Wednesday, December 16, 2020

Find higher fertility among men with higher cognitive ability among Swedish men using taxation & conscription registers; low income & low cognitive ability independently predict low fertility & high childlessness

Do income and marriage mediate the relationship between cognitive ability and fertility? Data from Swedish taxation and conscriptions registers for men born 1951–1967. Martin Kolk, Kieron Barclay. Intelligence, Volume 84, January–February 2021, 101514.


• We find higher fertility among men with higher cognitive ability among Swedish men using population-level taxation and conscription registers

• Low income and low cognitive ability independently predict low fertility and high childlessness

• Strong evidence of lower access to marriage and reproduction among low status men in contemporary Sweden

Abstract: Recent evidence suggests a positive association between fertility and cognitive ability among Swedish men. In this study we use data on 18 birth cohorts of Swedish men to examine whether and how the relationship between cognitive ability and patterns of childbearing are mediated by income, education and marriage histories. We examine whether the expected positive associations between cognitive ability and life course income can explain this positive association. We also explore the role of marriage for understanding the positive gradient between cognitive ability and fertility. To address these questions we use Swedish population administrative data that holds information on fertility histories, detailed taxation records, and data from conscription registers. We also identify siblings in order to adjust for confounding by shared family background factors. Our results show that while cognitive ability, education, income, marriage, and fertility, are all positively associated with each other, income only explains a part of the observed positive gradient between fertility and cognitive ability. We find that much of the association between cognitive ability and fertility can be explained by marriage, but that a positive association exists among both ever-married and never-married men. Both low income and low cognitive ability are strong predictors of childlessness and low fertility in our population. The results from the full population persist in the sub-sample of brothers.

Keywords: FertilityChildlessnessCognitive abilityIncomeSweden

6. Discussion

In our paper, we show that while income is strongly associated with cognitive ability, men with below average cognitive ability have fewer children, even after adjusting for income. We also find that these differences are magnified for childlessness, and are also very strong for entry into marriage. Consistent with previous research, we find that income and fertility are very strongly associated (Chudnovskaya, 2019Kolk, 2019), but that the relationship between cognitive ability and fertility persists net of the mediation of income. This is particularly true at lower income levels. Men with low cognitive ability who are above the median in cumulative income between age 18 and 45 have approximately the same number of children as men who score highly on cognitive ability. However, men with low cognitive ability are much less likely to find themselves in the top half of the cumulative income distribution. Among ever-married individuals, the association between cognitive ability and fertility is strongly attenuated, and only really suggests lower fertility among men with the lowest scores on cognitive ability. When comparing full biological brothers with each other, we find a strong positive fertility and cognitive ability gradient even after adjusting for income. Overall, our results indicate that the primary reason that we observe low fertility among men with lower cognitive ability is because of their failure to attract a partner for stable unions for childbearing. In addition to confirming previous findings on cognitive ability and fertility in Sweden (Kolk & Barclay, 2019), the findings of this study provide evidence for the importance of partnership formation for fertility, as well as showing that the intelligence-fertility association persists even after taking cumulative income into account.

Another intriguing empirical pattern that we have observed is that, although men with high cognitive ability have more children overall, men of average cognitive ability had more children than men with high cognitive ability scores among those who never married. These never-married men with high cognitive ability are too few to affect the population-level intelligence-fertility gradient, but may indicate a sub-population that either voluntarily abstains from childbearing and marriage, or in other ways have life trajectories that are associated with high education and income but not traditional patterns of family formation. In our full population analyses, when adjusting for cumulative income, we find that the men with the highest cognitive ability scores have slightly lower fertility and higher childlessness than men with median cognitive ability scores. After adjusting for income we observe slightly lower fertility among high IQ never-married men (left-panels of Fig. 6Fig. 7), as well as slightly lower fertility among men with high cognitive ability (see Fig. 1). However, our finding that higher cognitive ability men have higher childlessness and lower fertility than men with similar incomes but average cognitive ability is not replicated in sibling comparison models.

In our sibling comparison models we consistently observe lower fertility among men in the bottom half of the cognitive ability distribution. The difference between our population level models and the sibling models is intriguing. Although the results from our population level models are key to understanding how cognitive ability may be distributed in the following generation (though without data on women we cannot speculate about this), the sibling comparison models effectively adjust for all factors shared in the family of origin. It is certainly possible that the results in the full population are confounded by factors that are jointly associated with both cognitive ability as well as fertility outcomes, for example negative experiences in early adulthood such as health shocks or substance abuse.

