Uninterested, disenchanted, or overwhelmed? An analysis of motives behind intentional and unintentional news avoidance

Motives behind news avoidance

Prior literature has identified a broad range of motives for news avoidance. First, the cognitive load of news impacts news use behavior. Song, Jung, and Kim (2017) find that the high-choice environment can lead to perceived news overload (“I’m overwhelmed by news”) (see also Fletcher, Kalogeropoulos, and Nielsen, 2020; Ji, Ha, and Sypher, 2014), which in turn can result in a lack of understanding (“I don’t often understand what is said on the news”) or news fatigue (“I’m tired of news”). In a similar line, some people say that their social contacts inform them about important news (“Someone else informs me anyway”), while few people indicate that they do not decide themselves what they consume (“Someone else decides what I end up watching or reading”) (Palmer and Toff, 2020; Schrøder and Blach-Ørsten, 2016). In both cases, the cognitive load is reduced as responsibility is shifted to someone else. While news overload can also be associated with emotional burden, the main aspect connecting these motives is their relation to a (perceived) cognitive burden of news that can be lifted by avoiding the news.

At the intersection of cognitive and emotional motives lies the disenchantment with certain news topics (“I want to avoid certain topics”) and a resulting avoidance, or as Metag and Arlt (2016) put it, the “cognitive and emotional resistance of recipients to a topic, so that they ‘do not want to see or hear anything more of it’” (p. 542; translated from German). Further studies tap into the emotional aspects of news avoidance, indicating that disengagement with news is a way to feel less emotionally strained. In Schrøder and Blach-Ørsten’s (2016) study, news-avoiders frequently agreed that they avoided the news because news induces uncomfortable feelings (“News depresses me”, “News makes me angry”). In qualitative interviews with intentional avoiders, Woodstock (2014) also identified the emotion-inducing function of news and its negativity as reasons for avoidance (“News is too negative”). Respondents reported that avoidance was a form of “self-preservation”. They preferred to avoid anger, anxiousness, or disturbance (Kalogeropoulos, 2017; Kalogeropoulos, Fletcher, and Nielsen, 2020; Palmer and Toff, 2020).

Some studies identify motives that are connected to political attitudes. Woodstock (2014) identified criticism of the news as a reason for avoidance; specifically, avoiders pointed to a lack of neutrality in reporting and the misrepresentation of information compared to how they view something (“News doesn’t represent my opinions”, “News is biased”). Similarly, not feeling like one could rely on the truthfulness of news was one of the most important reasons for intentional avoiders in Kalogeropoulos’ (2017) study (“I can’t rely on news to be true”). More recently, Kalogeropoulos et al. (2020) and Fletcher et al. (2020) showed that avoiders of Covid-19 news were motivated by their distrust of the news.

Other motives are more closely connected to preferences and the way people arrange their time. Schrøder and Blach-Ørsten (2016) found that respondents frequently agreed that “There is usually something more interesting to do”. This is in line with some people preferring entertaining content over news (Prior, 2005, 2007). Others indicated that they lacked the time to follow the news (“I’m too busy”) or did not see it as relevant (“News isn’t relevant to me or my life”) (Palmer and Toff, 2020; Schrøder and Blach-Ørsten, 2016; Toff and Palmer, 2019).

Lastly, Katz, Blumler, and Gurevitch (1973) expected respondents to be able to identify their own media use motivations: “[…] people are sufficiently self-aware to be able to report their interests and motives in particular cases or at least to recognize them when confronted with them in an intelligible and familiar verbal formulation” (p. 511). A similar recognition of motives can be expected for avoidance, as prior studies have shown. Nonetheless, it should be considered that respondents have no clear remembrance of their original motive for this behavior. This makes “I’ve always done it like this” a preferable answer to a “post hoc rationalization for the researcher” (Diddi and LaRose, 2006, p. 195).

Some of these studies focus solely on intentional avoidance, others do not differentiate based on intentionality. These different approaches play their part in exacerbating the comparison of the aforementioned findings (for a discussion of these different definitions, see Skovsgaard and Andersen, 2020). To summarize, first, prior research on news avoidance has identified several motives; however, it is necessary to include this range of motives in one single study to determine their relevance for different forms of news avoidance. Second, as indicated throughout the research review, some motives show commonalities: There are motives that are closely connected to cognitive or emotional burden, political criticism, how people spend their time, and habit (see Table 1). This systematization captures main themes between specific motives which are used as guides throughout the paper. However, it does not imply that motives cannot be connected to other aspects as well. To test whether this systematization can be utilized for further analysis, I will apply a confirmatory factor analysis (CFA).

RQ1: Which motives are indicated by intentional and unintentional news-avoiders?

RQ2: Can the systematization of motives for news avoidance be empirically underpinned?

