Public And Secret Opinion

Theoretically speaking, no matter what we think about the preconditions, circumstances and results of last week's elections, the conclusions are unfavorable. Either the voters' beliefs and will really did change very much in favor of the radical right wing, or, on the other hand, the Socialist Party of Serbia is much better organized than it seems. The first interpretation somewhat diminishes the dark glory of the radicals' "victory" because - it should be noted - at the last pre-war elections in Germany (1933), Hitler won the votes of 39 out of 44 million voters, that is, he won the support of 88 percent of the electorate. In the latter case, if this was a big electoral deceit, the "night elections" between December 21st and 22nd had to be organized by people who are much more disciplined and faithful to the oath of silence than, at least according to tradition, is usual for this environment. Because of the (socialist) rose and ("Seselj's") Serbian cross one could also bitterly joke about a 'Rosenkreuz' conspiracy. Those who studied the public opinion are also surprised by the "miscalculations" in their forecasts. Or, as the director of the "Partner " Agency, Miroslav Sutic, said: "We have already analyzed, with our colleagues from other agencies, the possibility of a mistake in the applied methodology. Since most of the forecasts made on the basis of public opinion polls were very close or similar, one should take into account 'external' factors, especially because of a 'suspicious' concordance of figures". As of September, "Partner" devoted special attention to the category of the undecided voters, estimating that there were around 1.7 million of them, around 1.1 million of whom could opt for the opposition. Instead of this, added Sutic, according to the published (incomplete) electoral results, in regard to the 'pouring over' of votes "there exists only a mathematical, but not a real possibility that literally all the undecided voted for the radicals. There is a lot of doubt, because despite the "precise adding up", there were too many other imprecisions and ridiculous and unbelievable violations of the electoral procedure". In the night when the votes were counted, the director of the "Partner" agency convincingly defended the results of the polls that included 1370 interviewees at 60 at random selected voting places throughout Serbia (except Kosovo), according to which, in a dead heat, Slobodan Milosevic and Milan Panic (47:47) would enter the second electoral round. "Before the publication of the results, we corrected the results of the polls three times very rigorously in favor of Slobodan Milosevic. An American expert inspected the procedure and agreed that a mistake in the forecast could be plus-minus 3 to 3.5 percent at the most. In that regard, the announced difference of 15 percent is absolutely beyond the margin which we determined by three inspections and a rigorous correction of the results of the polls". The Belgrade Institute of Social Sciences, that is, its Institute for Public Opinion Research, enjoys the reputation of an institution with a long tradition in this field. Ljiljana Bacevic who was in charge of the pre-election research, first of all points out that the forecasts were based on around six million voters (without Kosovo) with the assumption that between 70 and 75 percent of the voters would turn out. Taking into account that part of the electorate (5 to 5.5 million voters), the forecast was 36:35 for Slobodan Milosevic. The forecast was controlled through the inclusion "additional questions" posed to the polled citizens and it was adapted to the determined fact that Milan Panic was suffering from "the Ante Markovic syndrome", that is, that he enjoyed the general, broad - but "shallow" support, and that unlike Markovic, he did not have political support in the form of "his own" party. This is what could be concluded (and Panic's electoral headquarters did not conceal it in it election campaign) from the answers to the set of questions in which the polled saw Panic's advantage in his abilities and capabilities in the field of the economy, in regard to the lifting of the sanctions, putting an end to the war, democratization and a "better future", in short, in everything except the protection of the national interest (51:26 for Milosevic). Ljiljana Bacevic also points to the fact that most of their (and other) public opinion polls were published in periods that were "bad" for the Serbian Radical Party, during certain "truces", when according to the logic of things, public support to the extreme right wing is not pronounced, and that in the last ten or so days - when there were no polls - the pressure aimed at increasing the fear of an international military intervention was considerably stepped up. "Regardless of that", said Ljiljana Bacevic, "in all three polls there was an unexpectedly high percentage of those who claimed that they would not vote, and in the last poll (10th-12th December) the number of those who refused to answer our questions increased from the usual one fifth to half, and somewhere even more. We ascribed this to fear, restraint, the general atmosphere and perhaps, too easily accepted the explanation that these are people who say that nothing interests them and that, therefore, they would not turn out at the elections. It now seems to me that this was precisely where the 'radicals' reserve' was hidden". The message to all those who study the public (and secret) opinion, considers Ljiljana Bacevic, is that an explanation of unusual situations, such as a high percentage of refusals to answer questions, the systematic underestimation of certain parties and so on, must not stop - at the first logical interpretation. "We cannot expect logical trends in illogical circumstances", said Ljiljana Bacevic. "One cannot count on the growth of pacifism in a pre-war situation". Srdjan Bogosavljevic, director of the Federal Statistics Bureau, points to yet another unexpected concordance between the forecasts and the (announced, incomplete) results of the elections: "When one puts together the parties' pre-election recommendations to voters concerning the presidential elections, one gets - almost precisely - the announced outcome of the elections. In the same way there is also concordance between the forecast and the published collective results of the Serbian Radical Party and the Democratic Movement of Serbia, with votes going over, in an unusual way, to the Serbian Radical Party. This can point to completely opposite conclusions: either that there was no 'wheeling and dealing', or that everything was organized and prepared with such unbelievable precision and in the utmost secrecy". Mr. Bogosavljevic considers that one must not neglect the trends in the last ten or so days before the elections when the state media, especially Radio and Television Serbia worked, with full steam, in favor of - Seselj. He gives the example of the cut up statement by Lawrence Eagleburger (throwing out names of potential war criminals on the other sides, and adding the name of Slobodan Milosevic to "ours"). However, stories about cheating and various "circumstances", the propaganda and the television, move the matter to a completely, non-statistical field. In the field of forecasts and results it is enough to cite Mihajlo Mihajlov and his sincere wish: "Let us hope that it was a deceit, because - if it wasn't - it means that Serbia is becoming Nazi-oriented."

