Literature Review
In this particular study a systematic prediction of opinion Poll in predicting the political results are analyzed. The traditional made of physical criteria was never a useful guide for successfully predicting the political results. Due to which the prominent case of understanding and conventional climbing of prediction is becoming a complex problem each day. In this situation the export is forecast in little more even sometimes less than the novels of the night statistical models might show. In case of election for casting scenario the aspects are continuously meeting such conditions of predicting the political results through opinion polls and so on.
In this particular case the research study is aiming on identifying the success rate of predicting political results through opinion Poll based data.
The significant research objectives for this particular research paper are stated below.
- To analyze the success rate of predicting political results from polling data
- To identify the inclusion of machine learning methods in predicting the election results from polling data
- To analyze the possibility of predicting election outcomes through polling data
- To measure the effectiveness of machine learning methods in predicting election results from polling data
The significant research question in this particular research study is stated below.
- What is the possible relevance of machine learning methods in predicting the election results through polling data?
There are particular sub research questions included in the study through which the entire research process will be conducted in the future. The sub research questions are considered below.
- What is the success rate of predicting political results through opinion polls?
- How can machine learning methods be included in predicting political results?
- What are the possibilities that the prediction of election outcomes will be accurate through opinion polling data?
Attaining the appropriate information regarding the relevance of political results and pulling data are largely supported by the public opinions and the opinion polling data. A considerable part of this particular section includes the continuous and rapid change in the political landscape that is succinctly enabling an organization to play the major role in gathering such opinion polling data from the public before any election has taken place.
- H1: Opinion polling data provides accurate political prediction of the election results.
- H2: Polling data is considered to be a significant part before setting the strategies for political aspects.
- H3: Machine learning methods provide successful prediction of the election results through polling data.
- H4: Machine learning methods have a better success rate of opinion polling data prediction for election results then social media polling and survey.
This particular chapter will entirely perform a detailed literature review on The chosen topic where the concert will be entirely based on predicting polling data through opinion polls and the success rate of machine learning method implementation in this prediction process (Smith and Gustafson 2017). The literature review will mainly focus on concept of predicting political results, the role of opinion Poll data in predicting the election results and the role of machine learning methods in the prediction process. It will refer to the knowledge where sharing the significance of theoretical frameworks will help to speak about the different literature gaps seen through the literature review section.
The concept of political forecasting directly focuses on forecasting the political event-based outcomes and results. These political events can be segregated into a number of events which can be either diplomatic decisions or political leader-based actions and other politicians and political institutions generated through the election method (Kennedy, Wojcik and Lazer 2017). In this particular case the area of political forecast entirely concerns a high population of a mass market audience. The political forecasting methodology can be done through frequent use of statistics, mathematics and data science. Poll damping is another method which is used when incorrect indicators of the public opinion are not being used within a forecasting model (Jain and Kumar 2017). In case of campaigns polls are considered to measure voters’ future choices due to which the poll results are closer to the election when the accurate prediction.
The idea of forecasting the election results can either involve skin in the grain crowdsourcing via the prediction markets on the basis of the theory where people are more honest in evaluating and expressing their true perception with money at stake (Valentino, King and Hill 2017). The prediction markets continuously show accurate forecasts of the election outcomes in an example during presidential elections from 1988 to 2004 of the United States 964 election polls were compared with five presidential elections of the United States. This particular data shows that electronic markets topped the poles by 74% of that time. The comparison of damped polls to the forecast done by Lowa electronic markets show that damped polls out performed all other more available models within the market segment.
The Role of Opinion Polls in Predicting Election Results
Opinion polls are simply considered to be a survey or poll where human research service and people or public opinion is the particular sample. Opinion polls are considered to be usually designed in order to represent the opinions of a public or the entire population through conducting a series of interview questions and exploring their general ideas within the confidence intervals (Wang and Gan 2017). The person who is conducting poles is referred to as the pollster. Most of the political scientists and economists have developed their own quantitative model of forecasting the election results. Most of these models are entirely based on retrospective voting purposes which are continuously assuming that the voters punish and reward incumbent party based on their performances (Jungherr et al. 2017).
For the current era fundamentals of elections are entirely based on forecasting and calculating the polling average which can be combined in order to compare the opinion polls and the exit polls to identify the accuracy of the opinion polling data (Beauchamp 2017). The combination of forecasting polling data exports and fundamentals are calculated through the weighted average individual expert forecast, the pulling average of that day and the fundamentals-based forecasting system (Leemann and Wasserfallen 2017). The opinion polls play a great role in impacting the strategies of different political parties because it helps them to recreate a better identification of significant voting strategies on the matters of public policies.
There have been previous outcomes where it was witnessed that the traditional opinion polling system in predicting the presidential election outcomes have failed to provide accuracy. These failures can be the outlining of different reasons where monitoring is one of the greatest factors along with the idea of inaccurate data collection methods and changes of populations’ minds (Jennings and Wlezien 2018).
The fascination of predicting the future is becoming one of the most intent and desired concept individuals and insisting on getting. The individuals and the different companies are continuously providing their effort innovating new ways to credit the election results beforehand. There are several works such as data mining and machine learning methods through which the prediction can be identified with the help of social media as well as opinion polling data (Warren 2018). In Spain during 2012 delta conducted a prediction political tendency through Twitter that revealed almost 140 characters about the political sentiments that did not match with the mining reviews of the tweets due to which the artificial intelligence was considered in making the computational procedures of automatic programming systems so that the sophistication could be increased (Oliveira, Bermejo and dos Santos 2017). The classification of the opinion polling data was entirely processed when an unsupervised machine learning process due to which the clustering of the data took place and the data mining also failed to provide appropriate data regarding election results.
