Resilience Of Social Networking Groups: Analysis Of Content Using Grounded Theory Method

Identification of the discursive characteristics of social networking group resilience

The identification of discursive characteristics was carried out by 4 psycholinguistic experts using psycholinguistic and functional-semantic analysis of the text corpus of 12 social networking groups. These are the group “Together” for charity fundraising on the Dirty platform (https://together.d3.ru/); Telegram messenger chat “The residential compound South Bittsa” (https://t.me/uznaabitca); the group of the BJD doll club artists (https://bjdclub.ru/index.php). They also include the news and educational blog “Varlamov” in LiveJournal (https://varlamov.ru/3809040.html); the forum of the charitable foundation for helping homeless animals “DOG and CAT” (http://pesikot.org/forum/index.php); the forum thread on rutracker.org dedicated to comics published by Bubble Comics (https://rutracker.org/forum/viewtopic.php?t=4238138&start=1980). There is also the chat in Telegram messenger for those who are interested in Antarctica (https://t.me/AntarcticaOnLineqwest); the chat in ALES group in social network VK dedicated to pop culture (https://vk.com/alesworld); the group dedicated to copyright infringement “League of Lawyers” on Pikabu (https://pikabu.ru/community/justice). That also involves news feed of an IT resource - discussion of technical problems of the latest version of Detroit Become Human (https://overclockers.ru/softnews/show/101719/pc-versiya-detroit-become-human-stala-znachitelno-menee-trebovatelnoj-k-processoru ); forum on the tactics of searching for missing people with the neural network program “Lakmus” https://lizaalert.org/; group “Autoimmune paleo diet” on the website of the Paleo Planet (https://paleoplanet.ru/help).

Discussion and agreement of extraction methods and encoding was carried out during 4 expert sessions within the framework of the grounded theory method (Glaser, 1978; Glaser & Strauss, 1967). The experts identified discursive features of various levels: a separate post or commentary, consistent remarks, discourse as a whole. In terms of form, the features were assigned to three groups: lexical-grammatical, procedural and semantic. In total, 97 initial codes of the social networking group resilience were identified. Thus, focused codes that provide lasting, ongoing, self-reproducing and self-sustaining communication, as well as positive relationships in the social networking group, manifested themselves in such initial codes as “statements saturated with optimism”, “expressions of solidarity about ...”, “support of positive emotions, maintaining a good mood”, etc. The ability to cope with problems with the increasing of the competence level was expressed in the initial codes “statements that participation in the social networking group allows you to navigate in the environment”, “thanks to the social networking group, proactive and active intervention in events is possible”, etc. The ability of group to learn and adapt to changes was reflected in such initial codes as “unusual and original messages about how to live differently”, “messages that everyone in the group is welcome, there is room here for a discussion of otherness”. The ability to recover from failures, conflicts and threats manifested in the “confidence that everything will return to its original state” and in messages like “we remember how good it was and we'll get to that”. Both lexical and semantic categories, which did not receive a specific form of implementation, were important in identifying initial codes, but were reliably identified by experts in terms of meaning: the controversial nature of communication and a variety of argumentation in disputes, solidarity and consistency in assessing events, discussion of new opportunities and their constant search. The discovered initial codes of the social networking group resilience were corrected during expert sessions and were grouped into 8 basic categories at the last session:

1) general resilience (basic signs associated with the existence of a group and its ability to develop in various conditions);

2) involvement (satisfaction from joint life activity, joint activity);

3) control (the belief of group members that joint activity affects the result);

4) risk taking (the belief that everything that happens in the group brings a positive experience);

5) plasticity (the ability to adapt after failures and losses, using the remaining resources);

6) resilience (the ability to recover from failures and losses, after the structure and functions of the group have been disrupted);

7) antifragility (the ability to become stronger, more varied, more extensive, better after difficulties);

8) self-development (development in accordance with the mission.

Each basic category functionally and structurally revealed a certain aspect of the social networking group resilience. Due to the fact that experts had disagreements and doubts about the acceptability of some initial codes for assessing resilience, the primary list was reduced to 44 initial codes.

The structure of the social networking group resilience based on analyzing their discourse

Five experts, specialists in the field of social and general psychology, were engaged in the study of the resilience of collective and individual subjects, participated in the final session to identify and describe the discursive characteristics of the social networking group resilience for review with their content and manifestations in texts. After that, they assessed the contribution of each of the 44 initial codes to various resilience characteristics, described by K. McEwen and K. Boyd, as its various indicators (Masten, 2011; McEwen, 2016; McEwen & Boyd, 2018). In total, 52 resilience characteristics were used: “the ability of the group to remain effective in difficult conditions”, “participation in the group helps to increase competence”, “gives optimism and confidence in the future”, “promotes the adoption of innovations”, “promotes prudence”, “allows positive perceive emerging problems”, etc.

