Introduction

All the sources of information used in this annotated bibliography are published within the last five years. They are reliable resources which have been peer-reviewed. I found these articles to be quite interesting and useful since these articles directly relate to my research area on the use of big data in political campaigns. These resources were found on online databases such as ACM Digital Library, ProQuest and using Primo. EndNote x9.2 was used to record the references and they have been included in APA7 referencing style.

Annotated Bibliography

Article 1

Bethu, S., Madhavi, K., Rupa, B., Hanuman, A. S., Soujanya, R., & Babu, B. S. (2019). A Survey Report on Data Analytics as a Tool in PoliticalCampaign Proceedings of the 2019 8th International Conference on Educational and Information Technology, Cambridge, United Kingdom. https://doi.org/10.1145/3318396.3318452

The article discusses about political marketing campaigns, especially related to the 2012 US Presidential election which had a significant influence through the use of big data. The use of internet and sending of customised messages enabled microtargeting. This article is relevant to the research since it explores how big data was utilised to carry out the political campaigns especially in the US and the Indian elections. A strength of this article is that it is adequately technical as it explains the models and algorithms used. However, it does not evaluate the experiment results to a sufficient level.

Article 2

González, F., Yu, Y., Figueroa, A., López, C., & Aragon, C. (2019). Global Reactions to the Cambridge Analytica Scandal: A Cross-Language Social Media Study Companion Proceedings of The 2019 World Wide Web Conference, San Francisco, USA. https://doi.org/10.1145/3308560.3316456

The article analyses cross language Tweets in English and Spanish related to the Cambridge Analytica scandal to determine the extent of data privacy concerns and issues that have arisen. Data was collected through a Tweeter API and appropriate cleansing was done to improve the data quality. One of the areas my research covers is related to data privacy concerns related to big data and this article examines this area thoroughly. The collected data on Tweets have been thoroughly analysed giving interesting findings and knowledge discovery which are applicable to contemporary political campaigns. A limitation of this article is that its limited to English and Spanish Tweets and data was collected in a general manner covering only certain areas.

Article 3

Grimaldi, D., Cely, J. D., & Arboleda, H. (2020). Inferring the votes in a new political landscape: the case of the 2019 Spanish Presidential elections. Journal of big data, 7(1), 1-19. https://doi.org/10.1186/s40537-020-00334-5

The article examines how Tweeter data was used to predict the 2019 Spanish presidential elections. Data was collected through the Tweeter streaming API and pre-processed before classifying them through the use of Machine learning. Then predictions were made on unlabelled tweets following a similar process to classify them. This is related to my research since it discusses about social media data gathering and analysis to predict presidential election campaign outcomes. This research has adopted a proper data collection and mining process, machine learning techniques and comprehensive analysis. A major limitation is that the paper has not taken into consideration the importance of analysing demographic data and understanding its relevant in predicting results.

Article 4

Konitzer, T., Rothschild, D., Hill, S., & Wilbur, K. C. (2018). Using Big Data and Algorithms to Determine the Effect of Geographically Targeted Advertising on Vote Intention: Evidence From the 2012 U.S. Presidential Election. Political communication, 36(1), 1-16. https://doi.org/10.1080/10584609.2018.1467985

The article looks at the effect of spending on political advertising on different media like Tweeter, Television, etc. Data was collected from multiple methods including proprietary sources and surveys covering a wider demographic area. This paper reveals interesting insights from the 2012 US presidential election campaigns which are relevant to my research. One of the areas include the real gain for a political party by spending more on a form of advertising and the extent to which they are effective. Unlike in couple of the other articles included in the bibliography, this paper looks at various forms of platforms and their relationships to effective political campaigns. However, this article has not explained the technical aspects of the algorithms used in a detailed manner.

Article 5

Laaksonen, S.-M., Nelimarkka, M., Tuokko, M., Marttila, M., Kekkonen, A., & Villi, M. (2017). Working the fields of big data: Using big-data-augmented online ethnography to study candidate-candidate interaction at election time. Journal of information technology & politics, 14(2), 110-131. https://doi.org/10.1080/19331681.2016.1266981

The article looks at how candidates interact with others on social media platforms. It explores the citizen to citizen interaction through posts and comments on platforms such as Facebook and more specifically related to the Finland elections. A mixed method strategy is adopted to gain data which involves online ethnography and data science. The article will help my research since it details out the mechanisms, effectiveness and impact of social interaction on determining success of election campaigns and the final results. The paper has detailed analysis since its based on both qualitative and quantitative research methods. Relevant examples have been quoted which makes it easier for the reader to comprehend. However, the paper has not discussed on data privacy concerns in collecting relevant data.