We believe that our study highlights the importance of examining and interpreting gross associations between cognitive ability and fertility by taking account of the associations between cognitive ability and mediating dimensions of social status and partnership formation. The sociological and demographic literature suggests great variation across the West in the associations between income and fertility, and education and fertility. The findings from our Swedish data may not generalize to other countries. In other high-income countries, the interrelationships between education, income, marriage, and fertility, differ in important ways from Sweden, and our results may to some extent be contingent on the aggregate positive relationships between status and marriage and family formation in Sweden. Nevertheless, we think that the fertility disadvantage of very low cognitive ability men is likely widespread across OECD contexts and that using datasets where such individuals are fully included is important if researchers are to be able to make population-level inferences. Future research on cognitive ability and fertility is therefore advised to pay careful attention to contemporary research in family sociology, demography, and economics on the overall relationship between status and fertility in the focal society. Importantly, the associations between income and fertility and education and fertility typically differ by gender. Unfortunately, we cannot examine any gender differences in the intelligence-fertility gradient in Sweden as we have male-only conscription data.

Our findings also contribute towards the increasing evidence for social polarization of childbearing in many Western countries. We find that the proportion childless and the proportion that never-marry is very substantial among men with lower cognitive ability. We find large separate effects where both low income and low cognitive ability are each strongly associated with a higher probability of childlessness and low completed fertility. When a man has both low income and low cognitive ability, fertility is even lower. This corresponds to the findings from a growing literature that shows that men with low income, low levels of education, worse health, and low cognitive ability, are much less likely to find a childbearing partner in Scandinavia (Barclay & Kolk, 2020Jalovaara et al., 2019Jalovaara & Fasang, 2020Kolk, 2019). Fertility in Scandinavia has traditionally been characterized by relatively small social differences between groups. Our findings of differences by cognitive ability in probabilities of childlessness and ever-marriage of 20 to 30 percentage points clearly show that partnership and childbearing are increasingly less likely for many men with low cognitive ability in contemporary Sweden.

Global trends in the prevalence and incidence of depression: There is a predominant increasing trend in its prevalence over time

Global trends in the prevalence and incidence of depression: a systematic review and meta-analysis. Darío Moreno-Agostino et al. Journal of Affective Disorders, Volume 281, 15 February 2021, Pages 235-243.



• There is a predominant increasing trend in depression prevalence over time

• The increase is not attributable to methodological changes within or across studies

• Studies on depression incidence change with similar designs over time are needed


Background: There is mixed evidence regarding the change in the prevalence of depression in the general population over time. This study aimed to synthesise the evidence on studies that use equivalent approaches in equivalent populations across different time points.

Methods: A systematic review was conducted to identify studies focused on the change over time in depression incidence and prevalence in the general population. A random-effects meta-analysis was performed to obtain a pooled effect for the change in the prevalence estimates between the first and last time points considered. Subgroup and meta-regression analyses were used to ascertain differences in the effect sizes by gender, age group, prevalence type, elapsed time between cross-sections, and depression operationalisation.

Results: 19 studies provided information on the change in depression prevalence over time, whereas none provided such information regarding incidence. The pooled odds ratio (OR) and confidence interval (CI) were estimated by using 17 studies: OR=1.35 (95% CI: 1.14, 1.61). Similar pooled effects were obtained for females and males, separately. The high heterogeneity across studies was not explained by any of the design variables considered. No evidence for publication bias was found.

Limitations: The review included published articles up to August 2018, and the information of studies with more than two time points was summarised in a single estimate of change.

Conclusions: There is a predominant increasing trend in the likelihood of experiencing depression over time that seems not to be explainable by study design differences or publication bias alone.

Keywords: depressionmeta-analysismood disorders

People tend to like interpersonal behaviors that are similar to their own and become bothered by behaviors that are the opposite of their own, consistently across different levels of relationship closeness

An Interpersonal Approach to Social Preference: Examining Patterns and Influences of Liking and Being Bothered by Interpersonal Behaviors of Others. Tianwei V. Du, Katherine M. Thomas and Donald R. Lynam. Journal of Personality Disorders, December 2020.

Abstract: Personality disorders are rooted in maladaptive interpersonal behaviors. Previously, researchers have assessed interpersonal behaviors using self-ratings of one's own behaviors and third-person ratings of dyadic interactions. Few studies have examined individuals’ perceptions of others’ interpersonal behaviors. Using a sample of 470 undergraduate students, the authors examined patterns of interpersonal perception as well as influences of these patterns on psychological functioning. Findings showed that people tend to like interpersonal behaviors that are similar to their own and become bothered by behaviors that are the opposite of their own. Such a pattern is particularly characteristic on the warmth dimension and is consistent across different levels of closeness of the relationship. The authors also found small but significant effects of interpersonal perception on personality and general psychological functioning, above and beyond effects of individuals’ own interpersonal traits. Such findings highlight the importance of including perceptions of others in investigating interpersonal dynamics when understanding personality disorders.

Teams of mostly women performed better than teams of mostly men, when negative relationships existed among team members

When Can Negativity Mean Success? Gender Composition, Negative Relationships and Team Performance. Bret Bradley, Sarah Henry, Benjamin Blake. Small Group Research, December 11, 2020.

Rolf Degen's take:

Abstract: Teams are often plagued by internal discord, such as negative relationships, which can impede successful team performance. While most teams eventually encounter negative relationships, we have limited knowledge on how teams manage this negativity. In this article, we expand scholarship on teams by exploring an inherent team characteristic, gender composition, to assess its role in how teams cope with negative relationships. On the one hand, social role theory suggests that teams comprised of more women will perform better in the presence of negative relationships. On the other hand, theories and evidence on personality and individual differences suggest that teams comprised of more men will perform better in the presence of negative relationships. We studied 151 student project teams, and found that teams of mostly women performed better than teams of mostly men, when negative relationships existed among team members. We discuss the implications of these findings for research and practice.

Keywords: gender composition, negative relationships, team performance, social role theory

Social information is immensely valuable, but lots of it is wasted (we fail to give such information its optimal weight); egocentric discounting is a pervasive effect, but we have single unifying explanation for it

Morin, Olivier, Pierre O. Jacquet, Krist Vaesen, and Alberto Acerbi. 2020. “Social Information Use and Social Information Waste.” SocArXiv. December 10. doi:10.1098/rstb.2020.0052

Abstract: Social information is immensely valuable. Yet we waste it. The information we get from observing other humans and from communicating with them is a cheap and reliable informational resource. It is considered the backbone of human cultural evolution. Theories and models focused on the evolution of social learning show the great adaptive benefits of evolving cognitive tools to process it. In spite of this, human adults in the experimental literature use social information quite inefficiently: they do not take it sufficiently into account. A comprehensive review of the literature on five experimental tasks documented 45 studies showing social information waste, and 4 studies showing social information being over-used. These studies cover “egocentric discounting” phenomena as studied by social psychology, but also include experimental social learning studies. Social information waste means that human adults fail to give social information its optimal weight. Both proximal explanations and accounts derived from evolutionary theory leave crucial aspects of the phenomenon unaccounted for: egocentric discounting is a pervasive effect that no single unifying explanation fully captures. Cultural evolutionary theory’s insistence on the power and benefits of social influence is to be balanced against this phenomenon.

The Role of Subjective and Objective Social Status in the Generation of Envy: Those who were the most respected in the eyes of others were envied more than the richest ones

The Role of Subjective and Objective Social Status in the Generation of Envy. Henrietta Bolló, Dzsenifer Roxána Háger, Manuel Galvan and Gábor Orosz. Front. Psychol., December 15 2020.

Abstract: Envy is a negative emotion experienced in response to another person’s higher status. However, little is known about the composition of its most important element: status. The present research investigates the two main forms of social status (objective and subjective) in the generation of envy. In Study 1, participants recounted real-life situations when they felt envious; in Study 2 we examined whether the effect was the same in a controlled situation. We consistently found that those who were the most respected in the eyes of others were envied more than the richest ones. Furthermore, perceived deservingness of the superior other’s success differentiated between benign and malicious envy. Although previous studies focused on material comparisons when investigating envy, our results indicate that envy is rather a subjective social status related emotion. Not material, but social advantage of the superior other causes the most painful envy and future studies should put more emphasis on this type of social comparison in envy research.

General Discussion

According to the social-functional approach to envy, the goal of envy is to lessen the social status gap between the self and a superior other (Van de Ven et al., 2009Lange and Crusius, 2015a,b). Previous research on envy was more focused on material inequalities (Carter and Gilovich, 2010Fiske, 2011Lin and Utz, 2015Lin et al., 2018); the present research aimed to investigate the subjective facet of social status as well, taking into account the role of deservingness. Our findings suggest that SSS intensifies feelings of envy more than OSS and that deservingness helps differentiate between benign and malicious envy. A potential explanation for the prominent role of SSS in envy is that social factors are more related to our identity and cause more frustration, which can result in envy (Salovey and Rodin, 1984DeSteno and Salovey, 1996Lin et al., 2018). It has to be noted that previous studies (Salovey and Rodin, 1984DeSteno and Salovey, 1996) confirmed this relationship in the case of jealousy, although recent research gave empirical evidence for the role of self-relevance in the case of envy as well (Lin et al., 2018). Although envy and jealousy have some overlap regarding hostility, lowered self-esteem, and sadness, they are two distinct emotions (Parrott and Smith, 1993). Envy is more focused on inferiority and therefore can be characterized by self-diminishment and resentment, while jealousy is more focused on the threat of loss of another’s fidelity and can be characterized by anxiety, distrust, and anger (Parrott and Smith, 1993).

There are two contradicting theories regarding the role of material things in envy. Some scholars suggest that, as material possessions are easily comparable, individuals compare themselves more frequently in this domain, and that consequently envy is experienced more in relation to material possessions (Carter and Gilovich, 2010). In contrast, others suggest that envy is most intense when social comparison is important for a person’s identity (Salovey and Rodin, 1984Bolló et al., 2018).

Furthermore, although previous studies indicated that individuals tend to exaggerate the importance of OSS in hypothetical situations (Bolló et al., 2018), this study did not confirm this finding. In Study 2 respondents were asked to evaluate their feelings in a hypothetical situation, but SSS still played a more prominent role. However, in Study 2 respondents were asked to imagine that they were in the role of the envier, while in previous studies they were either the envied one (Lin et al., 2018) or the comparison affected their own status (Bolló et al., 2018). The findings of this study therefore suggest that there is a discrepancy between what individuals believe others are envious of and what they themselves are envious of, which can be a direction for future research.

Furthermore, the present research study replicated previous findings about the role of deservingness in envy (Parrott and Smith, 1993Lange and Crusius, 2015bCrusius and Lange, 2017Crusius et al., 2017). Benign envy was more likely to be expressed when the superior other’s outcome was deserved and malicious envy was more likely when it was seen to be (Study 1) or characterized as (Study 2) undeserved. In the present research we applied the value theory of deservingness by Feather (1992). By definition, deservingness refers to whether the outcome is contingent with the situation: if there is a fit between the situation and the outcome it is deserved, otherwise it is undeserved (Feather, 1999). Based on Feather (1992), in Study 2 we characterized deserved advantage by positively valued behaviors (hard work in the case of OSS and being dependable in the case of SSS) and undeserved advantage by a negatively valued behavior (“cozying up” to others) and both were followed by the same positive outcome. In the present study, attributing the superior other’s success as undeserved (negatively valued behavior followed by a positive outcome) promoted malicious envy, while attributing the other’s advantage as deserved (positively valued behavior followed by a positive outcome) promoted benign envy. According to attribution theory (Hareli and Weiner, 2002), balanced structures (Heider, 1958) carry the possibility of controllability (namely that hard work pays off) and individuals will have the motivation to work hard and become as successful as the superior other.

In summary, the findings indicate that SSS and OSS play different roles in the generation of envy. SSS is more relevant in upward social comparisons leading to benign and malicious envy, and material possessions do not motivate people to move up the social hierarchy to the same extent.

Limitations and Future Studies

Although this study has important implications in relation to envy, there are a number of limitations which should be taken into account. Firstly, females were over-represented in the sample, which may lead to biased results. Previous studies suggest that women are more likely to avoid socially comparative situations that men (Niederle and Vesterlund, 2007Rand, 2017), which can have an effect on envy.

Secondly, both studies were cross-sectional and no behavioral measures were used. Future studies should apply longitudinal or experimental design with behavioral measures.

Thirdly, Study 2 was a situation evaluation task with an imaginary scenario, so participants’ reactions in this imaginary situation may differ from their reactions in a real-life scenario. Although applying a vignette method in Study 2 can lead to interesting and informative contributions, there are limitations, especially for examining potentially less desirable emotions like envy (Van Dijk et al., 2006). For example, in the undeserved conditions of Study 2 participants might draw negative evaluations of not only the outcome but also the person themselves and it might also affect their answer about envy, added to “deservedness,” as Crusius and Lange (2014) pointed out previously that malicious envy biases attention toward the envied person rather than the advantage of this person. In future studies, this bias can be treated with manipulating not only the deservingness of the outcome but also the characteristics of the person, and then investigating their interactions.

However, vignette studies have long been used in experimental emotion research, offering the possibility to systematically control for other factors by providing identical information to respondents, thereby increasing their internal validity (Powell et al., 2008). In addition, there is empirical evidence that vignette studies can be highly generalizable to real life behavior, while overcoming the ethical, practical, and scientific limitations associated with alternative methods (Evans et al., 2015). Furthermore, respondents were assured about their anonymity and were encouraged to answer honestly. In sum, despite the limitations of online hypothetical methods, they are widely used in envy research (for example, Parrott and Smith, 1993Lange and Crusius, 2015bPoelker et al., 2019) and there is also empirical evidence that people do not seem more reluctant to report envy than other negative social emotions (Hareli and Weiner, 2002).

There are several potential directions for future studies, but the most important is that more emphasis should be placed on social factors instead of material inequalities in envy research. SSS is a broad conception but investigating its elements, such as respect or influence, could be a fruitful area. Furthermore, future studies should investigate possible mediating variables between envy and social status. Some possible mediators may be status maintenance strategies, prestige, and dominance. Previous research indicates that prestige is related to SSS (Bolló et al., 2018) and benign envy (Crusius and Lange, 2017).