The assessment was carried out on a 5-point Likert scale, which describes the extent to which a discursive feature determines a specific characteristic of resilience. Thus, 5 matrices (44 * 52) were obtained – initial codes characteristics of resilience. To summarize the scaling results, an assessment of the consistency of expert opinions was carried out. For each scale, Cronbach's alpha coefficient of concordance α was calculated. If α for 3 or more experts turned out to be> 0.7, then the results were considered consistent.

For 13 characteristics, it was not possible to obtain sufficient agreement, and they were excluded from the analysis. The agreed data were averaged over experts and subjected to the cluster analysis procedure (Hierarchical Cluster Analysis, Squared Euclidian distance measurement method, Between-groups linkage clustering method). Figure 1 shows a dendogram.

The clusters were extracted taking into account the maximum distance when combining objects into clusters and the clarity of interpretation with meaningful generalization of the features included in the cluster.

See Full Size >

6 clusters were identified. A meaningful interpretation of the discursive features of resilience included in the clusters is shown in Figure 2. In fact, the resulting structure of resilience is a reconstruction of the structure of implicit ideas about the social networking group resilience by experts in social and general psychology who study the properties of collective subjects involved in various forms of online activity and understand the features of discourse which are characteristic of the mass media.

See Full Size >

The proposed components of resilience correlate to a certain extent with the characteristics of the resilience of real social groups and teams (Nestik & Bagrationi, 2018); they differ from them to some extent. Such a component of the social networking group resilience as “Detection and compensation of undesirable tendencies” is quite consistent with the ideas of Vogus and Sutcliffe (2007) about the resilience of teams, including measures to deal with possible risks even before its actual onset. However, in the social networking group, this component is mainly associated with the nature of communication between its members. Another component of resilience “Change monitoring”, concerns the discussion of negative events in the group and their impact on the life of the group. It is also presented as a resilience construct for offline organizations, and both of these components form such an important characteristic of collective resilience as group reflexivity. If in real teams, group reflexivity presupposes a willingness to participate in the regular analysis of joint experience (Nestik, 2016). Then in social networking groups the focus of reflection is on the nature and rules of communication, less often on discussing the rules of moderation and following them.

Most researchers of group resilience emphasize the importance of the innovative potential of team members and their ability to create and use opportunities for development (Hamel & Valikangas, 2003). Resilience is clearly promoted by the creativity of the project participants (Blatt, 2009), discussion of alternative points of view, continuous learning and development (Soon & Prabhakaran, 2016). In the social networking group, the creative potential of the participants manifests itself in a variety of discursive features that reflect the main idea: a crisis is a reason for improvement. Difficulties and problems provide an opportunity for self-improvement and improvement of the group as a whole, strangeness and uncommonness is a virtue and potential for development. In fact, this component reflects such a characteristic as “Antifragility”, introduced by Taleb (2014). It denotes the ability to benefit from failures, losses, mistakes; the ability to temper, develop and become stronger when faced with chaos (Taleb, 2014). The hyperbolized meaning of otherness is probably associated with the reduced responsibility of members of the social networking group and the illusion of security arising among users of social networks.

The “positive attitude and satisfaction with participation in the group” is perhaps the most obvious characteristic of resilience. This factor, referred to as “positive organizational behaviour” (Youssef & Luthans, 2007), or as “positive relationships: trust and psychological safety” (Soon & Prabhakaran, 2016), or as “positive interaction” (Xiao & Cao, 2017), is most widely presented when describing the phenomenon of group resilience. In the social networking group, it is naturally associated with the peculiarities of communication and involves the creation of a friendly atmosphere, support and approval of friends, solidarity and consistency in assessing various events.

The last two components, “Activity as a way of overcoming difficulties” and “Overcoming through self-development”, in fact, represent such characteristic as resilience, modified in the conditions of social media. Thanks to its resilience, the social networking group seeks not to change under external shocks, everyone “suffers to the end”; the significance of “external strikes” is ignored. Resilience prevents the emergence of internal tension in the group, contributes to the perception of stress as insignificant. The most typical way of resilient response is to show typical activity in the hope of overcoming the difficulties that arise. Such manifestations of resilience are quite adequate and typical for real off-line groups. Thus, Soon and Prabhakaran (2016) point to “persistence despite difficulties” as a marker of resilience. The resilience of a group is associated with the readiness to respond to challenges from the external and internal environment (Bhamra et al., 2011). The views of Goldstein (2011), who consider the ability of an organization to recover from unexpected and unfavourable external influences as a manifestation of the resilience of the organization as a whole, are quite consistent with idea about group resilience.

