Alexandre Avaliani (MA)

PhD student

Assistant (School of Social Sciences)

GIPA – Georgian Institute of Public Affairs

Tbilisi, Georgia


The paper aims to contribute to laying the foundation for the scientific, evidence-based and systematized Georgian measurement instruments of ideology and political orientation. As a very small number of such tools (e.g., questionnaires, scales etc.) that exist at the moment are either philosophical-artistic artifacts based on a priori, rational reasoning of the authors or are simply (inappropriately adapted) translations of a foreign analogues, it is especially important to create a scientifically supported foundation for the given field. It is also important to systematize existing information. The main goal of this article is to determine what issues are key to measuring ideology or political orientation in Georgia and what is the optimal structure and format of the assessment tool, which would employ these issues. The present paper contributes to the cause by qualitative analyses of existing literature and inventories. It outlines the optimal model of conceptualizing ideological issues, as well as the hierarchy of importance of the issues and constructs that would be used in such models. Other important recommendations are also presented.

Keywords: Georgia, Ideology, Political Orientation questionnaire/inventory/test/survey, qualitative analysis  

Introduction and Literature Review

Ideology is both a prism, through which our perceptions and opinions are colored, and an engine that propels our actions. It is one of the most fundamental frameworks, which shapes descriptive, normative, and behavioral parts of our mind. It guides and reflects almost every aspect of life both at an intrapersonal (individual) and an interpersonal (societal) level. Thus, it’s no wonder, that this concept is and has been for a long time a hot topic across various disciplines, thus attracting professionals from various backgrounds (including philosophers, experts of political science and international relations, sociologists, psychologists, journalists, historians, literary scholars and so on). Moreover, these themes engage not only professionals and experts, but laypeople too, who find their everyday life enormously saturated with ideological discourse.

Though ideology is most often thought of as an instrument of influence, it is also an analytical tool, which helps us understand the vastly complex world. By providing simplified (though, on the other hand, apparently distorted) representations, it produces intelligible concepts, patterns and systems, which are necessary in understanding relationships between the countless variables they represent.

This aspect and function of ideology is even more evident in the case of political orientation, as it is even more simplified version of just one aspect (namely, political) of it. Political orientation[1] aims to reduce the incomprehensibly infinite variability of political (mostly ideological) issues to the few factors/constructs[2].

Thus, the importance of having a good measurement instrument[3] to assess political ideology and/or political orientation cannot be overstated, as effective assessment tools provide researchers, policymakers, and the general public with valuable insights into the complex landscape of political ideologies, enabling informed decision-making and facilitating a deeper understanding of societal dynamics. Reliable and valid instruments are crucial for comparability, accumulation of knowledge, prediction, and decision-making (including design and implementation of better policies).

Despite this, from a psychometric perspective, relevant research and practice falls significantly short of the benchmark. In a recent analysis conducted by Azevedo and Bolesta (2021), examining 400 scientific publications spanning from 1930 to 2020, they identified 358 unique ideological instruments, wherein they found:

  • “Substantial variance in scoring and scale type even within identical scales” with “high frequency of incomplete reporting of the items used”;
  • Validity evidence, statistical or psychometric technique, extraction or rotation methods being sparingly reported;
  • „most instruments being either on-the-fly measures (18.16%) or an ad-hoc combination of items (30.17%) present in existing, publicly available surveys“;
  • Weak overlap in topics of 10 most popular scales (with Jaccard index of .2 and .33);
  • And that these problems do influence variability in results.

Georgian instruments are even more problematic. As of now, only 5 such scales have been identified (by the contributors of this study).

The oldest one in the list is the questionnaire used in a nationally representative survey (Kachkachishvili & Mezvrishvili , 2003). The instrument was comprised of only 7 items, which were intended to measure 3 proposed constructs. Though, the close investigation of these constructs and the variables reveals that its content may (almost exclusively) be explained by just one variable – left-right politics (mostly – 92% – related to economics). This is also indicated by the fact that the questionnaire also included only one unidimensional (left-right) self-placement scale. The model is questionable, not only from a logical perspective, but also from empirical as indicated by the authors themselves and also in another study (Avaliani, 2018).

