The article presents the grounds and principles of modelling with the example of the analysis of publications in the special issue of the journal “Political Science” devoted to the changing world orders (No. 2, 2024). A distinction is made between modelling in life and various scientific studies. The role of political modeling of research questions, subject, method and motivation of research is highlighted.
A distinction is made between phenomena and their concepts, their reification and dereification, as well as the reality, actuality and imaginativeness of modeled phenomena. Particular attention is paid to the miscalculations caused by the uncontrolled acceptance of the myth of the given and the myth of the framework. The processes of purification of saturation and rooting of models, the connection of these processes with simplex-complex transformations and transdisciplinary organon integrators – metretics, morphetics and semiosics – are considered. Alternative possibilities of modelling are discussed (in particular: the construction of dynamic models that take into account the synchronization and dissynchronization of political processes). The problem of forming a theoretical and methodological approach to development is discussed. The authors propose to develop not an abstract theory, but a transdisciplinary research program focused on mastering the extended evolutionary synthesis (evo-devo).

The struggle for the new world order is currently becoming more active and evident. The study of various aspects and dimensions of this struggle, as well as the forecasts based on this analysis, can be important and relevant both in theoretical and practical aspects. Among the many dimensions of the struggle for world order, for example military-strategic, geopolitical, economic, technological, etc., the demographic dimension of this struggle usually receives insufficient attention. Meanwhile, although demographic changes are long-term, they are undoubtedly basic, radically and usually irreversibly changing the image and capabilities of certain countries and actors. But the demographic factor is also extremely important in global terms, given the sharp decline in fertility in the Western world and even in Asia and its high growth in Africa, as well as the rapid global population ageing.
This article uses a research method that can be called the aspect-factor method. Its methodological value lies in the fact that it allows us to identify certain more closelyrelated phenomena in a large flow of different and interrelated processes, causes and factors. It makes it possible to reveal important regularities, trends, tendencies and details.
The article examines the various directions of the demographic dimension of the struggle for the world order (including the constant deterioration in the recruitment in the armies of the United States and Western countries), as well as the forecasts of the impact of changes in the demographic balance of power on the transformation of the world order.

Contemporary academic literature has focused on technology adaptation and innovation in politics and public administration. At present, there is a large number of studies focusing on actively developing phenomenon as public digital services. Despite the diverse range of theoretical approaches, there is no established methodological tradition of modelling perception of digital services. In addition, many studies are based on the premise that the user’s actions are voluntary. We assume that in the absence of voluntariness, the classical modelling form of decision-making – ‘yes or no’, ‘to act or not to act’ – loses much of its meaning. Mandatory environment gives rise to a number of important consequences, some of which require new modelling solutions.

To fill this gap, the author presents in this paper a computational model of the formation of attitudes towards public digital services under the mandatory conditions. The reason behind the choice of this area is the magnitude of the coercive power of the state and its ability to apply it to the widest range of social groups.

The main properties of agents in the model are trust in government and digital skills. Public services differ in technical requirements and, most importantly, in sensitivity – the scope of users’ rights and obligations that the service affects, and the volume of personal data it requires. The model shows how these variables interact under the conditions of voluntary and mandatory service use.

Computational experiments demonstrated that mandatory conditions give rise to a polarization of users’ attitudes when the service provided is sensitive. The polarization is intensifying when dissatisfaction with the associated risks affects the level of trust in government. Involuntariness also serves as a ‘catalyst’ for the attitude formation processes: in its presence, both the scale of changes and the leverage of other factors (sensitivity in the first place) increase.

The influence of online communications on the dynamics of protest activity holds one of the central places in modern political communication research. Internet continues to play an important role in protest mobilization, and authorities of different countries employ a broad repertoire of strategies aimed at reducing its effectiveness as a protesters’ communication channel.
This paper aims to evaluate whether internet shutdowns reduce or increase protest activity in the short term. Recent theoretical and empirical accounts provide conflicting evidence on the matter, and we attempt to resolve some of these debates. Relying on fine-grained data on internet shutdowns and protest episodes from Indian districts in 2016–2022, the authors model the effects of shutdowns at the district i and day t on the expected number of protest episodes at the same district and day t + 1 by employing negative binomial regression.
The findings provide compelling evidence that shutdowns increase the number of protest events in a district, casting doubt on shutdowns’ potential to reduce protest activity. Nonetheless, the authors also encourage researchers to take caution in interpreting these findings, as the authors don’t have the data on the actual protest attendance. The effects of shutdowns might be more nuanced, as they may lead to a greater number of smaller protest episodes and lower overall participation.

