Predicting the Results of the 2019 Indonesian Presidential Election with Google Trends

Analysis of Accuracy, Precision, and Its Opportunity

DOI

10.2991/assehr.k.210531.001How to use a DOI?

Keywords

Google Trends, Political Choice, Accuracy, precision, 2019 Presidential Election

Abstract

In electoral competitions, voter behavior research is conducted to develop effective campaign strategies, but survey methods tend to be expensive. By contrast, in the internet era where political campaigns evolve by using digital data and channels, Google Trends offers free services that display search interest indexes for certain keywords and topics measured through search volumes on Google search engines over a period of time and certain areas. Thus, Google Trends could be a tool that increases the efficiency of voter behavior research. By using the election real count data released by the General Election Commission on May 21, 2019, this study conducted an accuracy and precision analysis of the Google Trends topic query index that represented presidential candidate number 01 and number 02 in predicting the real count of the two candidates’ votes. The analysis indicated that using Google Trends as a tool to predict the results of the 2019 Indonesian Presidential Election produced imprecise results. However, Google Trends could better predict election outcomes by adding features for sentiment analysis and a more representative understanding of users’ search activity.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Ali Ar Harkan
AU  - Eriyanto
PY  - 2021
DA  - 2021/05/31
TI  - Predicting the Results of the 2019 Indonesian Presidential Election with Google Trends
BT  - Proceedings of the Asia-Pacific Research in Social Sciences and Humanities Universitas Indonesia Conference (APRISH 2019)
PB  - Atlantis Press
SP  - 1
EP  - 9
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.210531.001
DO  - 10.2991/assehr.k.210531.001
ID  - Harkan2021
ER  -