What Arouses Evangelicals? Cultural Schemas, Interpretive Prisms, and Evangelicals’ Divergent Collective Responses to Pornography and Masturbation

What Arouses Evangelicals? Cultural Schemas, Interpretive Prisms, and Evangelicals’ Divergent Collective Responses to Pornography and Masturbation. Samuel L Perry. Journal of the American Academy of Religion, Volume 87, Issue 3, Sept 2019, Pages 693–724, https://doi.org/10.1093/jaarel/lfz024

Abstract: This study elucidates the puzzle of evangelical grievance selection by comparing evangelicals’ divergent collective responses to pornography use and solo-masturbation. Drawing on eighty in-depth interviews and content analyses of fifty-five evangelical monographs, I show how internal and external influences shape evangelicals’ evaluations of and responses to the two issues. Internally, evangelical cultural schemas of biblicism and pietistic idealism necessitate that grievances be connected directly to the Bible and believers’ “hearts.” Pornography is more aptly linked to explicit biblical proscriptions against heart-lust and consequently perceived collectively as a moral threat, compared with masturbation, which is neither directly addressed in the Bible nor unambiguously connected to lust. Externally, the growing influence of psychology within evangelicalism heightened concern about pornography’s harms while debunking myths associating masturbation with mental illness. These cultural influences provide “interpretive prisms” through which evangelicals differentially perceive the two issues, resulting in fervent anti-pornography activism and relative ambivalence toward masturbation.

He is a Stud, She is a Slut! Meta-Analysis on the Continued Existence of Sexual Double Standards for decades - but the effect is small

He is a Stud, She is a Slut! A Meta-Analysis on the Continued Existence of Sexual Double Standards. Joyce J. Endendijk, Anneloes L. van Baar, Maja Deković. Personality and Social Psychology Review, December 27, 2019. https://doi.org/10.1177/1088868319891310

Abstract: (Hetero)sexual double standards (SDS) entail that different sexual behaviors are appropriate for men and women. This meta-analysis (k = 99; N = 123,343) tested predictions of evolutionary and biosocial theories regarding the existence of SDS in social cognitions. Databases were searched for studies examining attitudes or stereotypes regarding the sexual behaviors of men versus women. Studies assessing differences in evaluations, or expectations, of men’s and women’s sexual behavior yielded evidence for traditional SDS (d = 0.25). For men, frequent sexual activity was more expected, and evaluated more positively, than for women. Studies using Likert-type-scale questionnaires did not yield evidence of SDS (combined M = −0.09). Effects were moderated by level of gender equality in the country in which the study was conducted, SDS-operationalization (attitudes vs. stereotypes), questionnaire type, and sexual behavior type. Results are consistent with a hybrid model incorporating both evolutionary and sociocultural factors contributing to SDS.

Keywords: sexual double standards, meta-analysis, gender, sexuality, social cognitions

Discussion

In line with evolutionary theory (Buss & Schmitt, 1993; Trivers, 1972) and biosocial theory (Wood & Eagly, 2002, 2012), this meta-analysis demonstrated clear evidence for traditional SDS in studies assessing differences in people’s evaluation, or expectation, of men’s and women’s sexual behavior, although the effect was small. People expected behaviors associated with high sexual activity more from men than from women, and behaviors associated with low sexual activity more from women than from men. Similarly, people evaluated highly sexually active men more positively (or less negatively) than highly sexually active women, and low sexually active women more positively (or less negatively) than low sexually active men. In contrast, the overall set of studies using Likert-type-scale questionnaires for assessing SDS did not yield evidence of SDS.

We found some significant moderator effects in one or both sets of studies. First, existence of traditional SDS was behavior specific. Second, stereotypes about SDS were more traditional than attitudes about SDS. Third, studies using the “sexual double standard scale” (SDSS; Muehlenhard & Quackenbush, 1998) reported more traditional SDS than studies using the “double standard scale” (DSS; Caron et al., 1993) which demonstrated reversed SDS. Fourth, higher levels of gender equality in a country were associated with less traditional SDS. Participant gender and age, publication year, and study design were not significant moderators.

