Available via license: CC BY 4.0
Content may be subject to copyright.
105
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
THE RELATIONSHIP BETWEEN GOOGLE SEARCH VOLUME AND FLOOD
EVENTS IN INDONESIA: A GOOGLE TRENDS ANALYSIS
*Resti Kinanthi1, Mochammad Iqbal2
*1Graduate School – Universitas Gadjah Mada, Indonesia
2Departement of Communication - Universitas Pasundan, Indonesia
E-mail: restikinanthi@mail.ugm.ac.id
Submission
Revision
Publication
July, 12th 2023
September, 26th 2023
October, 06th 2023
Abstract
The increasing number of floods every year brings a wide impact; the
average loss caused by floods every year reaches 4.64T (Rp).
Information about floods is very important to improve the
preparedness of people living in disaster-prone areas. If information
can be provided when people have a high interest in learning, it can
have a profound effect on increasing public knowledge. To carry out
effective risk communication in flood disasters, knowledge about flood
disaster patterns is needed, as well as knowledge about community
behavior in seeking information. One of the people's behavior in
searching for information can be seen from the Google Trends
analysis. The purposes of this study are to analyse the relationship
between public interest in finding information through google trends
with the keyword "flood" with flood events in Indonesia for 10 years.
This research is a quantitative descriptive study by processing data
obtained from Google Trends and BNPB flood event data using the
Spearman Rank Correlation statistical test with SPSS software. The
results showed that Google Trends with the keyword "Flood" and the
pattern of flood events had a strong positive relationship with a
correlation coefficient of 0.64, this indicates that the increase in the
incidence of disasters causes an increase in public interest in seeking
information about flood disasters. Risk communication that was
carried out before the disaster occurred was not yet optimal in
reaching the community. Risk communication needs to be carried out
before the months that are predicted to be the peak of flood events.
Risk communication is carried out to form the preparedness of the
Indonesian people so that the risk of flood disasters can be reduced.
Keywords: Google Trends, Flood, Disaster Risk Communication
Introduction
Floods are a disaster that has the highest incident intensity in Indonesia. In
2022, there were 1,175 hydrometeorological natural disasters during the period
106
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
from 1 January to 3 April 2022 which were dominated by floods with a total of
428 incidents. As a result of the flood disaster, 80 people died, 10 people were
missing, 1.67 million people were displaced, and 605 people were injured. Flood
disasters in Indonesia tend to increase from year to year, with the highest
occurrence in 2020 with 1518 events (BNPB 2022).
Figure 1. Floods Events In Indonesia 2012-2021
Source: Results of data analysis, 2022
The increasing intensity of floods from year to year causes huge losses, the
average loss caused by floods every year reaches 4.64T (Kebijakan and
Kementerian Keuangan 2018). Flood disaster is a disaster with recurring
occurrences in Indonesia, public awareness of flood disaster along with prevention
and preparedness measures needs to be possessed by the community.
Information is very important to increase the preparedness of people living in
disaster-prone areas. When an individual has more experience with disasters, that
experience will shape their perception, and when there is a lack of experience with
a disaster, it is likely that the assessment of a disaster is based on information
obtained through the media, own intuition and information from social networks
owned. Information about floods is very important to improve the preparedness of
people living in disaster-prone areas. If information can be provided when people
have a high interest in learning, it can have a profound effect on increasing public
knowledge. To carry out effective risk communication in flood disasters,
knowledge about flood disaster patterns is needed, as well as knowledge about
community behavior in seeking information. One of the people's behavior in
searching for information can be seen from the Google Trends analysis. The
purposes of this study are to analyse the relationship between public interest in
finding information through google trends with the keyword "flood" with flood
events in Indonesia for 10 years.
Literature Review
Information from the media about natural hazards can also affect
perceptions of an event (Alcántara-Ayala and Moreno 2016).The form of public
awareness that they are in a disaster risk area is having efforts to increase their
107
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
disaster knowledge, one of which can be done by searching for information online
through the Google search engine.
According to the We Are Social report there were 204.7 million internet
users in Indonesia in January 2022. This number increased by 1.03% compared to
2021. The internet penetration rate in Indonesia reached 73.7% of the total
population in early 2022 (Total population of 277.7 million) (We are Social
2022). According to (Hampton & Goulet, 2011; Schroeder, 2018) people use the
internet for two main activities, namely to communicate (socialization) and to
search for information. For the purposes of searching for information via the
internet, people use search engines such as Google, Bing, Yahoo, etc. The most
popular search engine (has a market share of more than 92.03% of users) of the
world's population is the Google search engine. The Google search engine is also
the most popular source of information search in Indonesia. Google's market share
in August 2021 in Indonesia reached 97.65% or 197.83 million users (Statcounter,
2021). Big Data is generally considered as information that can be linked, with
large volumes of data and complex data structures (Khoury and Ioannidis 2014).