Next one in the list is the Political Compass (Ramishvili, [ca 2007]), published on the web domain of Free University, which seems to have no scientific or any other kind of article or report describing its psychometric characteristics (or any kind of quality). The questionnaire is too long (96 items[4]), while (supposedly) measuring only 2 constructs. In addition, the face validity of this structural model seems fairly controversial.

One of the instruments is based on a master’s thesis (Avaliani, 2018). Although it is partially based on questionnaires (Henningham, 1995; Everett, 2013; Wilson & Patterson, 1968), which are the most established in the field (Azevedo & Bolesta, 2021), it lacks reliability and validity (both internal and external), because its psychometric properties were studied with neither ideal methodological rigor nor on the ideal sample; Most of all, the process of translation and adaptation was inconsistent with the standards.

The fourth in the list is the Europe-Georgia Institutes “Political Compass” (2021). The associated article of this instrument was not found either[5]; Moreover, even an article dedicated to the American original (Individual Differences Research – IDRlabs, 2017) from which it is translated and adapted could not be found. This questionnaire also has a very questionable theoretical structure, as its factors are quite different from other more established instruments. Additionally, in the Georgian version of the instrument, the Diplomacy “value/axis” (factor), which measures juxtaposition of nationalism and globalism, was replaced with Antiwesternism-Pro-westernism, a new dimension called “Social”, which is a strange fusion of economic and public equality, was added, and the number of items was nearly halved. In addition to diverging from the original, it is noteworthy that the axis of this Foreign Policy, in my opinion, has almost nothing to do with ideologies. Furthermore, it is unclear what is the difference between Social issues and the axis of the Economy, as both are clearly about equality and equity. Discriminatory validity of the Public and Civil axes is also vague: usually progress and freedom, as well as tradition and control are closely linked, both philosophically as well as (if not more) empirically – with the psychometric investigation of surveys. This instrument is much shorter than the previous one (consists of 39 items), but still far from ideal in terms of conciseness.

The last one in this list is “Electoral Compass” (International School of Economics at TSU (ISET), CRRC Georgia, Georgian Institute of Politics (GIP), 2020; Papava & Tevdoradze, 2020; Kakhishvili, Keshelava, Papava, & Sichinava, 2021), which is a Georgian version of Kieskompas. The original of this questionnaire is of Dutch origin, although it has been translated into many languages and is used in at least 40 countries (ABOUT KIESKOMPAS, 2021). As for the Georgian version, it is clearly superior to the rest of the Georgian counterparts from a scientific point of view, although it still does not meet the desired standards. Let’s start by saying that although its sample is huge (19,000 respondents), which is certainly admirable, it was conducted in a non-probability manner, which limits the generalizability of the data, strictly speaking, to the number of respondents themselves. Also, this paper is characterized by one oddity: it presents two factorial structures of issues; one of them (which seems to have been obtained by non-scientific/non-psychometric methods) is used in the largest part of the paper, and the other (which was obtained by factor analysis) is used in the remaining, very small, cases. In addition, it should also be said that this factor analysis is not done properly, in my opinion.

Furthermore, such tools are very sensitive to time and culture. A clear example of this is that most of the issues from what was (and still is) apparently the most popular and well-established questionnaire for the study of ideology (Wilson & Patterson, 1968) were found to be completely obsolete after a couple of decades (Everett, 2013). Therefore, it is imperative to regularly update them and adapt them (to align with the peculiarities of the given culture).

As noted above, the purpose of political orientation is to devise a comprehensive model (with few intelligible constructs), that would explain a big part of the political variables. This can be done in two ways. One way is more qualitative and philosophical, based on a priori investigation of content; this approach mostly entails examination of face validity, which (preferably, mostly) is based on the experts’ opinions (e.g., philosophers, historians, or political scientists), about what does any given ideology or political orientation encompass, and what kind of structure would best characterize it[6]. The second approach is empirical, relying on actual evidence of how the ideology is manifested in a (broader) population at a given time. The clearest manifestation of the latter is the creation of constructs and questionnaires with psychometric methods. One big, yet not so outwardly apparent, difference between these approaches should be noted: experts often recognize a consistency between issues where ordinary people do not see them, which leads to inconsistencies between concepts derived from these different methods.