The machine learning model helps in predicting the different organizations about the accurate election outcomes where question based historical data will be analyzed. Machine Learning method identifies various activities the customer may consider with other aspects through which the tangible business values are considered (Lauderdale et al. 2020). Data report AI cloud platform also allows users to develop their own models through which highly accurate predictions can be done. It also streamlines the data science process entirely through which the users get high quality predictions in a matter of time without taking the traditional methods and waiting for the opinion Poll or survey results. This allows the people to be more quickly involved with the prediction process and it only impacts the bottom line of the public through its innovative advanced technology-based criteria (Atanasov et al. 2017).
Machine Learning Methods in the Prediction Process
The Research design provides the idea of a standard Framework which is chosen for a specific research study that helps in answering all the research questions and Research objectives. The entire aim and goal of the Research design is to depict the research problem effectively. The Research design layer of the Sanders research onion is divided into 3 significant categories which are explanatory Research design, exploratory Research design, and conclusive Research design.
For this particular research study the exploratory Research design has been chosen as a researcher is willing and aiming at investigating the entire concept of predicting election outcomes through opinion polls and the relevance of using machine learning methods in predicting such results (Karami, Bennett and He 2018). The exploratory Research design will help in a means to explore the available research problem so that better understanding of the nature and issue of the research can be identified properly without only providing conclusions with the broad concept of using the research method to identify search issues that can be the future research focuses.
The data collection process is a systematic approach in which way the researchers will collect data to accomplish their research process. In this particular research study and primary data collection method will be considered. The primary data will be based on collecting the opinions of people regarding effectiveness and roles of opinion polls in predicting election results and the additional concept of machine learning method in predicting the election outcomes (Hanias and Magafas 2011). The primary data will be collected through interviews and service will be entirely based on period review journals, research papers, books, websites and online documents where referral sources will be considered. The academic papers will be collected through credible databases which will provide previously research evidence on the particular topic.
The data analysis techniques are based on two primary methods such as qualitative and quantitative data analysis techniques.
In this particular research study, the researcher has chosen a mixed analysis method through which the quantitative data analysis will be focused on providing descriptive and statistical analysis on the basis of numerical data collected from the surveys and interviews. Different graphical representations will be provided based on the opinions of the sample participants (Warren 2018). On the other hand, qualitative data analysis will be conducted through code and theme based thematic approach. In this data analysis process, different information and data collected through journals and articles will be used to create a pattern which will help in creating thematic approach to cover the research topic, questions and objectives.
The ethical considerations are based on implementing moral solutions for the data collection purposes. In this case, the idea of considering the data collection process is the primary focus as the researcher will involve different living human beings who will directly provide their personal information and the researcher is obliged to maintain the confidentiality of the data in case of collecting data for the primary analysis from the participants. In case of secondary data and qualitative data analysis technique the researchers should properly cite the work as per reference sections (Beauchamp 2017).
Figure 1: Timeline of the project
(Source: Created by author)
References
Atanasov, P., Rescober, P., Stone, E., Swift, S.A., Servan-Schreiber, E., Tetlock, P., Ungar, L. and Mellers, B., 2017. Distilling the wisdom of crowds: Prediction markets vs. prediction polls. Management science, 63(3), pp.691-706.
Beauchamp, N., 2017. Predicting and interpolating state?level polls using Twitter textual data. American Journal of Political Science, 61(2), pp.490-503.
Hanias, M.P. and Magafas, L., 2011. Chaos theory in predicting election results. Journal of Engineering Science and Technology Review, 4(3), pp.286-290.
Jain, V.K. and Kumar, S., 2017. Towards prediction of election outcomes using social media. International Journal of Intelligent Systems and Applications, 9(12), p.20.
Jennings, W. and Wlezien, C., 2018. Election polling errors across time and space. Nature Human Behaviour, 2(4), pp.276-283.
Jungherr, A., Schoen, H., Posegga, O. and Jürgens, P., 2017. Digital trace data in the study of public opinion: An indicator of attention toward politics rather than political support. Social Science Computer Review, 35(3), pp.336-356.
Karami, A., Bennett, L.S. and He, X., 2018. Mining public opinion about economic issues: Twitter and the us presidential election. International Journal of Strategic Decision Sciences (IJSDS), 9(1), pp.18-28.
Kennedy, R., Wojcik, S. and Lazer, D., 2017. Improving election prediction internationally. Science, 355(6324), pp.515-520.
Lauderdale, B.E., Bailey, D., Blumenau, J. and Rivers, D., 2020. Model-based pre-election polling for national and sub-national outcomes in the US and UK. International Journal of Forecasting, 36(2), pp.399-413.
Leemann, L. and Wasserfallen, F., 2017. Extending the use and prediction precision of subnational public opinion estimation. American journal of political science, 61(4), pp.1003-1022.
Oliveira, D.J.S., Bermejo, P.H.D.S. and dos Santos, P.A., 2017. Can social media reveal the preferences of voters? A comparison between sentiment analysis and traditional opinion polls. Journal of Information Technology & Politics, 14(1), pp.34-45.
Smith, B.K. and Gustafson, A., 2017. Using wikipedia to predict election outcomes: online behavior as a predictor of voting. Public Opinion Quarterly, 81(3), pp.714-735.
Valentino, N.A., King, J.L. and Hill, W.W., 2017. Polling and prediction in the 2016 presidential election. Computer, 50(5), pp.110-115.
Wang, L. and Gan, J.Q., 2017, September. Prediction of the 2017 French election based on Twitter data analysis. In 2017 9th Computer Science and Electronic Engineering (CEEC) (pp. 89-93). IEEE.
Warren, K.F., 2018. In defense of public opinion polling. Routledge.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