Forecast, Realization

Forecast December 19th; Realization 24th 1992

MF AGENCY: (UN)EXPLAINABLE

by Miladin Kovacevic

The MF agency forecast the results of the republican presidential and parliamentary elections. Its forecast for the presidential elections was 46:41 percent in favor of Milosevic, in the first round. The collective votes for Vuk Draskovic (who, at the time, was still a candidate) and Milan Panic accounted for 32.8 percent which is very close to the results of the elections. However, when making the forecast, the point of departure was the assumption that there would be a proportional "distribution" of the undecided and that the "abstentions" would account for between 15 and 20 percent. At this moment (December 24th) we can compare the percentages of the votes won at the elections for federal parliament with the forecast. One can notice a high decree of concordance in regard to all the mentioned lists, except for that of the Democratic Movement of Serbia and the Serbian Radical Party. The concordance in regard to the Serbian Radical Party and the Democratic Movement of Serbia is greater when one relies only on the results of the polls, without corrections in the forecasts. We also find very interesting the surprising arithmetic concordance between the forecast collective percentage of the Democratic Movement of Serbia and the Serbian Radical Party and their collective results in the elections. At first glance, there are two possible explanations for this. The first one is that, in the last ten days of the election campaign, the Democratic Movement of Serbia lost 7 to 8 percent of the votes which went to the Serbian Radical Party. The second is that a high percentage of those who voted for the Serbian Radical Party said in the poll that they would vote for the Democratic Movement of Serbia. I never encountered a similar situation in all the numerous polls for which I designed the samples in the past 15 years (Slovenia, Croatia, Serbia). The deviations were always smaller and explainable by evident or very probable reasons.

"Statlab": Calculation or Deceit

The STATLAB of the Faculty of Organizational Sciences made, on the basis of a telephone poll, an assessment of the electorate's mood prior to the elections. The forecast was done for the population without the Albanians, that is for 88.2 percent of the voters. It was only the Socialist Party of Serbia that was in the right place, on the basis of the public opinion poll. In the case of other parties there was an unexplainable 'pouring over' of votes, even though the campaign led by Television Serbia was, during the whole time, and especially in the last days, oriented towards diabolizing the Democratic Movement of Serbia, the Democratic Party and Milan Panic, presenting them as traitors, so that the counter-weight could have been the radicals' pronounced militant patriotic approach. However, the collective results of 'impossible coalitions' which have close programs are absolutely unbelievable. The monarchist option opposed to the republican one was very well represented in the poll (this was also the case in some other polls). The seemly logical - but in practice high unlikely change in the mood of the voters - opens many questions, even the question of the regularity of the elections. Things will be clearer when the collective results of the local elections are known. The electorate's mood should be consistent there too and follow the trend of the federal and republican elections. The sympathizers of the Serbian Radical Party and the Socialist Party of Serbia certainly voted for Milosevic, as did the sympathizers of certain smaller parties (the Serbian National Renewal, "The Serbian opposition" - which has more votes than Milan Paroski etc.) All in all, Milosevic had to win those votes, that is a little over 58 percent, while the parties of the "Panic-orientation" had around 32 percent together. Here, the calculation confirms the results of the elections. Or perhaps this is a deceit of larger proportions.