The simplest, oldest, and the most widespread model of this is the one-dimensional left-right depiction of ideology. An example of this is the most popular and established Wilson and Patterson Conservatism Scale (1968) and its updates and variations, which also occupy leading positions in popularity (Everett, 2013; Henningham, 1995).

Despite the establishment and persistence of this tradition, the limitations of this model are increasingly evident in contemporary times, as the issue of multidimensionality of political orientation becomes more and more acute. The most minimal step in terms of the abundance of dimensions, which is widely recognized by most modern scholars, is the division of the aforementioned dimension into economic and social issues[7]. Occasionally, even researchers who support and use a one-dimensional model present evidence and arguments in favor of the left-right spectrum consisting of two (relatively independent) variables/factors. For instance, Jost, Glaser, Kruglanski, and Sulloway (2003) suggest that the core characteristics of conservatism[8] are resistance to change and acceptance of inequality[9], which they consider to be quite independent.

The separation of these two dimensions is particularly crucial in Georgia, because in this country, they are not merely sufficiently independent, but even exhibit a moderate negative correlation (Kakhishvili, Keshelava, Papava, & Sichinava, 2021). This pattern holds true not only in Georgia but also in Eastern European, and especially post-Soviet countries (Rovny & Edwards, 2012; Marks, Hooghe, Nelson, & Edwards, 2006).

There are many other proposed factors and axes (e.g., pacifism-militance, multiculturalism-assimilationism, multilateralism-unilateralism, free trade vs protectionism, etc.), but they don’t come close to significance of the 2 abovementioned factors for several reasons. Firstly, these 2 explain an immeasurably bigger variance than others. Secondly, they map onto political ideologies better[10]. Thirdly, they are relatively universal (across many variables; e.g. time and space)[11]. Lastly, these 2 constructs are mostly reflective by nature (as opposed to formative), and, thus, can be represented by axes/continuums with polar opposites, which a big advantage (or, in many cases, a necessity) for accurate and convenient measurement. This provides a sound rationale for why these 2 factors have been, are, and should be prioritized. Nevertheless, adding more factors than 2 or breaking them down into more concrete ones does make an instrument more precise and extensive in scope, though at a price of conciseness and/or comprehensibility.

Meanwhile, despite all this, political orientation (frequently, mistakenly even called “ideology”) is very often measured by single item self-report/placement scales, requiring participants to rate how right-wing/conservative or left-wing/liberal they are. Besides structural inadequacy, this approach has other serious problems; namely, this method implies that: 1) the ideological labels that represent latent political constructs (e.g., “conservatism” or “liberalism”, to name the most prevalent ones) are consistently defined and operationalized (at least, throughout relevant literature experts), and 2) respondents are proficient enough to understand the meaning of these labels. These assumptions are apparently false. Even the most prominent theorists disagree about the nature of these concepts; moreover, such inconsistencies are prevalent within the works of a single author, or even individual texts themselves. As for the second assumption, its falsehood is even more evident especially in Georgia (Kachkachishvili & Mezvrishvili , 2003; Avaliani, 2018), where general population (including both parties and the political elite) is widely uninformed about ideological matters (Barkaia, Kvashilava, Gogoladze, Kobalia, & Chkhikvadze, 2020; Institute of Social Studies and Analysis (ISSA), 2016; Tavakarashvili, 2018; Jibladze, 2019; Melikidze, 2017; Andrea, et al., 2021; Kachkachishvili & Mezvrishvili , 2003; Tsitsishvili, 2011; The Caucasus Research Resource Centers (CRRC) Georgia, 2023; Kakhishvili, Keshelava, Papava, & Sichinava, 2021).

Goals and Objectives

To address the problems stated above (concerning ideology research in Georgia), the main goal of the study is to create a scientifically rigorous foundation and research framework that will pave the way for elevating the scientific level of ideology research. This should include both a descriptive section supported by solid evidence, as well as specific guidelines and recommendations for conducting relevant research. To achieve this goal, ideological[12] instruments will be analyzed qualitatively; this will be done on both level of factors (critiquing, comparing, and integrating various factorial models) and items (integrating all items into the appropriate thematic groups). As a first step, optimal models and practices on an international level will be derived from the relevant literature. This framework will guide and shed light on the subsequent phase, which will focus on Georgian context. Lastly, information accumulated from all sources will be consolidated and summarized.


