Перейти на главную страницу





главная страница | наши сотрудники | фотобанк | контакт
 



  Цели и задачи Центра  
  Текущий комментарий  
  Тема  
  Автор дня  
  Социология и политика  

  Аналитика  
  Социологические исследования  
  Публикации и интервью  
  Новости  


Sociological Survey as a Tool for Control of the Results of Elections and Referendums

[31.10.03]

Vladimir Paniotto, Doctor of Philosophy, Director of the Kiev International Institute of Sociology, Nataliya Kharchenko, Deputy Director of the Kiev International Institute of Sociology

Abstract

The conditions at which sociological surveys can be used to control falsification of the elections’ and referendums’ results are analyzed in the paper. On the data of surveys conducted by the Kiev International Institute of Sociology before and after referendum of April 16, 2000, there is proved that the referendum results regarding questions about two-chamber parliament and about taking part in the referendum have been forged.

"Polling - Now more accurate than the election itself!"
The winner of the 2001 T-Shirt Slogan Contest of the American
Association of Public Opinion Research

Development of social surveys and forecasting of elections and referendums results has been viewed by the international community as a significant factor and indicator of democracy of the society. It has been believed that sociological surveys could be used to control infringements in election procedures. If data of surveys significantly differ from results of elections and referendums, it could testify to falsification of these results.

Is it really true? Is this approach sufficiently substantiated? Sociological survey use sampling, i. e., their results are of a probabilistic character. Moreover mistakes could happen not only due to unreliable sampling, but also due to deficient questionnaires, poor training of interviewers, mistakes at data input, etc. Therefore it seems to be more essential to check all stages of the sociological survey against the results of referendums and elections and not vice versa. If there are discrepancies between sociological forecasts and results of elections, first logical assumption is that the data of sociological survey are not reliable.

But we still believe that the international community has been right and that sociological surveys could be a useful tool for assessment of reliability of results of elections and referendums in addition to traditional tools for public control over falsifications (such as presence of observers at election district stations, parallel counting of votes, etc.). The purpose of the present article is to present our observations pertaining to this subject, based predominantly on the results of April Referendum.

Our first assumption is that if results of sociological surveys and elections coincide, it means that whether there was no falsification or the scope of falsification was within the normal deviation for sociological surveys (3-4%). Let’s analyze data of the exit poll 1 conducted by the Kiev International Institute of Sociology, SOCIS company and by the Social Monitoring Center on the first day of the first tour of presidential election of 19992. As it is shown in the Table1deviation between election results and data of the survey does not exceed 3%, which testify that there was no significant falsification (over 3%). Diagram shown below represents data of Table 1 and it demonstrates practically complete conformity between data of the survey and data of the elections.

Diagram. Results of the exit poll versus results of the first tour of presidential elections of 1999

The same pattern relates to the second tour of the presidential elections - the exit poll data predicted 60% for L. Kuchma and around 35% for P.Simonenko - in actuality for Kuchma voted 3% less and 3% more for Simonenko (it is possible that the difference could be ascribed not to the sampling error, but to the fact that some respondents did not want to admit that they voted against acting president). Thus there are no grounds no accuse the authorities of falsification (unless there were more that 3% of those unwilling to admit)

Table 1
Comparison between data of the sociological survey and results of the first tour of presidential elections of 1999

Thus when results of elections (or referendum) are in conformity with results of the sociological survey, it could be assumed that results of elections (referendum) have not been falsified.

Presidential candidates

Survey

results
Election results

(% of actual bulletins)
Deviation
A. Baziluk0.20.10.1
N. Vitrenko11.511.40.1
N. Gaber0.10.10
Yu. Karamzin0.30.4-0.1
V. Kononov0.20.3-0.1
Yu. Kostenko2.62.30.5
L. Kuchma40.238.02.2
E. Murchuk8.98.50.4
A. Moroz9.111.8-2.7
V. Onopenko0.50.50
A.Rzhavskiy0.20.4-0.2
P. Simonenko21.623.2-1.6
G.Udovenko1.91.30.6
Supported no one2.51.90.6
Total1001000

Now let’s analyze the case when data of the sociological survey differ from results of elections (the term ‘election’ also covers referendums in our case). Should one assume that these deviations could be perceived as a sign of falsification of results of elections? This case is much more complicated then the previous one. The answer depends on who conducted the survey, how it was conducted and what was the deviation with elections results.

