ჯგუფური ქცევის გავლენა ინდივიდუალურ ეკონომიკურ გადაწყვეტილებებზე

Group behavior is a multifaceted and noteworthy phenomenon. The purpose of this study
is to spark more interest in the economic effects of group behavior. Due to its specificity, the
study synthesizes the results of social psychology, sociology and economics studies. An
exhaustive list of key factors affecting group behavior is provided, as well as a Bayesian model
with bounded rationality for describing the phenomenon. This model eliminates problems
inherent in other group behavior models. The model takes into account the impact of both
informational and normative conformism. The formation of group behavior is a different
process for different groups, as it is influenced by group size, concensus, composition,
confidentiality, member responsibility, accountability, culture, self-esteem and security.
An experiment was conducted within the framework of the study, confirming the
widespread prevalence of group behavior. The experiment also confirmed the advantages of
the proposed theoretical model over earlier models and the need to incorporate social
psychology studies to better explain economic processes. As it turned out people often resort
to heuristics instead of rational decision making when comparing two alternatives, though
some people do not take into account others' decisions at all.
Group behavior analysis cannot be complete without considering interconnections between
individuals that can be described with networks. A person can only observe actions of a certain
part of society because human resources and capabilities are limited. It is therefore important
to study social networks in terms of different network structures and their parameters. Taking
into account different indicators observed on the network enables one to determine how
strong an individual’s influence is on other members of the network. In this respect, it is
important to note the popularity, relevance (connections), reach and brokerage/bridging of
members. Conducting computer simulations on four types of networks - random, single-hub,
multi-hub, and two-component networks - showed that all situations are individual and that
different information is distributed across different networks. Moreover, information in the
same network can be disseminated in different ways as well. The more closely connected
individuals are, the sooner a decision is made within those networks. The large difference in
the number of connections hampers the distribution of information to some extent, even
though one member may be playing a very important central role in the network.