As alluded to previously, one way to identify the key issues for measuring ideology or political orientation and fit them into convenient models and taxonomies is to examine existing alternatives. In this case, we rely more on the opinion of experts. This step is essential when creating or adapting instruments to establish face (and, partially, content) validity, but it is particularly important in this context because ideology is a complex issue and requires a high level of competence to understand it correctly. This has an even greater impact in Georgia, where the corresponding knowledge at a level of general public is very limited.

Establishing content validity is most commonly done through expert panels or surveys. However, analyzing questionnaires and similar tools presents a superior approach for several reasons. Firstly, questionnaires inherently incorporate the step of considering expert opinions, resulting in a broader range of expert input when multiple instruments are analyzed compared to a single study or paper. Additionally, pre-existing models often undergo additional validation stages such as peer review, or other type of feedback from colleagues and experts, further enhancing their refinement.

This method is widely regarded as a reliable and convenient source of information for such research. A notable illustration of its effectiveness is evident in case of the most well-established taxonomy of personality – the Big Five Factors – and the measurement instrument widely recognized for assessing these factors, the NEO-PI(-R), both of which are mostly derived using this method (Costa Jr. & McCrae, 2008).

Particular research methods will be presented at the outset of each section.

Factor-level analysis

This section will focus on investigating and integrating the factorial structure of existing instruments. Integration will ultimately result in a taxonomy, structured according to the following hierarchy: Taxons > Sub-Taxons > Topics/Categories/Issues (> sub-categories) > items[13].

The analysis and summary of an


 accumulated knowledge in the field will be based on the research briefly reviewed in the introduction (Azevedo & Bolesta, 2021), in which 400 scientific works (1930-2020) from 92[14] countries were reviewed.

This paper presents two relevant lists developed by analyzing the 10 most important questionnaires: 1) 32 categories[15] formed by open coding[16], and 2) 16 topics extracted by deductive category formation. For intents and purposes of the current study these lists are too concrete (with regards to the level of abstraction) and, thus, too lengthy[17]. Because of this, they were grouped into broader categories, following the same rules stated above. The merged taxonomy (at the broadest level of abstraction) of both methods[18] looks like this:

  1. Social issues[19] (35/140)[20]
    1. Equality-equity (109)
      1. Cultural and national (40)
      1. Universal (19)
      1. Sexual (18)
      1. Gender & sex related (8)
      1. Racial (7)
    1. Freedom (31)
      1. Purity of body (14)
      1. Freedom of speech and expression (5)
      1. Attitudes towards deviants (5)
      1. Prohibited substances and objects (4)
      1. Education (3)
  2. Economic issues (29/62)
    1. labor rights (36)
      1. Employer rights (22)
      1. Employee rights (14)
    1. Welfare (28)
  3. Foreign Policy (11/16)
  4. Domestic Politics (20/4)
  5. Other (7/3)

Besides the obvious, this analysis once again reinforced the notion stated previously, that constructs other than economic and social right and left[21] are mostly formative[22], as the issues included in those were clearly uncorrelated. Moreover, almost all of the topics included in these 3 taxons can be, and thus should be (for the reasons stated in the literature review), included in one of the first 2 taxons, making these 3 almost completely meaningless; This is less pronounced in the case of foreign rather than domestic policy.

As for the


 instruments, one extension was made (beyond the models given by authors); namely, the factorial structure of the Election Compass was investigated independently from the authors, with the open data provided by them. First 2 factor analyses were more confirmatory in nature[23]. The results did not support the models proposed by authors. The 3rd analysis was completely explanatory and resulted in a significantly different structure. This model explained 51.879% of variance, with 5 factors[24]. First 2 factors were one again economic and social freedom-etatism. The third one was conceptualized as in-group bias[25], as it spanned across various thematic groups, but offered a consistent semantic motif of opposition to other (e.g., ethnic or religious) groups and endorsement of the group one belongs to. 4th factor was a mix of 3 judiciary independence[26] and 2 economic freedom items; as the latter 2 did not offer any conceptual novelty, only the former part was retained. 5th factor did not offer any unique variance (besides one item with .55 loading).