The article is devoted to the study of individual methods of political profiling as a comprehensive tool for analyzing the behavior of political actors in modern conditions. The evolution of this direction from classical psychological profiling of political leaders to modern interdisciplinary methodology integrating methods of psychology, political science and strategic analysis is considered. The key components of political profiling are analyzed, including various methodological approaches, the use of a wide range of primary data and the use of complex analysis models. Special attention is paid to the prospects for the development of political profiling, taking into account mathematical modeling and the reflexive approach formulated by V.A. Lefevre. Methodological challenges and ethical dilemmas in the application of this toolkit, problems of validity of the data used and their interpretation are discussed. The issue of adapting the methods of political profiling for use in other spheres of human activity is touched upon. The study attempts to systematize approaches to political profiling, summarizing some existing methods of analyzing and predicting the behavior of political actors.

Censorship in various forms is a widespread method of controlling the information space. Two different censorship strategies are possible. The first of them focuses on preventing criticism of the authorities, the second one allows criticism, but prevents content that promotes collective protest actions. Recent studies show that the first strategy is ineffective, since content distributors, on the one hand, and its consumers, on the other, find ways to circumvent censorship restrictions. In the absence of direct assessments of the efficiency of the second strategy, the question is: Can it be more effective? In other words, if censorship is ineffective in combating criticism of the government and its policies, is it capable of preventing content that promotes collective actions? For what reason can censorship strategies have different effectiveness despite the fact that circumvention methods are universal? In order to study this issue, mathematical modeling is used in this article. A dynamic model is constructed in the form of a system of four equations with discrete time. Numerical experiments were conducted with it, which showed that the use of censorship slows down the distribution of content. In the case of criticism of the authorities, this slowdown does not play a significant role, since public interest in topics such as corruption or economic inequality is permanent. In contrast, content such as, for example, a call to participate in a collective action is relevant only for a short time. Therefore, slowing down the distribution of this kind of content is critical. Thus, the simulation shows that the use of censorship in relation to content that promotes collective actions is more effective than the use of censorship in relation to criticism of the government.

The article examines the phenomenon of affective political polarization, which is operationalized through the fact of detected hate speech. In this study we have made two assumptions. Firstly, it is believed that during political mobilization, affective polarization (hate speech) will be significantly stronger than during the non-protest period. Although, hate speech will be detected for both periods. Secondly, we believe protest significantly affects cleavages. During the period of protest mobilization, political cleavages become more important than social ones. To verify the assumptions, we examined data collected on the VK social network in two periods: July–September 2019 (Moscow protests) and March–May 2019 (without protests). The results of the analysis, based on the automatic mapping (ruBERT) and on the human mapping of independent encoders, confirm both assumptions. We prove both assumptions via cross correlation analysis. Moreover, present approach to data collection and mapping allows us to create predictive models (SARIMAX) to predict hate speech in a social network.

The article introduces the concept of an current political agenda, which includes a set of issues that generate the greatest debate and maximum polarization of positions among political actors. It is proposed to look for these issues in an inter-party discussion, since it is typical for parties to exploit the agenda in order to increase electoral capital.

The article describes the project “Current Political Agenda of Russia in InterParty Discussion”, which aims to monitor the current political agenda and is accompanied by the following tasks: identifying issues that polarize the political space and cover a significant part of the participants; thematic classification of these issues; assessment of actors’ positions; analysis of these positions using quantitative methods; building a dynamic model of the country’s political space based on identifying dominant confrontations. A methodology for selecting and analyzing issues is described, which involves using the tools of theories of issue dimensions, agenda setting and issue ownership, but filling these tools with original content.

The results of calculations for the summer and autumn of 2023 are summarized, from which, in particular, it follows that foreign policy issues dominate Russia’s current agenda, domestic policy is an outsider, and the socio-economic and ideological spheres are almost on a par with each other. It is also noted that of the parties participating in the Duma elections, the Communist Party of the Russian Federation is in first place in terms of activity level, with a large gap from the rest, followed by A Just Russia – For Truth, LDPR and United Russia.

A factor analysis of party positions on a set of issues on the current agenda identifies three main issue dimensions, the first of which is most associated with the confrontation between Westerners and originalists (worldview sphere) and “hawks” and “doves” (foreign policy), the second one – with the confrontation between loyalists and oppositionists in the internal political, socioeconomic and worldview spheres, the third one – with various sub-dimensions of the socio-economic field and the confrontation between soviet traditionalists and progressives.

This study is devoted to the examination and measurement of cohesion effects in the context of a political crisis. Traditionally, the process of social solidification has been defined either as the unification of the entire society during periods of exogenous shocks or as the intergroup cohesion of initially close communities. We hypothesize that these types of cohesion are interrelated and exhibit common developmental trends in pre-crisis and post-crisis periods. An important component of the study is the proposed approach to measuring group consolidation through network characteristics. Based on data from over 1,000 political Telegram channels, using machine learning and network analysis methods, we examined the dynamics of group consolidation during the four weeks before the Special Military Operation and the four weeks following it in three networks: one based on links between political channels, and two others constructed around social motivations of anger and belief in success. To assess the cohesion effect, key indicators of community partitioning – modularity and the number of communities – were used. It was found that the link-based network (baseline scenario) reflects a short-term effect of overall group cohesion, but in the long term, the effect of group consolidation diminishes. In the network built around anger-driven social motivation, communities, on the contrary, became more structured after the start of the military operation, indicating cohesion only among initially close channels. The motivation of belief in success does not manifest before the crisis but influences social solidification in the long term. The results provide deeper insights into the mechanisms of social community formation during periods of political instability and their network structures. The study contributes to understanding how digital platforms shape political behavior.

Counter-protests have become a common response to protest movements around the world. Understanding this process is often based on collective identities formed in opposition to an out-group. In the context of the Black Lives Matter (BLM) movement, the initial image of the “other” was mainly attributed to the authorities. However, the emergence of the All Lives Matter (ALM) counter-protest movement has changed the way how collective identities are constructed – now the image of the “other” is also attributed to participants in the counter-protest movement, introducing a new layer of complexity into the dynamics of identities and raising questions about how these identities are transformed when faced with counter-protest. Applying a new approach in BERT topic modeling, the author analyzes thematic shifts in the posts of BLM protesters between 2013 and 2014 on the social network Twitter (X). The application of the BERT topic modeling model allowed to conduct a detailed analysis of online messages, capturing contextual dependencies and revealing the complexities of the transformation of collective identity in response to the emergence of the ALM counter-protest movement. The study contributes to the existing literature by filling two gaps: demonstrating the methodological advantage of using BERT on empirics and exploring how the collective identities of protest movement participants are transformed in the context of protest and counter-protest dynamics. BERT topic modeling is a powerful tool for studying the transformation of collective identities, offering a promising methodological basis for further research at the intersection of protest, counter-protest, and collective identities in the digital age.

The study focuses on the application of modern machine learning methods for analyzing textual data in the context of the dynamics of ideological polarization in Russian-language political Telegram channels during the first half of 2022. This work proposes an approach to classify text messages basing on ideological orientation – conservative, liberal, and communist – allowing researchers to utilize resources more efficiently.

Based on the developed approach, an ideological orientation classifier using ChatGPT was created, demonstrating a high level of consistency in responses between humans and the large language model by evaluating the ideological stance of texts. This indicates that the proposed approach can reduce resource expenditures when conducting textual data analysis.

In the next phase, a sample of 559 popular political Telegram channels, which published 50,000 messages, was analyzed for the dynamics of ideological polarization following the onset of the special military operation. Several models were compared: changes in opinion distribution, group composition, and changes in the proportionality of ideological texts within channels. We concluded that following the initiation of the special military operation, there was a change in ideological polarization, primarily manifested in the strengthening of conservative views, and to a lesser extent, liberal views. Communist views are virtually absent from the popular Telegram space.

This work not only captures the dynamics of ideological polarization but also offers a method for analyzing complex socio-political processes in the Russian-language online environment using large language models. This method is suitable for studying polarization as well as for analyzing other processes based on textual data, significantly reducing the costs of research that require a large number of expert evaluations.

The authors study the geopolitical code and symbolic images of key geopolitical actors (China, the USA and the European Union) in the consciousness of the population of the Kaliningrad region. The interest in this case study is conditioned by the exclave position of the region, which aggravated the socio-economic problems caused by the growth of international tensions. The methodological basis of the presented research is the concept of geopolitical code of C. Flint. It provides answers to the following research questions: which countries, in the opinion of the region’s residents, are Russia’s current and potential allies/enemies? How to keep allies and attract new ones? How to resist enemies and prevent the emergence of new ones? The empirical basis of the study was the results of a mass survey (n=979) and eight focus groups with residents of the Kaliningrad region (n=61). Based on the generalisation of the obtained data, the authors identified similarities and differences of the geopolitical code inherent in the region’s residents belonging to different age groups. The authors also discovered the peculiarities of symbolic images of key geopolitical allies and adversaries in the perception of the residents of the Kaliningrad region. The image of China is predominantly positive, including such characteristics as strength, kindness and wisdom. The images of the USA and the EU are predominantly negative. The image of the USA is perceived as strong and aggressive, the image of the EU – as weak and passive. The latter is conditioned by the widespread opinion about the subordinate position of the EU before the USA, as well as about the internal political inconsistency of its members’ actions.