What is remarkable is that, in both studies, those who lean left were more inclined to unfriend someone after a political disagreement than those who lean right

Neubaum G, Cargnino M, Winter S, Dvir-Gvirsman S (2021) “You’re still worth it”: The moral and relational context of politically motivated unfriending decisions in online networks. PLoS ONE 16(1): e0243049. https://doi.org/10.1371/journal.pone.0243049

Rolf Degen's take: https://twitter.com/DegenRolf/status/1348863984494120969

Abstract: Political disagreements in social media can result in removing (i.e., “unfriending”) a person from one’s online network. Given that such actions could lead to the (ideological) homogenization of networks, it is pivotal to understand the psychological processes intertwined in unfriending decisions. This requires not only addressing different types of disagreements but also analyzing them in the relational context they occur. This article proposes that political disagreements leading to drastic measures such as unfriending are attributable to more deeply rooted moral dissents. Based on moral foundations theory and relationship regulation research, this work presents empirical evidence from two experiments. In both studies, subjects rated political statements (that violated different moral foundations) with regard to perceived reprehensibility and the likelihood of unfriending the source. Study 1 (N = 721) revealed that moral judgments of a political statement are moderately related to the unfriending decision. Study 2 (N = 822) replicated this finding but indicated that unfriending is less likely when the source of the morally reprehensible statement is relationally close to the unfriender and provides emotional support. This research extends unfriending literature by pointing to morality as a new dimension of analysis and offers initial evidence uncovering the psychological trade-off behind the decision of terminating digital ties. Drawing on this, our findings inform research on the homogenization of online networks by indicating that selective avoidance (in the form of politically motivated unfriending) is conditional upon the relational context and the interpersonal benefits individuals receive therein.

Check also What Do We Fear? Expected Sanctions for Expressing Minority Opinions in Offline and Online Communication. German Neubaum, Nicole C. Krنmer. Communication Research, Jan 2018. https://www.bipartisanalliance.com/2018/01/expected-sanctions-for-expressing.html

General discussion

The present studies focused on the phenomenon of politically motivated unfriending as a consequence of a moral disagreement. To this end, it examined (a) whether moral judgments are at work when reading political statements on social networking platforms, (b) whether and which moral evaluations have predictive value for the decision to politically “filtrate” one’s online network, and (c) to what extent moral judgements of political statements are undertaken differently depending on the relational context and the social rewards the source of political statement offers.

Study 1 and 2 partly replicated previous findings on characteristics of who are political unfrienders: Those who use social networking sites (in this case: Facebook) more frequently might encounter political dissents among their network more often and, therefore, they are more likely to unfriend or block someone [5, 8, 23]. Likewise, those who are more interested in politics feel more involved in certain political issues and, thus, perceive political disagreements as more severe, leading them to unfriend or block the dissenter [7]. What is remarkable is that, in both studies, those who lean left were more inclined to unfriend someone after a political disagreement than those who lean right. The fact that this result was not obtained consistently in previous research [4, 5, 27] may be explained by the national context and prevailing issues that are predominantly discussed in online networks at the time the study was conducted. It seems conceivable that in Germany (the country this research was conducted in 2017 and 2018), the debate about the immigration of refugees–as a rather dominant topic in the public [61]–might have led left-leaning users to unfriend or block those ties who opposed the immigration. Moreover, in contrast to previous findings [5, 23], higher political extremity was not associated with the likelihood of unfriending someone. This additionally suggests that the predictive value of the political identity or orientation on decisions of network filtrations depends on the current political landscape and state of polarization in certain countries.

While the present studies replicated certain findings from previous research, these factors only explained a limited variance of unfriending behavior (approx. 5% in both studies). In contrast to this, the present approach of focusing on the moral and relational nature of an unfriending decision appears to have greater explanatory value: First, both studies offered compelling evidence that evaluating a political statement on SNS as morally wrong is a driver of the decision to unfriend or block someone. This finding supports the notion that political disagreements that lead to the post hoc modification of one’s online network are rooted in moral discrepancies. Following this logic, Study 1 showed that individuals are able to differentiate between two different sets of moral foundations that a political statement on SNS violates: When the foundations of care/harm and fairness are trespassed, individuals seem to be slightly more inclined to unfriend someone than when the moral domains of loyalty, authority, or purity are violated. While Study 2 did not indicate main effects of the type of moral foundations violated, it showed that perceiving trespasses of individualizing foundations were slightly stronger associated with unfriending than perceived violations of binding foundations. Likewise, considering past behavior, users reported that violations of individualizing foundations were more likely to lead to unfriending decisions than trespasses of binding foundations. Collectively, this evidence is indicative for the role that morality and its nuances play when it comes to shaping one’s online network by terminating a digital relationship. In line with previous research, it seems that the stronger adherence to individualizing foundations in Western countries [22, 57] relative to binding foundations is also reflected in the way individuals deal with political statements in their online network. Given the patterns observed in the present research, the question arises whether there is a likelihood that online networks become clustered by moral foundations, potentially leading to a disconnection between those who prioritize individualizing versus those who prioritize binding foundations. While previous empirical research did not corroborate the notion that social media users are captivated in politically like-minded cocoons [9–11], it seems worthwhile to assume a more complex view and examine whether sub-networks in online communication can be characterized by homogenous moral values. As suggested by Greene [62], the investigation of moral conflicts and their related emotional tensions could contribute to explaining modern tribalism, potentially also in online environments.