Behavioral Specificity of SDS

Regarding sexual behavior types, we found strongest evidence of SDS for being a victim of sexual coercion, followed by casual sex, and having an early sexual debut. SDS were less evident for sexual infidelity, level of sexual activity, other/mixed sexual behavior types, premarital sex, and being a perpetrator of sexual coercion. The findings for coercion and sexual encounters within a power or age hierarchy were partly in line with the predictions from biosocial theory that SDS would be most prevalent in sexual encounters where there is a power/status difference between men and women (Wood & Eagly, 2002, 2012). However, we only found double standards for victims of sexual coercion, and not for perpetrators. That we did not find differences in the evaluation of male and female perpetrators might be because both male and female perpetrators violate gender role expectations, with men violating chivalry norms, and females violating communal characteristics. Thus, male and female perpetrators might have been evaluated equally negative for their gender-role inconsistent behavior. Moreover, the double standards for victims of sexual coercion, found in person perception studies, indicate that sexual behavior within the context of a power hierarchy is evaluated more negatively (or less positive) for female victims (e.g., more condemned, more perceived damage to reputation) than for male victims (e.g., “positive” experience that will be evaluated by peers as cool; Zaikman & Marks, 2017). Thus, girls might be blamed for being a victim of sexual coercion (Weis, 2009), whereas boys’ experiences of sexual coercion might be trivialized (Weis, 2010). This is inconsistent with the idea that male victims of sexual coercion or rape might be perceived as powerless and not willing to have sex, which violates men’s (hetero)sexual agentic gender role (Weis, 2010). Because only a few studies examined the evaluation of both perpetrator and victim, replication of the existence and direction of SDS for coercion victims is necessary in future studies.

The finding that engaging in casual sex and having an early sexual debut were more expected and rewarded in men than in women, fits partly with predictions from evolutionary theory. In terms of reproductive fitness, men would benefit more than women from having casual sex and by having sex at an early age (Buss & Schmitt, 1993; Petersen & Hyde, 2010). However, similar beneficial effects would have been expected for sexual infidelity and high sexual activity with numerous partners, but for those sexual behaviors less traditional SDS were applied. The same was true for other sexual behaviors, such as premarital sex when engaged or when in love. Our findings are in line with previous narrative reviews concluding that premarital sex in particular has become accepted for both men and women (Bordini & Sperb, 2013; Crawford & Popp, 2003).

Cross-Cultural Differences in SDS

In line with predictions from biosocial theory (Wood & Eagly, 2002, 2012), and not with evolutionary theory’s perspective of obligate sex differences, SDS were less traditional in countries with higher levels of gender equality. According to biosocial theory, in cultures with bigger differences in the gender roles of men and women, men have more power than women, which translates in traditional SDS (Wood & Eagly, 2002, 2012). However, level of gender equality was only a significant moderator in the meta-analysis conducted on studies using Likert-type-scale questionnaires, and not in the meta-analysis on differential evaluation and expectation of the sexual behavior of men and women. This might be because there was less variation in level of gender equality in the latter meta-analysis, as most studies in that meta-analyses were conducted in the United States. The direction of effect, albeit nonsignificant, was in the same direction as the effect from the meta-analysis on Likert-type scales.

Changes in SDS Over Time

In line with evolutionary theory, and not with biosocial theory, time period in which the study was conducted was no longer significant when controlling for other moderators, which indicated that traditional SDS have existed for decades and are still present. This finding could indicate that stable gender differences in reproductive strategies are underlying SDS. Also, it appears that even though gender roles have become less strict in most modern Western societies (Eagly & Wood, 1999), this did not lead to less differentiation in the norms for the sexual behavior of men and women (Wood & Eagly, 2002, 2012). Possibly, it takes more time for egalitarian gender roles to permeate into the bedroom, than in other domains of life such as the work field, because sexuality is very much a private issue. Furthermore, the content of SDS may have changed over time, because most older studies focused on double standards in premarital sex in different relationship types, whereas newer studies more often focused on double standards in casual sex. Thus, changes in gender roles over time might only be reflected in changes in the behavior specificity of SDS.