Utilization of big data for academic purposes, for example, with social media data,
cell phone call records, commercial website data, individual geographic
information, search engine data, smart card data, and taxi route data (Liu et al.
2016). Big data analysis can help generate better solutions to address issues
related to the health, education, transportation, housing, aid, and inclusion sectors
of certain socially, demographically, and geographically disadvantaged
groups(Rodríguez, Palomino, and Mondaca 2017) .
Google Trends TM is one of the most widely used tools for this purpose.
The analysis of relative internet search volumes (RSV) provides information
regarding the extent of public attention (Effenberger et al. 2020). Google Trends
is a unique data set that can be examined for deeper insights into social search
behavior during extreme events such as floods, droughts, heatwaves, wildfires and
landslides (Thompson et al. 2022). The Google Trends offers a special chance to
gauge the "interest" in disasters. If information can be made available when
individuals are most eager to learn, this can have a significant impact on raising
the general public's awareness of recent and upcoming events. Knowledge can
replace ignorance, and understanding can replace fear (Linkov et al. 2010). The
purposes of this study are to analysed the relationship between public interest in
finding information through google trends with the keyword "flood" with flood
events in Indonesia for 10 years. To assess the association of public interest to
flood, we used Google Trends Analyses. Google Trends is a public website owned
by Google Inc. and offers data based on Google Search that shows how often
certain search terms are entered compared to all other search terms in different
regions and languages (Google, 2021). Google Trends does not display the total
number of searches over time, but provides population-adjusted data that reflects
the popularity of searches at any given time (Pullan and Dey 2021). The data is
anonymous as no name or other personal user information is disclosed. Google
Trends provides comparisons of up to five search terms or topics. The results for
each search term or topic are associated with a maximum value of 100 (Kaatz et
al. 2022). Prior to publication, data is anonymized, categorized, aggregated and
108
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
undergoes normalization to be rescaled and converted into a popularity index with
a value between 0 and 100 (Google, 2021).
Research Method
This research is descriptive quantitative research. Data were collected
from two sources. Google trends data obtained through settings on Google Trends
(https://trends.google.com/trends): period set for 10 years "March 2012-April
2022", region selected as "Indonesia" and "all categories" and "web search". Data
on flood events was obtained from the BNPB website by setting an event period
of 10 years from March 2012-April 2022. Spearman's Rank Correlation Analysis
was carried out using SPSS software to find out the relationship between Google
search trends and flood events over a 10 years period (March 2012-April 2022).
Results and Discussions
Google Trends “Flood” in Indonesia
Google trends with the keyword "flood" with the topic "disaster" during the
research period (10 years) had a high trend at the beginning of the year (January-
February) and had the highest trend in January 2014. The following presents a
Google Trends chart for the keyword “flood” during the period March 2012-April
2022. Figure 2. Google Trends "Flood" for the period March 2012-April 2022
Source: Results of data analysis, 2022
Data over a 10 years period is presented by monthly trend averages to see
patterns from Google Trends. Search trends have gradually increased starting in
109
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
August, decreasing in October. The peak of the search trend is experienced in
January, remains high until February, and gradually decreases in March-July.
Figure 3. Google Trends Monthly Average “Flood” for the period March 2012-
April 2022
Source: Results of data analysis, 2022
3.2 Floods in Indonesia
Flood events were accessed from BNPB data starting in 2012-2022. The
data shows that Indonesia has a high intensity of flood events with a total of 8285
incidents. February 2020 is the month with the highest flood events, namely 283
events. August is the month that has the lowest number of flood events compared to
other months. A graph of flood events in Indonesia over a 10 years period is
presented in the following figure.
Figure 4. 10 Years Floods events in Indonesia
110
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
Source: Results of data analysis, 2022
To see trends in flood events, flood data obtained from BNPB over a 10
years period are presented with monthly average events. From the graph, it can be
seen that flood events began to increase in September and reached a peak in
January-February. In March-August the incidence of flooding has decreased. The
lowest number of flood events occurred in August, then increased linearly towards
the end of the year. Graph of average monthly flood events over a 10 years period is
presented in the following figure.