Predictors for news avoidance motives

After identifying relevant motives for intentional and unintentional news avoidance and exploring commonalities between motives, the question ensues about the roots of such motives or, put differently: Are there specific characteristics that lead individuals to their reasoning of why they avoid the news? The LCM indicates two broad categories: cognitive and emotional responses to media stimuli (Lang, 2009). Furthermore, prior research has identified different political characteristics as relevant for news selection behavior; thus, predictors of news avoidance motivations will be discussed here in terms of these three categories.

Cognition

As the LCM states, humans’ information processing capacity is finite; therefore, the availability of cognitive resources impacts message processing. Whether cognitive resources can be allocated depends on individual differences. These differences are, among others, expertise or familiarity within a field (Lang, 2009). Education can equip recipients with the ability to put (news) content into context, making it easier to process and, as news often focuses on political information, expertise can also be understood as political knowledge. This indicates that people with low education or little political knowledge could have greater difficulties understanding news content than their more highly educated and more knowledgeable counterparts. In addition to education and political knowledge, perceptions of one’s own political ability (internal efficacy) can impact news use behavior, as suggested by Palmer and Toff (2020), and could work similarly to education and knowledge. Although these relations seem theoretically plausible, empirical evidence is missing. Hence, I refrain from formulating hypotheses and instead ask whether people with low educational and political knowledge and internal efficacy levels display motives connected to cognitive ability (see Table 1) for avoidance more strongly than other groups of the sample. The research question builds upon the results of RQ2, that is, the empirical test of common themes in the motives (motive groups).

RQ3: Does the level of (a) education, (b) political knowledge, and (c) internal efficacy impact the probability to name motives related to cognition for news avoidance?

Emotion

The LCM indicates that the cognitive and motivational systems are interrelated (Lang, 2009). The motivational system is activated by either positive (appetitive system) or negative (aversive system) stimuli. Individual characteristics decide whether a person has an inclination for the so-called “positivity offset”, that is, the tendency to show a greater activation of the appetitive system when confronted with media stimuli and/or leaning to a “negativity bias”, such as a tendency to activate the aversive system (Lang, 2009).

News content tends to display negative information, such as natural disasters and their devastating effects on communities, as a means to inform and warn its viewers (for an overview of negativity in news, see Lengauer, Esser, and Berganza, 2011). In line with the proposed positivity offset or negativity bias, on the one hand, some studies suggest a “request” for negativity in news among specific groups (Trussler and Soroka, 2014). On the other hand, some recipients feel upset or angry when watching what is going on in the world, or, as Palmer and Toff (2020) put it, they see news as an “emotionally draining chore” (p. 1642). MacKuen, Marcus, Neumann, and Miller (2010) infer that “personality does condition our individuals’ emotional responses to the news” (p. 7). Here, one’s negativity proneness^[1] stands out, that is, how strongly one reacts when confronted with negative content (Sharma, 2013). People with a tendency to strongly react to negative content are expected to avoid the news because its negative content might induce strong negative feelings. Similarly, empathy describes the propensity to understand and share another person’s feelings (Davis, 1980). This makes highly empathetic individuals more vulnerable to negative news content, especially when the impacts on real people are implicated or displayed. In contrast, the need for affect is the “motivation to approach or avoid emotion-inducing situations” (Maio and Esses, 2001, p. 583), making those with high values of need for affect less likely to avoid news because of its bias for negative content.

To summarize, each of these three variables is connected to negativity and expected to impact the probability to indicate news avoidance motives related to individuals’ emotional capability (see Table 1). Again, the research question builds upon the results of RQ2.

RQ4: Do the levels of (a) negativity proneness, (b) empathy, and (c) need for affect impact the probability to name motives related to emotion for news avoidance?

Political motivation

News usage has been linked to several political variables. Research indicates that, among others, ideology, party identification, and efficacy can (de)motivate exposure (Gil de Zúñiga, Diehl, and Ardévol-Abreu, 2017; Hollander, 2008; Skovsgaard, Shehata, and Strömbäck, 2016). Different political orientations have been found to impact the perception of biased media and media trust. Some studies show strong conservatives and right-wing oriented citizens leaning toward mistrust (Lee, 2005; Livio and Cohen, 2016; Schultz, Jackob, Ziegele, Quiring, and Schemer, 2017; but see Arlt, 2018 for no connection between a right-wing political orientation and media trust), thinking that they cannot rely on the truthfulness of news (Kalogeropoulos, 2017) or seeing news as unfair (Fletcher, 2021). Others find similarities in hostile media perceptions and their effects between conservatives and liberals (Kaye and Johnson, 2016; Perryman, 2019). The question ensues whether political orientation (left or right of the political spectrum) also wields an influence on reasons for news avoidance, especially such related to political aspects. Building upon the results from RQ2, I ask:

RQ5: Does political orientation impact the probability to name political motives for news avoidance?

