Recognizing Emotions in the Discourse of Women in Power Positions in Social Networks via Sentiment and Content Analysis

Abstract

Understanding people’s emotions is one of the biggest challenges today. From the last decades, the study of emotions has been one of the most emerging research areas from social science to computer science. Indeed, emotions are recognized as a vital element in the relationships between people and computers. Human computer interaction (HCI) is the discipline that deals with the design, evaluation, and implementation of the interaction between people and computing systems. For an adequate interaction between the computer and people, the computer must be able to have some perception of the emotional state of the human being with whom it interacts. The computer be only able to obtain the perception of people emotions through the lexical expressions communicated in his speech. Emotions are present in most forms of communication and can change the meaning of the message transmitted. In this work, we take different Instagram accounts from well-known mass media (CNN, Deutsche Welle among others) to analyze the emotions contained in the representations of women in positions of power. We combine sentiment and content analysis to help you understand sentiment in briefings, including irony and mixed language. Interesting results were obtained to understand the feelings involved in information about women in positions of power.

1 Introduction

Women’s equal participation and leadership in politics and public life are critical to achieving the Sustainable Development Goals by 2030. Although the gender gap in the workplace continues to narrow on a daily basis, women who manage to reach top executive positions must face not only the challenges of being a leader in a company, but sometimes also the challenge of managing men who are not accustomed to working with women in positions of power [1]. However, data show that women are underrepresented at all levels of decision making around the world. Therefore, gender parity in politics is still far from being achieved.

In this article we will focus on all those women who have managed to overcome the obstacles and have been able to reach positions of power and are made visible by the media. The representation of women in the mass media has been a topic of discussion and debate for a long time. We cannot deny that there have been advances in the inclusion of women in the media, however, there are challenges and barriers to be overcome.

One of the main concerns is the way in which the female gender is represented in the media. Information about this gender is stereotyped and limited, focusing on their physical appearance rather than their abilities and achievements. This situation is of great concern, as it can perpetuate gender stereotypes and limit women’s aspirations. Another concern is the lack of representation of women in leadership positions in the media.

For this purpose, news about women in positions of power published on the Instagram accounts of CNN, Deutsche Welle and El Mundo in January and February 2023 will be studied. Based on Content and Sentiment Analysis, this research analyzes the media’s representation of women in positions of power. In addition, we will be able to analyze whether in representing the female gender it is done through positive, neutral or negative emotions.

Content analysis is a research technique used to analyze and study the content of texts, images, audio and video. This research technique performs the identification and categorization of significant elements in the study material. In the case study, (texts and images from social networks) themes, keywords, language patterns, tone, style and other important elements can be identified [2].

This paper is organized as follows: Sect. 2 presents the problem, results, and discussion. Conclusions and some lines of future directions are given at the end. Please note that the first paragraph of a section or subsection is not indented. The first paragraphs that follows a table, figure, equation etc. does not have an indent, either.

2 Discussion and Results

In this research they analyzed all the posts between January and February of CNN Chile, DW_es and El Mundo. The first thing that stands out is the scarce presence of women in positions of power in the media. In CNN Chile, the presence of the female gender in the case studied was only 3.8%, while in DW_es it was 2.14%. The presence increases slightly in El Mundo, where it was 6.82%.

We could observe that women working in the political area had a 52% presence, while those working in the artistic area had a 43% presence. Only 5% of them did so from a sports sphere.

It is noteworthy that there were no women in managerial positions in private companies.

On the DW_en site, the situation of the presence of women in positions of power is quite similar. From the political sphere, 44% were represented, 44% from the artistic sphere and 11% in managerial positions in companies. However, we should point out that the 11% above referred to the fact that the CEO of Youtube was leaving her position. In the case of El Mundo, the situation is practically the same, except that the presence of women in positions of power within the artistic sphere was notoriously higher with 70.37% and in the political sphere only 29.63%.

In short, we could observe that the analyzed media only publish news about women in positions of power if they belong to the artistic or political sphere. Sports or company directors, it seems, are still relegated to the male gender for the analyzed media.

We also observed that in some of the information published, reference was made to women’s clothing. This situation occurred in the political sphere. In the artistic sphere, women were repeatedly mentioned due to the couple conflicts they were experiencing.

It should be noted that there was no diversity in the representation of women in the media. Women in positions of power from different ethnic, cultural, religious and socioeconomic backgrounds were marginalized and excluded from the media.

After applying the Content Analysis, a sentiment analysis was carried out on the headlines of the analyzed news (using SentiStrength). Positive sentiment strength ranges from 1 (not positive) to 5 (extremely positive) and negative sentiment strength from −1 (not negative) to −5 (extremely negative). Through sentiment analysis, we can recognize if a text is positive, negative or neutral (Table 1).

Table 1. Polarity statistics in the news of women in positions of power

If we apply emotional computing to the titles of the information on Instagram, we can observe that 55% of them were rated as negative. 20% were positive captions and the remaining 25% were neutral.

3 Conclusions

With the application of content analysis, we were able to understand and interpret the different texts of the analyzed Instagram accounts. We were able to identify a trend: the spheres where the analyzed media report on women in a position of power were mainly artistic and political. The sentiment analysis showed that 55% of the headlines had a negative connotation.

It is important that the mass media work on the inclusion of women in all areas, from politics, economics, and sports. It is also crucial to pay attention to how women are portrayed and to promote fair and realistic representation.