The small effects of political advertising are small regardless of context, message, sender, or receiver: Evidence from 59 real-time randomized experiments

The small effects of political advertising are small regardless of context, message, sender, or receiver: Evidence from 59 real-time randomized experiments. Alexander Coppock, Seth J. Hill and Lynn Vavreck. Science Advances  Sep 2 2020: Vol. 6, no. 36, eabc4046. DOI: 10.1126/sciadv.abc4046

Abstract: Evidence across social science indicates that average effects of persuasive messages are small. One commonly offered explanation for these small effects is heterogeneity: Persuasion may only work well in specific circumstances. To evaluate heterogeneity, we repeated an experiment weekly in real time using 2016 U.S. presidential election campaign advertisements. We tested 49 political advertisements in 59 unique experiments on 34,000 people. We investigate heterogeneous effects by sender (candidates or groups), receiver (subject partisanship), content (attack or promotional), and context (battleground versus non-battleground, primary versus general election, and early versus late). We find small average effects on candidate favorability and vote. These small effects, however, do not mask substantial heterogeneity even where theory from political science suggests that we should find it. During the primary and general election, in battleground states, for Democrats, Republicans, and Independents, effects are similarly small. Heterogeneity with large offsetting effects is not the source of small average effects.

Check also Persuasive Effects of Presidential Campaign Advertising: Results of 53 Real-time Experiments in 2016. Alexander Coppock, Seth J. Hill and Lynn Vavreck. Prepared for presentation at the 2019 meeting of the American Political Science Association, Washington, D.C., August 23, 2019. https://www.bipartisanalliance.com/2019/10/persuasive-effects-of-presidential.html

And Le Pennec, Caroline, and Vincent Pons. "Vote Choice Formation and the Minimal Effects of TV Debates: Evidence from 61 Elections in 9 OECD Countries." Harvard Business School Working Paper, No. 20-031, September 2019. https://www.bipartisanalliance.com/2019/10/tv-debates-small-effect-in-voters.html

And Testing popular news discourse on the “echo chamber” effect: Does political polarisation occur among those relying on social media as their primary politics news source? Nguyen, A. and Vu, H.T. First Monday, 24 (5), 6. Jun 4 2019. https://www.bipartisanalliance.com/2019/10/testing-popular-news-discourse-on-echo.html

And Right-Wing Populism, Social Media and Echo Chambers in Western Democracies. Shelley Boulianne, Karolina Koc-Michalska, Bruce Bimber. New Media & Society, presented, in review. Sep 2019. https://www.bipartisanalliance.com/2019/10/echo-chambers-usa-overall-we-find-no.html

And Kalla, Joshua and Broockman, David E., The Minimal Persuasive Effects of Campaign Contact in General Elections: Evidence from 49 Field Experiments (September 25, 2017). Forthcoming, American Political Science Review; Stanford University Graduate School of Business Research Paper No. 17-65. American Political Science Review. https://www.bipartisanalliance.com/2017/11/the-best-estimate-of-effects-of.html

Liberals Report Lower Levels of Attitudinal Ambivalence Than Conservatives, despite conservatives been shown to be positively correlated with intolerance of ambiguity & negatively correlated with openness to new experiences

Liberals Report Lower Levels of Attitudinal Ambivalence Than Conservatives. Leonard Newman & Rikki Sargent. Social Psychological and Personality Science, July 22, 2020. https://journals.sagepub.com/doi/abs/10.1177/1948550620939798

Abstract: Political conservatism has been shown to be positively correlated with intolerance of ambiguity, need for closure, and dogmatism and negatively correlated with openness to new experiences and uncertainty tolerance. Those findings suggest that conservatism should also be negatively correlated with attitudinal ambivalence; by definition, ambivalent attitudes are more complex and more tinged with uncertainty than univalent attitudes. However, little published research addresses this issue. The results of five studies (total N = 1,049 participants) reveal instead that political liberalism is negatively associated with ambivalence. This finding held for both subjective and potential (i.e., formula-based) measures of ambivalence and for both politicized and nonpoliticized attitude objects. Conservatives may prefer uncomplicated and consistent ways of thinking and feeling, but that preference might not necessarily be reflected in the actual consistency of their mental representations. Possible accounts for these findings are discussed.

But... Political conservatives exhibit greater judgment & decision-making confidence than liberals: They exhibit a greater motivation to make rapid & efficient judgments & are more likely to “seize” on an initial response option

The confident conservative: Ideological differences in judgment and decision-making confidence. Benjamin Ruisch & Chadly Stern. Journal of Experimental Psychology: General, August 13 2020. https://www.bipartisanalliance.com/2020/09/political-conservatives-exhibit-greater.html

Political conservatives exhibit greater judgment & decision-making confidence than liberals: They exhibit a greater motivation to make rapid & efficient judgments & are more likely to “seize” on an initial response option

The confident conservative: Ideological differences in judgment and decision-making confidence. Benjamin Ruisch & Chadly Stern. Journal of Experimental Psychology: General, August 13 2020. https://read.qxmd.com/read/32790460/the-confident-conservative-ideological-differences-in-judgment-and-decision-making-confidence

Abstract: In this research, we document the existence of broad ideological differences in judgment and decision-making confidence and examine their source. Across a series of 14 studies (total N = 4,575), we find that political conservatives exhibit greater judgment and decision-making confidence than do political liberals. These differences manifest across a wide range of judgment tasks, including both memory recall and “in the moment” judgments. Further, these effects are robust across different measures of confidence and both easy and hard tasks. We also find evidence suggesting that ideological differences in closure-directed cognition might in part explain these confidence differences. Specifically, conservatives exhibit a greater motivation to make rapid and efficient judgments and are more likely to “seize” on an initial response option when faced with a decision. Liberals, conversely, tend to consider a broader range of alternative response options before making a decision, which in turn undercuts their confidence relative to their more conservative counterparts. We discuss theoretical implications of these findings for the role of ideology in social judgment and decision-making.


But... Liberals Report Lower Levels of Attitudinal Ambivalence Than Conservatives, despite conservatives been shown to be positively correlated with intolerance of ambiguity, need for closure, and dogmatism and negatively correlated with openness to new experiences and uncertainty tolerance:

Liberals Report Lower Levels of Attitudinal Ambivalence Than Conservatives. Leonard Newman & Rikki Sargent. Social Psychological and Personality Science, July 22, 2020.
https://www.bipartisanalliance.com/2020/09/liberals-report-lower-levels-of.html

Intolerance toward ideological outgroups: We show that conservatives are more ideologically intolerant than liberals and that the more intelligent are more ideologically intolerant than the less intelligent

Ganzach, Y., & Schul, Y. (2020). Partisan ideological attitudes: Liberals are tolerant; the intelligent are intolerant. Journal of Personality and Social Psychology, Sep 2020.  https://doi.org/10.1037/pspi0000324

Abstract: In this article we examine intolerance toward ideological outgroups, conceptualized as the negativity of the attitudes of liberals and conservatives toward their ideological outgroup. We show that conservatives are more ideologically intolerant than liberals and that the more intelligent are more ideologically intolerant than the less intelligent. We also show that the differences between liberals and conservatives and the differences between the more and less intelligent depend on ideological extremity: They are larger for extreme than for moderate ideologists. The implication of these results to questions regarding the relationship between intelligence and ideological intolerance and regarding the relationship between ideology and prejudice are discussed.

