Prevalence and Impact of Hate Speech among Politicians in Switzerland

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I have been asked to revise the article “Prevalence and Impact of Hate speech among politicians in Switzerland. As the same author(s) are stressing, the article makes a contribution to the field by focusing on a relatively unstudied phenomenon, the one of hate speech targeting politicians. I personally think it is crucial to learn more about this argument so I would recommend the article for publication even though some revisions are to be taken into consideration.

Comments are presented below in order of appearance. I will try as much as possible to keep track of the comments using page and row reference.

Abstract: it is rather not clear the logical connection in this sentence: “hate speech poses a threat because it can impede freedom of expression”. Once I have read the paper I got what the author is referring to. However, it should be made more clear in this sentence. Indeed, it looks like it is hate speech the first cause of lack of freedom of expression. However, once I move forward in the paper, I understand that hate speech targeting politicians might be a driver of lack of freedom of expression. However, as far as I got, in the end are the same politicians that voluntarily decide not to express their opinion because they are targeted by hate speech. I know it might look as a nuance but the first time I have red the first sentence of the abstract I was a bit puzzled by the above-mentioned sentence.

Line 28: “it can impede freedom of expression”-> please consider the comment that I have written for the abstract.

Line 83- to 85: what do the author(s) mean by “the prevalence of hate speech remains largely unknown”? I do not think this sentence adds anything and could be easily removed

Page 3 (beginning): A minor point. I think that there might be some trouble in the pagination of the table. For now I only see half blank sheet.

Table at page 4: while the questions are clearly formulated and connected to the theoretical background of the study, the notes to the table are rather not clear. I am adding the comments here below:

-          Note 1 says that the overall experience is used to calculate the prevalence, but prevalence of what? Why do the author(s) use prevalence as a word. Prevalence of what?

-          Note 4: please consider using the lable “centrist” instead of middle in general when referring to political parties’ ideology.

Line 229 – 232: when you comment upon the fact that politicians are particular target of hate speech, besides the fact that this statement makes sense to me, it is still not completely proved by the study. A comparative study including various categories of targets of hate-speech (e.g.: politicians, civil society association exponents, other public figures) could shed some more light on “how much politicians are special targets of hate speech when compared to other personalities”. Having said that I would make the sentence a bit less strong in its message. Furthermore, you can stress in the paper this possibility as an avenue for further research in the field.

Line 249: please consider substituting the expression “parties that fall in the middle of the political spectrum with “centrist parties”.

Line 259: If it is true that sometimes odds ratio are difficult to be interpreted, I just would love to stress the fact that some differences between left and right wing politicians are to be detected . I was moreover just wondering if it could make sense to have a more fine grained left right scale. Indeed, future research in the field could add some cases/countries and could need a more fine grained left-right scale (e.g.: moderate left and moderate right is different from extreme left and extreme right). Indeed, it would be interesting to understand also the presence of differences within parties’ categories (e.g.: extreme right vs. moderate right/conservative parties, or extreme left vs moderate left).

Section 4.1 from line 272 onwards: please consider putting this really short “limitation section” together with the methods’ one. If not explicitly required by the journal, this section in that position does not add anything.

Section 4.2: even in this case, if it is not required a priori, please consider to merge this extremely small conclusion together with the discussion (just calling the section “Discussion and conclusion”

I have two further points that might be of some help for the investigator(s)

Firstly, the study does not point to any sort of follow up in the research. Having such a unique dataset and not further using it, it is really a pity. Besides this consideration, what is really puzzling is that the author(s) state how much important it is to study such argument since no proper systematic study is present. However, when it  should be the time to propose a future research agenda, nothing is actually presented. Some further considerations in this sense could really improve the quality of the study.

Secondly, the title talk about “impact” of hate speech however, I do not see any reference to the effective impact of hate speech in the study. Indeed, if it is true that the scholar(s) asks some questions about the effect of hate speech on politicians, what is also true is that we cannot consider it as an impact at least not in the long run. I would rather advice the author(s) either to milder the “impact” word in the title or to explain better how we should, as reader, understand “impact” in the text.

Comments on the Quality of English Language

I find the overall quality of the used English good. Some minor remarks have been reported in the previous comment box. No other issues are to be stressed. 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The article examines hate speech directed at politicians. The topic is timely and important, as hate speech targeted at politicians aims at silencing certain voices and thus eroding democracy. Although the article only examines the hate speech directed at politicians in a limited area, one Swiss canton (Zurich), the results can be generalized to some extent. The results also stress the need for measures to curb hate speech. The article is convincing even in its certain narrowness: the materials are sufficient and they are treated appropriately.

 

Hate speech is difficult to define, and the concept is constantly debated both academically and in public and political discussion. A very simple definition has been chosen in this article, presumably a choice dictated by practical conditions. The definition reads: "Hate speech is a harmful expression targeting a personal characteristic, such as skin color, language, nationality, religion, gender [...]" It emphasizes harassment directed at an individual, but it does not take into account, for example, hate speech directed at certain minorities or other groups of people. This may have affected the results for example by not allowing for visual expressions of hatred. The order of the list of traits that can be targeted with a "harmful expression" may also have some impact.

 

The manuscript is clear and well-written. The topic is relevant for research on politics and on hate speech, and the research is presented in a well-structured manner.

The cited references are recent and relevant for the study of hate speech.

The research design seems appropriate to study the prevalence and impact of hate speech (of the participants of the survey – which not necessarily a representative sample of Swiss or even Zurich politicians). The research in all seems well thought out and the results as such seem valid. However, about the respondents: there were significantly more male respondents (almost 60%). Is there such a big difference in the number of male and female politicians in Switzerland? And can it then be said based on the underrepresentation of women among the respondents that hate speech was not gendered? Also, the majority of respondents were over 50 years old. It is known from other studies that women receive significantly more hate speech than men. Similarly, it is known that young people encounter significantly more hate speech than middle-aged people (even more than 80% of young people encounter hate speech in their everyday life).

It is also noteworthy that the respondents were asked about personal experience – of themselves been targets of hate speech – not whether one has witnessed hate speech. This might have influenced the results (based on earlier research).

 

So, is it possible that the age and gender of the respondents affect the low experience of hate speech? Should this possibility be taken into account when analyzing the results? Although "neither gender nor age showed a significant effect" (p. 6) in the case of this group dominated by middle-aged men, it is important to consider whether it means that gender and age really do not matter when encountering hate speech.

 

As I am not a statistician, I do not judge the credibility of the statistical analysis. The tables are appropriate, and they describe the data properly. They are clear and easy to understand. The data is interpreted appropriately throughout the manuscript, but more emphasis should be given to the underrepresentation of young people and women among the respondents. Another issue is, do we know about the racial/ethnic background of the respondents? This might be relevant as we know, based on a variety of studies in the field of hate speech research, that it its women, young people and minorities that receive most of hate speech – and young racialized women and/or members of sexual minorities are many times more likely to become victims of hate speech

 

The conclusions should consider the above mentioned issues, when it comes to gender and age. This should also be taken under discussion in the Limitations section of the manuscript. Otherwise the arguments are sound.

  The authors state in the introduction that they are not aware of similar studies in which hate speech directed at politicians has been studied elsewhere than on social media. I would like to bring to the attention of the writers that there has been a study conducted in Finland, unfortunately only published in Finnish (Knuutila et al., 2019, see: https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/161812/57_19_Viha%20vallassa_Vihapuheen%20vaikutukset%20_Netti%20.pdf?sequence=8&isAllowed=y, abstract in English) in which hate speech against politicians was investigated using multiple methods and at different sites (online and offline).

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf