1 : Journey of a Sexist Tweet

Sexism

Sexism is prejudice or discrimination based on one's sex or gender. Sexism can affect anyone, but it primarily affects women and girls. It has been linked to stereotypes and gender roles, and may include the belief that one sex or gender is intrinsically superior to another. Extreme sexism may foster sexual harassment, rape, and other forms of sexual violence.

Types of Sexism

  • IDEOLOGICAL AND INEQUALITY : The text discredits the feminist movement, rejects inequality between men and women, or presents men as victims of gender-based oppression.
  • STEREOTYPING AND DOMINANCE : The text expresses false ideas about women that suggest they are more suitable to fulfill certain roles (mother, wife, family caregiver, faithful, tender, loving, submissive, etc.), or inappropriate for certain tasks (driving, hardwork, etc), or claims that men are somehow superior to women.
  • OBJECTIFICATION : The text presents women as objects apart from their dignity and personal aspects, or assumes or describes certain physical qualities that women must have in order to fulfill traditional gender roles (compliance with beauty standards, hypersexualization of female attributes, women’s bodies at the disposal of men, etc.).
  • SEXUAL VIOLENCE: Sexual suggestions, requests for sexual favors or harassment of a sexual nature (rape or sexual assault) are made.
  • MISOGYNY AND NON-SEXUAL VIOLENCE: The text expressses hatred and violence towards women.

PROBLEM STATEMENT

MOTIVATION

  • Sexism is a social issue that is widespread on different social media platforms such as twitter

  • Here, we aim to understand how sexism is responded by people on social media platforms by using a mixture of social media scraping and behavioral experiments (interviews/surveys)

Hypothesis

  • Sexist tweets garner more attention (in terms of number of likes and retweets) compared to non-sexist tweets.
  • Replies to sexist tweets are significantly more sexist than replies to non-sexist tweets.

RESEARCH QUESTIONS

  • Observing how much attention does a sexism tweet garner on social media platforms, how people react to it in general.
  • What kind of replies do people give to a sexist tweet, what proportion of user support and what proportion stand up against it.
    • What are the general public’s opinions regarding the people who post such sexist content on social media?
  • Understanding difference in reactions and opinion of people regarding their thoughts on replies to such sexist tweet in two cases: On social media When asked to express thoughts in a anonymous survey
  • Understanding difference in opinion of people on basis on age and gender

Public Opinion



We conducted survey in campus for approx 50 people. They were shown a sexist tweet and their reactions were studied.


EXPERIMENTATIONS

Data Collection

We have scraped Twitter (100000 tweets) for sexist tweets by using the keywords such as ‘sexism’ etc.
For each tweet, we got all the replies to the tweet. Each tweet has its public metrics (retweet_count, reply_count, like_count, quote_count), author, created_at, and conversation_id )



Methodology

Search for tweets based on particular tags. Employ a sexism detection model to find if a given tweet's text is sexist.
Analyze the text of the replies to a tweet, and number of replies to the tweet, based on the level of sexism
Preprocess the tweets and apply the model on each tweet and reply to get a sexism score/level from 0 to 1.
Perform analysis on the tweets and replies by splitting the dataset based on the sexism score

GENDER



MAXIMUM DEGREE LEVEL

             

CURRENT OCCUPATION

                                             

RESULTS

  • Group 0 represents non sexist tweets and group 1 represents sexist tweets. Graph shows that the number of likes on sexist tweets were less than that of non-sexist tweets which is obvious. However, on performing the permutation testing on this data, we found out that there was not much significant difference between the two groups.

  • Above graph shows the sexism level along with the likes obtained by tweets under the respective category. This sexism level was predicted by our AI model. The pearson correlation is -0.66 which is moderately good but the spearman correlation value is -0.26 which is quite low.

   

  • Here we observe the number of times a particular tweet was retweeted. This surprisingly is almost similar for both sexist and non-sexist tweets.
  • For the sexism level vs like obtained graph, the pearson and spearman correlation values are -0.4 and -0.26 which are both quite low.

 

  • In this case, we analyze how many users reply to both the categories of tweets. We observe that the sexist tweets have somewhat more replies than the non-sexist tweets.
  • For the sexism level vs like obtained graph, the pearson and spearman correlation values are 0.68 and 0.64 which are both moderately good in this scenario.

  • The average sexist replies of the sexist tweets are significantly high than the non-sexist replies.
  • For the sexism level vs like obtained graph, the pearson and spearman correlation values are 0.93 and 0.98 which are both very good in this scenario.

DISCUSSIONS

We conducted a survey among 50 students from IIITH campus. Following are our observations -

  • We asked them to fill a form which consisted of a sexist tweet and asked their opinion about the sexist level of tweet on the scale of 1 to 5 .
  • They were also asked to fill an appropriate reply to that tweet.
  • They also must categorize each tweet in one of the following categories:
    • Derogatory
    • Non-derogatory
    • Sexist
    • Non-sexist
    • Counter-sexist.
  • The survey results were transcribed and summarized for qualitative analysis.



ANALYSIS

PROPOSALS TO DEAL WITH IT

The good thing was most people did not find the tweet appropriate and were offended by the same but their thoughts were on the extreme end. Most people demanded banning them from the media. After giving them options, some of them reach the middle ground. They agreed to shadow ban them from social media. There were comments like -
Sorry to burst your bubble but these duties are not just for women but also for men too. Morning! PS: Its 21st century and not the eighties.. Well you shouldn't even have said morning men cause they would have been sleeping.
There were also some neutral comments It should be treated as a joke This isn’t the way of describing a gender Household responsibilities are for both men and women equally.

66% people think of this as a sexist tweet and 4% find it neutral. Almost 80% of the people were students and out of which 50% post-graduated.

CONCLUSIONS

  • We observe the number of times a particular tweet was retweeted was surprisingly similar for both sexist and non-sexist tweets.

  • We observe that the sexist tweets have somewhat more replies than the non-sexist tweets.

FUTURE WORK

  • We can extend our survey from interviewing 50 people to some thousands of people whether offline or with some forms.
  • We are just running our models based on some sexist keywords, but there might be some tweets which don't have those keywords and still can be highly sexist. We can include those in any further iterations.
  • We can also include image and video tweets in these experiments which will make the experiments more inclusive.

Comments

Post a Comment

Popular posts from this blog

23: Understanding the discourses around CAA NRC

Image
About CAA - NRC The Citizenship (Amendment) Act, 2019 aims to facilitate grant of citizenship to migrants belonging to Hindu, Sikh, Buddhist, Jain, Parsi or Christian communities from Pakistan, Bangladesh and Afghanistan who entered into India on or before 31st December, 2014 (1). It was enacted by the Government of India on 12 December, 2019. This particular move by the government of India sparked nationwide protests against CAA and the National Register of Citizens (NRC). The NRC is the official record of all the legal citizens of India. The main reason for the protests was the exclusion of Muslims and other communities who fled from the same or neighboring countries along with refugees from Sri Lankan Tamils in India, Rohingyas from Myanmar, and Tibetan refugees (4).  Due to the widespread national and overseas unrest from this act, it became a very important event in Indian history. Therefore, this led to an extensive media coverage in India. Furthermore, detailed accounts of t...
Read more

Archaeological Data Analysis on Harappan Civilization

Image
  Introduction Deriving conclusions from historical datasets is one of the hardest problems historians have to solve. The problems include verifying information, deriving correlations, and defending conclusions. The complexity of the problem results in increased time for a solution.  Through this project, we have proposed an accurate and faster way of concluding new inferences from historical data. This methodology primarily has 4 parts 1. Collecting and preprocessing the dataset 2. Analysing data using modern machine learning techniques. 3. Validate results using existing Information 4. Derive conclusions To showcase the working of the proposed methodology we have designed an experiment, in which we have derived cultural and economical relationships between ancient cities of the Indus valley civilization. Project Description Experimentation Data Collection (https://github.com/sidgupta2205/CSS_Project/tree/descriptors/Data) In the first phase, we worked on research and knowled...
Read more

14 : Misinformation Spread in Social Networks

Image
  Misinformation Spread in Social Networks Problem Statement In the Information Age, social networking sites have become a notable agent for the spread of misinformation, fake news, and propaganda. Misinformation on social media spreads quickly in comparison to traditional media because of the lack of regulation and examination required before posting. These sites provide users with the capability to spread information quickly to other users without requiring the permission of a gatekeeper such as an editor who might otherwise require confirmation of the truth before allowing publication. A 2018 study of Twitter determined that, compared to accurate information, false information spread significantly faster, deeper, and more broadly.  In this study, we aim to better understand misinformation flows in social networks. Related Work The spread of misinformation has been an ongoing research topic for some time now but before understanding the spread of misinformation, we need a cl...
Read more