Following the 2016 Presidential Election: Positive and Negative Mood Affect and the Impetus Towards Activism

Given the tenuousness of the U.S. presidential race and subsequent election, it appears that more individuals are becoming politically active. What remains unclear is how the current political climate has influenced citizens regarding their perceptions of feeling psychologically safe, the related mood affect, and their impetus towards activism. Additionally, the role of social media in today’s world is seemingly a key stimulus for individuals who were typically less involved in the past become more so now. Given there may be a number of reasons that individuals feel moved to engage in discourse and action that others might view as unconventional or rebellious, the intent of this study was to investigate activism and mood affect as mediated by social media use to better understand people’s motivation for becoming more socially engaged. While historically, there have been numerous movements to address perceived societal and governmental injustices, how safe individuals feel when becoming involved is still up for debate. Findings from this study showed that individuals experienced positive and negative mood affect in both the likelihood they will engage in a variety of activist behaviors in the future, as well as seeing themselves as being politically active. Furthermore, the role of social media also showed a relationship between participant’s positive and negative mood affect, as well as sense of activism, demonstrating the impact of social media on how people express their opinions today.


INTRODUCTION
Many in the United States would agree that the 2016 presidential election was tenuous at best. From the moment the candidates announced their intention to run, the presidential field was teeming with divisive banter. -18 FOLLOWING THE 2016 to find like-minded people more easily, compelling individuals into action or further engagement. This precarious balance of feeling safe, yet being called into action has limited understanding to date. Research by the Pew Institute indicates that people are "worn out" from the political content that they encounter and that 59% of people engaging with "those with opposing political views" find these interactions to be "stressful and frustrating" (Duggan & Smith, 2016). Due to the limitations of understanding, and the unusual social environment that the 2016 presidential election has perpetuated, the intent of this study was to measure the relationship between mood affect and activism, as moderated by social media use in the context of the election. The goal of the research was to forward our understanding regarding individuals' feeling of safety towards activism identity and behavior, and how the use of social media influenced these impulses.

Activism
Ozymy (2012) suggests that there is a certain "underlying assumption concerning the rationality of political participation" (p. 104) in that "individuals engage in political activity to pursue particular goals, and they decide to participate when the benefits of such activity outweigh the costs" (Leighley 1995, p. 7). Scholars have found that a number of factors are associated with increased political participation or activism, including having social sciences and humanities majors as undergraduates, coming from a higher socioeconomic familial background, and having highly educated parents, among others (Altbach, 1989;Astin, Astin, Bayer, & Bisconti, 1975). In addition, identifying as a person of color and/or woman and having beliefs that come from a marginalized background can also impel the need to become an activist (Linder & Rodriguez, 2012).
To understand the nuances of activism, Klar and Kasser (2008) proposed four dimensions: activist identity, commitment, behavioral intentions, and past behaviors. Activist identity was described as the extent to which one participates in activism and has been associated with one's social or collective identity (Stryker, Owens, & White, 2000). Commitment to activism has been widely discussed in social movement research (e.g., Klandermans, 1997), and findings have suggested that a strong predictor of activist behavior is a person's activist identity in their hierarchy of roles (Stryker, 2000). Music and Wilson (2008) propose that activism is a collective action for a collective good. Therefore, activists view the social structure as a target of intervention, rather than a framework within which to work. As such, activists seek to create change at the local, regional or global level. In addition to the societal benefits of challenging social structures, Adler con-tended that interest in fostering the welfare of others is a fundamental human endeavor and that an "expression of social interest was a prerequisite for psychological health" (as cited in Klar & Kasser, 2009, p. 756). Thus, community engagement has been linked to improved personal empowerment (Christens et al., 2011), and political activism and improved individual well-being connections have been found (Klar & Kasser, 2008).
In addition to the benefits of community engagement and activism, the impetus towards action may be explained through a number of lenses. For example, stress levels may play a role. Stanton, LaBar, Saini, Kuhn and Beehner (2010) concluded that individuals supporting a losing political candidate had significantly higher levels of cortisol (a hormone released in response to stress) than those who supported the winning candidate (p. 772). Another motivation includes the extent to which one participates in activism and the association with one's social or collective identity (Stryker, Owens & White, 2000). Stryker, Owens and White's (2000) research suggests that a strong predictor of activist behavior is the relative position of the activist identity in a person's hierarchy of roles. Motivational components are also widely discussed in social movement research (e.g., Klandermans, 1997).
Thus, there are a number of beneficial and motivational factors associated with the impetus towards activism. Based from a definition by Corning and Myers (2002), for the purpose of this research, we are defining activism as the behavior of advocating for some kind of political cause or causes (for example, human rights issues, protecting the environment, or protesting wars). These may range from acts such as posting on social media to more rebellious such as civil disobedience.

