In the post-war Kosova, the practice of public opinion polls has become more frequent. On its regular monthly polls, in June and July 2001, Index Kosova has asked citizens of Kosova questions emphasising their preference of political leaders, as well as their anticipation about the victory of a certain political party on the forthcoming November general elections. A special question was asked to understand the likelihood of citizens actually giving their vote to the same party they did on the last elections. Data obtained from these polls are not the data about where citizens' votes will go, but rather data about which party they think would win the elections. Knowing that, in a democratic society, public opinion goes through frequent changes in a relatively short period of time (as the comparison between data collected in two different months shows), the presented results, in this case, represent exclusively the present Kosovar opinion on possible future electorate behaviour.

Polls have been conducted in the entire region of Kosova - 1,100 Albanian citizens aged older than 18 were interviewed in June and 1,000 in July. Selection of households and respondents was done randomly, the sample being representative for the Kosovar Albanian population structure concerning gender, generations, social-professional status and residence.

On the question: "If there were direct Kosovo-wide elections for president in November, for whom would have you voted?" less than a half of the interviewed respondents (46.3 percent) in June, with a 3 percent decrease (43.3 percent) in July, have expressed their opinion, favouring Ibrahim Rugova. Almost every fourth respondent (22.8 percent) in June, however, with a more substantial decrease in July (15.8 percent), has favoured Hashim Thaçi. The leader of the third party - Ramush Haradinaj, keeps the third place, but with the slightest decrease of the three main Kosovar leaders (5.4 percent in June and 4.5 percent in July). "The best person for the job" - is the answer given by 5.2 percent of respondents in June and 5.7 percent in July. Despite the decrease in these percentages, in July, there is a more than a double increase on the "Don't know" answer (7.6 percent in June and 16.6 percent in July).
Looking at the age groups of respondents, a considerable diversity can be seen between preferences of different gender and age groups for different political leaders. While in June and July, too, Ibrahim Rugova was mostly favoured by males aged older than 50 (28 percent in June and 27.4 percent in July). Two remaining leaders, in June, were favoured by males of the third age group - aged 30-39 (32.7 percent Hashim Thaqi and 37.1 percent Ramush Haradinaj), however, in July, they were favoured by males of the fourth, respectively the fifth age group. Females of the third age group dominantly favour all three leaders in June and July, too.

For all three leaders, there are big differences between their male supporters depending on where they live. The differences are bigger in rural areas in comparison to urban ones, although this difference has been levelled by 5-6 percent in July. In females' case, this difference is more balanced.

Which party will win?

There is a certain parallelism between the results on the presidential question and the question: "Which party do you think is going to win on general elections in November?" All three parties that where categorised as likely winners by citizens are the same parties that won the most places in last year's local elections. Around half of the respondents (51.8 percent in October and 50.1 percent in July) said that LDK would win the elections, and in June every fifth respondent (20.6 percent) thought PDK would win the forthcoming elections. This opinion, concerning PDK, has changed a lot in July, as 13.2 percent of respondents think this party would win the elections. A slighter decrease is seen at the opinion favouring AAK. In June, 4.4 percent of respondents thought that this coalition would win the elections, and in July 2.9 percent think so. This decrease in percentages for these parties is accompanied by an increased percentage in the number of respondents choosing "Don't know". While in June, 19.7 percent of respondents chose "Don't know", in July, this percentage got to 30.4 percent.

The differences between people living in different regions are mostly stressed in the case of PDK and AAK. LDK, on the other hand, has more diverse respondents favouring the party, from both rural and urban areas. However, in the case of PDK and AAK, the number of respondents favouring them and living in rural areas is twice as high as the number of the ones living in urban areas.

Will the vote be repeated?

What is Kosovar citizens resolution in voting for the same party they did last year can be seen from results from the question: "How likely is actually that you are going to repeat your party vote from the local elections last year?" There was a high degree of resolution in both waves of research, although in July it is one percent lower. Three fourths of the respondents, in both polls (75.9 percent in June and 74.9 percent in July) claim that it would be "very likely" they would repeat their vote and vote for the same party they did last year.

Looking closer into the structure of the supporters of this opinion, it reveals that there are more females than males giving this answer in the first, second and third age groups. The fourth age group (aged 40-49) was well balanced in June, however, males dominate in July. In the fifth age group (aged 50 and older) the likelihood of repeating the vote is much more expressed among males than females (59.3 percent M / 40.7 percent F in June and 60.4 percent M / 39.6 percent F in July). Males of rural regions are less flexible in comparison to males from urban areas in giving their vote to a certain party. In June and July, too, much more villagers than people from cities claim that it is very likely they would repeat their vote (62 percent rural / 38 percent urban in June and 60.9 percent rural / 39.1 percent urban in July). On this matter, males and females with high-school education and those with University education are more likely to change their opinion than the ones with elementary or no elementary education. In comparison to June, in July marked a decrease in respondents with high school and/or University education saying they would most likely give the vote to the same parties they did on the last elections.

It is a well-known fact that pubic opinion is very sensitive and flexible. Momentary highs and lows on different popular public issues cannot be measured easily, but Index Kosova in partnership with BBSS Gallup International will continue to follow the monthly trends of changes of public opinion in Kosova.