Figure 5. 10 Years Average Floods events in Indonesia
Source: Results of data analysis, 2022
3.3. Correlation between Google Trends and Flood Events in Indonesia
Spearman's correlation analysis was carried out to determine the
relationship between Google trends on the keyword "flood" and the number of
flood events for 10 years. The results of the correlation analysis show that the
google trend on the keyword "flood" and the number of flood events for 10 years
has a significant correlation (sig 0.000), and a positive correlation coefficient of
0.640 (strong correlation). This shows that the google trend on the keyword "flood"
and flood events for 10 years has a strong positive relationship, the high number of
flood events will be followed by an increase in the google trend on the keyword
"flood". The results of Spearman's correlation analysis presented as bellow.
111
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
Figure 5. The results of Spearman's correlation analysis
Correlations
Google_Trend
_Banjir
Kejadian_Ban
jir
Spearman's
rho
Google_Trend_
Banjir
Correlation Coefficient
1.000
.640**
Sig. (2-tailed)
.
.000
N
118
118
Kejadian_Banjir
Correlation Coefficient
.640**
1.000
Sig. (2-tailed)
.000
.
N
118
118
**. Correlation is significant at the 0.01 level (2-tailed).
Flood events in Indonesia have experienced a positive trend increase, this
shows that more and more flood disasters occur in Indonesia every year. This
phenomenon shows that the community's flood preparedness must always be
improved because an increase in the intensity of flood events will increase the risk
of disaster to the community. Flood events are a patterned annual cycle in
Indonesia. The results of data analysis show that the intensity of flood events
began to increase in December and experienced a peak in February. This shows
that
Overall, the analysis shows that there is empirical evidence regarding the
relationship between the intensity of flood events and the search volume for
"flood" on the Google search engine. This analysis confirms that public interest
increases when there is an increase in the number of flood events. The increase in
public interest in searching for the keyword "flood" on the Google search engine
has a positive and strong correlation with the number of flood events in Indonesia.
This finding is in line with the results of a study (Bakshy et al. 2012) which states
that Google Trends tracks search interests during the flood. There is a trend of
increased interest in events that have received widespread media coverage or
where deaths have been reported. This is not surprising given that post-flooding
messages have appeared online and netizens are likely to see them on their
timelines. Additionally, when flood reports are published online, users are more
likely to share them on social media, further driving search interest in the topic
(Bakshy et al. 2012). Thompson et al. (2022) also found results similar to this
study. Research (Thompson et al. 2022) shows that there is a significant
correlation between monthly average search interest and monthly rainfall in
Kenya and Uganda over a five-year time period.
Public interest in searching for information on Google has a positive
correlation with extreme events such as increased rain intensity and flood event
112
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
intensity. This is in line with the statement (Alcántara-Ayala and Moreno 2016)
that information is very important to increase the preparedness of people living in
disaster-prone areas (Alcántara-Ayala and Moreno 2016).
The correlation between Google search trends and flood events in
Indonesia shows that people are starting to be interested in finding information
about flood disasters when there is an increase in incidents. Communities still
tend to be responsive to disasters. When there was an increase in flood incidents,
the mass media began to cover news about the chronology, victims, losses, and
fundraising. Social media also began to react to the disaster that occurred, causing
an increase in traffic regarding the flood disaster and increasing public interest in
seeking information about the flood disaster.
Disaster risk communication is an effort to increase community
preparedness through the provision of information on hazards quickly and
accurately, as well as information on the response needed to reduce disaster risk.
The role of disaster risk communication is fundamental in disaster communication
management. Mass media support in communication or coordination management
activities can reduce disaster risk or reduce the level of vulnerability and danger
due to disasters (Asteria 2016). Public awareness of the risk of floods can be an
illustration for determining strategic steps for reducing the risk of floods in
Indonesia.
During the critical period before the peak months of flood events, it is
necessary to quickly and accurately provide information regarding flood disasters
to the public through various existing media. This is done as an effort to increase
community preparedness in dealing with disasters and reduce disaster risk.
Conclusions
Google's trend regarding floods has increased gradually starting in August
and has decreased in October. The peak of the search trend is experienced in
January to February, and falls in March-July. Flood events have increased gradually
starting in September and reaching a peak in January-February. Flood events began
to decrease and experienced the lowest number of incidents in August. Google
Trends with the keyword "Flood" and patterns of flood events have a strong
positive relationship with a correlation coefficient of 0.64, this shows that an
increase in disaster events has led to an increase in public interest in finding
information about flood disasters. Risk communication needs to be carried out
before the months that are predicted to be the peak of flood disasters. Risk
communication is useful in shaping the preparedness of the Indonesian people to
reduce the risk of flood disasters.