Who conducted the survey is important in regard of one aspect only - has it been conducted professionally, or not? If not, deviations could be any and they stand for nothing. It should be noted that it is not difficult to differ between professional and unprofessional firms - professional firms are members of professional associations and international organizations (e.g., Sociological Association of Ukraine, European Association of Public Opinion and Marketing Surveys, etc.), they support certain standards for data publication, have, as a rule, a long-term experience and are independent enough (predominantly these are private firms with only a part of their budget allocated for political research); there are about 5-6 firms of this type in Ukraine, not more.

Surveys could also be differentiated in regards of time they have been conducted: before elections, exit polls and those conducted after elections. Data of the surveys conducted before elections (as a rule a couple of weeks before elections, because a standard term for conducting of sociological survey is a couple of weeks) may differ from results of the elections due to changes in the situation that take place in between the survey and the election. Besides, a majority of respondents has not set their mind yet and the sociologists have to make some assumption 3 as to voting and these forecasts are to a major extend uncertain. As it has been shown by the data of our surveys around 15% of the voters make their choice during the last week (see Table 2).

Direct and indirect agitation during the last week before elections are of significant importance (e.g., TV debates with participation of the main candidates).

All this makes it very difficult to use polling conducted before elections for detecting of falsifications. Exit polls are the most efficient means for monitoring of breaches in election procedures. Polls conducted within a couple of weeks after the elections are also an effective tool for control, because we deal with actual numbers. But with this type of survey certain inaccuracy could occur due to peculiarities of the memory of respondents and due to impact of the published results on responses ("observer’s effect", when a part of respondents - 2-3%, do not want to admit they voted for a losing candidate4).

Table 2
Exit poll data on timing of decision making and choices of candidates for the second tour of 1999 presidential elections

When have you made your choice? % of respondents
You always knew for whom to vote50.3
More than 3 months before13.4
More than 1 months and less than 3 months before7.0
More than 1 week and less than 1 month before12.3
Within the last week12.1
Decided today3.2
Difficult to say/ do not know/no answer1.6
Total100

As for deviation rate, 2-3% or even 4-5% could hardly be taken as an evidence for significant manipulations with elections results. And it is not only statistical value of the deviations, but also the possibility of systematic errors. If deviations exceed 7-10% (here we refer to professional sociological surveys) the probability that here we deal not with errors of the survey, but with manipulation of the results is rather high.

Thus, our next assumption is that significant deviations between the data of sociological surveys conducted by professional sociological firms on the day of elections or after and the results of elections could rather testify to possible falsification of results of elections and not only to problems pertaining to the survey. Especially it relates to the cases when several sociological firms generate close results.

We would like to illustrate our assumption with analysis of the situation of the last referendum.

Background On Dec. 7, 1999 in the town of Zhitomir there was a meeting and a group of 26 persons was setup that put forward the initiative of the referendum. On Dec. 23 after short debates the Central Election Committee registered this initiative. More than 300 initiative groups set up by the Democratic Union Party collected 3.3 million signatures from all regions in support of the all-Ukrainian referendum. On January 15 Leonid Kuchma signed an enactment on all-Ukrainian referendum based on public initiative. On March 29 the Constitutional Court recognized the presidential enactment as in its major part conforming to the constitution. But out of 6 questions proposed for referendum the first and the sixth pertaining to dissolution of the parliament by the president in order to overrule veto of the parliament and to adoption of the mew constitution through the referendum were defined as unconstitutional. These questions were not put for the referendum. The remaining four questions (see Attachment) were put on referendum that took place on April 6, 2000. Hrn 30 million (about USD 5.5 million) were allocated to prepare and conduct the referendum from the national budget.

Short political history of modern Ukraine has been accompanied by vivid discussions of procedural breaches and falsification of election results. In regard of the last referendum the complaints may be classified in the following way:

- Outdated and deficient Law on Referendum, lack of formalized procedures stipulating agitation for and against the referendum, absence of responsibilities and sanctions applied against those who was involved in malpractice.

- Inadequacy of questions put for referendum.

- Significant influence exercised by the executive branch of power to manipulate public opinion during election campaign, at voting including forcing to participate in preliminary voting and blunt falsification of voting lists.

