposal. This does not show the importance of the issue itself, but provides the subjective opinion of those addressing the proposal. Now, the reasons for adopting a proposal certainly do show the importance of the proposal, but the feature that defines the salience is not the number of those reasons, but their validity and quality.

When it comes to media coverage, I personally perceive it with a lot of skepticism. The whole purpose of media is to impose a particular viewpoint on the public. Therefore, I do not think this can be a valid measure of the salience. There is often a surplus of media coverage of particular issues, this does not necessarily mean that this is the viewpoint of the whole population. In fact, issues, opinions, evaluations covered and addressed on different channels, in different journals, newspapers are very subjective, and may depend not only on the audience, but also on the viewpoint of the journalist himself, or even on the viewpoint of the owner of the channel, newspaper, ect. Thus, there may be no demand for information about the issue, but the huge amount of supply we get from the media misrepresents its importance, and improperly influences the Council’s decision-making process.

I think that the measure of demand for a particular issue may be a good example of measuring salience. The only way I can think of doing it is by measuring the number of internet searches. At the end, all the bills and issues are accepted, processed or addressed for the benefit of the population, and therefore, I believe that it is the population who has to determine whether the issue is important or not.

 

 

 

 

 

 

 

Work Cited

Andreas Warntjen. “Measuring salience in EU legislative politics.” European Union Politics, 13

December 2011.

 

 

 

 

 

 

Measuring Salience in EU Legislative Politics

 

By

Tamara Ikonnykova

 

Professor

Marquita Bowers

 

28 Mar, 2014