Acknowledgements
Financial supports from the Indonesian Education Endowment Fund
(Lembaga Pengelola Dana Pendidikan-LPDP), and Center of Education Financial
Services (Pusat Layanan Pembiayaan Pendidikan-PUSLAPDIK) are gratefully
acknowledged.
113
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
Bibliography
Alcántara-Ayala, Irasema, and Ana Rosa Moreno. 2016. “Landslide Risk Perception
and Communication for Disaster Risk Management in Mountain Areas of
Developing Countries: A Mexican Foretaste.” Journal of Mountain Science 13
(12): 2079–93. https://doi.org/10.1007/s11629-015-3823-0.
Asteria, D. 2016. “Optimalisasi Komunikasi Bencana di Media Massa sebagai
Pendukung Manajemen Bencana.” Jakarta.
http://www.penanggulangankrisis.depkes.go.id/,.
Bakshy, Eytan, Itamar Rosenn, Cameron Marlow, and Lada Adamic. 2012. “The Role
of Social Networks in Information Diffusion.” In WWW’12 - Proceedings of the
21st Annual Conference on World Wide Web, 519–28.
https://doi.org/10.1145/2187836.2187907.
BNPB. (2020). Data & Informasi Bencana Indonesia. (Online). https://dibi.bnpb.go.id.
7 April 2022
Effenberger, Maria, Andreas Kronbichler, Jae Il Shin, Gert Mayer, Herbert Tilg, and
Paul Perco. 2020. “Association of the COVID-19 Pandemic with Internet Search
Volumes: A Google TrendsTM Analysis.” International Journal of Infectious
Diseases. Elsevier B.V. https://doi.org/10.1016/j.ijid.2020.04.033.
Google. (2022). Google Trends Help, in, Google Inc.
https://support.google.com/trends/answer/4365533?hl=ko&ref_topic=6248052
Kaatz, Martin, Steffen Springer, Roger Schubert, and Michael Zieger. 2022.
“Representation of Long COVID Syndrome in the Awareness of the Population Is
Revealed by Google Trends Analysis.” Brain, Behavior, & Immunity - Health 22
(July): 100455. https://doi.org/10.1016/j.bbih.2022.100455.
Kebijakan, Badan, and Fiskal Kementerian Keuangan. 2018. Strategi Pembiayaan Dan
Asuransi Risiko Bencana.
Khoury, Muin J., and John P.A. Ioannidis. 2014. “Big Data Meets Public Health.”
Science. American Association for the Advancement of Science.
https://doi.org/10.1126/science.aaa2709.
Linkov, Faina, Ali Ardalan, Meredith Hennon, Eugene Shubnikov, Ismail Serageldin,
and Ronald Laporte. 2010. “Using Google Trends to Assess Interest in Disasters.”
Prehospital and Disaster Medicine. Cambridge University Press.
https://doi.org/10.1017/S1049023X00008608.
Liu, Jianzheng, Jie Li, Weifeng Li, and Jiansheng Wu. 2016. “Rethinking Big Data: A
Review on the Data Quality and Usage Issues.” ISPRS Journal of
Photogrammetry and Remote Sensing. Elsevier B.V.
https://doi.org/10.1016/j.isprsjprs.2015.11.006.
114
Wasilatuna: Jurnal Komunikasi dan Penyiaran Islam
Vol. 06, No. 2, 2023.
Pullan, Samuel, and Mrinalini Dey. 2021. “Vaccine Hesitancy and Anti-Vaccination in
the Time of COVID-19: A Google Trends Analysis.” Vaccine 39 (14): 1877–81.
https://doi.org/10.1016/j.vaccine.2021.03.019.
Rodríguez, Patricio, Norma Palomino, and Javier Mondaca. 2017. “Using Big Data and
Its Analytical Techniques for Public Policy Design America and the Caribbean.”
http://www.iadb.org.
Statcounter. (2021). Search Engine Market Share Worldwide Aug 2020 - Aug 2021.
Retrieved on 8 April 2022, from https://gs.statcounter.com/search-engine-market-
share
Thompson, Joshua J., Robert L. Wilby, Tom Matthews, and Conor Murphy. 2022.
“The Utility of Google Trends as a Tool for Evaluating Flooding in Data-Scarce
Places.” Area 54 (2): 203–12. https://doi.org/10.1111/area.12719.
We are Social. 2022. “Digital 2022 Indonesia February 2022.”