- Breaches of procedures that have become traditional such as giving several bulleting to one person, casting votes outside voting cabins, voting without IDs, casting of votes after voting stations have been closed, etc.

We leave the task of analysis of this "blacklist" to political analysts and to legal experts and instead we would try to compare other irregularities with results of sociological surveys.

Wording of the questions put for referendum. Exact wording of the questions is presented in the Attachment. Sociological analysis of the wording of the questions shows the following 6:

- the questions are too long (question one has 93 words, question 2 - 41, question 3 - 41, question 4 -33 words);

- the questions consist of complex sentences and include grammatically difficult words (many syllables, suffixes and prefixes, etc.)

- three questions out of four have references to an additional source (Constitution of the Ukraine) with the presumption that the source is known;

- the respondents have to express their opinion on the problems where majority of them has no expertise;

- the questions include abstract phrases like "interests of the regions of Ukraine", "a constant parliamentary majority", "adequate amendments to election legislation", etc;

- presence of complicated and obscure logical structures within the questions;

- proposed answers do not correspond to wordings of the questions and cannot be considered as attributes of agreement or disagreement;

- use of closed types of questions that exclude additional alternatives such as " do not know", " do not understand the question", etc.

It is interesting that despite the evident difficulties of interpretation of the questions and quite specific patterns for filling the bulletins (if you agree, cross out ‘no’), the number of invalid bulletins was surprisingly low (see Table 3).

Table 3
Comparative analysis of invalid bulletins at referendum (16.04.2000) and at presidential elections (31.10.199 and 14.11.1999)

Items in question



Actual number of votersInvalid bulletins
Number%
Question 1 of referendum296980473936691.3
Question 2 of referendum296947053707631.2
Question 3 of referendum296948613665141.2
Question 4 of referendum296953804168241.4
1st tour of presidential elections2630519810387293.9
2nd tour of presidential elections282124847051612.5

It is difficult to believe that more people made mistakes answering a simple question pertaining to choice between two opposing candidates than answering complicated and obscure questions of the referendum.

Election campaign and mass media. In March - first half of April 2000 the Committee of Ukrainian Voters conducted analysis of publications in the government-controlled mass media pertaining to preparation of the all-Ukrainian referendum. The objective of the analysis was to check whether the principle of equal coverage of different opinions about the referendum was adhered to and also to evaluate coverage of legal aspects of referendum and of voting procedures. Totally 160 editions from 13 oblasts have been analyzed Results of the analysis have been presented as subjective evaluation of the Monitoring group of the Committee of Ukrainian Voters (see Table 4).

Table 4
Coverage of preparation to the All-Ukrainian Referendum in the Government-controlled Mass Media

Coverage Percentage analysis
Number of referencesTotal area allocated for coverage (sq. cm) 5.
Positive71.762.6
Neutral25.924.5
Negative2.312.9
Total100%

N=2852
100%

N=31196

As Table 4 shows that positive coverage of the referendum dominates significantly. A predominant majority of the government-controlled mass media actively supported the idea of the referendum and carried out agitation to answer all the questions positively. Left wing press was in majority in opposition and called to ignore the referendum or to answer the questions negatively. It is interesting to note that very little coverage was given to explanation of the questions put for the referendum and to procedures of voting. Typical headings - "The Referendum is here to Solve Critical Problems", "Ask the People", "It is for the People to Make their Judgement", The Referendum is our Chance to Clean it Up", "Should the MPs Pack their Suitcases?", "It will Suffice to Have 150 MPs", etc.

Level of activity of the electorate and its motivation. Compared to all previous elections starting from 1991 the referendum could be characterized by exceedingly high turnout (see Table 5). As extensive experience of polling shows, after the first surge in political activity during the independence referendum (presidential elections took place together with the referendum then), turnout is usually around 60-75%, and the increase in the level might be achieved through use of some incentives.