Women report lower subjective well-being in areas with more female-biased sex ratios, but males’ well-being was unaffected; I suggest that results may be due to their decreased bargaining power in the dating market

Richardson, Thomas. 2020. “The Effect of the Adult Sex Ratio on Subjective Well-being: Evidence from Europe.” PsyArXiv. September 7. doi:10.31234/osf.io/sb8w6

Abstract: In recent years researchers studying subjective well-being have found that ecological factors may underpin societal differences in happiness. The adult sex ratio, the number of males relative to females in an environment, influences many behaviours in both humans and non-human animals. However, the possible influence of the sex ratio on subjective well-being has received little attention. I investigated the relationship between the adult sex ratio and subjective well-being in over 29000 respondents 133 regions of Europe. I find that women report lower subjective well-being in areas with more female-biased sex ratios, but males’ well-being was unaffected. I did not find that the sex ratio influences sex specific probability of marriage or marriage rates overall. I find that increased population density is associated with lower well-being. Drawing from sociological and evolutionary theories, I suggest that results may be due to their decreased bargaining power in the dating market.

Getting Fewer “Likes” Than Others on Social Media Elicits Emotional Distress Among Victimized Adolescents

Getting Fewer “Likes” Than Others on Social Media Elicits Emotional Distress Among Victimized Adolescents. Hae Yeon Lee  Jeremy P. Jamieson  Harry T. Reis  Christopher G. Beevers  Robert A. Josephs  Michael C. Mullarkey  Joseph M. O’Brien  David S. Yeager. Child Development, September 6 2020. https://doi.org/10.1111/cdev.13422

Abstract: Three studies examined the effects of receiving fewer signs of positive feedback than others on social media. In Study 1, adolescents (N = 613, Mage = 14.3 years) who were randomly assigned to receive few (vs. many) likes during a standardized social media interaction felt more strongly rejected, and reported more negative affect and more negative thoughts about themselves. In Study 2 (N = 145), negative responses to receiving fewer likes were associated with greater depressive symptoms reported day‐to‐day and at the end of the school year. Study 3 (N = 579) replicated Study 1’s main effect of receiving fewer likes and showed that adolescents who already experienced peer victimization at school were the most vulnerable. The findings raise the possibility that technology which makes it easier for adolescents to compare their social status online—even when there is no chance to share explicitly negative comments—could be a risk factor that accelerates the onset of internalizing symptoms among vulnerable youth.

Culturally learned first impressions occur rapidly and automatically & emerge early in development; automaticity, rapid access, & early emergence are not evidence that first impressions have an innate origin

Culturally learned first impressions occur rapidly and automatically and emerge early in development. Adam Eggleston Jonathan C. Flavell  Steven P. Tipper  Richard Cook  Harriet Over. Developmental Science, Jul 20 2020. https://onlinelibrary.wiley.com/doi/abs/10.1111/desc.13021

Abstract: Previous research indicates that first impressions from faces are the products of automatic and rapid processing and emerge early in development. These features have been taken as evidence that first impressions have a phylogenetic origin. We examine whether first impressions acquired through learning can also possess these features. First, we confirm that adults rate a person as more intelligent when they are wearing glasses (Study 1). Next, we show this inference persists when participants are instructed to ignore the glasses (Study 2) and when viewing time is restricted to 100 milliseconds (Study 3). Finally, we show that six‐year‐old, but not 4‐year‐old, children perceive individuals wearing glasses to be more intelligent, indicating that the effect is seen relatively early in development (Study 4). These data indicate that automaticity, rapid access, and early emergence are not evidence that first impressions have an innate origin. Rather, these features are equally compatible with a learning model.

COVID-19: Higher media volume was associated with higher perceived knowledge, but not with higher actual knowledge; perceived threat was linked to perceived knowledge, but not to actual knowledge

Granderath, Julia S., Christina Sondermann, Andreas Martin, and Martin Merkt. 2020. “The Effect of Information Behavior in Media on Perceived and Actual Knowledge About the COVID-19 Pandemic.” PsyArXiv. September 7. doi:10.31234/osf.io/3y874

Abstract: The COVID-19 pandemic poses a global health threat that has dominated media coverage. However, not much is known about how individuals use media to acquire knowledge about COVID-19 under conditions of perceived threat. To address this, this study investigated how perceived threat affects media use (i.e., media volume and media breadth), and how media use in turn affects perceived and actual knowledge about COVID-19. In a German online survey, N = 952 participants provided information on their perceived threat and their media use to inform themselves about COVID-19. They further indicated how well they are informed about COVID-19 (perceived knowledge) and completed a COVID-19 knowledge test (actual knowledge). The results indicated that individuals who felt more threatened by COVID-19 used media more often to inform themselves (i.e., media volume), but focused on less different media channels (i.e., media breadth). Higher media volume was associated with higher perceived knowledge, but not with higher actual knowledge about COVID-19. Further, exploratory analyses revealed that perceived threat was linked to perceived knowledge, but not to actual knowledge. The association of perceived threat and perceived knowledge was mediated by increased media volume. Finally, a smaller media breadth was linked to higher perceived and actual knowledge.

Disgust Toward Interracial Couples may only emerge under certain conditions; current research offers limited support for the hypothesis that disgust response is exclusively linked to interracial unions

Disgust Toward Interracial Couples: Mixed Feelings About Black–White Race Mixing. Shoko Watanabe & Sean Laurent. Social Psychological and Personality Science, Jul 22 2020. https://journals.sagepub.com/doi/abs/10.1177/1948550620939411

Abstract: Three studies further explored Skinner and Hudac's (2017) hypothesis that interracial couples elicit disgust. Using verbal and face emotion measures (Study 1), some participants reported more disgust toward interracial couples than same-race White and Black couples. In Study 2, only people higher in disgust sensitivity tended to “guess” that rapidly presented images of interracial (vs. White) couples were disgusting. Study 3 used a novel image classification paradigm that presented couples side-by-side with neutral or disgusting images. Participants took longer to decide whether target images were disgusting only when interracial (vs. White) couples appeared next to neutral images. Greater sexual disgust heightened this difference. Mixed evidence suggesting an association of disgust with Black couples also emerged in Studies 2 and 3. Thus, the disgust–interracial romance association may only emerge under certain conditions, and the current research offers limited support for the hypothesis that disgust response is exclusively linked to interracial unions.

The deficit model states that incorrect opinions are a result of a lack of information; the cultural cognition model states that opinions are formed to maximize congruence with the group that one affiliates with

How others drive our sense of understanding of policies. NATHANIEL RABB, JOHN J. HAN & STEVEN A. SLOMAN. Behavioural Public Policy, September 2020. https://doi.org/10.1017/bpp.2020.40

Abstract: Five experiments are reported to compare models of attitude formation about hot-button policy issues like climate change. In broad strokes, the deficit model states that incorrect opinions are a result of a lack of information, while the cultural cognition model states that opinions are formed to maximize congruence with the group that one affiliates with. The community of knowledge hypothesis takes an integrative position. It states that opinions are based on perceived knowledge, but that perceptions are partly determined by the knowledge that sits in the heads of others in the community. We use the fact that people's sense of understanding is affected by knowledge of others’ understanding to arbitrate among these views in the domain of public policy. In all experiments (N = 1767), we find that the contagious sense of understanding is nonpartisan and robust to experimental manipulations intended to eliminate it. While ideology clearly affects people's attitudes, sense of understanding does as well, but level of actual knowledge does not. And the extent to which people overestimate their own knowledge partly determines the extremity of their position. The pattern of results is most consistent with the community of knowledge hypothesis. Implications for climate policy are considered.