Article 6

Lepore, J. (2020). Scientists use big data to sway elections and predict riots — welcome to the 1960s. Nature (London), 585(7825), 348-350. https://doi.org/10.1038/d41586-020-02607-8

The article is based on a company called Simulmatics and the adoption of computer simulation in the 1960 US Presidential election campaign. The paper has looked at historical data, especially around the 1960’s and how it has contributed to the modern day data analysis. This is relevant for my research as it helps to understand and explain the history and evolution of using big data techniques in political campaigns. The article reveals rare insights from the past which can be correlated to prediction of contemporary election campaigns. The overall structure of the article can be further improved with more references and a solid conclusion.

Article 7

Mavragani, A., & Tsagarakis, K. P. (2019). Predicting referendum results in the Big Data Era. Journal of big data, 6(1), 1-20. https://doi.org/10.1186/s40537-018-0166-z

This paper looks at how Google Trends have been used in the form of big data analysis to predict EU referendum results. The research methodology involved downloading data from Google Trends and then normalizing them over a selected period of time in order to carry out the required comparisons. This data can be considered to be timely and accurate hence enabling more accurate predictions. This article directly contributes towards my research since Google Trends related to election campaigns is one of the big data mining approaches. This paper contains numerous graphs which assist the reader in easily understanding the results of various analyses. The overall analysis is limited to Good Trends only and does not consider other social platforms such as Facebook.

Article 8

Nickerson, D. W., & Rogers, T. (2014). Political Campaigns and Big Data. The Journal of economic perspectives, 28(2), 51-74. https://doi.org/10.1257/jep.28.2.51

This paper discusses about utilising data driven political campaigns in modern day election campaigns. It also details out the mechanism of predictive scores and how they are used. Data is collected through various means including via voter databases. This data is then analysed to develop predictive scores which relate to the voters’ behaviours and reasons for such attitudes. This article contributes greatly towards my research since its related to data analysis on contemporary political campaigns and the introduction of predictive scores. The paper has explained the predictive models in great detail including the development of predictive scores. However, the article is limited to the US presidential elections only.

Article 9

Pawełczyk, P., & Jakubowski, J. (2017). Political marketing in the times of big data. Przegląd Politologiczny, 33-44. https://doi.org/10.14746/pp.2017.22.3.3

This paper discusses about the development of tools and approaches in political marketing over the past few decades leading up to the 2016 US Presidential elections. Data has been primarily collected through the internet for this research. The advancement of technology, especially through utilising the internet enabled a radical shift in political marketing and in relation to Poland. This paper will assist my research through providing information on the transformation of political marketing campaigns. This paper has covered several key areas however most of the examples indicated in the paper are general and does not have in depth study or analysis performed.

Article 10

Trish, B. (2018). Big Data under Obama and Trump: The Data-Fueled U.S. Presidency. Politics and governance, 6(4), 29-39. https://doi.org/10.17645/pag.v6i4.1565

This article has a section that discusses about microtargeting in political campaigns mainly related to the Obama campaigns. This refers to campaign personnel directly getting in contact with voters through various modes of communication. Technological advancements have contributed towards harnessing the benefits of utilising big data in election campaigns. The article contains numerous examples that I can quote in my research. The paper has also discussed on a wide range of topics in great detail. The paper also discusses on two other areas that involve the use of big data analysis however they are not directly related to the research being studied.

Article 11

Xie, Z., Liu, G., Wu, J., & Tan, Y. (2018). Big data would not lie: prediction of the 2016 Taiwan election via online heterogeneous information. EPJ data science, 7(1), 1-16. https://doi.org/10.1140/epjds/s13688-018-0163-7

This paper discusses about the collection of time-series data from mainstream social media platforms and feeding them into the Kalman filter (a statistical model) to predict the vote shares between the political candidates in the 2016 Taiwan general elections. This article will further support my research since it explores the utilisation of big data techniques in a country like Taiwan. Most of the other papers that were published relate to the US presidential elections. This article has depicted the statistical models and visual representations in a useful manner. However, the limitations of utilising the Kalman model has not been discussed adequately.