Table 5
Turnout of Ukrainian voters in 1991-2000 elections

ElectionsTurnout
Presidential elections 01.12.1991 84.9
Presidential elections (1ST tour) 26.06.9468.0
Presidential elections (2nd tour) 10.07.9469.3
Parliamentary elections 28.03.9863.8
Presidential elections70.2
Presidential elections73.8
Referendum 16.04.200081.2

Activity of the electorate on the eve of April referendum was stimulated by a strong administrative interference into free expression of public will. The Committee of Ukrainian Voters detected a lot of instances of interference by the executive branch of power into the process and breaches of principle of voluntary participation in the elections. Employees of the utilities, schoolmasters and finally police car with loudspeakers were involved in agitation. Most pressure was applied to organized groups of voters (militia, army, prisoners, students; because servicemen and prisoners are not traditionally covered by sociological surveys it is difficult to assess their influence on total results of elections). But the data of the survey show that this pressure had no decisive character. Table 6 shows that less than 55 of respondents consider pressure to be an important reason for their participation in the referendum.

Table 6
Motivation for participation in the referendum

Motives for participation in the referendum% of respondents
Desire to influence decisions important for the country 40.6
I always participate in voting38.1
Desire to express dissatisfaction with MPs 22.4
Agitation from superiors, teachers, etc.4.0
Mass media campaign1.1
Miscellaneous4.0

The use of uncontrolled preliminary voting in the referendum was an interesting strategic move. Pursuant to Clause 35 of the law of Ukraine "On National and Local Referendums" and decree of the Central Election Committee No. 4 of January 2000 if citizens change their place of residence, they are entitled to cast their votes from April 6 to April 15. But many instances of applying pressure to force people to participate in preliminary voting, made the Committee of Ukrainian Voters to issue a special resolution condemning this practice. According to CUV’s data in different regions of Ukraine 8-12% of voters had cast their votes before April 10, 2000 (see table 7).

Table 7
Central Election Commission data on timing of voting at the All-Ukrainian referendum of April 16

Time of votingNumber of voters% of voters
As of 8.00 (preliminary voting)1373995437.5%
As of 12.001984049354.2
As of 16.002455238167.0
As of 20.002886075078.8
Total (registered voters)36639876100

As it could be seen from Table 7, according Central Election Commission data approximately every third voter has cast its vote during preliminary elections. These data are in conflict with the data of the sociological survey conducted by KIIS after the referendum (see Table 8)

Table 8
Have you participated in All-Ukrainian Referendum of April 16, 2000?

AnswersDistribution% from total number of voter
Voted on April 16105457.06
Cast my vote during preliminary voting1357.4
Members of my family voted for me432.4
Did not participate in the referendum59632.6
Do not know, difficult to say10.0
Total1829100

Even under assumption that all voting on behalf of the family member took place during preliminary voting and with adjustment for sampling error the share of those who participated in preliminary voting amounts top 5-15%, which differs significantly from the data of the Central Election Committee.

Now we would like to analyze dynamics of voting at the referendum (see Table 9)

Table 9
Summary of Central Election Committee data on voting at referendum

Time of votingTurnout at the referendumVoting on 16.04.2000 (without votes cast during preliminary voting)
Preliminary voting1373995447.6%- -
8.00-12.00610053921.1%610053940.4%
12.00-16.00471188816.3%471188831.2%
16.00-20.00430836914.9%430836928.5%
Total actual number of voters28860750100%15120796100%

As we can see half of the voters has cast their votes during preliminary voting. Another fact that draws attention is uncharacteristic for Ukraine high turnout in the afternoon. For the sake of comparison we would like to present data on turnout during other elections (see Table 10)

Table 10
Comparison between turnout at referendum of 16.04.2000 (without preliminary voting) and at presidential elections (31.10.1999 and 14.11.1999)

Time of voting1st tour of presidential elections2nd tour of presidential electionsReferendum of April 16, 2000
8.00-12.0043%38%40%
12.00-16.0038%41%31%
16.00-20.0019%21%29%
Total registered number of voters100%100%100%

There is practically no difference in the share of disciplined voters that cast their votes in the morning. In the afternoon a significantly smaller number cast their votes compared to previous elections. But in the evening there was an uncharacteristic surge in activity during referendum. Certainly one could assume that tireless gardeners decided to fulfill their civic duty upon return from their dachas. But this hypothesis is in contradiction with the data of the KIIS sociological survey conducted after the referendum at which only 10%±%5% of the respondents stated that they cast their votes at 16.00 or later.

