Citizens & Public Opinion
The effect of emotions experienced during COVID-news consumption on attitudes towards governmental COVID-measures moderated by gender
Abstract
Throughout the pandemic, governments across the world, to different extents, struggled to keep their citizens calm, satisfied and safe at the same time. Citizens over the world have been protesting against their government’s COVID-measures (BBC News, 2021). In order to understand how attitudes towards COVID-measures are shaped, it is important to look at emotions, because they significantly affect citizens’ evolution regarding different topics (Frederickson, 2004). Yet, when it comes to emotions, gender is an essential moderator to considers, due to men’s tendency to under-report their negative emotions (LeFrance & Banaji, 1992, p. 189, 191), as well as the different experiences of the pandemic (Galasso et al., 2020; Ryan & Ayadi, 2020). This paper looks into this triangle through analysing the data of a survey design. In contrast to the findings of Dai et al. (2021), this study has revealed that while desire (as a positive emotion) decreases the support towards the measures, disgust (as a negative emotion) actually increases the support for more measures. This might be an interesting finding for both, policy-makers and researchers.
Introduction
Contemporary studies have found that the situation of the pandemic has increased citizens’ negative emotions, which leads to lower levels of compliance with governmental COVID-measures (Dai et al., 2021). It is clear then, that emotions experienced during the pandemic have an impact on attitudes towards COVID related governmental measures. This phenomenon is in line with the psychological broaden-and-build theory, which argues that emotions (whether negative or positive) have a function in human evolution. While positive emotions broaden our worldview – meaning, learning from positive experiences lead to our open-mindedness –, negative emotions do the opposite. They narrow our worldview and make us focus solely on issues directly affecting us (Frederickson, 2004). Therefore, emotions experienced during the pandemic will highly affect citizens’ evolution when it comes to their sight. Consequently, it is scientifically important to understand which emotions dominate public opinion during the pandemic, in order to predict, and potentially, control citizens’ support towards- and compliance with the measures against the spread of the virus. For governments, it is also important to understand how to target people with different emotions, in order to keep them calm and satisfied. For example, people feeling disgusted will have a different trigger to go on a COVID-hotspot protest, or get vaccinated, than those feeling anxious. Since news are influential on our perception of topics (Christne & Huberty, 2007), it is best to ask:
RQ1: How do emotions experienced during COVID-news consumption affect attitudes towards the government’s COVID-policies?
Based on the findings of (Dai et al., 2021), we can expect that:
H1: The more negative the emotions primarily experienced during COVID-news consumption, the more negative the attitudes towards the government’s COVID-policies.
H2: The more positive the emotions primarily experienced during COVID-news consumption, the more positive the attitudes towards the government’s COVID-policies.
Gender is well-known for being a moderating factor when it comes to self-reported experience of emotions (LeFrance & Banaji, 1992, p. 189, 191).
RQ2: How does gender affect attitudes towards the government’s COVID-policies in relation to emotions experienced during COVID-news consumption?
H3: Women experience more negative emotions during COVID-news consumption, than men, which makes their attitude towards the government’s COVID measures more negative.
Theoretical framework
We know that the pandemic has increased the negative emotions of citizens, leading to lower levels of compliance with governmental COVID-measures (Dai et al., 2021). Importantly, the academia has uncovered that gender is a factor that shows significant differences – between men and women – when it comes to emotions. It has been found that men, in general, are less likely to self-report the true level of experiencing negative emotions. In other words, men tend to be less honest and report a lower level of negative emotion, than what they actually experience (LeFrance & Banaji, 1992, p. 189, 191).
Second, due to the effect of negativity bias – that is, citizens tend to focus more on negative information, than on positive ones –, we can assume that negative emotions have a greater impact in people, than positive ones. Negativity bias is linked to three main reasons: 1) due to the diagnosticity of the information coming through the COVID-related news (negative information goes against people’s expectations, so it is more outstanding), 2) as a consequence explained by the prospect theory (negativity bias occurs as people try to avoid making a bad decision), and 3) credibility (that is, negative information seems more honest to people) (van Noort & Willemsen, 2012). Under-reporting such emotions is therefore, problematic and needs our attention.