In sum, it has to be said once again, that individually all of the structural models were more or less flawed in regard to both logical consistency and empirical verification. Nevertheless, somewhat consistent patterns could be observed. Some models tend to overgeneralize, while others are inclined to the opposite bias; If were to sum them up though, these tendencies would cancel each other out and result in a balanced structure, that would look like the one presented below.

Combined constructs/taxons of all instruments:

  1. Economic Freedom[27]
  2. Social Freedom
  3. Foreign Policy: Pro-western vs Independence and/or Pro-Russian[28]
  4. In-group Bias
  5. Judiciary

In sum, analysis revealed that, the most basic dimensions clearly appear to be social and economic freedom. These factors are present in all models in one form or another. In addition, as we have seen, these factor(s) can combine big part of the other dimensions’ variance and content.

Additionally, for greater accuracy (although at the cost of sacrificing a large dose of brevity and intuitiveness), other categories can be added as well. We must remember that they are (mostly) formative constructs, meaning that we cannot form any common dimension/continuum with the issues within them, many separate scales have to be constructed for each sub-taxon. Furthermore, the vast majority of these appear to be highly variable over time, ideologically peripheral[29] and unconnected to each other. The most common construct from these was foreign policy, followed by ingroup bias (which is fairly tightly connected to the former), and judiciary (which, semantically, is mostly covered by the social freedom taxon).

Item-level analysis: Inductive Category Formation

At the next (most important) stage, the Georgian instruments were analyzed inductively by forming categories (aka open coding) from the lowest level of abstraction – items. 189 items were analyzed, resulting in the following taxonomy[30]:

  1. Domestic Policy (80)
    1. Authoritarianism/Centralisation-Democracy/Freedom (71)
      1. Economics (28)[31]
      1. Environmental Issues (10)
      1. Freedom of speech, expression, opinion and belief/religion (8)
      1. Family and Upbringing (8)
      1. Abortion (3)
      1. Other (14)
    1. Other (9)
  2. Orthodoxy and Dogmatism (58)
    1. Religious O. (17)
      1. Secularism (8)
        1. Education (4)
        1. Other (4)
      1. Abortion (3)
      1. Sexuality (2)
      1. Other & Universal (4)
    1. Traditions and Customs (8)
    1. Lawfulness (8)
    1. Freedom of speech, expression, opinion and belief/religion (8)
    1. Ethnic-national (4)
    1. Other (13)
      1. Abortion (3)[32]
      1. Other and Universal (10)
  3. Equality-Equity (38)
    1. Economic E. (32)
      1. Labor Rights/Policies (9)
        1. Employer Rights/Support (7)
        1. Employee Rights/Support (2)
      1. Welfare (8)
      1. Fiscal Politics (7)
      1. Cultural and Educational Policy[33] (3)
      1. Other and Universal (5)
    1. Social E. (4)
    1. Other and Universal (2)
  4. Foreign Policy (32)
    1. Prowesternism-Antiwesternism (9)
    1. Neutrality/Independence (6)
    1. Pro-Russian vs Anti-Russian (4)
    1. Other (13)
  5. Deviance (26)
    1. Judiciary (21)
      1. Severity (10)
      1. Other (11)
    1. Abortion (3)
    1. Other (2)
  6. In-group bias (23)
    1. Nationalism and Ethnocentrism (18)
      1. Traditions (4)
      1. Language (4)
      1. Economic Protectionism (3)
      1. Other & Universal[34] (7)
    1. Religious N. (4)
    1. Other/self-sacrifice (1)

Once again, many categories were interrelated and most of this was connected to the economic and social freedom taxons. Only the most noteworthy connections will be discussed. In-group bias and Deviance (to a lesser degree) were tightly associated to Orthodoxy and Dogmatism. In turn, Orthodoxy/Dogmatism along with Domestic Policy could be mostly explained by other taxons (mainly Social Equality-Equity aka S. Freedom). These relationships are largely acknowledged in the field in general too.