Now we would like to compare the data of sociological survey and those of referendum. The survey conducted by KIIS a couple of week before the referendum (from March 17 to March 27) allowed us to make an assumption that the turnout for the referendum would be around 60% and that all the questions proposed for the referendum would be supported (for the first three we envisaged strong support) The results of the referendum surprised, first of all by a very high turnout and, second of all, by the fact that an obscure question of two-chamber parliament was supported not by 60%, which would have been in agreement with our forecasts, but by more than 80% of voters.

Table 11
Share of turnout and distribution of responses for the questions of the referendum, %

Questions for referendumForecastActualDifference
1. Preliminary cessation of authority of the parliament7785-8
2. Suspension of MPs’ immunity8889-1
3.Downsizing the number of MPs97907
4. Two-chamber parliament6382-19
Total turnout6181-20

Since it has been established KIIS participated in forecasting of dozens of elections and referendums (referendum of 1991, the first and the last presidential elections, parliamentary elections). As a rule, maximum deviation from actual does not exceed 3-4% (with the exception of the second tour of the presidential elections of 1996, where the difference of the survey completed a week before the elections was 8% and an additional survey was undertaken to explain the difference). That is why deviation of 20% could hardly be ascribed to errors of the survey.

To explain the difference we put forward three hypotheses:

1) our forecasts were based on the assumption that voting would take one day and not two weeks, which happened in reality;

2) within the couple of weeks between the survey and the referendum the electorate had been mobilized with the use of certain tools, starting from a TV campaign and finishing with administrative pressure at the enterprises;

3) results of the referendum have been falsified.

In order to check the hypothesis we conducted a special survey (it was carried out in all oblasts of Ukraine and in the Crimea from May 15 to May 26 as a multistage sampling, random for all stages with sampling size 2101). We asked our respondents to answer if they participated in the referendum and how they voted for two-chamber parliament. Table 12 compares results of the referendum with data of the survey.

Table 12
Comparison between results of the referendum and data of survey conducted before and after it

Categories of electorateForecast, %Referendum, %Survey after referendum, 5
Those that participated in referendum6181.267.4
Those that put ‘yes’ for question 4 (about two-chamber parliament)6381.761.3

As we can see the first two hypothesis could not fully explain the difference of 13.8% on the basis of the share of those that participated in referendum (even though increase in 6.4% in the share of those who answered that they participated in the referendum compared to the forecast could be explained by "prolonged" time of the referendum and by administrative pressure). As for the support to idea of two-chambered parliament these hypothesis could provide explanation for only a small fraction of 20.4% difference between the results of referendum and data of the survey after the referendum.

Thus data of the survey after the referendum give ground to support the third hypothesis - falsification of the results of referendum concerning two-chamber parliament and concerning turnout

We would like to present data on referendum turnout provided by survey and by the Central Election Commission separate for every oblast of Ukraine and also for the city of Kiev and for the Autonomous Republic of Crimea (see Table 13).

Table 13
Share of those that participated in the referendum

Administrative divisionsResults of referendumSurvey after referendumDifferenceSignificance of deviation
Zakarpatskaya oblast96.7765.1331.64¨
Donetskaya oblast86.4155.9830.43¨
Khersonskaya oblast70.3048.6621.64-
Odesskaya oblast76.9456.0620.88¨
Chernovitskaya oblast78.6057.7520.85-
Sumskaya oblast87.4269.8217.60-
Dniepropetrovskaya oblast85.2069.9315.27¨
Kievskaya oblast78.9065.8013.10-
City of Kiev69.6557.7411.91-
Kirovogradskaya oblast78.1166.3411.77-
Autonomous republic of Crimea61.1449.4911.65-
Ivano-Frankovskaya oblast88.4377.8810.55-
Lvovskaya oblast82.2772.519.76-
Luganskaya oblast83.9074.619.29-
Cherkasskaya oblast74.6666.08.66-
Rovenskaya oblast86.8478.957.89-
Kharkovskaya oblast78.6374.843.79-
Nikolaevskaya oblast80.8077.243.56-
Zaporozhskaya oblast66.1363.242.89-
Khmelnitskaya oblast71.4969.172.32-
Zhitomirskaya oblast76.2475.260.98-
Vinnitskaya oblast75.3774.630.74-
Poltavskaya oblast75.3875.64- 0.26-
Chernigovskaya oblast70.2870.72- 0.44-
Volynskaya oblast78.8079.44-0.64-
Ternopolskaya oblast87.2488.56- 1.32-

¨ - difference between the referendum and the survey exceeds 99% interval of reliability (that is the probability of deviation equals 0.99)

The analysis shows that in four oblasts (Zakarpatskaya, Donetskaya, Odesskaya and Dniepropetrovskaya) the difference between the results of referendum and the data of survey has reliability rate of 99%. This means that occurrence of the difference in these oblasts is possible only in one out of 100 surveys.

