Surveys are not the end-all and be-all of political campaigns.
Although polls can disclose trends of voters’ preference, their real value lies in giving campaign managers a snapchat of voters’ sentiments at the time the survey was conducted for the sole purpose of creating and adjusting campaign strategies when necessary.
In fact, even SWS President Mahar Mangahas noted in an interview that “voters’ sentiments are hardly constants” and that the “2022 election outcome is far from set in stone.”
But to get a full glimpse of voters’ preference and behaviors, there are other quantitative and qualitative metrics that can be used in conjunction with election surveys.
These metrics are the following: 1) Social Media Sentiment Analysis, 2) Google Trends and 3) Voters’ Enthusiasm
Social Media Sentiment Analysis
Social Media Sentiment Analysis uses data points from Twitter, Facebook and/or Google to analyze candidates’ engagement, voters’ sentiments and potentially predict election results.
It is a combination of textual and content analysis of social media chatters and Big Data Analytics to generate actionable insights which can be helpful in election campaigns.
There have been multiple researches on the use of social media and Big Data Analysis in election forecasting.
Some of these works found that sentiment analysis and its variants (ie. volumetric and social network analysis) were able to predict in varying extent the election results in 2009 Germany Federal election (Tumasjal et. al. 2010), 2011 Dutch Senate election (Skoric et. al, 2012) , 2011 Irish General Election (Bermingham and Smeaton, 2011), United Kingdom general election 2010 (Boutet et. al., 2012), American republican presidential election 2011 (Shi et. al., 2012), 2013 Pakistan General Election (Ahmed and Skoric), Indian 2014 General Election (Khanua et. al., 2015), general elections in Netherlands, Greece and Germany (Tsakalidis et. al., 2015), 2013 General Elections in Malaysia and Pakistan (Jaidka et. al., 2014), and 2014 General Election in India (ibid.), 2016 Taiwan General Election, 2016 US General Election (Heredia et. al., 2016) and Gujarat State Assembly 2017 in India (Bose et. al., 2019), among others (cited in Chauhan et. al., 2021).
Just recently, Xia et. al. (2020) predicted the results of the 2020 US Presidential election using sentiment analysis (cited in Zhou et. al., 2021).
In their conference paper, they concluded that “sentiment analysis can be used as a low-cost and easy alternative to gather political opinion.” (ibid.)
In the Philippines, the use of social media sentiment analysis has been around since the early 2010’s but confined only within advertising and market research.
Its applications in politics are in its infancy.
But recently, Wilson Chua in an article in Manila Bulletin proposed the use of social media sentiment analysis in the coming Philippine general election.
Using his proprietary ADDS Sentiment Analysis, Chua tracked the engagement and net sentiments generated by each of the presidential candidate for the period between January 1 to February 27, 2022 and found out that the resulting Sentiment Analysis Chart favors a two-horse race between Ferdinand Marcos Jr (BBM) and Vice President Leni Robredo.
However, Chua also mentioned the limitations of the use of Data Analytics and Facebook Social Media Sentiments which include the following:
1) Not all voters are users of Facebook
2) Not all Facebook users are registered Philippine voters
3) Not all positive sentiments will translate to votes
4) Not all reactions are authentic since Data privacy restrictions prevent verification of Facebook accounts.
Although Social Media Analysis as applied in election forecasting is still under-developed and under-researched in the Philippine setting, Chua’s preliminary exploration of its use in the coming election is worth looking into noting that 82.4% of the Philippine population are social media users at the start of 2022.
The use of Google Trend is also part of Big Data Analytic approaches in election forecasting.
Just like Social Media Sentiment Analysis, there is a handful of research that positively correlates the results of Google Trends and elections.
Some of these researches include the work of Prado Roman et. al. (2020) who proposed a method of using Google searches that predicted the winners of the past four elections in the United States and past five elections in Canada.
Lui et. al (2011) also found that Google Trends volume was predictive of 2008 and 2010 US Congressional Election.
Google data points were also used to predict the result of the 2016 Taiwan General Election (Xie et. al., 2018).
In the Philippines, there is dearth of systematic study on the predictive capability of Google Trends in a general election, but a quick look at Google Trends data and the outcomes of 2010 and 2016 Presidential and Vice Presidential Race suggests strong correlation between the two variables as shown in the following graphs.
For the 2022 election, a netizen shared on Twitter that Vice President Leni Robredo is presently leading over Bongbong Marcos in Google Trend results.
The tweet generated buzz and insights from other netizens.
A netizen noted that the searches about Leni Robredo were mostly positive which may indicate strong voter interest to her campaign.
Another netizen, however, cautioned that many voters might have no access to the internet.
In an article in the Journal of Electoral Studies, Hill (2015) defined voter’s enthusiasm or enthusiasm gap as the measure of the level of mobilization of the supporters of a political party as compared to the others.
This may further refer to the behavior of core party’s supporters and get-out-the-vote activities which are positively correlated to voter’s turnout.
For Marcus and Mackuen (1993), the ability of political leaders to generate enthusiasm stimulates interest and political involvement in the campaign.
