Desy Nurhayati, Jakarta – Opinion surveys will not greatly influence the decision of voters in the April 9 legislative election, a discussion in Jakarta heard Tuesday.

Attendants of the event held by Institute for Studies on the Free Flow of Information (ISAI) agreed that the surveys only spark a bandwagon effect and barely impact a voters' preference.

Dodi Ambardi from the polling company Indonesian Survey Institute (LSI) said that the behaviour of voters is determined more by a parties' prominent figures as well as their personal, peer group and family's preference than by survey results.

"There has not been any proof that voters choose a certain party just because the party ranks high in many surveys," Dodi told the discussion.

"People justify the great influence of surveys on the preference of voters' just because of the bandwagon effect, which is when, in some cases, voters "follow the mainstream" in deciding their choice of party."

To some extent, he added, there was an underdog effect that led voters to vote for parties with less support. Results of surveys predicting that one party or another will get the most are often criticized by the party's competitors.

They question the credibility of the surveys, saying they are intended to legitimize a party's unprecedented win and that the survey institutions are manipulated by the winning party.

Most recent surveys have constantly ranked President Susilo Bambang Yudhoyono's Democratic Party as the leader, ahead of the country's two largest parties, Golkar and the Indonesian Democratic Party of Struggle (PDI-P).

Dodi maintained that survey institutions are independent and that their clients do not influence survey results.

"Our business is all about trust. Who will trust us if we modify the results upon the request of clients," he said.

"In LSI, indeed we receive many sponsors and funding, but we do not know who finances which survey because all of the money goes to a separate division."

The ISAI recently published the results of its research on media coverage of this year's election.

The research sampled 13 print media outlets between January and Mar. 20 and 10 national TV stations between Mar. 1 and 6.

ISAI research coordinator Ahmad Faisol said that the research was aimed at ensuring the media's ability to provide information about elections that the public needed, given its power to contribute to voters' participation in the elections.