Unhappiness is hill-shaped in age and the average age where the maximum occurs is 49 with or without controls

Unhappiness and age. David G. Blanchflower. Journal of Economic Behavior & Organization, Volume 176, August 2020, Pages 461-488. https://doi.org/10.1016/j.jebo.2020.04.022

Abstract: I examine the relationship between unhappiness and age using data from eight well-being data files on nearly 14 million respondents across forty European countries and the United States and 168 countries from the Gallup World Poll. I use twenty different individual characterizations of unhappiness including many not good mental health days; anxiety; worry; loneliness; sadness; stress; pain; strain, depression and bad nerves; phobias and panic; being downhearted; having restless sleep; losing confidence in oneself; not being able to overcome difficulties; being under strain; being unhappy; feeling a failure; feeling left out; feeling tense; and thinking of yourself as a worthless person. I also analyze responses to a further general attitudinal measure regarding whether the situation in the respondent's country is getting worse. Responses to all these unhappiness questions show a, ceteris paribus, hill shape in age, with controls and many also do so with limited controls for time and country. Unhappiness is hill-shaped in age and the average age where the maximum occurs is 49 with or without controls. There is an unhappiness curve.

3. Discussion

There appears to be a midlife crisis where unhappiness reaches a peak in mid-life in the late forties across Europe and the United States. That matches the evidence for a nadir in happiness that reaches a low in the late forties also (Blanchflower, 2020a). In that paper it was found that, averaging across 257 individual country estimates from developing countries gave an age minimum of 48.2 for well-being and doing the same across the 187 country estimates for advanced countries gives a similar minimum of 47.2.

Table 14 summarizes the results obtained by solving out the age at which the quadratic fitted to the data reaches a maximum. There are sixteen without controls that average at 47.4 and twenty-eight with controls with the maxima averaging out to 49.1, and 48.6 years overall for the forty-four estimates. This is very close to the finding in Blanchflower (2020a) that the U-shape in happiness data averaged 47.2 in developed countries and 48.2 in developing. The conclusion is therefore that data on unhappiness and happiness are highly consistent at the age when the low point or zenith in well-being occurs.

[table]

I add to the growing list of unhappiness variables that have hump shapes in age with or without controls. I find a broadly similar hill or hump shaped curve in twenty measures of unhappiness including being many not good mental health days; being stressed, unhappy; anxious, sad, sleepless; lonely; tired; depressed, tense, under strain; having bad nerves; phobias and panics and being in pain, feeling left out of society and several more. I also found the hump shape for a more general measure relating to the respondent's belief that the country 'is getting worse'. It doesn't seem to matter much how the question about unhappiness is phrased or coded or which country the question is asked or when we get similar results.

A referee has noted that if you look at the graphs, you see wave-like patterns (sadness, panics), hump-shaped patterns (sleep, stress), and increasing-to-a-plateau-like patterns (pain and worry with limited controls). No matter the exact shape of the plots in the various charts, it is clear that there is a peak somewhere in mid-life. I don't claim the patterns are all identical, but their broad similarity is striking, with a peak in prime age. There is a clear consistent pattern in the unhappiness and age data.

Blanchflower and Graham (2020) showed that the drop in measured happiness from youth to the mid-point low of the U-shape is quantitatively large and was not "trivial" as some psychologists have claimed. Indeed, they show the decline in well-being was about the equivalent of that observed from losing a spouse or a job. The results on unhappiness are similar. For example, in the Gallup USDTP averaged across the years 2008–2017 the probability of depression in the raw data rose from 12% at age 18 to 21% at age 58. The proportion of the employed who were depressed was 12% versus 24% for the unemployed. In addition, 12% of the married were depressed yesterday versus 19% of the widowed. In the raw data from the BRFSS the proportion who said they had 20 or more bad days in a month was 6.6% at age 18 and 8.4 at age 47, the peak. Among the married the rate was 5.5% versus 8% for the widowed. The rise in unhappiness to the mid-life peak, is thus large and comparable in magnitude to major life events.

So, what is going on in mid-life? In Blanchflower and Oswald (2008) we suggested three possibilities. First, that individuals learn to adapt to their strengths and weaknesses, and in mid-life quell their infeasible aspirations. Second, it could be that cheerful people live systematically longer than the miserable, and that the nadir in happiness in mid-life thus traces out in part a selection effect. A third is that a kind of comparison process is at work: I have seen school-friends die and come eventually to value my blessings during my remaining years. Stone et al. (2010) suggest that "it is plausible that wellbeing improves when children leave home, given reduced levels of family conflict and financial burden" (p.9986, 2010).

The finding of a nadir in well-being in midlife likely adds important support to the notion that the prime-aged, and especially those with less education, are especially vulnerable to disadvantages and shocks.27 The global Covid-19 pandemic, which is disproportionately impacting marginal workers will likely make matters even harder to deal with for many at a well-being low point (Bell and Blanchflower, 2020). Some especially defenseless individuals might face downward spirals as age and life circumstances interact. Many will not be getting the social/emotional support they need as they are isolated and lonely, in addition to the first-order effects of whatever they are coping with in normal times. Lack of health care coverage in the US may well be a compounding factor where there is also an obesity epidemic. A midlife low is tough and made much harder when combined with a deep downturn and a slow and weak recovery. Peak unhappiness occurs in mid-life. There is an unhappiness curve.

Men Fake Orgasms Too - Research suggests that men fake orgasm in one out of four sexual encounters

Men Fake Orgasms Too - Research suggests that men fake orgasm in one out of four sexual encounters. Berit Brogaard. Psychology Today, Sep 04, 2020. https://www.psychologytoday.com/intl/blog/the-mysteries-love/202009/men-fake-orgasms-too

Check these:

Brewer G. (2019) Deceiving for and During Sex. In: Docan-Morgan T. (eds) The Palgrave Handbook of Deceptive Communication. Palgrave Macmillan. https://www.bipartisanalliance.com/2019/05/almost-no-information-assessing-womens.html

Abstract: Romantic and sexual relationships form an important part of the social landscape. These relationships are however vulnerable to deception, which may occur prior to intercourse (in order to obtain sex) or during sex (for a range of reasons including enhancement of relationship satisfaction). The current chapter details the use of deception to obtain sex, e.g., the use of ‘false advertising’ to attract a partner and the use of deception during sex such as pretending to experience orgasm and infidelity. Throughout the chapter, important differences between men and women are highlighted.