Gender Differences in SDS

Regarding gender, we did not find differences between men and women in their cognitions about SDS. In light of male control theory and female control theory (Baumeister & Twenge, 2002), these findings could indicate that both male control and female control contribute equally to the existence of SDS. This means that SDS might provide evolutionary and sociocultural advantages for both genders that they would like to control. Advantages for men that arise from SDS could be improved certainty about paternity (Buss, 1994), patriarchal power over women, prevention of sexual chaos, and reduced male insecurity (Hyde & DeLamater, 1997). The advantages of SDS for women are the high value of sexual favors that they can trade for lower valued favors from men, such as economic provision, monogamous relationships, and parental investment.

Age Differences in SDS

Regarding participants’ age, we did not find support for the predictions of the gender-intensification hypothesis (Hill & Lynch, 1983). It appears that adolescence is not necessarily a period that is characterized by increased gender role pressure and intensification of people’s social cognitions about gender. However, it should be mentioned that most studies were conducted with high-educated college samples, mostly including emerging adults. It may be possible that the relatively small number of studies conducted with adolescents and adults, decreased the power to detect effects of age on SDS.

Implications for Evolutionary Theory and Biosocial Theory

In sum, some of the above findings are in line with evolutionary theory (Buss & Schmitt, 1993; Trivers, 1972) whereas others are in line with biosocial theory (Wood & Eagly, 2002, 2012). This converges with the findings of a recent theory-based narrative review, which demonstrated some support for predictions of both evolutionary theory and biosocial theory about the behavioral specificity of SDS, and for predictions of biosocial theory about cultural differences in SDS (Zaikman & Marks, 2017). Each theory suggests a different mechanism that underlies SDS, but these mechanisms might be intertwined. We therefore propose that a hybrid model explaining SDS from the interplay between biological predispositions and sociocultural pressures is most appropriate (Lippa, 2009). According to biosocial theory, different norms for the behavior of men and women may have arisen from societies’ division in gender roles that expects men to be assertive, dominant, and powerful, and women to be submissive, caring, and kind (Wood & Eagly, 2002, 2012). However, the division in gender roles may have a biological or evolutionary origin (Wood & Eagly, 2012), because there are gender differences in adaptive reproductive strategies leading people to view (sexual) behaviors in men and women differently. Also, the predictive power of evolutionary and sociocultural gender role pressures to explain SDS appears to depend on the sexual behavior or context under consideration. Gender roles may have more predictive power in a sexual context characterized by power/status differences. Yet, evolutionary processes might play a larger role in sexual behaviors that increase successful reproduction.

Conceptualization and Measurement of SDS

We also looked at the effects of moderators related to conceptualization and measurement of SDS. Regarding SDS conceptualization, effect sizes were significant for both stereotypes and personal attitudes. This suggests that both stereotyped beliefs about the sexual behavior of men and women and people’s personal attitudes in response to sexual behavior that violates expectancies are underlying SDS. Yet, traditional SDS were more prevalent in collective or personal expectations about the sexual behavior of men and women (i.e., stereotypes) than in people’s personal evaluation of the sexual behavior of men and women (i.e., attitudes). This finding is in line with the idea that people can have knowledge of collectively shared stereotypes with regard to SDS or personal stereotypical expectations about the sexual behavior of men and women, although they do not apply these stereotypes personally when evaluating other people’s sexual behavior (Milhausen & Herold, 2001; Signorella et al., 1993). It has been argued that knowledge of collective stereotypes is strong, stable, and does not depend on one’s experience with other people, but on culturally shared and generalized social beliefs (López-Sáez & Lisbona, 2009). Indeed, research in children as well as adults showed that content of collective gender stereotypes has not changed over time, whereas gender attitudes did become more egalitarian (e.g., Ruble, 1983; Signorella et al., 1993).