Comparison of the resilience assessed by discursive features and by external evaluation indicators

To compare the resilience, assessed by discursive features and resilience, assessed by external evaluation indicators, a composite score was calculated, including objective and subjective indicators. Objective indicators included information on the duration of the existence of group and the dynamics of its numbers. Groups that ceased functioning and reduced their size received minimal scores, while groups that existed for a long time and whose size increased, received high scores. Subjective indicators included expert assessments of “the constancy of topics discussed in the group” and “increasing frequency of remarks in discussions”.

The texts of the social networking groups were marked by 4 psycholinguistic experts by highlighting initial codes. Resilience was calculated as the relative frequency of discourse markers found in the texts of social networking groups. Conversations lines and separate sentences were used as units of analysis. Resilience indicators were calculated in accordance with the cumulative-additive model with winsorizing, assuming 5% reduction of the highest and lowest values, followed by convolution of the arithmetic mean. This is the model of the manifestation of psychological constructs in initial codes was identified as the most adequate (Voronin et al., 2019b). In accordance with the structure of the social networking group resilience, 7 indicators were used: detection and compensation of undesirable tendencies, change monitoring, antifragility, positive attitude and satisfaction, overcoming through self-development, activity as a way to overcome difficulties and general resilience.

The values of various indicators of the social networking group resilience, as well as their resilience level, assessed by external evaluation indicators (3 - high, 2 - medium, 1 - low) are presented in Table 1. Social networking groups in the table are sorted according to the resilience level according to external evaluation indicators.

The Jonkhier-Terpstra test was used to test the hypothesis about the ordering of the shift in the number of discovered discursive features of the social networking group resilience, depending on the external manifestations of resilience. Analysis of the data in the table suggests the ordering of the level of resilience indicators depending on the level of resilience, assessed by external criteria. To test this hypothesis, we used Jonckheere-Terpstra test. The results of comparisons of shifts are presented in Table 2. The direction of the shift for all resilience indicators is the same and direct. All indicators of resilience increase with increasing resilience, assessed by external evaluation indicators. However, the shifts in the indicators “Overcoming through self-development” and “Activity as a way to overcome difficulties” are not significant. Most likely, the manifestation of activity to overcome emerging difficulties is not significant for the duration of the existence of groups.

The indicator of general resilience as an integral indicator unambiguously determines the social networking group resilience: the more often various discursive features of resilience are found in the content of the group, the longer its lifetime and the broader its audience. Actually, a similar result was obtained in the study of the subjectivity of social networking groups: the variety and frequency of discursive features is a clear sign of the sought-for psychological characteristic (Voronin, 2019; Voronin & Kovaleva, 2019). Among other indicators of resilience, antifragility seems to be the most significant. The idea that any crisis is a reason for improvement is a priority and ensures the longevity of the group. The new things, innovations, alternatives, everything unusual are actively discussed in groups, this is the reason why people join them and live in them. No less important factors are “Detection and compensation of undesirable tendencies” and “Change monitoring”. These indicators reveal another reason for a constant presence in social networking groups and, accordingly, for increasing its resilience - reducing uncertainty and anxiety, which are integral attributes of our constantly changing world. The resilience indicator “Positive attitude and satisfaction with participation in the group” is characteristic both for social networking groups and for real social groups and teams. Friendly communication in the circle of friends, the possibility to criticize strangers, including backbiting, all this makes it possible to compensate for the lack of real communication and often serves as a very reliable basis for participation in the life of social networking groups. Non-binding virtual communication either guarantees a comfortable pastime, or creates the illusion of being involved in joint activities. Both have a positive effect on the duration of the existence of social networking groups and the expansion of their numbers.

There are the features that determine the indicators “Overcoming through self-development” and “Activity as a way to overcome difficulties” in the social networking group discourse. Theу are clearly not associated with increasing resilience and it is most likely explained by the thematic orientation of the content of some groups. This is indirectly confirmed by the themes of coping in the Antarctic and Paleo Planet groups. The group “Antarctic” discusses the difficulties of sober life in “Antarctica”, and the group “Paleo Planet” discusses the challenges of dieting. In the group “Detroit”, an extremely critical and aggressive discourse is built around the technical problems of the latest version of Detroit Become Human. The indicator “Overcoming through self-development”, most likely, complements “Activity as a way to overcome difficulties”, which is especially clearly seen for groups from the category with high resilience according to external evaluation indicators. Probably, this kind of general factor suggests the manifestation of activity in response to difficulties and manifests itself in the resilience of real teams as persistence despite difficulties in the terminology of Soon and Prabhakaran (2016).