Muehlenhard, C. L. & Shippee, S. K. (2010). "Men's and Women's Reports of Pretending Orgasm," Journal of Sex Research 46, 1–16. https://www.tandfonline.com/doi/abs/10.1080/00224490903171794

Abstract: Research shows that many women pretend or “fake” orgasm, but little is known about whether men pretend orgasm. The purpose of this study was to investigate (a) whether, how, and why men pretend orgasm and (b) what men's and women's reports of pretending orgasm reveal about their sexual scripts and the functions of orgasms within these scripts. Participants were 180 male and 101 female college students; 85% of the men and 68% of the women had experienced penile–vaginal intercourse (PVI). Participants completed a qualitative questionnaire anonymously. Both men (25%) and women (50%) reported pretending orgasm (28% and 67%, respectively, for PVI-experienced participants). Most pretended during PVI, but some pretended during oral sex, manual stimulation, and phone sex. Frequently reported reasons were that orgasm was unlikely, they wanted sex to end, and they wanted to avoid negative consequences (e.g., hurting their partner's feelings) and to obtain positive consequences (e.g., pleasing their partner). Results suggest a sexual script in which women should orgasm before men, and men are responsible for women's orgasms.

Séguin, L. J. & Milhausen, R. R. (2016). "Not all fakes are created equal: examining the relationships between men's motives for pretending orgasm and levels of sexual desire, and relationship and sexual satisfaction," Sexual and Relationship Therapy 31, 2: 159-175. https://www.tandfonline.com/doi/abs/10.1080/14681994.2016.1158803

Abstract: Limited research on feigning orgasm, particularly among men, exists, and even less investigates motivations for doing so. Further, whether feigning orgasm, and motivations for feigning orgasm, is associated with sexual and relationship satisfaction and sexual desire is unknown. Thus, the purpose of the current study was to examine these relationships in a sample of 230 men (18–29 years old) having pretended orgasm with their current relationship partner at least once. Participants were recruited on Amazon Mechanical Turk. On average, participants reported feigning orgasm in approximately one-fourth of sexual encounters in their current sexual relationship, most commonly during vaginal sex. Feigning orgasm for reasons related to a poor sexual experience or to poor partner choice was the strongest predictor; associated with lower levels of desire and sexual and relationship satisfaction. Feigning orgasm to support a partner's emotional well-being was associated with higher levels of desire. Feigning orgasm because one was intoxicated, having undesired sex, or out of a desire to improve the quality of the sexual encounter was associated with higher levels of sexual satisfaction (though these variables accounted for little variance). This research indicates men do feign orgasm, and motivations for doing so are associated with sexual and relational outcomes.

The tallest men have the most children & men in the lowest two deciles of height have significantly lower fertility; the strong associations persist even among men who married

The Influence of Health in Early Adulthood on Male Fertility. Kieron Barclay  Martin Kolk. Population and Development Review, August 25 2020. https://doi.org/10.1111/padr.12357

Abstract: Despite the large literature examining predictors of fertility, previous research has not offered a population‐level perspective on how health in early adulthood is related to male fertility. Using Swedish population and military conscription registers, we study how body mass index (BMI), physical fitness, and height are associated with total fertility and parity transitions by 2012 among 405,427 Swedish men born 1965–1972, meaning we observe fertility up to age 40 or older. Applying linear regression and sibling fixed effects, we find that these anthropometric measures are strong predictors of fertility, even after accounting for education and cumulative income. Men with a “normal” BMI and in the highest decile of physical fitness have the most children. Men who were obese at ages 17–20 had a relative probability of childlessness almost twice as high as men who had a “normal” BMI, and men in the bottom decile of physical fitness had a relatively probability of childlessness more than 50 percent higher than men in the top decile. In sibling comparison models the tallest men have the most children and men in the lowest two deciles of height have significantly lower fertility. Further analyses show that the strong associations persist even among men who married.

Discussion

Using population register data, we have examined how several anthropometric measures are associated with fertility for men in Sweden. We find remarkably strong patterns in our data. We observe a clear monotonic pattern where men who were less physically fit have substantially lower fertility, with the least fit men having 0.31 fewer children and a relative probability to be childless over 50 percent higher than the most fit men. The results for BMI were even more striking: those underweight, overweight, or obese at ages 17–20 also have substantially lower fertility, and were more likely to be childless, with men who were obese having more than 0.5 fewer children and an estimated probability to be childless 86 percent higher than men with a “normal” BMI, even after adjusting for educational attainment and cumulative income. In the full population of Swedish men born 1965–1972, the results for the relationship between height and later fertility show a curvilinear pattern where both the tallest and shortest men have lower fertility, consistent with previous research (Stulp et al. 2012), though in our sibling comparison analyses only the shortest men have lower fertility.

We suggested that there are two primary channels by which height, physical fitness, and BMI should influence later fertility, which were fecundity and desirability as a potential partner, with the latter channel also allowing for indirect pathways such as the effects of health on socioeconomic attainment, which is itself strongly associated with fertility. To test whether the association was mediated by socioeconomic attainment, we both adjusted for educational attainment and cumulative income by age 40 and examined interactions, but this made very little difference to the results, despite the fact that educational attainment and cumulative income were independently strongly associated with the fertility outcomes in our results. Although height, physical fitness, and BMI have been shown to influence socioeconomic attainment, which is itself strongly associated with fertility (Jalovaara et al. 2019), our results suggest that our anthropometric measures influence fertility by a channel other than socioeconomic attainment, such as desirability for a healthy partner. This is particularly clear in our interaction analyses and sibling comparison analyses: even after comprehensively adjusting for all early life factors shared by brothers, and looking within levels of attained education and cumulative income, the relationship between our anthropometric measures and fertility persists in both direction and magnitude.

As an indirect way of examining whether the association between height, physical fitness, BMI, and fertility is related to how these anthropometric factors affect finding a stable romantic partner, we examined the associations between the anthropometric measures and fertility among men who had ever‐married. Although our anthropometric measures are strongly associated with entrance into marriage, we also find that the relationship between physical fitness, BMI, and fertility persists even among ever‐married men. These findings suggest that the observed relationship between BMI, physical fitness, height, and fertility is not simply attributable to never‐partnering. Our findings indicate that height, physical fitness, and BMI do influence desirability as a potential partner, but they also suggest that BMI and physical fitness influence fecundity because the probability of childlessness was much higher among those with worse health, even among the men who had ever‐married.

Although the strong associations between these anthropometric measures and fertility among ever‐married men are striking, we want to highlight several important limitations of these analyses. First, nonmarital fertility in Sweden has accounted for over 50 percent of childbearing since the 1990s, and most of this nonmarital fertility occurs in stable cohabiting relationships. Therefore, most childbearing in Sweden occurs outside of marriage today. Second, our analyses of fertility among men who ever‐married do not condition on childbearing within marriage, they only condition on the men having become married at some point by age 40, and the childbirths could have occurred before or after marriage, or even after a subsequent divorce. Nevertheless, men who have ever‐married have in some fundamental way demonstrated that they can develop a long‐term relationship. Never partnering is the dominant pathway to childlessness in the Nordic region (Jalovaara and Fasang 2017; Saarela and Skirbekk 2020), and over 93 percent of men who ever‐married in the cohorts that we study did have children at some point. We therefore believe that the results from these analyses of ever‐married men allow some insights into the extent to which the relationship between the anthropometric measures that we study and fertility are attributable to never‐partnering, and the extent to which they are attributable to physiological aspects of fecundity.