Article 12

Younus, A., Qureshi, M. A., Saeed, M., Touheed, N., O’Riordan, C., & Pasi, G. (2014). Election trolling: analyzing sentiment in tweets during pakistan elections 2013 Proceedings of the 23rd International Conference on World Wide Web, Seoul, Korea. https://doi.org/10.1145/2567948.2577352

This paper discusses about predicting 2013 Pakistan election results through the analysis of data taken from Twitter. More precisely, the research examines “election trolling” in Tweeter. The data is collected primarily through Twitter posts and messages between potential voters and candidates. This article contributes to my research as it details out how big data analysis techniques have been used on Tweeter data in in a country like Pakistan. The article is very short and the research is limited only to Twitter. The authors could have discussed more on the evaluation and results. The volume of data collected is relatively lower as a proportion of the entire population.

Self-Evaluation Report on Originality (100-150 words)

The sample of 1000 words tested on Turnitin for originality showed a similarity of zero percentage (refer appendix). I have written each of the annotated bibliographies using my own words without relying on direct quotes from the articles read. Key discussion points and themes have been summarised preserving the original meaning and intended ideas. There was no breach of academic integrity or plagiarism involved in this work. I shall continue to do the same in the final report which involves expanding this literature review. In the annotated bibliography all sources have been acknowledged clearly and included under the reference section. Its important that everyone avoids any form of plagiarism and respects the original work of others.

Annotated Bibliography Reflection

Some of the keywords include microtargeting, voter persuasion, Tweeter Data Mining, voter online behaviour.

These articles have discussed the mechanisms and the effectiveness of using big data techniques in political campaigns. Most of the researches conducted had utilised data collected through mainstream social media platforms such as Facebook and Twitter. Some researchers analysed the behaviour and attitudes of online users through Google Trends. All these approaches have been successful to a greater extent in accurately predicting the outcome of elections and referendums across the world. Most articles relate to the US Presidential elections in 2012 and 2016, however there were papers related to other countries such as India, Pakistan, Taiwan and Finland and also in relation to the EU Referendum. The researchers have explained various statistical models that were used to analyse the data and arrive at the predictions. Machine learning techniques were adopted and some of the common models used were clustering algorithms. There appear to be growing concerns over data privacy and ethical issues.

Each individual research and methods proposed were based on and applicable for a particular country. However, there is potential to find out about new models that can be developed and tested which would apply universally. This would allow in applying the model in different countries by changing few parameters of the model. Also, there are high risks of Governments and other political campaign personnel accessing private data on platforms such as WhatsApp and Viber. Unlike data mining on publicly available data, accessing and analysing private data has various negative implications that needs to be analysed and discussed further.

The next step involves collating the relevant information from each of the articles and then combining them. Then discuss on the gaps identified in the above paragraph with supporting theories and evidence.

References

González, F., Yu, Y., Figueroa, A., López, C., & Aragon, C. (2019). Global Reactions to the Cambridge Analytica Scandal: A Cross-Language Social Media Study Companion Proceedings of The 2019 World Wide Web Conference, San Francisco, USA. https://doi.org/10.1145/3308560.3316456

Grimaldi, D., Cely, J. D., & Arboleda, H. (2020). Inferring the votes in a new political landscape: the case of the 2019 Spanish Presidential elections. Journal of big data, 7(1), 1-19. https://doi.org/10.1186/s40537-020-00334-5

Lepore, J. (2020). Scientists use big data to sway elections and predict riots — welcome to the 1960s. Nature (London), 585(7825), 348-350. https://doi.org/10.1038/d41586-020-02607-8

Mavragani, A., & Tsagarakis, K. P. (2019). Predicting referendum results in the Big Data Era. Journal of big data, 6(1), 1-20. https://doi.org/10.1186/s40537-018-0166-z

Nickerson, D. W., & Rogers, T. (2014). Political Campaigns and Big Data. The Journal of economic perspectives, 28(2), 51-74. https://doi.org/10.1257/jep.28.2.51

Pawełczyk, P., & Jakubowski, J. (2017). Political marketing in the times of big data. Przegląd Politologiczny, 33-44. https://doi.org/10.14746/pp.2017.22.3.3

Trish, B. (2018). Big Data under Obama and Trump: The Data-Fueled U.S. Presidency. Politics and governance, 6(4), 29-39. https://doi.org/10.17645/pag.v6i4.1565