In sum, it was once more confirmed that the Economic and Social Freedom/Equality are the most important possible factors. Other categories may be considered in particular situations.


If we were to summarize all kinds of analyses and information given in the article, the most important and robust finding would be, that economic and social freedom axes are by far the most fundamental constructs for measuring political orientation. After these the following categories may be included, considering peculiarities of given circumstances: foreign policy, deviance (including judiciary), orthodoxy/dogmatism, In-group biacs, and domestic policy.

In the case of domestic and foreign policy, it should be clearly emphasized that the circumstances in this direction are changing very quickly, which is why it is impossible to ask questions (or even conceptualize constructs) that will be relevant for a long time.

It should be emphasized, that it is recommended to ask Georgians questions related to any ideology and political philosophy in the most understandable, simple form (avoiding jargon and relatively unknown words and phrases). The reason for this is that the Georgian population in general, as it was said, is quite inexperienced and uneducated in ideological issues.

It probably follows from this that self-reported information about ideology does not work among the Georgian population. Though it would be helpful to include such scales along with more reliable and valid measures to check its validity again, because the evidence of the opinion given above is not very reliable and also the situation in this regard does change in time.

There are several ways to solve this problem. One way is to simply continually update the questionnaire(s) and instruments. Another is to ask only (or mostly) questions that don’t get old that quickly, and then philosophically and/or empirically deduce attitudes about more volatile, concrete issues from them.

It should be noted, that research once again shed light on the well-established rule of thumb among scholars of many disciplines (though not widespread among other fields), that many variables that even the very knowledgeable would assume are part of some broader category, turn out to be uncorrelated with them[35]. Such problems often arise, when the instruments are not properly adapted to the circumstances (e.g., culture or time) in which it is intended to be used. Therefore, proper adaptation and regular updates to instruments, as well as the underlying scientific or philosophical foundation(s), are generally necessary, particularly following significant events.


ABOUT KIESKOMPAS. (2021, 1 22). Retrieved 4 11, 2023, from Kieskompas:

Andrea, V., Tobias, B., Werner, K., Pola, L., Theres, M., Sven, R., . . . Lisa, Z. (2021). The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR). . doi:

Avaliani, A. (2018). განწყობის ტიპები და პოლიტიკური ორიენტაცია. Master thesis, GIPA, Tbilisi.

Azevedo, F., & Bolesta, D. (2021). Measuring Ideology: Current practices, its consequences, and recommendations. Retrieved 2022, from Flavio Azevedo Web site:

Azevedo, F., & Bolesta, D. (2021). Measuring Ideology: Current practices, its consequences, and recommendations. Retrieved 2022, from Flavio Azevedo Web site:

Barkaia, Z., Kvashilava, B., Gogoladze, N., Kobalia, I., & Chkhikvadze, I. (2020). The Political Landscape of Georgia (second ed.). Tbilisi: Eastern European Centre for Multiparty Democracy (EECMD).

Costa Jr., P. T., & McCrae, R. R. (2008). The revised NEO personality inventory (NEO-PI-R). In G. M. G. J. Boyle, The SAGE handbook of personality theory and assessment (Vol. 2. Personality measurement and testing, pp. 179–198). London: SAGE Publications Ltd. doi:10.4135/9781849200479.n9

Europe-Georgia Institute. (2021, 6 15). EGI Political Compass. Retrieved 3 21, 2023, from ევროპულ-ქართული ინსტიტუტი – Europe-Georgia Institute:

Everett, J. A. (2013). The 12 Item Social and Economic Conservatism Scale. PLoS ONE.

Henningham, J. P. (1995). A 12-item scale of social conservatism. Personality and Individual Differences.