“When voters respond to a candidate with enthusiasm, they are not merely evincing passive sympathetic reactions but sharing convictions and commitment to common endeavors. Rather than stopping, looking, and listening, enthusiasts throw themselves into the cause.” (ibid)
Although considered an important determinant of election results (Hill, 2015), voters’ enthusiasm is hard to quantitatively measure especially in the context of Philippine election.
In the United States, Gallup measures voter’s enthusiasm through survey questionnaires.
No such similar data is available in the Philippines.
However, voter’s enthusiasm can be qualitatively assessed through grounded theory or ethnography, or a mixture of both methodologies.
While this article does not aim to proffer an academically exhaustive theory of measuring voters’ enthusiasm in the Philippine setting, I suggest that it is possible to have a glimpse of the level of Filipino voters’ enthusiasm through cultural analysis of 1) campaign rallies and sorties particularly the behavior and attendance of supporters of presidential candidates during these events and 2) the distinct cultural practice of Filipino voters of attaching the posters of their favored candidate in the gates or façade of their homes.
Election Campaign as Fiesta and Sabong
The whole gamut of Philippine election campaigns particularly the sorties, rallies and motorcades can be likened to a quintessential Filipino fiesta (Tatel Jr., 2013) which functions as a Bakthinian carnival where the ordinary configurations of power relations within society are subverted momentarily.
In this festival, the politicians who ordinarily wield power become the jesters and performers while the crowd are momentarily treated as fastidious masters who must be pleased at all cost.
Aside from fiesta, Philippine election can also be likened to a sabong where voters root or bet for their manok which can either be a llamado (incumbent) or dehado (newbie candidate) (ibid.).
Unfortunately, this sabong analogy can also partly explain the propensity of Filipino netizens to engage in trollish behaviors in social media during election season.
Similar to their religious devotion and beguilement to sabong, Filipino voters’ enthusiasm is expressed by the level of their devotion or obsession to attend, participate and invest in campaign rallies or sorties of their favored candidate whom they treated as their (political) patron or manok (team).
In the present election campaign, political parties and supporters have been mobilizing for the campaign rally and sortie of their preferred presidential candidates.
At this front, it can be readily observed that among the Presidential candidates, it is the campaign of Vice President Leni Robredo which is showing strong performance in generating enthusiasm and ground momentum as evidenced by massive attendance and more festive atmosphere of her campaign rallies and sorties.
Such groundswell of support for Robredo may be attributed to her strategy of adopting a People’s campaign.
Another cultural practice that shows the level of voters’ enthusiasm is the Filipino practice of attaching the posters of their chosen candidate at the gates or façade of their houses.
In the days before the internet and social media, such practice was one of the clear yardsticks of the level of support of Filipinos to their preferred candidates.
There is certain level of intimacy when Filipino families decide to attach the posters of their preferred presidential candidates at the facade of their homes.
Just like the old cultural practice of putting the professional titles of family members in the front wall of houses, the posters of political candidates attached in balconies, gates or walls of Filipino houses are symbols of family pride and values.
It signals to their entire community the kind of values that their family holds dear as reflected by their chosen political candidate.
Thus, in the cultural worldview of Filipinos, this behavior ranks as political endorsement par excellence which is unmistakably an indicia of Filipino voters’ enthusiasm.
On this score, it can be casually observed that among the various supporters of presidential candidates, it is the supporters of Vice President Leni Robredo or the Kakampinks who usually attach the campaign posters of the Vice President in their homes.
Of course, for some, this may easily be dismissed as mere virtue signalling or futile attempt to campaign a losing candidate, but a deeper cultural analysis of this practice reveals a more profound facet of Filipino voting attitude and behavior.
By looking at the results of the three (3) alternative election metrics, it can be readily concluded that Vice President Leni Robredo is performing better in these areas (Social Media, Google Trend and Voters’ Enthusiasm) than the rest of the presidential candidates as of this writing.
Of course, just like election preference surveys, these metrics have their own limitation and are susceptible to errors and biases.
It can also be readily admitted that there is a clear disjunction between the results of election survey and the metrics discussed.
But this variance can simply be explained by the fact that each of the election metrics focuses on different data sets and data gathering period.
What is essential is whether the final trends of each of the metrics (Survey, Social
Media Sentiment Analysis, Google Trend and Voters’ Enthusiasm) will correctly predict the final outcome of the election.
Admittedly, surveys are traditionally thought of as a more empirical enterprise, but any campaign managers worth their salt and even casual observers of Philippine election cannot simply discount the actionable insights that can be derived from these qualitative observations and Big Data election analysis.
In other words, although currently favorable to Vice President Leni Robredo, the insights from Big Data Analytics and ground assessments can also be used by other presidential candidates as complement to election surveys to help them recalibrate their respective campaign strategies specifically in areas where they are performing poorly as shown by these metrics.
Besides, with roughly two months left before the May 09 general election, many things can still happen; voting trends and preference can definitely change.
[READ our latest article: [OPINION] Surveys are not infallible but Robredo’s campaign must go back to fundamentals: TV & Radio Air War]