Navigating the Digital Sphere: Exploring Websites, Social Media, and Representation Costs—A European Union Case Study

Round 1

Reviewer 1 Report

Dear Author(s),

The understanding interaction of representation in the web and social networks and the costs on representation is a significant contribution to the understanding in and prospects of the information society. Thus, the article “Navigating the Digital Sphere: Exploring Websites, Social Media, and Representation Costs - A European Union Case Study” appears to be an important attempt to evaluate the relationship between websites, social platform presence and presentation costs on the example of participants in the European Union Transparency Register. However, the article needs a minor revision; the main comments include the following.

In the Section 1. “Introduction”, the interaction between lobbying groups and lobbying interests in social networks and platforms is substantiated. However, in the following, only organizations from the Transparency Register are used in the context of lobbying. Here it would be worth clarifying the concept of lobbying or expanding the context of the trend under consideration as a representation in the information field as a whole. The Register is not always a platform for the promotion of lobbying interests to the government, for example, for some professional consultants, self-employed individuals, or academic institutions. It also follows from the form presented on the Register’s website that some organizations may not pursue any direct interests. If this Register is the primary source of lobby group classification or reporting, it should be mentioned in the abstract, keywords and references.

In the Section 1.1. “Interest group categorization and EU Transparency Register”, it is presented the different categories of interests: all interests, promotes their own interest or the collective interest of their members, advances interest of their clients, do not represent commercial interest. It should be clarified which of these interests the author(s) chose.

The author(s) use words “robust” in the Section 3.2. “Statistical methodology” and Section 3.4. “Ethical consideration”, however, it is not shown that any reliable methods are applied to the data, methods or results. This word should be deleted or confirmed.

In Table 2. “Frequency and percentage of websites by type of group of interest”, in the first row, the sum of participants with and without websites should be equal to 100%. In addition, the table indicates that most organizations do not have websites. For the first row, 99.5% of academic institutions do not have websites. May be the authors changed the titles of the columns in this Table?

In the Section 4.3. “Social media usage by interest groups”, there are some contradictories by figures. The author(s) write that “In the table 3 Academic institutions show a high presence on social media platforms, with 97.16% of the analyzed groups having a website”. However, these figures are not shown in Table 3. Further, “Among the social media platforms, Twitter has the highest frequency of use, with 185 groups having an account.”. In Table 3 it is 166. May be the author(s) should convert this Table to percentages?

It is needed to change the title of Figure 2. “Figure 2. This is a figure. Schemes follow the same formatting.”. In Figure 2, the author(s) should also give abbreviations of organisations.

In the Section 4.4 “Relationship between social media usage and representation costs of interest groups” or in the paragraph regarding limitations, it is should be noted that there are some limitation on the data. Since representation costs relate to the Register, and not to social networks, that is, these be different areas of representation costs. In addition, the registration form in the Registry does not contain verified information about the amount expenses and therefore may not be accurate. 

In the sentences 509, 515, the authors write about the R-squared value of 0.001, in the Table it is 0.009 in the Table. This is a very small value and can be explained by above mentioned or other limitations.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This research piece has promise, but it needs reworking to be more informative.
The strength of this piece lies in the descriptive findings.  The mapping of the persuasive infrastructure of groups lobbying the EU on policy is important in and of itself.  Table 1 would become even more informative if you put in the mean of the budgets for each category.  This would be very informative as it would give us additional understanding of the amount of resources deployed by each sector in the lobbying efforts.  For instance, do the NGOs and labor organizations have the resources to effectively compete with the lobbying efforts of corporations and business associations?  In the discussion highlight findings such as these.  Mapping the territory is important in exploratory research.

On the second research question, note you have research questions in this piece as opposed to hypotheses, you find that a very high percentage of lobbying organizations have web sites.  The descriptive statistics in Chart 2, provide additional confirmation that there is not much variation across the different sectors (e.g., organizations, NGOs etc.).  These are important preliminary findings as to how the different kinds of groups use their web sites in conjunction with their social media.  Your results provide the baseline for more in-depth content analyses of website content and social media content.  I suspect that there is considerable overlap between the two.

Chart 2 needs to be resorked a bit.  You have thw headings on the columns switched-has a website, does not have a website.  In addition, I would recomend removing the column that has the percentages of the organizations in a sector that do not have a website.   Your readers are numerate enough that they can easily figure out what the remainder is.

In chart 3,  please report the percentages of organizations in each sector that employ each of the social media.  The absolute frequencies give me little idea of how the use of social media may varies or does not vary across the sectors you are comparing-reporting the percentages of organizations in each category that have each type of social media would be very helpful-like you report in Table 2 for web site usage.  In addition, you might consider an additional chart before chart 3, that simply gives the frequencies of organizations that have different types of social media.  The rank order of the the relative usage of social media all types of of lobbying organizations should be discussed by itself.  The comprehensive descriptive data that you have in this piece is both comprehensive AND valuable.  In addition, dispense with six figures that repeat the information in chart 3.

The multiple regression results and interpretation for research question 3, are of limited value.  You describe the regression findings between social media employment and budget as robust.  They absoutely are not.  You confuse statistical significance with practical significance.  You have a multiple r-square of 1%, which by any conventional definition is trivial.  With large sample sizes even trivial relationships between variables willend up as statistically significant. 

I recommend doing another analysis where you look at the relationship between budget size of an organization and the breadth of social media that the organization employs.  Your dependent variable would be the number of social media that the organization employs.  Create an index that simply adds up how many of the social media the organization employs.  Your range would be from zero to six-six types of media.  Then run the regression of budget this dependent variable.  Here, you can hypothesize that larger budgets do positively correlate with greater breadth of social media usage.  One of the advantages of this approach is that you can include the Tik Tok frequencies in constructing your index of social media breadth.

In summary, spend more time discussing your descriptive findings, do an additional multiple regression analysis.

The English is fine.  The writing, however, needs to trim the number of adjectives in the paper that inflate the description of the methodology and the results.  We are repeatedly told that you investigated the 'intricate ptterns" of relationships, that you were "rigorous" in your methods, and that the multiple regression analyses were "robust". 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

You provided a thoughtful and thorough response to my suggestions.  Thanks.