We believe that the results from this study may have important implications for understanding a large related literature examining how reproductive history affects the postreproductive health of mothers and fathers. Previous research has shown that childless men and women, as well as those with many children, tend to have higher mortality (see Högnäs et al. 2017, for a review and meta‐analysis). Although previous research on the relationship between reproductive history and postreproductive health has included careful adjustment for socioeconomic confounding (Barclay et al. 2016), research on this topic has generally not controlled for health in early adulthood. Given the strong association between physical fitness, obesity, and mortality (Blair et al. 1995; Stokes and Preston 2016), our results suggest that health in early adulthood may be an important explanatory factor that explains why childless men and women, as well as those with many children, have higher postreproductive mortality. Indeed, in this study we observe that obese and overweight men, and men with the lowest aerobic fitness, are overrepresented both among the childless and those who have four or five children.

Although this study has many strengths, there are certainly limitations. First, it must be highlighted that we have measures of BMI, physical fitness, and height from ages 17 to 20, and we do not have dynamic information on changes to these anthropometric measures over time. Although this does not matter for height, research shows that people tend to gain weight and to become less physically active as they age (Seefeldt, Malina, and Clark 2002; Malhotra et al. 2013). As such, we do not know the BMI or physical fitness of the men that we study at the time of partnership formation or childbearing, unless these transitions occur at a similar time to our measurements. Previous research indicates that although people tend to gain weight as they age, this is largely an additive effect of age where individuals stay in roughly the same rank order on BMI within their cohort (e.g., see figure 2 in Malhotra et al. 2013). In terms of physical fitness, interage correlations in dimensions of physical fitness tend to range from 0.3 to 0.6 (Seefeldt, Malina, and Clark. 2002). Although it would be very useful to have measures of BMI and physical fitness over the life course, a strength of having these measures at ages 17–20 is that we generally avoid potential concerns about reverse causality in the relationship between BMI, physical fitness, and fertility.

Another important limitation is that we only had data on height, physical fitness, and BMI for men, and it is difficult to know the extent to which these results could be generalized to women. The relationship between height and fertility would almost certainly be different for women, but it is possible that the patterns for physical fitness and BMI might be similar. A related limitation is that we did not have information on the anthropometric characteristics of the female partner of the men that we study. Due to assortative mating, it is very possible that part of the lower fertility of men who are less physically fit or who are overweight or obese could be attributable to having a partner with similar characteristics. As such, the lower fertility of these men might be attributable to having a partner with lower fecundity (Ramlau‐Hansen et al. 2007), which we also know is more common than would be expected by chance due to assortative mating (Chen, Liu, and Wang 2014).

In this study we examine men born 1965–1972. The prevalence of being overweight or obese in childhood or early adulthood, and sedentary behavior, has become much more common in more recent birth cohorts in most high‐income countries, and it is well established that the prevalence of obesity has increased dramatically across the world over the past several decades. Indeed, global obesity is estimated to have tripled between 1975 and 2016 (Jaacks et al. 2019). Given the research that demonstrates that being overweight or obese, or having a largely sedentary lifestyle, has a negative effect on fecundity (Hammoud et al. 2008a), it is plausible that rising obesity and decreasing fitness may depress fertility. Given secular trends in BMI and sedentary behavior, further research is needed to better understand how these developments are influencing fertility, and particularly childlessness, in Sweden as well as other countries.

Causes of Variation in Food Preference in the Netherlands: Our study convincingly showed that genetic factors play a significant role in food preference

Causes of Variation in Food Preference in the Netherlands. Jacqueline M. Vink  et al. Twin Research and Human Genetics, Volume 23, Issue 4, pp. 195-203, Aug 2020. https://doi.org/10.1017/thg.2020.66

Abstract: Our current society is characterized by an increased availability of industrially processed foods with high salt, fat and sugar content. How is it that some people prefer these unhealthy foods while others prefer more healthy foods? It is suggested that both genetic and environmental factors play a role. The aim of this study was to (1) identify food preference clusters in the largest twin-family study into food preference to date and (2) determine the relative contribution of genetic and environmental factors to individual differences in food preference in the Netherlands. Principal component analysis was performed to identify the preference clusters by using data on food liking/disliking from 16,541 adult multiples and their family members. To estimate the heritability of food preference, the data of 7833 twins were used in structural equation models. We identified seven food preference clusters (Meat, Fish, Fruits, Vegetables, Savory snacks, Sweet snacks and Spices) and one cluster with Drinks. Broad-sense heritability (additive [A] + dominant [D] genetic factors) for these clusters varied between .36 and .60. Dominant genetic effects were found for the clusters Fruit, Fish (males only) and Spices. Quantitative sex differences were found for Meat, Fish and Savory snacks and Drinks. To conclude, our study convincingly showed that genetic factors play a significant role in food preference. A next important step is to identify these genes because genetic vulnerability for food preference is expected to be linked to actual food consumption and different diet-related disorders.

Sexual dimorphism in body clocks: The repeated pattern of dimorphic rhythmicity observed in humans & animal models suggest that these differences are not attributable simply to societal pressures on either sex

Sexual dimorphism in body clocks. Seán T. Anderson, Garret A. FitzGerald. Science  Sep 4 2020, Vol. 369, Issue 6508, pp. 1164-1165, DOI: 10.1126/science.abd4964

Abstract: Circadian rhythms, or the body clock, confer temporal structure on human behavior and physiology to align homeostatic processes with anticipated changes in the environment. Disruption of these rhythms can influence health and well-being. Chronobiological research has often failed to consider how this temporal organization may be affected by sex. The few studies that do consider how these rhythms differ between sexes suggest a dimorphism that warrants further investigation. Recent findings from both humans and animal models illustrate how the systems that generate circadian rhythms diverge between the sexes, which has potential consequences for health and resilience to changes in sleep pattern.

---

The repeated pattern of dimorphic rhythmicity observed in humans and animal models suggest that these differences are not attributable simply to societal pressures on either sex. Consistent with the findings that female mice show enhanced entrainment to phase shifts, studies in rodents have shown that females tend to be more resistant to genetic and environmental circadian disruption. In ClockΔ19/Δ19 mutant mice, in which mutation of the Clock protein interferes with transcriptional regulation by the BMAL1-CLOCK heterodimer and leads to lengthening of the circadian period, females do not develop any detectable cardiac dysfunction until 21 months of age, despite male ClockΔ19/Δ19 mice showing cardiac hypertrophy and dysfunction after 12 months (14). However, in ovariectomized ClockΔ19/Δ19 mice, cardiometabolic function was impaired relative to ovariectomized controls by 8 months of age, highlighting the protective effect of estrogen.

One possible reason for the resilience to circadian disruption in females relates to their biological imperative. Resistance to the negative consequences of circadian disruption coupled with improved sleep, even when experiencing nocturnal disturbances, may facilitate their adaptation to frequent nocturnal awakenings over a sustained period, given their predominant role in nurturing offspring. The early-activity chronotypes seen in women before menopause also align with those in children.

Circadian rhythms are influenced by sex, and this interaction is remolded throughout life. In the healthy state, females often show higher-amplitude oscillations with an earlier peak in gene expression. Dimorphism can also shape the response to circadian misalignment and the downstream consequences of disruptions to normal rhythms. A chronic disruption to human circadian rhythms is shiftwork, which is associated with cardiometabolic disease and cancer. Studies have sought to clarify whether this risk is affected by sex (15), but the results are constrained by a lack of longitudinal data. There are large differences in the rhythmic regulation of the liver transcriptome between males and females, but it is unknown whether other organs show similar differences or how faithfully this translates to protein expression and function. In humans, well-controlled, longitudinal analyses of the impact of misalignment will be necessary to address the hypothesis that females are more resilient than males to the disruption of circadian function caused by shiftwork and repeated long-distance travel.

Prenatal substance exposure, socioeconomic adversity, caregiver distress/depression, & adverse family functioning and birthweight/gestational age, physical illness, uniquely predicted more risky sexual & aggressive behavior in youngsters

Early external‐environmental and internal‐health predictors of risky sexual and aggressive behavior in adolescence: An integrative approach. Bruce J. Ellis  Nila Shakiba  Daniel E. Adkins  Barry M. Lester. Developmental Psychobiology, August 31 2020. https://doi.org/10.1002/dev.22029

Abstract: External predictive adaptive response (PAR) models assume that developmental exposures to stress carry predictive information about the future state of the environment, and that development of a faster life history (LH) strategy in this context functions to match the individual to this expected harsh state. More recently internal PAR models have proposed that early somatic condition (i.e., physical health) critically regulates development of LH strategies to match expected future somatic condition. Here we test the integrative hypothesis that poor physical health mediates the relation between early adversity and faster LH strategies. Data were drawn from a longitudinal study (birth to age 16; N = 1,388) of mostly African American participants with prenatal substance exposure. Results demonstrated that both external environmental conditions early in life (prenatal substance exposure, socioeconomic adversity, caregiver distress/depression, and adverse family functioning) and internal somatic condition during preadolescence (birthweight/gestational age, physical illness) uniquely predicted the development of faster LH strategies in adolescence (as indicated by more risky sexual and aggressive behavior). Consistent with the integrative hypothesis, the effect of caregiver distress/depression on LH strategy was mostly mediated by worse physical health. Discussion highlights the implications of these findings for theory and research on stress, development, and health.

We find that school districts in locations with stronger teachers’ unions are less likely to reopen in person; also, no evidence that measures of COVID-19 risk are correlated with school reopening decisions

DeAngelis, Corey and Makridis, Christos, Are School Reopening Decisions Related to Union Influence? (September 1, 2020). SSRN: http://dx.doi.org/10.2139/ssrn.3684867

Abstract: The COVID-19 pandemic led to widespread school closures affecting millions of K-12 students in the United States in the spring of 2020. Groups representing teachers have pushed to reopen public schools virtually in the fall because of concerns about the health risks associated with reopening in person. In theory, stronger teachers’ unions may more successfully influence public school districts to reopen without in-person instruction. Using data on the reopening decisions of 835 public school districts in the United States, we find that school districts in locations with stronger teachers’ unions are less likely to reopen in person even after we control semi-parametrically for differences in local demographic characteristics. These results are robust to four measures of union strength, various potential confounding characteristics, and a further disaggregation to the county level. We also do not find evidence to suggest that measures of COVID-19 risk are correlated with school reopening decisions.

Keywords: Collective Bargaining, Coronavirus, COVID-19, Reopening, School Closures, Unions
JEL Classification: I28, I20

Who are the Conspiracy Theorists? Demographics and conspiracy theories

Who are the Conspiracy Theorists? Demographics and conspiracy theories. Steven M. Smallpage, Hugo Drochon, Joseph E. Uscinski, Casey Klofstad. In Routledge Handbook of Conspiracy Theories, eds. Michael Butter, Peter Knight. Feb 2020. https://www.taylorfrancis.com/books/e/9780429452734/chapters/10.4324/9780429452734-3_1

Abstract: This chapter aims to address conspiracy theorising across cultural contexts. It provides the only cross-nationally-representative data focusing on conspiracy theories currently available. The chapter shows how cultural context conditions conspiracy theorising and as a consequence may frustrate attempts to study conspiracy theories cross-culturally. Conspiracy beliefs are individuals’ acceptance of specific conspiracy theories as likely true. Many researchers skip the step of systematically measuring conspiracy beliefs or thinking, and instead rely on impressions. Conspiracy beliefs are partially dependent on demographic characteristics and group memberships because they cast one’s own group as a victim of other groups. If conspiracy thinking is related to feelings of marginalisation, anomia and helplessness, then higher income levels should negatively predict belief in conspiracy theories. Studying conspiracy theories is important for understanding contemporary political life. Much of the theorising of who counts as a conspiracy theorist has taken place in the USA and other English-speaking areas.

Emophilia is a trait characterized by falling in love fast, easily, & often; is associated with ignoring red flags and rushing into relationships; the Dark Triad personality traits are interpersonally toxic but considered attractive to some

Emophilia and other predictors of attraction to individuals with Dark Triad traits. Jacqueline Lechuga, Daniel N. Jones. Personality and Individual Differences, Volume 168, 1 January 2021, 110318. https://doi.org/10.1016/j.paid.2020.110318

Highlights
• Emophilia is a trait characterized by falling in love fast, easily, and often.
• Emophilia is associated with ignoring red flags and rushing into relationships.
• The Dark Triad personality traits are interpersonally toxic but considered attractive to some.
• Emophilia was the best predictor of attraction to individuals high in Dark Triad traits.

Abstract: The current research investigated the association between different relationship orientations (e.g., insecure attachment, sociosexuality, emophilia) and attraction to individuals high in different Dark Triad traits (Machiavellianism, psychopathy, narcissism). Although research has focused on general attraction to antisocial partners, less attention has been paid to traits that affect that attraction. One understudied construct in predicting relationship-relevant attraction is emophilia, which is the tendency to fall in love fast and often. Across two studies (N = 452), we found that emophilia had a strong and unique relationship with attraction to individuals high in dark personality traits. In Study 1, participants completed an assessment of the Dark Triad as they would want their “ideal romantic partner” to complete it. Participants in Study 2 rated the attractiveness of different dating profiles generated by people who varied in the Dark Triad. Participants in both studies filled out self-assessments of emophilia, sociosexuality, and attachment styles. Across both studies, people high in emophilia were the most attracted to any target, including targets high in the Dark Triad traits, even when controlling for other relationship-relevant traits. The findings have implications for relationship research, personality research, and potential risk factors for becoming involved with antisocial partners.

Keywords: AttractionDark TriadAttachment stylesEmophiliaSociosexualityMatingRelationships

6. General discussion

In two different studies, the current research investigated the association between relationship variables (emophilia, sociosexuality, attachment styles) and attraction to the different Dark Triad traits (Machiavellianism, psychopathy, narcissism). Of interest was emophilia, the tendency to fall in love fast and often. We hypothesized that emophilia would correlate with attraction to the different Dark Triad traits. In Study 1, participants who scored higher in emophilia, avoidant attachment, and sociosexuality reported increased preference for “ideal romantic partners” who scored higher in the different Dark Triad traits. However, in Study 2, using mock dating profiles, only emophilia was significantly associated with generalized attraction. In particular, individuals high in emophilia reported higher attraction to all profiles, regardless of the nature of the profile or who wrote it. In contrast, sociosexuality was associated with only the narcissistic profile.

Previous research has suggested that individuals high in the different Dark Triad traits are attractive (Carter et al., 2014; Grosz et al., 2015), especially men (Jonason et al., 2009). However, these findings may largely depend on individual differences in different relationship orientations. Among these individual differences, emophilia emerged as the strongest predictor of attraction to ideal partners high in the different Dark Triad traits, and attraction to dating profiles in a non-discriminating fashion. Thus, they recorded a generalized attraction to others, consistent with the construct definition of emophilia (e.g., Jones, 2015).

7. Implications

Attraction to Dark Triad traits is not universal among women or men. Instead, certain individuals find the characteristics and presentation of the Dark Triad traits attractive. Specifically, individuals high in different Dark Triad traits may be especially appealing to people high in emophilia. Although, for some, there may be an initial attraction to Dark Triad individuals, people high in any of the Dark Triad traits struggle with retaining partners (e.g., Webster et al., 2016). Thus, future research should examine how long individuals high in emophilia actually stay in a relationship with individuals high in different Dark Triad traits. The data also speak to the possibility that individuals high in emophilia may have a type of reproductive strategy that lends itself to short-term mating. Thus, future research should examine the relationship between emophilia and life history strategy (Figueredo, Vásquez, et al., 2006). Future research should also determine mechanism. It could be that individuals high in emophilia are attracted to dark personality partners. However, it could also be that individuals high in emophilia ignore the red flags that might indicate a partner could be problematic in the future.

8. Limitations and future directions

There are also limitations to the present research. For example, all participants were obtained from Amazon's Mechanical Turk. Objections to the usage of Amazon's Mechanical Turk often include a lack of representation of the general population (Paolacci & Chandler, 2014), repeated participation of workers (Kees et al., 2017), and workers' motivation to participate for monetary compensation (Kees et al., 2017). However, it is worth noting that our research included attention checks. Further, some workers participate in studies for an innate interest in the study (Kees et al., 2017), and overall data collected from Amazon's Mechanical Turk has demonstrated to be of good quality (Hauser & Schwarz, 2016; Kees et al., 2017; Paolacci & Chandler, 2014). Another limitation is the absence of actor-partner data. Specifically, we did not study self and other ratings of individuals who are in relationships. Additionally, most of our participants were not single. However, research has found that individuals can have ideal conceptualizations of a romantic partner's personality that is uncorrelated with their actual romantic partner. Further, the best predictor of actual relationship pair-bonding is mate quality, which includes physical and non-physical traits (Figueredo, Vásquez, et al., 2006). Therefore, individuals can still gauge the attractiveness of individuals outside of a relationship.

Another limitation to consider is the usage of mock profiles to investigate attraction to the Dark Triad. Although vignettes have previously been used successfully to assess attraction to the Dark Triad (Carter et al., 2014; Qureshi et al., 2016; Rauthmann & Kolar, 2013), it is possible that these reported preferences may not replicate on actual choices (Figueredo, Vásquez, et al., 2006; Jauk et al., 2016; Eastwick et al., 2011). Further, the differences among the profiles were subtle. For example, narcissistic interests were: “I like the outdoors, cycling, MMA, hotrods, tattoos, swimming, music shows, and roller derby. Pretty much anything fun.” These interests reflected sensation seeking and social visibility, which are consistent with narcissism. Further, these interests differed from low Dark Triad interests, which were: “Music, photography, hiking, cycling, traveling, history, movies, learning, reading, computers, and gadgets.” Nevertheless, the subtle nature of these differences likely attenuated the ability to make conclusions about attraction specifically to individuals high in different Dark Triad traits. However, the attraction observed to all four profiles among those high in emophilia, does support further validity of the construct of emophilia and the assessment. Future research should focus on a more naturalistic study design to investigate attraction for the Dark Triad.

Although we do not have any theory as to why male participants high in emophilia would respond differently, one potential limitation was the focus on women's attraction to Dark Triad profiles in men. Just as there are gender differences in Dark Triad scores (Jonason et al., 2009), there are gender differences in emophilia worth noting. Specifically, men are higher on emophilia than are women, and among men, emophilia has a higher correlation with anxious attachment (Jones, 2017). Nevertheless, we focused on women's potential attraction to Dark Triad traits in men in Study 2 because that has been the traditional focus of research on Dark Triad and mate attraction (e.g., Brewer et al., 2018; Carter et al., 2014; Marcinkowska et al., 2015; Marcinkowska et al., 2016; Qureshi et al., 2016). Further, Dark Triad traits also mediate the relationship between gender and short-term mating efforts (Jonason et al., 2009). Nevertheless, future research should delve deeper into the potential gender differences within emophilia. Finally, future research should focus on a more naturalistic study design to investigate attraction for the Dark Triad.

Future research should also examine whether emophilia is associated with attraction to other problematic personality constructs such as borderline personality or vulnerable narcissism (Miller et al., 2010). It may be the case that the charm, risk-taking, and agentic nature of individuals high in the different Dark Triad traits is uniquely attractive to individuals high in emophilia. It may also be the case that the vulnerable nature of traits such as borderline personality is also attractive to those high in emophilia. At the very least, the presence of these traits may not interfere with emophilia-based attraction. Future research should also examine emophilia-based attraction to everyday traits such as extraversion and openness to experience. Finally, research should examine the mechanisms behind emophilia-based Dark Triad attraction. For example, although superficial charm is a likely candidate, this needs to be more directly tested.

Although not assessed in the current study, vulnerable traits such as borderline personality may also be associated with increased attraction to different Dark Triad traits. Because emophilia and borderline personality have a positive correlation (i.e., r = 0.17–0.22; Jones, 2011a), there may be similar but distinct processes driving the attraction of individuals with borderline characteristics to individuals high in different Dark Triad traits.

No significant evidence that people with high trait neuroticism access social media particularly frequently or for long periods of time; also, failed to find a relationship between status update frequency & neuroticism

Understanding neuroticism and social media: A systematic review. Thomas Bowden-Green, Joanne Hinds, Adam Joinson. Personality and Individual Differences, Volume 168, 1 January 2021, 110344. https://doi.org/10.1016/j.paid.2020.110344

Abstract: Despite people spending nearly 10% of each day on social media platforms, many also now appear to be anxious to limit the intrusion of social media into their everyday lives. Given the known link between mental anguish and trait neuroticism, understanding how personality relates to social media usage has become an important area of study in recent years. As a result, there is an abundance of articles exploring neuroticism and social media across a variety of contexts. This article synthesizes and reviews the existing work, based on a systematic search that identified 159 studies. Our findings highlight that this current research is highly contradictory, for example people with high trait neuroticism report excessive use despite appearing to use social media infrequently. We discuss the key trends across existing studies to date, and we consider the apparent difficulties experienced by people with high trait neuroticism in exploiting the full benefits of ‘social’ media, discussing the importance of considerations for both designers and users of social media platforms.

Keywords: NeuroticismPersonalityBig FiveSocial media

4. Discussion

4.1. Major findings and implications

Our review highlights a number of key trends across the vast amount of research attempting to understand how neuroticism relates to social media. First, there is no significant evidence that people with high trait neuroticism access social media particularly frequently or for long periods of time. Likewise, research has failed to find a relationship between status update frequency and neuroticism. There is also no consistently reported link between reacting to the content of others through commenting or liking and neuroticism. These findings contrast with the perceptions of their own usage.

Our findings also highlight numerous contradictions within the current research. For instance, although there appears to be a clear link between neuroticism and excessive usage of social media, often identified through self-report measures (such as Hawi and Samaha, 2019), this isn't necessarily reflected in duration or frequency of use findings. Some researchers have raised concerns about self-report measures as a valid and reliable means of determining technology addiction (Coyne et al., 2020; Ellis, 2019). Ellis (2019) also hypothesized a potentially causal correlation between ‘anxiety’ and self-reported addiction. A logical conclusion to be therefore drawn from these apparently conflicting findings (excessive yet infrequent use) is that people with high trait neuroticism worry about their usage more than their actual usage warrants, reflecting the anxiety facet associated with trait neuroticism (Tackett & Lahey, 2017). In turn, their anxiety to reduce perceived ‘excessive’ use could lead to lower usage in terms of time (e.g Kuo and Tang, 2014) and/or more ‘passive’ use. Ironically though, even passive use has previously been found to reduce well-being (Verduyn et al., 2015). The concern among people with high trait neuroticism about their social media use, may actually lead to a behaviour that is further contributing to their anxiety.

In respect of the initial goal to understand how personality traits relate to the apparent harmful effects of social media (Waterson, 2019), we find a wealth of evidence to support a link between self-report measures of excessive use and neuroticism as discussed above. Therefore, given that personality traits relate to reports of problematic use, the evidence suggests that responsivity to reduction efforts may also correspond with traits. Modern ‘microtargeting’ methods (Barbu, 2014) for example, especially in this social media environment, provide the potential to tailor specific intervention messages to specific personality traits (Matz et al., 2017).

Another prominent trend is the apparent motivation of those with high trait neuroticism to use social media to socially interact (such as Eşkisu, Hoşoğlu and Rasmussen, 2017). Again, the research we identify indicates that people with high trait neuroticism have a comparatively small online social network (such as Noë, Whitaker and Allen, 2018), appearing at odds with this socialising motivation. Yet, one explanation requiring further research may be a difference in perception regarding the number of friends/contacts required to ‘socialise’. Are those with high trait neuroticism content with a smaller network? Given that trait neuroticism is generally associated with feelings of dissatisfaction towards a social group (such as Dehle & Landers, 2005), people with high trait neuroticism do not seem to particularly value networks or strive to achieve a large online social network. Another explanation could be their usage of social media in terms of content creation and reaction. The design of social media platforms/algorithms requires users to be ‘social’ and to interact through creating and sharing content. Our findings suggest that those with high trait neuroticism generally do not provide significantly high quantities of status updates (such as Cheevasuntorn et al., 2018) or react to others' content through ‘likes’ or ‘comments’ (such as Lee, Ahn and Kim, 2014). Therefore, it is unsurprising that this ‘passive’ use (Ryan and Xenos, 2011) does not attract a large following or friendship base. Furthermore, the valence of the content that is shared by people with high trait neuroticism may well play a part in their experience of social media (and vice-versa – through the virtuous circle discussed below). Previous research indicates that positive content is more attractive to online audiences (Berger & Milkman, 2012), yet the overwhelming evidence is that people with high trait neuroticism share negative valence content (such as Shen, Brdiczka and Liu, 2015). Infrequent posting of ‘positive’ content that might otherwise attract peers is therefore a third potential causal explanation of the small network size.

In some ways, there appears to be an injurious circle or self-fulfilling prophecy, with potential dissatisfaction in a network (such as Dehle & Landers, 2005) leading to a negative experience which is reflected in negative sentiment and emotion (such as Shen, Brdiczka and Liu, 2015), again limiting the extent to which those with high trait neuroticism socialise on social media. A similar pattern may be occurring with the concern for self-presentation leading to extreme caution or anxiety about the posts created and therefore a smaller quantity of posts. It seems likely that this low level of interaction may lead to a smaller network size, which may intensify the concern for self-presentation.

Ultimately, if individuals with high trait neuroticism struggle to present themselves or interact with others properly on social media, they can be left craving social interaction (such as Eşkisu, Hoşoğlu and Rasmussen, 2017) and seeking favourable self-presentation (such as Seidman, 2013), yet limited in their success with either due to facets within their personality. People with high trait neuroticism may be unable to consistently present a positive online persona that would otherwise enable them to socialise successfully. As research on self-disclosure shows, there is perhaps no ‘filter’ (or not enough of a filter) on revealing facts about themselves (such as Peluchette et al., 2015) or presenting their true (i.e. negative) feelings. Therefore, as also demonstrated offline (Lincoln et al., 2003; Russell et al., 1997; Shurgot & Knight, 2005), those with high trait neuroticism do not enjoy interactions with others and are left unsatisfied with their social support network (De Jong et al., 1999; Dehle & Landers, 2005; Suurmeijer et al., 2005; Tong et al., 2004).

In sum, these findings cast considerable doubt on the ability of users with high trait neuroticism to exploit the full benefits of ‘social’ media. This is important given the increasing reliance (or assumed reliance) on social media to disseminate information and gather viewpoints. Thus, developers of social media platforms, businesses and organisations using social media should consider these aspects when designing social media spaces that are inclusive to all types of people.

4.2. Limitations and future directions

This research is not intended to offer quantifiable means of assessing the strength of effect sizes, and instead explores key trends and nuances across these studies. There is therefore an opportunity for future research to present meta-analyses of these trends. This would also explore in greater depth, for example, the differences in outcome when different data collection techniques are employed. Greater statistical analysis might also explore the effect size when traits combine.

In addition, further studies could shed further light on the apparent inconsistency between measures of compulsive, excessive or even addictive use, yet relatively low frequency of use. For example, data could be collected for a single sample on both frequency and some measure of self-reported problematic use, to further investigate this possible difference between perception and reality. Furthermore, a related area for further research is why those with high trait neuroticism self-report ‘problematic’ social media use. Are all participants accurate or does neuroticism affect self-perception? For instance, Davidson and Ellis (2019) have previously hypothesized a causal link to anxiety, a particular facet of the neuroticism trait.

Another important topic is the finding that people with high trait neuroticism appear to have fewer friends/contacts than their counterparts. This appears to contradict motivations for using social media and suggests that they may have a different perception of what it is to be ‘sociable’ online. This is an important topic for research because the nature of ‘social’ media requires users to establish and maintain networks. For example, although all recipients may believe they are participating as required, a social media campaign inviting users to share content is likely to be less effective when targeting people with high trait neuroticism recipients rather than those who are emotionally stable. An important question therefore is whether they realise this difference.

A final, and fundamental, topic for investigation is the relationship between content creation and network size. Our findings demonstrate that people with high trait neuroticism are, on average, less frequent in their posting of status updates i.e. they are ‘passive’ users of social media. In addition, they tend to post content that is negative in valence. Either, or both, variables could offer a causal explanation of the smaller-than-average network size.