Now let’s consider the question of introduction of two-chamber parliament (Table 14). As it is shown the difference there is even more significant. The differences for 16 oblasts has reliability rate of 95% (for 13 oblasts it reaches 99%).

Table 14
Share of those who voted for introduction of two-chamber parliament

Administrative divisionsResults of referendumSurvey after referendumDifferenceSignificance of deviation
Kirovogradskaya oblast82.8838.5144.37¨¨
Volynskaya oblast86.5048.9437.56¨¨
Chernigovskaya oblast77.56.40.0537.51¨¨
Dniepropetrovskaya oblast77.9542.2435.71¨¨
Rovenskaya oblast89.3156.2533.06¨¨
Zakarpatskaya oblast94.4862.3732.11¨¨
Zhitomirskaya oblast80.5850.4730.11¨¨
Odesskaya oblast85.2456.4728.77¨¨
Kievskaya oblast81.4755.2126.26¨¨
Vinnitskaya oblast79.1254.1324.99¨¨
Donetskaya oblast89.5264.8624.66¨¨
City of Kiev65.1740.5224.65¨¨
Cherkasskaya oblast82.8859.3323.55¨
Lvovskaya oblast85.3262.5322.79¨¨
Ivano-Frankovskaya oblast92.1569.8522.30¨
Kharkovskaya oblast78.4856.3622.12¨¨
Khmelnitskaya oblast79.5961.9117.68-
Zaporozhskaya oblast72.4356.6515.78-
Sumskaya oblast88.2572.7315.52-
Luganskaya oblast85.8675.0410.82-
Nikolaevskaya oblast80.2672.77.56-
Ternopolskaya oblasts91.6091.150.45-
Chernigovskaya oblast85.6686.60-0.94-
Autonomous republic of Crimea67.2468.57-1.33-
Poltavskaya oblast70.7180.82-10.11-
Khersonskaya oblast72.5390.76-18.23-
Total81.6861.3230.36¨¨

¨ - difference within 95-99% reliability interval.

¨¨- difference between the data of survey and the results of the referendum exceeds 99% reliability interval (that is the probability for difference to occur equals 0.99).

Thus it is possible to assume that in the said oblasts the results of the referendum relating to turnout percentage and percentage of those that support introduction of two-chamber parliament have been falsified.

There is a possibility that falsification also happened in other oblasts, but the size of sampling in these oblasts is insufficient to statistically substantiate this conclusion.

Our assumptions certainly could not be perceived as legal grounds for accusation in the breach of law. It is just a hypothesis with some statistical substantiation. Alongside with that the data of the survey show that there is no ground to doubt that more than 50% of population participated in the referendum (i.e., it has been valid) and that positive answers for the first three questions (and most probably for question 4) put for the referendum has been supported by the Ukrainian electorate.

Summing up all above said, one could reach a conclusion that sociological surveys (especially those conducted on the day of elections and after the elections) could be used as a tool for control of malpractice and breaches that take place during elections, and thus one of the mechanisms that contribute to democratization of our society.

Attachment

Results of voting at the all-Ukrainian referendum of April 16, 2000

Total number of citizens entitled to vote36629926100%
Number of citizens that received bulletins2972857581.15%
Number of citizens that cast their votes2886075078.77%

1. Do you support amending Article 90 of the Constitution of Ukraine with Clause 3 of the following wording: "The president of Ukraine is entitled to suspend the authority of the Supreme Rada of Ukraine before its term expires, if the Supreme Rada of Ukraine within one month fails to establish a stable parliamentary majority or if within three months the Rada fails to adopt the national budget duly drafted and submitted by the Cabinet of Ministers of Ukraine" which shall give additional grounds for dissolution of the Supreme Rada of Ukraine by the President of Ukraine and Paragraph 8, Clause 1, Article 106 of the Constitution of Ukraine shall attached with the following wording : ‘and for other instances envisaged in the Constitution of Ukraine"?