Social Media and Activism
One venue that has helped raise the collective consciousness is social media (Martinez, 2017). In the last decade, social media outlets have been a platform to facilitate a global-local orientation to and action the world (Sobré-Denton, 2016). Additionally, social media has been an avenue that has created a new dimension in becoming more socially engaged.
In the U.S. alone after the 2016 U.S. election, news outlets began reporting on various polls conducted on political activism. As one example, Survey Monkey (2017) indicated roughly two-thirds of respondents had been involved in politics or causes within their communities between the time period of January and March 2017, with 34% claiming to have shared their opinions on social media.
Other movements and instances of solidarity have tapped into social media for community building as well. For instance, the Iranian women's movement transformed cyberspace into a place for social action (Abbasgholizadeh, 2014). In 2005, cyber resistance was first utilized during the rise of Mahmoud Ahmadinejad's radical government (Abbasgholizadeh, 2014). Furthermore, Velasquez and LaRose's (2015) work showed that with college student's political uses of social media might influence their participation in individual (e.g. Bakker and De Vreese, 2011) or collective political actions (e.g. Enjolras et al., 2013). The use of social media to express political views was also explored in Karamat and Farooq's (2016) study of Pakistani graduate students. These authors found that approximately 50% followed political pages on both Facebook and Twitter, and that nearly 50% asked friends to participate in political events or actions (Karamat & Farooq, 2016). More significantly, Karamat and Farooq (2016) discovered that nearly 70% of the participants had not participated in political activism prior to their engagement on social media. Thus, "the tactical deployment of digital communications by activists has been portrayed as an upscaling of interest and participation in contentious politics" (Bastos, Mercea, & Charpentier, 2015, p. 322). These prior studies indicate that social media has a significant impact on users' movement toward activism.

Mood Affect and Psychological Safety
According to Hackman (1992), affect has been recognized as a critical motivation in group environments. Political scientists believe that rational individuals will view the costs and benefits of political activity through the lens of self-interest (Platt 2008). In particular, positive affectivity (PA) and negative affectivity (NA) are "differentially associated with the direction, duration, and intensity with which personal resources are invested in social contexts" (Carver & Scheier, 1990, p. 21). While PA reflects the propensity to be energetic and optimistic (Watson, Clark, & Tellegen, 1988), "NA increases the tendency to perceive and ruminate over unfavorable information regarding ones' self and others" (Watson et al., 1988(Watson et al., , p. 1064). Edmondson and Lei (2014) suggest that these differences are likely to have an influence on the extent to which group environments are perceived as discouraging or encouraging.
Linked to affectivity, psychological safety is generally used to describe a cognitive construct manifested through a perception of intrapersonal risk of individuals within a group or organization (Chen, Gao, Sheng & Ran, 2015). Psychological safety is defined as "the belief that a person can express himself or herself without negative consequences" (Appelbaum et al, 2016, p. 344). Team members have a high level of psychological safety if they feel secure in expressing opinions, being open and honest, and asking for feedback without fear of punishment, sanctions, retribution or ostra-158 JOURNAL OF HATE STUDIES Vol. 14:153 cism (Chughtai, 2016;Edmondson, 1999;Mayfield, Tombaugh & Lee, 2016). Low-status members can feel anxiety and fear in situations in which they are required to take interpersonal risks. To be psychologically safe, therefore, members of a group need to feel respected and to respect other members within the group (Edmondson, 1999). Additionally, studies have positively correlated the psychological safety felt by followers with a leader's inclusivity and support (Appelbaum et al, 2016;Chughtai, 2016;Edmondson, 1999). Furthermore, Chughtai's (2016) research found a relationship between psychological safety and the perception of supportive leadership. As a part of mood research, studies were conducted to determine positive (PA) and negative affects (NA) in various settings. These studies demonstrated that a positive affect "reflects the extent to which a person feels enthusiastic, active, and alert. . .[whereas] a negative affect is a general dimension of subjective distress" (Watson, Clark, & Tellegen, 1988, p. 1066. Findings indicated that PA, and not NA, was related to social involvement, gladness and a prevalence of pleasant events. Moreover, Hood, Bachrach, Zivnuska and Bendoly (2016) found that PA and NA were positively and negatively related to psychological safety respectively. Collectively, studies utilizing PA and NA scales relate mood dimension with behavior. These findings imply that mood affect then might be utilized to predict the impetus toward being more of an activist in social concerns.

