In 2019, not a single polling organisation declared Prabowo Subianto the winner, a marked change from 2014, when four pollsters said on election day that Prabowo had one. Photo by Puspa Perwitasari for Antara.


When the Constitutional Court announced on 27 June that it rejected Prabowo Subianto’s legal challenge against the 2019 election result, its unanimous decision was not only an endorsement of the work of the General Elections Commission (KPU).


The Court also affirmed that virtually all of Indonesia’s reputable polling institutes had correctly predicted the result in their quick counts on election day. More than 10 officially registered pollsters published quick count results, and all of them turned out to be within about 1 per cent of the official KPU result.


Given the track record of long-established polling institutes such as Saiful Mujani Research and Consulting (SMRC), the Indonesian Survey Institute (Lembaga Survei Indonesia, LSI), Indikator, Indo Barometer, Charta Politika, the Centre for Strategic and International Studies (CSIS), Cyrus Network, Poltracking, the Indonesian Survey Circle (Lingkaran Survei Indonesia, LSI – Denny JA), Populi Center and Litbang Kompas (Kompas newspaper’s research and development arm), it was no surprise at all that the top pollsters were more or less spot on with their quick counts.


What is noteworthy, however, is that there was not a single renegade pollster who publicly declared Prabowo the winner. Even a less reputable organisation such as Median, which in the past had faced allegations of publishing questionable survey results, announced that Jokowi had won.


This stands in stark contrast to the last election in 2014. Back then, no less than four pollsters – Lembaga Survei Nasional, Jaringan Suara Indonesia, Indonesian Research Center and Puskaptis – declared on election day that according to their quick counts Prabowo had won the presidential election.


The provocative announcements were broadcast live on television stations owned by media tycoons like Aburizal Bakrie and Hary Tanoesoedibjo, who in 2014 were allied to Prabowo. In 2019, however, both Bakrie’s Golkar Party and Tanoesoedibjo’s new Perindo Party supported Jokowi. Thus, it was hardly surprising that this time no mainstream media outlet was prepared to collaborate with dubious pollsters on election day.


Despite the overwhelming evidence that he had lost the election, Prabowo did not concede defeat. Instead, he pressed on with his narrative of electoral fraud, which he had already spread long before the election. Claiming that “internal” exit poll and quick count data showed that he, not Jokowi, had won the election, he accused the mainstream pollsters of lying and suggested they move to Antarctica to spread their lies to penguins.


The anti-pollster discourse then quickly moved from absurd to serious. Online abuse and threats escalated, forcing prominent pollsters like Burhanuddin Muhtadi (Indikator) or Yunarto Wijaya (Charta Politika) to file police reports. In June, police confirmed that Yunarto had even been targeted in an obscure murder plot.


In order to understand why pollsters could become targets of excessive abuse and even death threats it is important to remember that Indonesia’s polling industry is not only heavily fragmented but also politicised. Several of the most prominent pollsters, for example, have often been highly critical of Prabowo, including Yunarto, who once described a campaign speech by Prabowo as propaganda and later engaged in a polemic Twitter debate with Prabowo supporters in which he offered to move to a different country should the former general win the election.


While independent observers have no doubt about the professional integrity of these pollsters, they are by no means just neutral political commentators. In the context of the polarised political atmosphere that characterised Indonesia’s recent elections, these political preferences, combined with their prominence as public figures, made pollsters popular targets for Prabowo and his supporters.


Faced with these unprecedented levels of hostility, the polling industry responded with remarkable professionalism. In the immediate aftermath of the election, when the Prabowo team launched frequent attacks against the credibility of the pollsters, the Indonesian Public Opinion Survey Association (Persepi) held a press conference where eight of its member organisations opened their quick count data to the public and challenged Prabowo and his team to do the same with their internal data.


Persepi chairman Philips Vermonte subsequently posted a brief statement about the event on his Facebook account which was liked by more than 4,300 people and shared more than 2,000 times. Of course, the Prabowo team did not respond to the challenge, and the pollsters were eventually officially vindicated when the Constitutional Court declared Prabowo’s legal challenge null and void.


One may therefore argue that Indonesia’s polling industry emerged out of the 2019 presidential election with a renewed sense of strength and commitment to help improve the quality of Indonesia’s precarious democracy. The fact that on election day none of the KPU-registered polling institutes dared to fabricate results in favour of Prabowo also gives reason to hope that the differences over ethical standards within the industry could perhaps eventually be resolved.


To achieve this beyond the heavily scrutinised national level, however, a whole armada of local pollsters will also need to be brought into line. Moreover, a rapprochement between Persepi and its rival association Aropi (Indonesian Association for Public Opinion Research), where many of the less prominent pollsters and consultancies are organised, may be necessary if the industry wants to further consolidate its professionalisation.


The next litmus test to observe the trajectory of Indonesia’s polling industry is not far away. In September 2020, no less than 270 local elections (pilkada) will take place, and surveys about the electability of certain candidates are already circulating. In the past, local elections have at times proven to be more difficult to predict than national elections as pollsters faced some significant methodological challenges.


Buoyed by yet another strong performance at the national level, Indonesia’s top pollsters will no doubt be keen to show that they can do the same at the local level.


This is part of a series of posts by presenters at “The 2019 Indonesian Elections: Electoral Accountability and Democratic Quality” workshop, held at the University of Melbourne on 6-7 August.



, ,

We acknowledge and pay respect to the Traditional Owners of the lands upon which our campuses are situated.

Phone:13 MELB (13 6352) | International: +(61 3) 9035 5511
The University of Melbourne ABN:84 002 705 224
CRICOS Provider Code:00116K (visa information)