Individual Differences Research – IDRlabs. (2017). 8 Values Political Test. Retrieved 3 21, 2023, from Individual Differences Research – IDRlabs:

Institute of Social Studies and Analysis (ISSA). (2016). Readiness and pre-election attitudes of Georgian population regarding 2016 parliamentary elections, 2016. Institute of Social Studies and Analysis (ISSA). Retrieved 3 12, 2023, from

International School of Economics at TSU (ISET), CRRC Georgia, Georgian Institute of Politics (GIP). (2020, 19 19). Election Compass. Retrieved 6 7, 2022, from

Jibladze, S. (2019, April 22). რატომ ქრება საქართველოში მმართველი პარტიები ძალაუფლების დაკარგვის შემდეგ? . Retrieved fromრატომ-ქრება-საქართველოში-მმართველი-პარტიები-ძალაუფლების-დაკარგვის-შემდეგ-/29896577.html

Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political Conservatism as Motivated Social Cognition. Psychological Bulletin, 129, 339–375.

Kachkachishvili, I., & Mezvrishvili , L. (2003). ელექტორატის წინასაარჩევნო მდგომარეობის შემსწავლელი ეროვნული შერჩევის სოციოლოგიური გამოკვლევა . Institute of Social Studies and Analysis – ISSA.

Kakhishvili, L., Keshelava, D., Papava, G., & Sichinava, D. (2021). Georgia’s Political Landscape: Diversity, Convergence and Empty Spots. Tbilisi: Friedrich Ebert Foundation. Retrieved from

Marks, G., Hooghe, L., Nelson, M., & Edwards, E. (2006). Party Competition and European Integration in the East and West: Different Structure, Same Causality. Comparative Political Studies, 39(2), 155-175.

Melikidze, G. (2017). პოლიტიკური კონსულტირება საქართველოში (მრავალპარტიული საარჩევნო კამპანიის ოცწლიანი გამოცდილების თვისებრივი ანალიზი) . Ivane Javakhishvili Tbilisi State University, Faculty of Social and Political Sciences. Tbilisi: TSU press.

Papava, G., & Tevdoradze, L. (2020, October 27). მარჯვნივ თუ მარცხნივ: როგორია პარტიების ეკონომიკური ხედვა? Retrieved from Tabula:

Ramishvili, L. ([ca 2007]). პოლიტიკური კომპასი. Retrieved 3 21, 2023, from Free University:

Rovny, J., & Edwards, E. (2012). Struggle over Dimensionality: Party Competition in Western and Eastern Europe. East European Politics & Societies, 26(1), 56-74. doi:10.1177/0888325410387635

Tavakarashvili, A. (2018). პარტიების პოლიტიკური ლანდშაფტი პოსტსაბჭოთა საქართველოში . Tbilisi.

The Caucasus Research Resource Centers (CRRC) Georgia. (2023, March). NDI: Public attitudes in Georgia, March 2023. Retrieved 4 5, 2023, from The Caucasus Research Resource Centers (CRRC):

Tsitsishvili, D. (2011). პოლიტიკურ ელიტათა ცვლა პოსტ-საბჭოთა საქართველოში . Tbilisi University Publishing House. Retrieved 2 7, 2022, from

Wilson, G. D., & Patterson, J. R. (1968). A new measure of conservatism. British Journal of Social and Clinical Psychology, 264–269.


Alexandre Avaliani, MA., PhD student, Georgian Institute of Public Affairs (GIPA), or

[1] Often also called the political spectrum.

[2] Many synonyms and related concepts are used interchangeably in this field; while some (mostly psychologists and statisticians) prefer the abovementioned terms (“factors” and “constructs”) others use terms like, „dimensions“, „axes“, „values“, „themes“, „thematic groups“ etc. in this paper the most frequent terms to refer to these concepts will be “construct”, which will be used as the broadest category, “factor”, which will refer to psychometrically validated or founded constructs, and “axis”, which will be utilized in contexts where dimensionality is the core issue; other terms will be used only as a literary embellishment in the specific contexts, mostly referring to the lexical preference of the instrument being discussed.

[3] In this article terms “measurement/assessment instrument/tool” are used as a broad category, which encompass different types of measurement tools (e.g., inventories, scales, questionnaires, surveys, tests, quizzes & indices), which are often (mostly mistakenly) used interchangeably in the field (especially in non-academic settings).

[4] In this paper, term “item” refers to all kinds of verbal stimuli used in such research (including e.g. questions or issues).

[5] and probably couldn’t have even existed because, as the authors of the study told me, the data were collected only on the aggregated/ level to ensure the protection of respondents’ personal information.