Also, perceptions of the responsivity of the political system (external efficacy) might impact why people avoid the news. In qualitative interviews, Palmer and Toff (2020) reported that the idea of being unable to induce change in the political system was a common theme among news-avoiders (i. e., low external efficacy was frequently conveyed). Thus, it seems plausible that external efficacy, similar to political orientation, impacts the probability of agreeing to motives related to political aspects.

RQ6: Does external efficacy impact the probability to name political motives for news avoidance?

Defining news avoidance

As news avoidance has been defined differently throughout the literature, I want to first describe how (intentional) news avoidance was defined and operationalized before providing detailed information on the data. While literature on digital disconnecting points toward occasional news avoidance as part of a regular media diet (e. g., Ytre-Arne and Moe, 2021), in this study the focus lies on citizens that indicate to have not, or infrequently, used news in the long term. For the purpose of this study, news-avoiders were identified by an absolute cut-off. Respondents who indicated never using news or once a week were classified as news-avoiders, everyone else was classified as news users. This approach and cut-off point were chosen in line with prior research on habitual news-avoiders (see Shehata, 2016 for a similar approach; see Edgerly, 2021 for a similar cut-off).

Similar to general news avoidance, intentional news avoidance can be a strategy in any media diet. In this study, however, the focus lies on people for whom intentional avoidance is more than an occasional behavior and instead is a decisive part of their news-related behavior. The intentionality of avoidance was, for that purpose, measured using two questions. Respondents were asked whether (yes/no) and how frequently (on a scale of 1 to 5) they had actively avoided the news in the last four weeks. Previously identified news-avoiders who answered 3 “sometimes”, 4 “rather frequently”, or 5 “frequently” were then classified as intentional avoiders (see Kalogeropoulos, 2017 for a similar approach).

Data

This study is based on survey data collected in April and May 2019 in Germany. Respondents were both news-avoiders and users. Since previous studies pointed to a low number of news-avoiders in the general population (e. g., Blekesaune et al., 2012; Karlsen, Beyer, and Steen-Johnsen, 2020), recruitment focused on people with low news usage. Respondents were recruited via the market research institute respondi. Pre-sampling based on news use was not possible; therefore, education and age were used as proxies to target potential avoiders (see Blekesaune et al., 2012 for the connection between age and education with news avoidance). A subset of respondi panelists, especially people of lower age and education, were invited to take part in the study (opt-in). Thus, the sample is not representative of the wider public. The respondents filled in an online questionnaire. respondi incentivizes participation by rewarding points for completed questionnaires, which can either be converted into money or gift cards (respondi, 2018).

The complete sample consisted of 416 respondents. This study focuses on the subsample of news-avoiders. There were 175 respondents identified as news-avoiders, among them 64 intentional avoiders. That means that 63.4 % (N = 111) of news-avoiders are unintentional avoiders, and 36.6 % (N = 64) are intentional news-avoiders (see Table 2). Among the news-avoiders, ages ranged from 19 to 75 (M = 47.37; SD = 13.21). Additionally, 126 avoiders (72 %) were identified as having little or no education, 34 avoiders (19.4 %) indicated a medium education level, and 15 avoiders (8.6 %) had higher education.

News use and motives for news avoidance

Respondents were asked about their news use frequency in an average week; the scale ranged from never (0) to seven days (7). Participants were provided with a list of possible reasons for news avoidance (see Table 1 for sources) and were asked to indicate how strongly they (dis)agreed. The questionnaire then focused on intentional avoidance, asking respondents to indicate whether they had actively tried to stay away from the news and, if so, how frequent that behavior was (Kalogeropoulos, 2017).

Cognitive variables

Respondents indicated their education level (Roßteutscher et al., 2019). Leaning on the distinction in the Hohenheim inventory of political knowledge, half of the political-knowledge questions asked about general political knowledge and the other half about current political knowledge (Trepte, Loy, Schmitt, and Otto, 2017); the six items were retrieved from Rattinger, Roßteutscher, Schmitt-Beck, and Weßels (2009, 2010) and Trepte et al. (2017), where necessary wording was adapted to current political circumstances (e. g., changes in personnel). Two variables measured internal efficacy (Beierlein, Kemper, Kovaleva, and Rammstedt, 2012).

Emotional variables

Being negativity-prone was measured using four items of the big five inventory (short version) tapping into neuroticism (Rammstedt and John, 2005). The five empathy items focused on empathy with people in the media. The questionnaire contained only the sub-dimension “mediatized sympathy with real people” (Happ and Pfetsch, 2015). A 10-item scale measured individuals’ need for affect (Appel, Gnambs, and Maio, 2012).

Political variables

One question assessed placement on a political left-right scale (Rattinger et al., 2010). External efficacy was measured by two items (Beierlein et al., 2012).

Demographics

Respondents were asked to indicate gender (female, male, diverse) and age in years.

Motives for news avoidance

By looking at the mean values, relevant motives for news avoidance are identified (RQ1).^[3] To test if there is a significant difference between intentional and unintentional avoiders, the Mann-Whitney-U-Test was employed. The data show that intentional avoiders display higher mean values compared to unintentional avoiders on nearly all motives. Intentional avoiders argue that they are, foremost, tired of news. Interestingly, when looking at motives that avoiders do not identify with, we see a strong rejection of all other motives related to cognitive burden by both intentional and unintentional avoiders. Besides being tired of news there was no significant difference between the two groups.

At the same time, intentional avoiders agree with motives related to emotion, especially because news is too negative and they agree with mainly political motives in that they say news is biased and lacks reliability. For these emotional and political motives, intentional and unintentional avoiders differ significantly. People avoiding the news intentionally and unintentionally agree that there is usually something more interesting to do. At the same time, both do not feel too busy to watch or read the news (see Table 3).

Motive dimensions

With the help of a CFA, I assessed the grouping of the avoidance motives (RQ2). The R package lavaan version 0.6–7 (see Rossel, 2012) was used. Only respondents identified as news-avoiders were included in the analysis. There were no missing data. As data were not normally distributed and, in some cases, not linear, “maximum likelihood estimation with robust standard errors and a Satorra-Bentler scaled test statistic” (MLM) (Rossel, 2020, p. 31) was calculated and the robust indices interpreted. Nonetheless, conclusions need to be made carefully. Habit was excluded from the CFA, as it includes just one variable and might distort model fit.

Topic disenchantment can have both, a cognitive and an emotional component (Metag and Arlt, 2016); thus, to identify with which group it is connected more closely, I compared two models. In the first model, “I want to avoid certain topics” was grouped with cognitive motives and in the second model, with emotional motives. The model fit indices (Table 4) indicated that the second model fits better, that is, topic disenchantment is closer related to emotional than to cognitive avoidance motives and was therefore classified as emotional motive for further analysis. Yet overall, the model fit was not satisfying, and there were some variables that did not load well (below .4); thus, it was necessary to exclude some items from the model. By considering the exclusion of items, I stepped away from confirmative onto explorative territory. Therefore, I first considered the theoretical reasons why a specific motive did not load as expected and, if it was reasonable, proceeded by deleting the items from the model one after another, starting with the lowest loading item. The variable “I’m tired of news” (standardized factor loading .228) might have been understood as “I’m sick of news” by respondents, thus, taking away the cognitive component. Both variables “someone else informs me anyway” and “someone else decides what I end up watching or reading” were supposed to offer cognitive relief by shifting responsibility away from oneself. However, they might be more related to situational circumstances and out of the power of the respondent, which means that the respondent does not control or reinforce the behavior of the other person but merely accepts it. “I’m too busy” loaded low on the factor time allocation. The other two variables in this factor (“news isn’t relevant to me/my life” and “there is usually something more interesting to do”) were closer related with choice and interest in news than with opportunity to consume news, while it was the opposite case for the item “being too busy”. After deleting this variable, no low factor loadings remained (Table 4). The final model showed good model fit for each indicator. Based on the items in the final model, a mean score for each factor was created and used for further analysis.

Cognition, emotion, and political motivation

To answer RQ3 through RQ6, multiple regression analyses are employed. Only news-avoiders are included in the analyses. First, RQ3 will be addressed, that is, the influence of education (RQ3a), political knowledge (RQ3b), and internal efficacy (RQ3c) on cognitive motives for news avoidance. The regression analyses show that education does not have a significant effect on the agreement to the cognitive relief factor, while political knowledge (ß = –.196; p = .009) and internal efficacy (ß = –.257; p < .001) both have a significant negative effect (see Table 5). This means that the lower the political knowledge or internal efficacy of a person, the higher the probability that they will avoid the news because they are motivated by cognitive relief.

Next, I focus on the influence that negativity proneness (RQ4a), empathy (RQ4b), and the need for affect (RQ4c) might wield on emotional motives for avoidance. The analysis shows that the need for affect does not yield any influence on the agreement with emotional motives, but proneness to negativity (ß = .181, p = .027) and empathy (ß = .317, p < .001) do. The higher people score on the empathy scale (i. e., the stronger they feel for people in the media) or the stronger they are prone to negativity, the higher is the probability that they will say that they avoid news for emotional reasons.

The data show that political orientation wields no influence on political motives for avoidance (RQ5). External efficacy, however, shows a negative impact on political motives (ß = –.217, p = .004) (RQ6). This means that the stronger someone believes in the responsivity of the political system, the lower their probability of mistrusting and avoiding the news.^[4]