The threat that online communication leads to full moral clustering, though, appears rather unlikely as is suggested by Study 2 when the relational context is taken into account. In line with previous works [2, 5], in disregard of the moral violation, people were more unlikely to unfriend or block someone when this person was relationally close (compared to relationally distant). The findings also offer an explanation why: Since relationally closer ties are more likely to offer emotional support [30], individuals seem to be willing to tolerate moral violations and not terminate the digital tie, in return of maintaining the reception of social support. This result clearly shows the boundaries of the predictive value of a moral judgment in relation to a political statement and indicates the importance of a relational context for consideration. This research, thus, represents a first step to theorize the inner trade-off individuals go through when making a decision about digital interpersonal relationships in the face of political debates. In a nutshell, a specific type of social support, that is, emotional support, seems to be an inhibitor of terminating a digital connection when exposed to a political statement that violates a moral foundation. In light of the vivid debate on users’ active homogenization of their online networks in terms of “echo chambers” [14, 15], this novel theoretical link between political disagreements, moral judgments, and interpersonal context in contemporary communication technologies reveals that there are boundaries to users’ selective avoidance of dissents. These boundaries seem to come into play when individuals can gain something from certain relationships (i.e., network ties). A psychological view on users’ decision to unfriend or block someone offers a fruitful ground for the discussion about why political and probably also moral diversity emerges and prevails in individuals’ online networks [10, 15, 41].

The present findings need to be interpreted in light of this research’s limitations. First, although this work assessed individuals’ past unfriending and blocking behavior, it predominantly employed a scenario-based approach relying on hypothetical unfriending decisions. Both types of measures, though, seem to offer concurring findings: Those who indicated that they had unfriended because of a political disagreement in the past were significantly more likely to express a higher likelihood of unfriending in the hypothetical scenarios (on a five-point scale/ Study 1: M = 2.96, SD = 0.90; Study 2: M = 3.27, SD = 1.20) than those who did not unfriend because of political reasons in the past (Study 1: M = 2.67, SD = 1.07; Study 2: M = 2.53, SD = 1.25) (Study 1: t(301.62) = -3.49, p = .001, Cohen’s d = -.29 / Study 2: t(583) = -6.58, p < .001, Cohen’s d = -.60). Participants’ answers on previous politically motivated unfriending actions also corroborated the relative importance of individualizing moral foundations, the role of relational closeness and social support. Thus, both studies reveal a significant connection between what participants told they did in the past and what they would do. Objective observations on how online networks and their potentially moral clusters change over time or mobile experience sampling questionnaires on users’ smartphones, though, would be an informative complement to the present research.

Second, while the present work speculates about the motivation behind unfriending decisions, the psychological processes at work are still to be uncovered. For instance, it is unclear to what extent unfriending decisions are driven by cognitive or by affective processes. Given that moral conflicts are often fueled by emotions [62], it seems plausible to assume that not every unfriending decision is the result of a rational calculus of costs and benefits when dissolving this relationship. Future research could address the psychological mechanisms behind the creation and modification of one’s online political network in a systematic manner.

Third, the composition of our samples should be taken into account. The fact that female and left-leaning participants (see A4 and A10 Figs in S1 File) were overrepresented in Study 1 leads to the question of how these variables influence unfriending decisions. While the gender-balanced sample of Study 2 indicates that politically motivated unfriending does not occur more often among women than among men, χ2 (1) = .61, p = .434, both studies revealed that left-leaning individuals (compared to right-leaning ones) are more likely to unfriend others. It seems worthwhile to scrutinize whether a certain political ideology comes with enhanced involvement in certain topics which, in turn, could lead to less tolerance (and a higher unfriending likelihood) when disagreements come up or whether this finding was due to the specific national context and selected topics.

To conclude, the present work extends previous research by providing initial evidence for the importance of morality as a link between encountering politically challenging content online and actively banning this kind of information from one’s news feed. If users estimate another user’s political comment to be morally wrong, they will be more likely to terminate the digital relationship with this person. The power of morality, though, is limited when users are aware of the social resources, i.e., emotional support, that the person that is potentially to be unfriended can offer. This research, thus, presents a new level of analysis in online networks that could contribute (a) to understand the potential of social media communication to foster (moral) tribalism and (b) to identify the limits of this potential moral segregation in light of the social benefits human beings provide to one another.