Third, existing research has shown that gender is an influential factor when it comes to both, attitudes towards government and experiencing the coronavirus pandemic. For example, Lizotte (2015) found that mothers are more likely to support increase in governmental services and public goods (e.g., public schools, or in this case, COVID-vaccination). As during the COVID-19 pandemic some governments provided free COVID-testing (e.g., the Netherlands) (GGD Amsterdam, 2021), while others did not (e.g., Hungary) (Kormányzati Tájékoztatási Központ, 2021), some ensured care for all (e.g., Australia) (Haseltine, 2021) while others were unable to (e.g., India) (Menon, 2021), this gender difference on issue positions might be interesting to analyse when it comes to attitudes towards government’s COVID-policies.
Last but not least, scholars have found that women and men experienced the pandemic differently. For instance, women took it more seriously than men (Galasso et al., 2020). This might then very well affect attitudes towards COVID-related governmental measures, since taking an issue seriously affects response (Bishop, 1990), and therefore, opinion. These so-calles ‘issue publics’ (Krosnick, 1990), clusters of issues of interest may very-well be formulated by gender as well, since different issues and policies are experienced differently by men and women. This is also true in the case of COVID-19 as a disease and as an economic issue (Ryan & Ayadi, 2020).
Yet, due to the newness of the topic as well as a lack of attention, there is not yet research on how gender plays a role in emotions that affect attitudes towards the government’s COVID-policies. The effect of gender on public opinion regarding the government is important because women are more likely to turn out to vote than men, therefore a difference in their opinion on the government’s COVID-policies – due to their different emotional experiences – may affect outcomes at upcoming elections (Lizotte in Bos & Schneider, 2015).
Data and methods
As the study has been part of a larger group of students’ research at the University of Amsterdam, the original survey that has been distributed contained several more items of measurements (questions). In this paper, only the data relevant for the RQ1, H1, H2, RQ2 and H3, mentioned above, are analysed and discussed.
Sample
Data was collected between the 22nd October and the 8th November, 2021. The Qualtrics survey’s access-link has been distributed by the researchers in a snowball sample across their networks, out of convenience. This was because within the limited time-frame, random sampling, for an ideally socio-economically diverse population was not feasible. Although, this negatively affects both, the reliability as well as validity of the research and its findings. However, this method of sampling has allowed for a higher turnout within the very limited time-period. Also, the internationality of the researcher-team has supported us in covering a culturally, ethnically, and demographically still diverse population – although not age-wise, because the researchers distributing the survey were in their ‘student age’, with mostly student acquaintances. It is also worth noting that participants were not offered any incentives for their participation in the time-consuming survey.
From the participants, a total of 18 have been removed after the dataset from the survey has been created. Those who have been removed did 1) not respond to all questions measuring the dependent- or independent variables, or 2) spend an outstandingly high- or low amount of time with the survey (that is .505 or less-, 30000 or more seconds). No entry had to be removed due to being underaged – no person under the age of 16 has consented to being older than 16 and still filled out the survey. In addition, the other research groups’ questions have also been removed, for a clearer dataset and analysis. The final dataset analysed contained a total number of 216 participants (N=216). Among these, the average time of completing the survey was 33 minutes (M=1964.94, SD=2245.97).
The snowball sample still resulted in some socio-economic diversity. The average age was 30 years, the youngest being born in 2005 and the eldest in 1951. Participants were mostly women, (n=140), almost twice as many as men (n=72), while there were two non-binary participants (n=2) and further two who did not want to share their gender-identity (n=2) in the survey.
The survey has been filled out by people from 27 countries of residence, with 24 different nationalities. Due to the university’s location, the most represented country was the Netherlands (nationality 35.6%, residence 42.1%). Other most represented countries included the United States (US, nationality 19%, residence 17.6%) and Germany (nationality 17.6%, residence 13.9%). One continent has not been represented in the survey at all, namely Africa. The representation is biased towards European citizens (67.1%) and residents (74.5%) – therefore, towards the Global South. In terms of educational background, 42.6% of participants have obtained at least one bachelor’s degree and 35.2% a master’s degree. Regarding political self-placement, the mean value of 3.466 (where 0 = left, 10 = right) among respondents indicated a bias towards the left-side of the political axis. 75.2% positioned themselves on the left, while 16.5% on the right-wing, but only 2.3% choose the extremes (0 or 10), while 8.3% the central value (5).
Although English was not the primary language of 25% of participants, 90.7% indicated that their English skills were ‘somewhat good’, and only one respondent found his/her/their knowledge of the language ‘somewhat bad’. Therefore, the survey’s English language did not create a significant barrier, and consequently did not influence the responses by, for example, misunderstanding of questions.
Method
Participants of the cross-sectional survey design through the snowball sampling have filled out an online questionnaire. The target audience has included anyone of or over the age of 16, besides the post hoc filtering of items (questions), non-respondents and outliers, as mentioned above. Faulty scales (e.g., attitude towards COVID-measures ranging -3 to 3, importance of emotions ranging from 9 to 13) have been recoded, as well as date of birth into age, and countries of nationality and residence into continents (due to low representation per country). Those who failed the ‘banana-test’ (attention check), were kept in the final dataset, because the question was found too ‘tricky’ and hidden by the researchers, after the survey has been closed. Participants still gave reasonably varied answers – so we could conclude that they were not only went with the easiest available option (e.g., pressing the option on the top, leaving the indicator on the scale where it originally was).
Measures
Attitude towards governmental COVID-measures.
The dependent variable (‘What do you think about the measures that the government in your country might have enforced to reduce the spread of COVID-19?’) has been measured by a 7-point Likert-scale (from ‘not enough’ to ‘too much’) through eight popular COVID-measures adopted by many governments across the world (Ritchie et al., 2020), adjusted by the researchers. These included: mask mandate in public space, vaccine pass, curfew, social distancing, public health campaigns to promote vaccines, cancellation of large-scale physical events, restricted in-person education/work, and hospitality sector’s shutdown.
The reliability analysis of the eight items measuring attitude towards the measures showed an good internal consistency, with a Cronbach’s alpha of .80 (masks M=4.50, SD=1.64, vaccine M=4.29, SD=1.67, curfew M=3.37, SD=1.52, social distancing M=4.23, SD=1.46, campaign M=4.55, SD=1.60, events M=4.17, SD=1.63, education M=3.82, SD=1.46, hospitality M=3.88, SD=1.44). The factor analysis has also confirmed that all items were measuring the same factor: attitudes towards COVID-measures.
Emotions experienced during last recalled COVID-news consumption.
The independent variable (How did you feel the last time consuming COVID-related news?) has been enquired through a ranking scheme, whereby participants had to place the emotions in order, from the most- to the least experienced. These emotions were borrowed from Plutchik (2001): sadness, anger, fear, disgust, anxiety, happiness, desire, relaxation.
Gender.
The gender moderator, most important for this paper, has been measured through the question of ‘what is your gender?’ with pre-given answer options of male, female, other/non-binary, and prefer not to say. The sample contained 140 women, while only 72 men (M=.64, SD=.49, whereby 0=male 1=female).
Age.
Data about participants’ age, as a moderating effect, has been collected through asking about the year they were born in. The pre-given answer options ranged from 1920 to 2010 – to make sure, besides the consent, that those under 16 become visible in the dataset. Among participants (M=30.30, SD=12.92), the lowest age was 16 and the highest 70.
Nationality.
Participants were asked about their nationality through a list of 195 pre-given country-options (in case of multiple nationalities, they were asked to choose their primary one). The most occurring country has been the Netherlands (n=77), Germany (n=38) and the US (n=41), while other nationalities have been represented 10 or less times.
Country of residence.
Participants were asked about their country of residence– particularly important when it comes to governmental COVID-regulations – through a pre-given list of 195 country-options. Country of residence of participants has mostly been the Netherlands (n=91), Germany (n=30) and the United States, while other countries have been represented 10 or less times. Overall, the distribution of country of residence aligns with that of country of nationality.
Highest level of completed education.
The moderation of level of education has been assessed through a question, asking participants to indicate their highest level of completed education. Pre-given answer-options included no completed education, primary education, secondary education, high school or equivalent, post-obligatory technical education, undergraduate university degree (Bachelor), graduate university degree (Masters), doctoral degree (PhD, MD, JD) or equivalent, and other (‘please specify’). Among participants (M=6.23, SD=1.17, whereby 6=undergraduate- and 7=graduate university degree), the frequency was the highest for bachelor’s- (n=92), and master’s degree (n=76).
Political self-positioning.
Participants were asked to self-report their left-right political alignment through an 11-point scale (0 = very left, 10 = very right). Since citizens have generally been found to know the left-right axis and their own position on it well (Coughlin and Lockhart in Lesschaeve, 2017, p. 357), this item of measurement was reliable. On the ‘very left’ – ‘very right’ scale (M=3.466, SD=2.03), the most frequently occurring value has been 2 and 3 (n=54 respectively), followed by 4 (n=24), 1 (n=19) and 5 (n=17).
Personal importance of emotions.
It was also important to measure how important – influential – people generally perceived their emotions to be. For the question of ‘in general, how important would you say emotions are to you?’, they could choose form five pre-given answers, ranging from ‘not at all important’ to ‘extremely important’. Importantly for this study on the effect of emotions, participants perceived emotions to be moderately- to very important (M=3.80 SD=.88).
English language knowledge.
Finally, due to the distribution of the survey across nations, the potential barrier of the survey being provided in English had to be tested. For this, participants were asked whether their primary language was English (through a yes-no binary question). For those who answered no, a question about their level of ability to understand written and spoken English was added. They could choose from five options, ranging from ‘very bad’ (0) to ‘very good’ (4). Overall, those whose English was not their first language (n=162) were still ‘somewhat good’ to ‘very good’ in understanding written and spoken English (M=3.58, SD=.675). Therefore, language did not limit the reliability of the data gathered.
Results
Hypotheses 1 and 2 expected a positive main relationship between negative emotions and negative attitudes towards governmental COIVD-measures. Meaning, the more of the former should lead to more of the latter.
The first set of regression analyses (N=200) has revealed that mostly negative emotions are experienced during COVID-news consumption. Sadness (6.59) the most, followed by anger (5.99), anxiety (5.33), disgust (5.08), fear (4.96), desire (3.11), relaxation (2.68) and happiness (2.26). Although, only the regression analysis of disgust, F(7, 192) = 5.19, p < .001, and desire, F(7, 192) = 5.11, p < .001, have demonstrated significant effect on the average attitudes towards governments’ COVID-measures (model 0d – disgust: b= -.07, b*= -.14, t= -2.10, p= .037, 95% CI[-.14, -.00], model 0g – desire: b= -.07, b*= -.04, t= -1.99, p= .048, 95% CI [-.14, .00]). While in the other (significant) models, sadness (F(7, 192) = 4.53, p < .001, b= .03, b*= .04, t= .64, p= .522, 95% CI [-.06, .11]), anger (F(7, 192) = 4.61, p < .001, b= .04, b*= .07, t= .96, p= .337, 95% CI [-.04, .11]), fear (F(7, 192) = 4.82, p < .001, b= .05, b*= .10, t= -1.47, p= .144, 95% CI [-.02, .94]), anxiety (F(7, 192) = 4.48, p < .001, b= -.01, b*= -.03, t= -.36, p= .718, 95% CI[-.08, .006), happiness (F(7, 192) = 4.54, p < .001, b= .03, b*= -.05, t= -.69, p= .492, 95% CI [-.06, .12]), and relaxation (F(7, 192) = 4.62, p < .001, b= .03, b*= .07, t= .097, p= .331, 95% CI [-.04, .10]) remained insignificant in this sample.
Age, continent of nationality, continent of residence and level of education have been found to be non-significant (p > .05) influencing factors in the main relationship (see models 0a-0h in Table 1). Yet, political-self-placement has been proven to be a significant moderating effects of all emotions (anger: b=-.12, b*=-.25, t= -3.54, p < .001, 95% CI [-.18, -.05], sadness: b=-.12, b*=-.26, t=-3.66, p < .001, 95% CI [-.19, -.06], fear: b=-.12, b*=-.24, t=-3.48, p < .001, 95% CI [-.18, -.05], disgust: b=-.12, b*=-.26, t=-3.71, p < .001, 95% CI [-.19, -.06], anxiety: b= -.12, b*=-.26, t=-3.65, p < .001, 95% CI [-.19, -.06], happiness: b=-.12, b*=-.26, t=-3.66, p < .001, 95% CI [-.19, -.06], desire: b=-.11, b*=-.24, t=-3.43, p < .001, 95% CI [-.18, -.05], relaxation: b=-.20, b*= -.26, t=-3.73, p < .001, 95% CI [-.19, -.06]).
Although non-significantly, differences have been found among the continents. The one-way ANOVA tests showed that people with different nationalities experienced all the emotions to a different extent (see Table 2 in Appendix A), similar to based on their residency (more importantly regarding COVID-measures). In Europe (M=6.62) and North America (M=6.62) sadness was more experienced than elsewhere (M=6.36), similarly to anger (North America M=6.23, Europe M=6.02, other M=5.44), anxiety (North America M=5.80, elsewhere M=5.28, Europe M=4.88), and disgust (North America M=5.50, rest of the world M=5.15, Europe M=4.96). While fear showed the opposite tendency (rest of the world M=5.15, Europe M=5.00, North America M=4.70), similarly to happiness (rest of the world M=2.76, North America M=2.33, Europe M=2.16), and relaxation (rest of the world M=3.32, Europe M=2.68, North America M=2.25). Desire was outstanding in Europe (M=3.28) compared to North America (M=2.58) and other continents (M=2.92).
In terms of the dependent variable, the attitude towards COVID-measures was significantly different across continent of nationality. European citizens – who felt more sadness and desire –were more likely to perceive the implemented measures to be ‘too much’ (M=3.89), while in North America (mostly feeling anger and anxiety) they were rather implying that measures were ‘not enough’ (M=4.80). Only these two continents had a sample size that allowed for such conclusions about their citizens (North America 18.5%, Europe 74.5%). Since the country of residence is more important when it comes to the COIVD-measures being experienced, I have also tested this moderator. The results were somewhat reflecting the results by nationality (North America M=4.71, Europe M=4.15).
More importantly, gender has been found a significant moderator of disgust (b=-.25, b*=-.14, t=-2.00, p =.046, 95% CI[-.49, -.00]) and fear (b=-.25, b*=-.14, t=-1.97, p =.050, 95% CI [-.49, .00]) on attitudes towards COVID-measures. Gender has been making a difference in the distribution of emotions experienced as well. Going against the expectations, men reported more sadness (M=3.54, SD=1.42) and disgust (M=5.16, SD=1.95), but also happiness (M=2.46, SD=1.63), desire (M=3.37, SD=2.01) and relax (M=3.18, SD=2.13), than women (respectively M=6.55, SD=1.55, M=5.04, SD= 1.81, M=2.15, SD=1.35, M=2.96, SD=1.64, M=2.40, SD=1.60). Women felt more anger (M=6.09, SD=1.54), fear (M=5.16, SD=1.66) and anxiety (M=5.64, SD=1.79), than men (respectively M=5.82,SD=1.92, M=4.59, SD=1.01, M=4.75, SD=1.88). These numbers align with H3: women do experience more negative emotions.
Overall, H1 and H2 are mostly rejected because the opposite has been found: one negative emotion (disgust) increases-, while a positive emotion (desire) decreases support towards COVID-measures.
Hypothesis 3 has expected that, within the main relationship, gender moderates the effect of experienced emotions on COIVD-measure-attitudes, because women have experienced more negative emotions during the pandemic. Therefore, despite the negative emotions increasing negative attitudes, women should have more negative attitudes than men.
The second set of regression analyses (N=200) has revealed that no emotion had a significant effect on average attitudes towards COVID-policies, when the interaction-effect of gender is taken into account (sadness: b=-.04, b*=.06, t=-.75, p = .456, 95% CI [-.06, .13], anger: b=-.05, b*=.08, t=1.04, p = .301, 95% CI [-.04, .13], fear: b=.07, b*=.13, t=1.62, p = .106, 95% CI [-.02, .16], disgust: b=-.07, b*=-.13, t=1.53, p = .129, 95% CI [-.15, .02], anxietyb=.02, b*=.04, t=-.47, p =.637, 95% CI [-.07, .12], happiness: b=.05, b*=.08, t=.83, p = .409, 95% CI [-.07, .17], desire: b=-.08, b*=-.16, t=-1.94, p =.054, 95% CI [-.17, .00], relaxation b=.04, b*=.07, t=.86, p =.390, 95% CI [-.05, .11]).
While the socio-economic factors (age, nationality, country of residence, level of education) remained insignificant (p > .05, see models 1-8 in Table 1), political-self-placement has remained significant on all emotions (sadness: b=-.12,b*=-.26, t= -3.67, p < .001, 95% CI [-.19, -.06], anger: b=-.12, b*=-.25, t=-3.52, p < .001, 95% CI [-.18, -.05], fear: b=-.12, b*=-.25, t=-3.54, p < .001, 95% CI [-.18, -.05], disgust: b=-.12, b*=-.26, t=-3.71, p < .001, 95% CI [-.19, -.06], anxiety: b= -.13, b*=-.27, t=-3.75, p < .001, 95% CI [-.19, -.06], happiness: b=-.12, b*=-.26, t=-3.65, p < .001, 95% CI [-.19, -.06], desire: b=-.11, b*=-.24, t=-3.44, p < .001, 95% CI [-.18, -.05],relaxation: b=-.12, b*= -.26, t=-3.71, p < .001, 95% CI [-.19, -.06]).
Although, gender has lost its significance on all emotions (sadness: b=-.16, b*=-.09, t= -.82, p =.411, 95% CI [-.54, .22], anger: b=-.16, b*=-.09, t=-.87, p =384, 95% CI [-.03, .20], fear: b=-.16, b*=-.09, t=-90, p =374, 95% CI [-.50, .19], disgust: b=-.20, b*=.12, t=1.71, p =.088, 95% CI [-.02, .21], anxiety: b= -.05, b*=-.03, t =-.29, p =776, 95% CI[-.43, .32], happiness: b=-.15, b*=-.08, t=-.85, p =.396, 95% CI [-.50, .20], desire: b=-.27, b*=.16, t=-1.64, p =.102, 95% CI [-.58, .05],relaxation: b=-.22, b*= -.12, t=-1.40, p =.164, 95% CI [-.52, .09]), and none of its interaction-effects have been found to be significant (gender*sadness: b=-.01, b*=-.04, t= -.40, p =456, 95% CI [-.06, .13], gender*anger: b=-.01, b*=-.04, t=-.40, p =692, 95% CI [-.08, .05], gender*fear: b=-.03, b*=-.08, t=-.72, p =.471, 95% CI [-.10, .05], gender*disgust: b=-.01, b*=-.04, t=-.32, p =.752, 95% CI [-.09, .07], gender*anxiety: b= -.05, b*=-.15, t=-1.13, p =261, 95% CI [-.13, .03], gender*happiness: b=-.03, b*=-.05, t=-.47, p=636, 95% CI [.10, .38], desire: b=.02, b*=.05, t=.48, p =.635, 95% CI [-.07, .15], gender*relaxation: b=-.00, b*= -.00, t =-.02, p=.981, 95% CI [-.10, -10]). Therefore, gender on its own is not a significant moderator of emotions experience during COVID-news consumption, on attitudes towards COVID-measures. Therefore, H3 is rejected through the analysis of this sample, because gender does not moderate the effect of emotions towards COVID-policies. Yet, women have experienced more negative emotions than men (as found above) and were also slightly more against the measures (M=4.09) than men (M=4.12). Therefore, these assumptions of H3 are still visible. Even though, gender does not make a difference through emotions experienced, as expected, but it still affects attitudes towards COVID-policies, that is important for upcoming elections and public compliance with the governments’ regulations.
Discussion and conclusion
The results of the main relationship can be translated as follows. Besides the expectations of H1 and H2 – that negative emotions negatively influence attitudes towards COVID-measures –, the only emotions having significant impact were disgust and desire. One negative and one positive emotion, questioning the negativity bias. Also, socio-economic factors have mostly been found to be insignificant moderators in the relationship, besides political self-placement on all-, ang gender on the emotions of fear and disgust.
Therefore, negative emotions experience during COVID-news consumption have not been found to be influential on attitudes towards COVID-policies. Opposite to the expectations based on Dai et al. (2021), only the experience of more desire (a positive emotion) increases negative attitudes towards COVID-measures (seeing them as ‘too much’), while disgust has increased support for the measures (seeing them as ‘not enough’). 21% variance has been explained by disgust while 15% by desire, on attitude towards COVID-measures. Also, the emotions experienced are only affected by political self-placement, and fear and disgust by gender. Not any other socio-economic factor, assessed in this study.
The results of the moderation-relationship have revealed that gender, alone, has no effect on any of the emotions, in relation to attitudes towards COVID-measures, since its significance ceases to exist once its interaction-effect is being taken into account. Furthermore, applying the interaction-effect of gender in the model eliminates the significance of both, disgust and desire, which were found significant in the base model (models 0a-0h).
Potentially, with a larger sample, insignificant but visible effects – of emotions, gender, nationality, and residence – could have been revealed.
Limitations
Besides some relevant findings, this study also has its limitations. First, the self-reporting data-collection method gave space for social-desirability bias. Second, the sets of questions measuring emotions were in the end of the long survey, therefore participants’ attention might have declined. Third, the convenience sampling strategy has led to 1) the unequal distribution of men and women (creating a gender bias, that is particularly important because women, who were in 2/3 majority, report more negative emotions and experience the pandemic more negatively), and 2) a highly educated European population. This sample does not allow for generalisability. Fourth, the pandemic has been going on for a while, negative emotions (and their frequency) might have changed over time. Fifth, in the emotion-ranking question, ‘sadness’ was on top by default. Since this emotion became the most frequently ranked one as number one, we cannot exclude the possibility of participants’ inaction. Fifth, this item of measurement did not allow participants to give two or more emotions the same weight, nor to indicate if one or more of them they have never experience – the latter also being a problem on COVID-measures, varying country by country. In addition, the distribution of negative- and positive emotions was not equal. A final limitation of our research is that participants had to self-report their experienced emotions during the last time they consumed COVID-news. This can be problematic due to people’s limited capacity to recall (Lang, 2000).
Conclusion
Negative emotions have been found to be the most experienced emotions during COVID-news consumption. These emotions have an effect on public attitudes towards government’s COVID-measures. This report has revealed that, despite past research showing that negative emotions should be minimised by the government during a pandemic – in order to increase the public compliance with restrictive measures (Dai et al., 2021) – disgust is a negative emotion that actually increases the wanting of further measures. I have found that the emotions of desire and disgust – experienced during COVID-news consumption – have a significant effect on attitudes towards governmental measures against the virus. Again, against the expectations, people who felt more desire, believed the COVID-measures were too much – and wanted less. Even though, there were visible differences among men and women in experiencing emotions as well as attitudes on the measures, it has not been a significant moderator.
Although, the direction of the effect is not clear after this research – whether it is the emotion influencing attitudes, or the other way around. Future research could focus on this, while using larger sample-size that could potentially show the significance of the other (negative) emotions as well.
References
BBC News (2021, November 21). COVID: Huge protests across Europe over new restrictions. BBC. Retrieved from https://www.bbc.com/news/world-europe-59363256
Bishop, G. F. (1990). Issue involvement and response effects in public opinion surveys. Public Opinion Quarterly, 54(2), 209-218. https://doi.org/10.1086/269198
Christen, C. T., & Huberty, K. E. (2007). Media Reach, Media Influence? The Effects of Local, National, and Internet News on Public Opinion Inferences. Journalism & Mass Communication Quarterly, 84(2), 315–334.
https://doi.org/10.1177/107769900708400208
Dai, W., Meng, G., Zheng, Y., Li, Q., Dai, B., & Liu, X. (2021). The Impact of Intolerance of Uncertainty on Negative Emotions in COVID-19: Mediation by Pandemic-Focused Time and Moderation by Perceived Efficacy. International Journal of Environmental Research and Public Health, 18(8), 4189. https://doi.org/10.3390/ijerph18084189
Fredrickson, B. L. (2004). The broaden–and–build theory of positive emotions. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 359(1449), 1367-1377. https://doi.org/10.1098/rstb.2004.1512
Galasso, V., Pons, V., Profeta, P., Becher, M., Brouard, S., & Foucault, M. (2020). Gender differences in COVID-19 attitudes and behavior: Panel evidence from eight countries. Proceedings of the National Academy of Sciences, 117(44), 27285-27291. https://doi.org/10.1073/pnas.2012520117
GGD Amsterdam (2021, November 12). Coronavirus testing at GGD Amsterdam. GGD Amsterdam. Retrieved fromhttps://www.ggd.amsterdam.nl/english/coronavirus/testing-amsterdam/
Kormányzati Tájékoztatási Központ [Government Information Centre] (2021, August 6). Kormányrendelet szabályozza mely esetekben ingyenes a PCR teszt műtéti beavatkozás előtt [Governmental decree controls in what case the pre-surgery PCR test can be free of charge]. Kormányzati Tájékoztatási Központ. Retrieved from https://koronavirus.gov.hu/cikkek/kormanyrendelet-szabalyozza-mely-esetekben-ingyenes-pcr-teszt-muteti-beavatkozas-elott
Haseltine, W. A. (2021, March 24). What Can We Learn From Australia’s COVID-19 Response?. Forbes. Retrieved from https://www.forbes.com/sites/williamhaseltine/2021/03/24/what-can-we-learn-from-australias-COVID-19-response/?sh=6408005e3a01
Krosnick, J. A. (1990). Government Policy and Citizen Passion: A Study of Issue Publics in Contemporary America. Political Behavior, 12(1), 59–92. http://www.jstor.org/stable/586285
LaFrance, M., & Banaji, M. (1992). Toward a reconsideration of the gender-emotion relationship. Emotion and social behavior, 14, 178-201.
Lang, A. (2000). The limited capacity model of mediated message processing. Journal of Communication, 50(1), 46-70. https://doi.org/10.1111/j.1460-2466.2000.tb02833.x
Lesschaeve, C. (2017). The predictive power of the left-right self-placement scale for the policy positions of voters and parties. West European Politics, 40(2), 357-377. https://doi.org/10.1080/01402382.2016.1229088
Lizotte, M-K. (2015). The gender gap in public opinion. In A. L. Bos & M. C. Schneider Eds.), The Political Psychology of Women in U.S. Politics (pp. 51-70). Retrieved from https://books.google.nl/books?hl=en&lr=&id=DCglDwAAQBAJ&oi=fnd&pg=PA51&dq=gender+political+opinion&ots=VGK487JZXa&sig=Sawcm05PDFuZx_MjrQuM0kk3CqI&redir_esc=y#v=onepage&q=gender%20political%20opinion&f=false
Matthews, V., Stough‐Hunter, A., & Marazita, J. M. (2021). Attitudes towards social distancing in response to COVID‐19. Public Health Nursing (Boston, Mass.), 38(6), 1019–1029. https://doi.org/10.1111/phn.12954
Menon, S. (2021, May 7). India COVID: How bad is the second wave?. BBC. Retrieved fromhttps://www.bbc.com/news/56987209
Plutchik, R. (2001). The Nature of Emotions: Human emotions have deep evolutionary roots, a fact that may explain their complexity and provide tools for clinical practice. American Scientist, 89(4), 344–350. http://www.jstor.org/stable/27857503
Ritchie, H., Mathieu, E., Rodés-Guirao, L., Appel, C., Giattino, C., Ortiz-Ospina, E., Hasell, E., Macdonald, B., Beltekian, D. and Roser, M. (2020). Coronavirus Pandemic (COVID-19). Our World in Data. Retrieved from https://ourworldindata.org/coronavirus
Ryan, N. E., & El Ayadi, A. M. (2020). A call for a gender-responsive, intersectional approach to address COVID-19. Global Public Health, 15(9), 1404-1412. https://doi.org/10.1080/17441692.2020.1791214
Tskhay, K. O., & Rule, N. O. (2015). Emotions facilitate the communication of ambiguous group memberships. Emotion, 15(6), 812–826. https://doi.org/10.1037/emo0000077
Zimmermann, & Iwanski, A. (2014). Emotion regulation from early adolescence to emerging adulthood and middle adulthood: Age differences, gender differences, and emotion-specific developmental variations. International Journal of Behavioral Development, 38(2), 182–194. https://doi.org/10.1177/0165025413515405
Van Noort, G., & Willemsen, L. M. (2012). Online damage control: The effects of proactive vs. reactive webcare interventions in consumer-generated and brand-generated platforms. Journal of Interactive Marketing, 26, 131–140. https://doi.org/10.1016/j.intmar.2011.07.001
Appendix A
Tables & figures
| sadness | anger | fear | disgust | anxiety | happiness | desire | relax. |
N.American | 6.57 | 6.23 | 4.69 | 5.50 | 5.71 | 2.33 | 3.34 | 2.29 |
European | 6.74 | 6.02 | 4.93 | 4.89 | 5.21 | 2.20 | 2.64 | 2.74 |
rest o. W. | 5.77 | 5.44 | 5.58 | 5.54 | 5.36 | 2.50 | 2.54 | 2.96 |
Comments
Post a Comment