[6] The most appropriate term to name and describe this approach would be the deductive category formation method.

[7] Resulting in the axes of economic and social left and right, which are also called freedom, liberalism, libertarianism, authoritarianism, anarchism, collectivism and other related terms.

[8] In this case, “conservatism” (contrasted with “liberalism”, in the same American fashion) represents just the right wing of political spectrum.

[9] The resistance to change describes the dimension of social conservatism axis, which is also called traditionalism, and inequality is linked to economic equality-equity factor.

[10] One clear illustration of this is the fact, that very often “Conservatism” and “Liberalism” are even used as opposing labels of pollical spectrum (replacing right/left-wing or other such terms).

[11] Most other constructs (e.g. foreign and domestic policy) vary across time, culture or other factors much more.

[12] In some cases, in this paper, terms ideology, political spectrum and political orientation are used interchangeably (mostly in conjunction with words “instrument” and “tool”), though they differ significantly, as noted in the introduction, and must not be confused.

[13] To avoid confusion more prevalent terms like “themes”, “main categories” or “codes” were avoided, as there’s little agreement around the subordination of them.

[14] In the search carried out as part of the research, questionnaires from 92 countries were found, although the search was of a global level.

[15] Which are also called “(ideological) topics”, within the work, and may also be called issues.

[16] Same as inductive category formation

[17] A comprehensive factorial model should have few, broad categories to fulfill its main purpose – simplification.

[18] Both groupings resulted in a very similar taxonomy (the 5 main categories – taxons – were identical).

[19] This taxon along with the subsequent one (economic issues) is essentially the same construct as the 2 main factors in the political orientation research, discussed earlier. Briefly, they may be called social & economic freedom axes.

[20] The issues are rank-ordered by the items the encompass (which are indicated in the brackets).

[21] In this case, 3rd, 4th and 5th taxons.

[22] Moreover, this holds true even at a level of topics.

[23] They were conducted in a way (namely, limiting the numbers of factors manually) that would result in the closest replication of the model proposed by the authors.

[24] Same number as the one endorsed by the authors, but with different distribution of items.

[25] Popular term within social sciences (mostly psychology, sociology and evolutionary studies); also known as in-group(–out-group) favoritism, intergroup bias, or parochialism/parochial altruism.

[26] Items that were about democratization getting rid of top-down, authoritarian/totalitarian influence in this regard.

[27] Though left vs right wing and (to a lesser degree) conservatism vs liberalism are more common terms, “Freedom” was preferred for 2 reasons: 1) stylistic advantage (brevity and elegance), and 2) coherence, especially in Georgia, where economic & social freedom are positively correlated, while other labels result in the opposite.

[28] While all instruments (that included this dimension) placed prowesternism on one pole of the continuum, the other pole varied significantly (from independence and/or pro-Russian course to, plainly, antiwesternity). If we were to summarize though (as intended by the authors), we would get the opposition of prowesternism to all other courses.

[29] As opposed to central: not part of the core issues and values.

[30] Multicollinear items were coded in all related categories.

[31] This category is apparently related to the one in equality-equity taxon. The broad economic factor, which is one of the 2 most basic axis of political spectrum incorporates both; while sometimes the freedom aspect is more evident, the equity-equality aspect is more accented in other cases; this is evident even from its alternative names.

This category could be subdivided similarly to its counterpart, but was not, for brevity and comprehensibility.

[32] This is the same category as above (in religious orthodoxy/dogmatism). It was included in both themes, because even though many oppose abortion for religious reasons, still a big part of opposers are non-religious.

[33] This category encompasses financing and subsidizing entities like culture, sports, and education.

[34] “universal” categories entail semantics of a whole given broader category; e.g., item “One should always defend and justify the motherland, even when it is wrong” relates to all kind of ethnocentrism or nationalism and is not limited to its only one particular sub-category (e.g, language or traditions).

[35] E.g., in the Election Compass the items “Participation in the accumulative pension system must be voluntary” and “Privatization of state hospitals will reduce healthcare costs for consumers” were categorized as part of economic freedom axis, but factor analysis showed that they were completely unrelated (with factor loadings of .136 &.134).

Download PDF

Share this content: