INDICATORS OF INTERNATIONAL COMPARISONS

In the course of the opinion poll respondents were asked questions pertaining to international comparisons. For example, they were offered to compare situation in Russia and Belarus as well as in the EU countries (See Tables 1-3).

To what extent is assessment of living in Belarus vs. EU/Russia and of democratization in Belarus vs. Russia given by respondents is based on other answers, on their viewpoints of various sides of social and economic living? Since Belarusians don’t travel abroad much, this assumption is fairly reasonable – as a rule this or that assessment appears irrespective of the fact that a person went abroad and compared. More likely, it is based on evaluation of Belarusian reality.

This is a classic task of the discriminant analysis. Data in Tables 1-3 will stand dependent variables in our research. For simplicity and clearness of analysis we shall in each case exclude indefinite and interim answers while determining the choice of marginal answers: EU or Belarus for the variable v103 and Belarus or Russia for variables v118 and v125.

We chose eight other variables as explanatory and independent ones. Their values are given in Supplement. The variable "Age" in the current model is presented in the same way as in Supplement while variables v78 and v151 were transformed into dichotomy (v78: "Don’t want to move anywhere" – 1, other variant of answer – 0; v151: "Use the Internet daily / several days a week" – 1, other variant of answer – 0.) In the remaining five variables indefinite answers are given the value equaling to the average on definite answers.

As far as explanatory variables have only two values in the current model, the objective of discriminant analysis is finding linear combinations of explanatory features which separate (or discriminate) the best the values of explanatory variables.

We carried out the discriminant step-by-step analysis in this research with the SPSS data processing program. Its results are given in Table 4.

Data of the first two lines in Table 4 show adequacy of discriminant models: canonic correlations are the correlations between values of explanatory variables and values of discriminant functions; discrimination accuracy is the part of correct forecasts among the number provided by discriminant function. All the indicators except for canonic correlation for the variable v125 are fairly high. The third line of Table 4 shows, to put it simple, big positive and big negative values of discriminant function which correlate with the values of explanatory variable. The final part of Table 4 presents the contribution which separate variables make in discrimination. In the discriminant analysis, the weight of a variable in discrimination is determined by the absolute value of standardized coefficients – the bigger is the coefficient, the stronger a variable assigns observed elements to this or that group.

Before proceeding to the results of analysis, we should like to note that the variables excluded step by step from the final model are not necessarily having impact on the explanatory variable. Their influence is mediated and generated by the variables that remain in the model and exert the greatest influence on the explanatory variable.

As it follows from Table 4, previously independent variables appear important for each of the explanatory variables. Variable v78 showing attitude to emigration has the biggest in absolute value coefficient in the model for the variable v103. Interpretation lies on the surface – it is natural for the respondents willing to leave the country (the majority of them would like to move to the EU countries) to assess the living standard in Belarus as lower than in the countries they want to move to.

Variable v1 shows likewise material grounds of assessment – feeling of improving living standard increases disposition to Belarus while feeling of declining living standard prompts to the opposite evaluation.

Ideological assessments also make a great contribution – faith into right course for Belarus and satisfaction with A. Lukashenko’s rule considerably influences the preferences in favor of the Belarusian living as compared to European.

Noteworthy in this regards is the contribution made by variable v46 – assurance in that entrepreneurs serve for the good in Belarus takes the comparative assessment of welfare in favor of the EU while assurance that entrepreneurs only harm – in favor of Belarus.

Age doesn’t exert much influence and it appears quite predicted – the older is the respondent, the most probably he/she will give preference to Belarus and not to the EU.

When it comes to comparison of living standards in Russia and in Belarus, the number of significant variables sharply decreases to two. The greatest influence has the variable pertaining to purely political evaluation of A. Lukashenko’s governance. The second significant variable on this issue is again variable v78 (attitude to emigration). Interpretation is less obvious here – the large part of respondents willing to emigrate is not going to Russia. Apparently, if a respondent wants to move from Belarus, he/she doesn’t rate high the various aspects of living in Belarus as compared to the countries he/she would like to move to.

Ideological variable v17 exerts the greatest influence on variable v118 which is comparison of the democratization level in Russia and in Belarus: faith in the right course carried in Belarus generates higher estimates of the Belarusian democracy as against Russian democracy.

Variable v78 works in a similar way. Remarkably, variable v46 characterizing attitude to entrepreneurship shows up in the same way when comparing living in Belarus vs. the EU and democratization in Belarus vs. Russia: the opinion that entrepreneurs serve for country’s good generates preference in favor of foreign countries while belief that businessmen serve to country’s detriment disposes in favor of Belarus.

Variable v151 pertaining to the use of Internet appears very important for explanation of variable v118. Influence of this variable is obvious – those who often use the Internet give preference to Russian process of democratization. It should be noted that this factor appears insignificant in the model explaining variable v103 – comparison of Belarus vs. the EU. This is clear taking into account that these are not European or American but Russian web sites which are the most popular among Internet users. These people are deeply involved into the Russian informational space, which is one of Russia’s most democratic segments.

In our opinion, results of the discriminant analysis prove the assumption made in the beginning of this article that population looks at many practical sides of everyday life, in particular its advantages and disadvantages as compared to foreign countries, through ideological spectacles, and preferences given to these or those political situations within the country generate respective comparative assessments.

Back   Top