| Home > South-East Asia >> Indonesia | |
Puskaptis, JSI refuse to reveal data behind quick counts
Jakarta Post - July 17, 2014
The organization said the two member pollsters, which called the presidential election for the Prabowo Subianto-Hatta Rajasa ticket, were in breach of the organization's ethics by their irresponsible quick-count announcements to the public.
"It's not about whether or not their [quick-count] results match the real count by the KPU [General Elections Commission]. It's about their refusal to open their data to the public," Persepi's ethics council head Hari Wijayanto told a press briefing on Wednesday. The two pollsters have therefore been dismissed from the association.
Hari, a statistician at the Bogor Institute of Agriculture (IPB), explained that neither Puskaptis nor JSI had complied with an audit order issued by Persepi, which aimed to systematically examine the data and methodology used by member pollsters that had released their quick-count results of the presidential election, as well as their sources of funding.
Seven of Persepi's member pollsters called the election for Joko "Jokowi" Widodo and his running mate Jusuf Kalla, while only JSI and Puskaptis favored Prabowo-Hatta. The seven pollsters favoring Jokowi have been audited and their results have been declared valid.
According to Hari, instead of complying with the request to be audited, JSI and Puskaptis sent letters tendering their resignations as members of Persepi.
"Puskaptis in particular argued that the audit should take place after an official announcement by the KPU and that it should be transparent. [Puskaptis] regarded us, Persepi, as not being transparent," Hari said.
This contradicted Puskaptis executive director Husin Yazid, who had previously said that his institution was ready for an audit to ensure that its polling was accountable.
Puskaptis and JSI declared that Prabowo-Hatta beat Jokowi-Kalla by as much as 6 percent; but their results were met with public skepticism due to their poor track records.
Besides Puskaptis and JSI, other pollsters that also released their quick-count results favoring Prabowo-Hatta included the National Survey Institute (LSN) and the Indonesian Research Center (IRC), but they are not subject to auditing by Persepi as they are not members of the association.
Persepi's ethics council has validated the seven pollsters that called the election for Jokowi as they reported their data and fulfilled the probability-sampling standard in collecting their data.
They are Saiful Mujani Research and Consulting (SMRC), the Indonesian Survey Institute (LSI), Indikator Politik, the Centre for Strategic and International Studies (CSIS), the Cyrus Network, the Populi Center and the Pol-Tracking Institute.
SMRC, for example, used 4,000 samples while the LSI collected data from 2,000 polling stations. Puskaptis earlier said in a discussion that they only gathered data from 1,250 stations.
Ethics council member Hamdi Muluk, a political analyst from the University of Indonesia (UI), said the council would forward the result of its investigation to the KPU for further consideration.
"We need to set up a national board that has the authority to certify pollsters that conduct quick counts of regional as well as national elections. The confusion caused by questionable surveyors has damaged the credibility of other pollsters," Hamdi said.
In 2009, Persepi also investigated a survey institution, the Indonesian Research Institute (LRI), due to alleged election quick-count manipulation.
The LRI predicted that Susilo Bambang Yudhoyono of the Democratic Party would only get 33 percent of the vote, followed by the Golkar Party's Kalla with 29 percent and Megawati Soekarnoputri of the Indonesian Democratic Party of Struggle (PDI-P) with 20 percent. However Yudhoyono actually garnered 60.8 percent of the vote, winning in one round. The LRI was later banned.
See also:
 |