Essay on the Impact of the Twitter Use on US Presidential Election Results

Social media plays a critical role in studying the social life of people. Most countries have used different sites to analyze and predict presidential elections. In the United States, social media has been instrumental in influencing people’s behavior to vote, and many candidates tend to use it to present their manifestos to the public. Some of the common social media platforms that people use to campaign, launch their complaints, and present their views and opinions include Facebook, WhatsApp, and Twitter. Suciu (2020) indicates that by 2020, the number of registered voters in the US using social media comprised 72 percent of the total number of voters. Out of the 72 percent, 69 percent of them used Facebook for their social interaction. In the US alone, there exist more than 233 million social media users using major platforms. According to Nelson (2020), these users believe that they can influence election outcomes depending on their engagement.

I choose Twitter as the social media platform to explore the scenario. Though it has positive and negative consequences, the role of social media in deciding election outcomes is increasing. In other democratic nations where people have freedom of speech and communication, Twitter users are free to criticize their leaders and other aspiring candidates depending on how they relate with the people. People are also free to share the histories of their leaders in a manner that will convince people to vote in a person. Therefore, candidates for different election positions are also investing in the platform to convince people to vote for them. Considering that most people are now using Twitter, it is important to reach and convince them through these platforms.

I choose the impact of social media on election outcomes following my interest in understanding how its use impacts people’s thoughts and behaviors. I want to know how a social media message will make an individual change their mind and make contrary decisions than their original decisions. Therefore, using election will guide me to understand the issue better considering that it is a national event and its outcome is thus fundamental to the citizens of a given country, alongside their allies.

Data Collection Approach and Reflection

Twitter use is common among young voters within the United States. Young voters are influential and play a crucial role in determining the election outcome of a given nation. Also, most of them are very active in using social media, thus making the platform a suitable avenue for leaders to present their manifesto and convince them to vote. In 2016, the number of voters who were below the age of 30 in the US was estimated to be half the total number who registered in 2020. For this case, Nelson (2020) mentions that experts are convinced that social media played a fundamental role in convincing them to register as voters. The surge in the number can be attributed to the social media advertisements that people receive, constantly reminding them to cast their votes. In an earlier study, Holt et al. (2013) indicated that social media serve as s leveler in the US as it provides an avenue for studying political interests and participation of youths before an election. A reminder to vote is irresistible for social media users as pop-up messages are constantly recorded on these platforms.

I will review different hashtags within the Twitter platforms to check how people interact before an election. Understanding this level of interaction is crucial as it studies how their views and opinions influence an election outcome. The data sample comprises 10 hashtags that prevailed in the US before the 202 general elections. In this case, I will target hashtags that were promoted since the start of the year, up to July, when the US elections were about to start. The keywords for the search will be “US election,” “Donald Trump,” “Joe Biden,” “US general elections 2020.” After identifying these websites, I will use Microsoft Excel software to collect data that I find from the 10 hashtags identified that are related to the US general elections. I will then conduct a sentiment analysis using Microsoft Excel to establish the effect of social media on the election outcome.

In Sentiment analysis, artificial intelligence and machine learning are used to predict whether an answer is positive or negative. In Excel, Azure Machine Learning is a tool that conducts sentiment analysis. The tool uses MPQA Subjectivity Lexicon with a generic dictionary containing 2533 and 5097 positive and negative words respectively (Jelen, 2020).

My data collection has some strengths and weaknesses. The first advantage is that Microsoft Excel is a powerful data analytics tool that would enable me to conduct sentiment analysis and understand user behavior and its impact on the US elections. For example, it will enable me to understand people’s sentiments before the election and check whether it impacted the 2020 United States Election. The second advantage is that the Azure tool in Excel is capable of determining the polarity of each tweet, thus making it suitable to analyze short Twitter posts (Jelen, 2020). One of the weaknesses of using Microsoft Excel to conduct data analysis is that it is complex. To accomplish sentiment analysis, the analysts should be well equipped with advanced Microsoft excel use knowledge. Also, the software is good at handling quantitative analysis, and would therefore become challenging to incorporate people’s sentiments into it before analysis.

Research Method Design and Reflection

The study will apply sentiment analysis as a research design to study people’s behavior. The analysis is a form of qualitative research considering that no quantification of data is required. the researchers will only analyze the sentiments of voters before an election and check if they influenced the US election outcome. Sentiment analysis is crucial for this design considering that it involves the analysis of unstructured data. In a study conducted by Jeevanandam Jotheeswaran (2015) to analyze views and opinions of consumers and how such views influenced the market behavior, the researcher concluded that such sentiments enable producers to design products and services that meet the needs and preferences of consumers. From the perspective of an election, sentiment analysis is also crucial as it establishes the step that people will take during voting.

Therefore, to complete the analysis, the researchers will collect twitter tweets from 10 identified hashtags that were prevailing before the United States 2020 general election. In this case, Twitter posts from individual bloggers who have a large following on the platform will be used. These messages should only be restricted to the ones that address issues affecting the US 2020 elections. Then, the reaction of blogger’s followers on social media will be identified using Excel Azure. The procedure will be repeated for other posts within the 10 hashtags that will be identified by the researcher. For each hashtag, approximately five posts will be identified, making a total of 50 posts to be analyzed in the process. Using the Lexicon property in Excel Azure to check for the polarity of statements (Messias et al., 2017), researchers will then establish whether the views and opinions of the US people had an impact on the general election result. Using Azure software is useful in feeling assessment and opinion mining (Mehta & Pandya, 2020). The outcome from the study will be fundamental as it will enable establish the role of social media in presidential elections.

To ensure that these opinions and assessments meet the requirements of the target population, comprising of young United States voters, the researchers will check profiles of each of the social media users to establish collect this crucial data. The data will be crucial as it will enable the research to operate within its scope, such that the outcome and conclusion will only be used to make conclusions that would only affect the young population.

The research design will inform the research questions as its analysis and presentation of outcome should inform whether social media affects the opinions of others. Gonçalves et al. (2013) indicate that sentiment analysis using the lexicon-based approach is effective as it enables the researcher to study the emotions and moods of an individual. In this case, it will facilitate the study since studying the opinions of the people about political views, it will be easy to study the repercussions of these actions on the people. Therefore, it is crucial in establishing whether a candidate will be supported or cannot be supported within a given election. On the negative side, sentiment analysis may give the wrong impression about a given event. For example, in the 2016 US elections, different social media analyses showed that Hillary Clinton was most likely to win the election (Ramteke et al., 2016). Ramteke et al. (2016) used sentiment analysis, analyzing most social media platforms to predict the presidential results. Despite such an outcome, Donald Trump was declared the United States president during that period. When it provides a misleading conclusion about different phenomena within the market, sentiment analysis will push people to make decisions that would affect their social and financial lives.

Ethical Considerations

Different ethical issues are associated with social media research. Collecting information about the social media users may raise some ethical concerns which could impact both the user and the researcher. Kennedy (2012) indicates that studying people’s norms, behaviors, and sentiments could have some ethical implications considering that most of these people are not aware of these analyses. One of the key ethical issues is the use of the person’s data without their consent. For example, for the research design of the proposed study, researchers will visit different hashtags and check messages posted by different bloggers having a large following on their Twitter accounts. This information will be collected without their knowledge, and this is ethically wrong considering that researchers may have failed to consent to these bloggers. The second issue is that the researchers may collect sensitive data about the participants of these messages and share them across different social media platforms. Such actions are ethically wrong and may have adverse legal, social, and financial implications on the researcher. Kimball and Kim (2013) inform that as people interact through social media, they may be tempted to report information about others, solicit conversation, and present opinions through their social media domains, including Twitter. Researchers may also be tempted to behave in such a manner that they share personal information through their social media domains.