METHODS
The intent of this study was to determine the relationship between mood affect and activism, as mediated by social media use after the 2016 presidential election. To measure these variables three instruments were used to test the hypotheses. These instruments included the PANAS, developed by Watson, Clark and Tellegen (1988) to measure mood affect, as well as the Activism Orientation Scale (AOS) that measures one's behaviors as an activist (Corning & Myers, 2013), as well as the Activist Identity and Commitment Scale (AICS), by Klar and Kasser (2009) that measures activism identity. Additional questions were created regarding social media use as part of the research.

Sample
Using Facebook, participants were solicited to complete an online questionnaire using Survey Monkey. To gather responses, requests were made to forward the instrument to move beyond a sample of convenience towards a snowball effect. Ultimately, of the 380 individuals who began the Of the entire sample, the significant majority, 96%, stated they voted in the 2016 presidential election (see Table 1).

RESULTS
To analyze the data, statistical software was employed. Data analysis included simple statistics, Cronbach's alpha, factor analysis and correlation analysis to test for relationships (please see tables 7-10 in appendices for descriptive statistics). All of the instruments showed strong reliability ranging from .80-.95 (see Table 2). Initially, due to the construction of the social media instrument, a factor analysis was conducted on the 15 items. The data factorability was tested using Bartlett's test of sphericity (see Snedecor & Cochran, 1989), which was significant (p < .0001), suggesting that the data were suitable for factor analysis. An exploratory factor analysis (EFA) was initially conducted to approximate the number of factors to be retained. The factor communalities (i.e., the estimates of the variance in each variable accounted for by the factor solution) varied between .38 and .73. Based on the examination of the scree plot, it was determined that a two-factor solution should be retained, explaining 46.64% of variance. Descriptive statistics were then run on the four instruments including mean, standard deviation, skewness and kurtosis (see Tables 3-6).

Hypotheses One
Based from the literature and the inconsistent findings regarding the mood affect of becoming active (i.e. see Klar & Kasser, 2008;Stanton, et al., 2010), the first hypothesis was; There is a relationship between Positive Affect (PA) and Negative Affect (PANAS) and activism (AOS and AICS).
To answer this hypothesis, a regression analysis was ran that included PANAS positive affect (PA) and negative affect (NA) as predictors, and AOS as the dependent variable. The regression model explained 47.8% of variance in the AOS scores, with adjusted R2 = 0.478. The standardized â coefficients for PA and NA were â = .43 and .53 respectively (p < .0001).
When entered as a sole predictor of AOS, the PA accounted for 19.6% of variance and the standardized â coefficient = .44, p < .0001 while the NA accounted for 29.3% of variance in the AOS scores and the standardized â coefficient = .54, p < .0001.
A second regression included PANAS subscales (positive and negative) with the AICS as the dependent variable. Analysis showed that the two PANAS sub-scores were not significantly correlated to one another, thus multicollinearity was not an issue. The regression model between the PANAS and AICS explained 42.3% of variance in the AICS scores, with adjusted R2 = 0.423. The standardized â coefficients for PA and NA were â = .38 and .52 respectively (p < .0001). When entered as a sole predictor of AICS, PA accounted for 15.2% of variance and the standardized â coefficient = .39, p < .0001 while NA on its own accounted for 28.0% of variance in AICS scores and the standardized â coefficient = .53, p < .0001. Thus, hypotheses one was accepted.

Hypotheses Two
Due to implications of Amichai-Hambuger, Gazit, Bar-Ilan, Perez, Aharony and Dyne (2015), and Rau, Gao and Ding (2008) regarding mood and the use of social media, the second hypothesis was: There is a positive relationship between Positive Affect (PA) and Negative Affect (PANAS) and social media use. For this analysis, a linear regression equation was tested, with PA and NA scores as the predictors and the social media score as the dependent variable. This model explained 30.9% of variance in social media use, with adjusted R2 = 0.309. The standardized â coefficients for PA and NA were â = .26 and .48 respectively (p < .0001). When entered as a sole predictor of social media use, PA accounted for 7.5% of variance and the standardized â coefficient = .27, p < .0001 while NA on its own accounted for 23.6% of variance with a standardized â coefficient = .48, p < .0001. Thus, hypotheses two was accepted.

Hypotheses Three
As previous research has shown correlations between the use of social media in activism recently (e.g. see Bastos, et al., 2015;Karamat & Farooq, 2016), the third hypothesis was: There is a relationship between activism (AOS and AICS) and social media use. A linear regression predicting social media use from AOS and AICS revealed that 43.7% of variance in social media use, with adjusted R2 = 0.437. The standardized â coefficients for AOS and AICS were â = .23 and .46 respectively (p < .0001). (The correlation coefficient between AOS and AICS was r(349) = .78, p < .0001 and -18 FOLLOWING THE 2016 thus the effect size would be inflated if these two variables would be entered in two separate regression models). Thus, hypotheses three was accepted.

DEMOGRAPHIC ANALYSIS
Additional statistical analysis was then conducted on related demographic information. First, a t-test was run to compare males and females regarding their activism (AOS and AICS) scores, and social media scores. Significant differences were obtained for the AOC, t(360) = 4.26, p < .0001 (mean females = 2.36, mean males = 1.98); AICS, t(344) = 4.95, p < .0001 (mean females = 2.40, mean males = 1.97); and social media use, t(344) = .74, p < .0001 (mean females = 2.68, mean males = 2.30). This indicates that females viewed themselves significantly higher in regards to being an activist, or displaying activist tendencies on social media then males did.
To continue demographic analysis, an MANOVA was used to explore significant differences between age, political party affiliation, U.S. region, and education groups regarding their activism scores. The multivariate analyses were followed by univariate analyses and Tukey post-hoc analyses. Significant multivariate differences were obtained for political affiliation, F(4, 288) = 4.11, p < .05. Univariate analyses also revealed significant differences between political affiliation groups regarding both AOS scores, F(2, 176) = 4.38, p < .05 and AICS scores, F(2,176) = 1.36, p < .05.

DISCUSSION
The results of our study show several new insights, as well as align with previous research findings. For example, the participants, while overwhelmingly women, showed a relationship between activism and both positive and negative mood affect. This seemingly supports Calazza's (2005) findings that levels of perceived safety have a more significant influence on women's civic engagement than they do on men's. For women as a group, a sense of perceived safety is strongly related to involvement in the community, while a lack of perceived safety is linked to disengagement. These findings also indicate the complicated dynamic of being concerned with the times (NA), having an activist identity (AICS), as well as feeling the impetus to do something about one's trepidations (AOS). Furthermore, the correlation between both the identity of being an activist (AICS) as well as the intent of the perceived behavior of being active (AOS) supports Klar and Klasser's (2009) work. In addition, the relationships between positive mood affect and activism confirmations Zimmerman's (1995) work on empowerment that emphasized the situation-specific nature of this dynamic, sug-162 JOURNAL OF HATE STUDIES Vol. 14:153 gesting that focused activism, in this case via social media, could improve one's sense of control and safety. Previous research has also shown that there is growing "convergence in the knowledge that networked communication on social media is germane to protest participation" (Bastos et al., 2015, p. 322). Moreover, the prevalent interest and behavior conducive to political protest (e.g., using social media for news consumption) or embodied in political protests-for example, tweeting hash tagged action updates or expressing emotional support has also been documented (Bastos et al., 2013), and could be forwarded by mood affect, or feeling safe. As Martinez (2017) reflected, as a "Chicana feminist professor-scholar. . .I purposefully tuned in to mass and social media in search of competing interpretations of campaign rhetoric, sometimes seeking an array of perspectives to stay well informed. . .other times seeking solidarity in my sameness silos of like-minded Others" (p.146).

FUTURE RESEARCH
While the findings of this study show an interesting relationship between positive and negative affect and activism, further research is needed. This study looked at the relationship between these factors, but why individuals, specifically women, feel more of a tendency towards activist identity and characteristics has not yet been answered. Perhaps given the uptick in political dialogue on social media, women are finding a more comfortable venue for asserting their perceptions and concerns. Another premise may be the sense of community that social media has created (Amichai-Hamburger et al., 2015), and this feeling of belonging helps individuals overcome their negative mood affect. Future research could include a qualitative version of this study, inquiring why individuals feel called to become more active, as well as the forums in which they do so. Other research could include a longitudinal account of activism and mood affect, to see if this is a surge of activism that will be maintained, or if this is just a specific place-in-time event.    CONCLUSION Given the nature of activism and the tenuousness of the 2016 U.S. presidential race and subsequent election, it was our premise that more individuals are becoming socially active. This may be due to several distinctive factors; the high use of social media to both inform and organize gatherings, identifying and behaving as an activist, as well as the impetus toward being active, even when one's mood affect is tenuous and uncomfortable.
The intent of this study was to explore several potential factors that may better understand people's impetus for becoming more socially engaged. Findings indicate that participants, while experiencing both positive and negative mood affect, were still willing to engage as an activist, as well as identify as such. As shown, the use of social media had a relationship to participant's positive and negative affect implying that social media is providing a venue for people to voice more of their political perspectives. While previous research had showed this relationship in youth, this study had a predominantly adult population. The findings in this study are consistent with previous research; however, these results also help forward understanding in that previous research had not discussed the connection between mood affect and activism mediated by social media use. Invite a friend to attend a meeting of a political organization or event.
2.59 1.08 Serve as an officer in a political organization.
1.87 .08 Engage in a political activity in which you knew you will be arrested.

.85
Attend an informational meeting of a political group.

.84
Give a lecture or talk about a social or political issue.