Total number of citizens that participated in the referendum36 629 926 (100%)
Number of citizens that received the bulletins29 728 575 (81.15%)
Cast their votes:

29 698 047
"Yes"

25 177 984

(84.69%)
"No"

4 126 394

(13.89%)
Invalid bulletins:

393 669 (1.33%)

2. Do you support the necessity to limit immunity of the people’s deputies of Ukraine and suspension of Clause 3, Article 80 of the Constitution of Ukraine: "People’s deputy of Ukraine shall not be criminally prosecuted, detained or arrested without consent of the Supreme Rada of Ukraine"?

Total number of citizens that participated in the referendum36 629 926 (100%)
Number of citizens that received the bulletins29 728 575 (81.15%)
Cast their votes:

29 694 705
"Yes"

26 461 382

(89%)
"No"

2 862 560

(9.64%)
Invalid bulletins:

370 763 (1.33%)

3. Do you support downsizing of total number of people’s deputies of Ukraine from 450 to 300 and introduction of a corresponding amendment to Clause 1, Article 76 of the Constitution of Ukraine - words: "four hundred and fifty" shall be changed for "three hundred" and also introduction of corresponding amendments to the election legislation?

Total number of citizens that participated in the referendum36 629 926 (100%)
Number of citizens that received the bulletins29 728 575 (81.15%)
Cast their votes:

29 694 861
"Yes"

26 730 432

(89.91%)
"No"

2 597 915

(8.75%)
Invalid bulletins:

366 514 (1.23%)

4. Do you support the necessity to establish two-chamber parliament in Ukraine, with one of the chambers representing interests of the regions of Ukraine and facilitating their implementation and introduction of corresponding amendments to the Constitution of Ukraine and the election legislation?

Total number of citizens that participated in the referendum36 629 926 (100%)
Number of citizens that received the bulletins29 728 575 (81.15%)
Cast their votes:

29 695 380
"Yes"

24 284 220

(81.68%)
"No"

4 994 336

(16.82%)
Invalid bulletins:

416 824 (1.4%)

Data sources:

1. Web server of the Committee of Ukrainian Voters

2. Web server of the Central Election Commission of Ukraine

3. KIIS public opinion surveys.


1Exit poll – polling of voters right after they cast their votes.

2The poll was contracted by the Democratic Initiatives Fund with technical assistance of QEV Analytics (Washington D.C.) and funded by Charles S. Mott Fund (USA). 6000 respondents were interviewed in all provinces of Ukraine and in the Crimea (multi-staged, random sampling).

3One of the assumptions we use is that with a large number of candidates (in the first tour) those uncertain would be voting same as those who set their mind, while in the second tour those uncertain would be voting according to their orientations.

4Comparison of the results of KIIS surveys conducted at the eve and after presidential elections of 1994 and 1999 clearly show absence of this effect.

5the data are given for 7 oblasts only: Cherkasskaya, Khmelnitskaya, Rovenskaya, Khersonskaya, Chernovitskaya, Ternopolskaya, and Kharkovskaya.

6the analysis of the questions has been conducted by Yu. V. Gavrilenko in his qualification thesis for the degree of Master of Sociology “Factors that Influence the Quality of Procedures, Tools and Results of the Sociological Survey”


Социология и политика

Рубрики: 





Рейтинги партий - [25.10.2008]



Подрубрика: Реакция на текущие события. - [17.08.2007]



Результаты эксперимента с техникой secret ballot в изучении электорального поведения украинских граждан - [26.05.2004]



Проблемы "рейтингологии" - [26.05.2004]



Sociological Survey as a Tool for Control of the Results of Elections and Referendums - [31.10.2003]



A Spatial Analysis of Ukraine`s 1998 Parliamentary Elections - [30.10.2003]



Загадка рейтингов - [11.10.2003]



Что могут и чего не могут социологи - [30.05.2002]















Copyright © 2002-2012 Киевский центр политических исследований и конфликтологии
Copyright © 2002-2012 Центр эффективной политики

При использовании материалов сайта ссылка на источник обязательна.






bigmir)net TOP 100