The legal implications associated with such ethical issues is that if the owner of the blog learns about the research, they may file a lawsuit against the researcher. Lawsuits are costly as one is required to appear before the court and defend themselves. For an ethical breach, it is hard for an individual to defend themselves considering the rights and privacy of each person are protected by the court of law. Therefore, if it is concluded that the research breached the rights of the bloggers and other social media users, they may be fined, and this will affect their financial lives. also, Schroeder (2017) mentions that studies that use social media to study the behaviors of others are manipulative and go against the electronic information privacy of a person. For the success of such studies, researchers are required to acquire the consent of each social media user before collecting their engagement and using them to make important decisions.

Addressing ethical challenges associated with conducting social media research requires a vast understanding of these issues and their implications on one’s life. Taylor and Pagliari (2018) indicate that the greatest challenge to a researcher is to understand between the public and private space of an individual’s social media. Therefore, one issue of getting rid of challenges associated with the sentiment analysis is to conduct a comprehensive analysis of one’s social media to check if is a private or public account. Such information will enable the researcher to evade the legal implications associated with the use of one’s privacy. Also, Jouhki et al. (2014) indicate that most social media users fail to seek social media user consent before acquiring their information, and thus sharing their information brings potential privacy implications which affect them. Therefore, researchers can avoid breaches of social media user’s privacy rights by seeking their consent before using part of their data for research purposes. Seeking this consent is crucial as it enables the researcher to conduct their research in an ethically correct approach that would not interfere with them later on.

References

Gonçalves, P., Araújo, M., Benevenuto, F., & Cha, M. (2013, October). Comparing and combining sentiment analysis methods. In Proceedings of the first ACM conference on Online social networks (pp. 27-38).

Holt, K., Shehata, A., Strömbäck, J., & Ljungberg, E. (2013). Age and the effects of news media attention and social media use on political interest and participation: Do social media function as leveller?. European journal of communication, 28(1), 19-34.

Jeevanandam Jotheeswaran, D. S. K. (2015). Sentiment Analysis: A Survey of Current Research and Techniques. International Journal of Innovative Research in Computer sand Communication Engineering, 3(5).

Jelen, B. (November 19, 2020). Excel 2020: Perform Sentiment Analysis in Excel. https://www.mrexcel.com/excel-tips/excel-2020-perform-sentiment-analysis-in-excel/

Jouhki, J., Lauk, E., Penttinen, M., Sormanen, N., & Uskali, T. (2016). Facebook’s emotional contagion experiment as a challenge to research ethics. Media and Communication, 4.

Kennedy, H. (2012). Perspectives on sentiment analysis. Journal of Broadcasting & Electronic Media, 56(4), 435-450.

Kimball, E., & Kim, J. (2013). Virtual boundaries: Ethical considerations for use of social media in social work. Social Work, 58(2), 185-188.

Mehta, P., & Pandya, S. (2020). A review on sentiment analysis methodologies, practices and applications. International Journal of Scientific and Technology Research, 9(2), 601-609.

Messias, J., Diniz, J. P., Soares, E., Ferreira, M., Araújo, M., Bastos, L., … & Benevenuto, F. (2017). An evaluation of sentiment analysis for mobile devices. Social Network Analysis and Mining, 7(1), 20.

Nelson, L. (2020). The Good, the Bad, and the Ugly. Techné: Research in Philosophy and  Technology, 24(1/2), 195-217.

Ramteke, J., Shah, S., Godhia, D., & Shaikh, A. (2016, August). Election result prediction using Twitter sentiment analysis. In 2016 international conference on inventive computation technologies (ICICT) (Vol. 1, pp. 1-5). IEEE.

Schroeder, R. (2014). Big Data and the brave new world of social media research. Big Data & Society, 1(2), 2053951714563194.

Suciu, P. (October 26, 2020). Social Media Could Determine the Outcome of The 2020 Election. Forbes. https://www.forbes.com/sites/petersuciu/2020/10/26/social-media-could-determine-the-outcome-of-the-2020-election/?sh=6a4c1c6a26f6

Taylor, J., & Pagliari, C. (2018). Mining social media data: how are research sponsors and researchers addressing the ethical challenges?. Research Ethics, 14(2), 1-39.