However, our findings with regard to social cognition type need to be interpreted with caution, because the vast majority of studies examined personal SDS attitudes or a mix of stereotypes and attitudes. In studies examining a combination of stereotypes and attitudes, evidence for a reversed double standard was found, a finding that is difficult to disentangle because of the muddled operationalizations of SDS in these studies. Furthermore, in the small number of studies examining stereotypes, it was not possible to distinguish between descriptive and prescriptive aspects, or between personal stereotypes and knowledge of collective stereotypes. Yet, these distinctions are important for future research. For example, knowledge of collectively shared stereotypes is less predictive of one’s own behavior toward men and women than personal stereotypes (Stangor & Schaller, 1996). Furthermore, prescriptive stereotypes (e.g., perceptions of how men and women should behave sexually) might be particularly relevant in the context of SDS as they have been associated with negative evaluations and backlash for people who behave in stereotype-inconsistent ways (Burgess & Borgida, 1999). Indeed, gender stereotypes in general are highly prescriptive in nature (Prentice & Carranza, 2002) and more predictive of people’s personal evaluation of men and women (i.e., attitudes) than descriptive stereotypes (Gill, 2004).

As expected from dual-process models of social cognition (Gawronski & Creighton, 2013), studies using explicit Likert-type-scale questionnaires did not yield evidence for traditional SDS. Yet, studies using more implicit within- or between-subjects designs did yield evidence for SDS. The Likert-type-scale questionnaires often include items such as “It’s worse for a woman to sleep around than it is for a man” in which male and female sexual behavior is explicitly contrasted to each other (Muehlenhard & Quackenbush, 1998). Therefore, in studies using such questionnaires it might have been more clear to participants that personal cognitions about SDS were assessed, leading to social-desirable responding (Greenwald et al., 2009). In between- and within-subjects designs, the focus on SDS is more implicit than in explicit self-report questionnaires. This is because in a between-subject design researchers assessed cognitions about women’s and men’s sexual behavior with separate items or vignettes that they randomly assign to participants, who are generally unaware of the presence of other vignettes presented to other participants. Or in a within-subject design researchers administered separate vignettes or items about women’s and men’s sexual behavior in a counter-balanced way to participants (Jonason & Marks, 2009; Reid et al., 2011; Weaver et al., 2013). Thus, this finding suggests that traditional SDS might only be present at a more implicit level. Previous research indeed showed that implicit assessments are less prone to social-desirable responding (Gawronski & Bodenhausen, 2006) and more likely to suggest existence of traditional gendered cognitions (Endendijk et al., 2013).

However, SDS were not different between studies using between- or within-subjects designs, or between studies using extensive vignettes/scenarios versus studies using questionnaires with different items about the sexual behavior of men and women. This indicates that social desirability and demand characteristics might not necessarily play a larger role in within-subject research on SDS than in between-subject research (Marks & Fraley, 2005; Milhausen & Herold, 2001). Also, this finding suggests that study designs that have only a slightly less explicit focus on SDS (i.e., not contrasting male and female sexual behavior in the same items) can yield evidence for the existence of traditional SDS. This argument is consistent with one study that specifically examined differences in implicit (i.e., under divided attention) and explicit (i.e., under full attention) SDS-cognitions, showing that traditional SDS were only present at an implicit level (Marks, 2008). However, between-subjects designs, like the study by Marks (2008), have been criticized for measuring single standards (because there is no comparison with how an individual would rate another target) instead of double standards (i.e., contrasting evaluation of male vs. female target; Crawford & Popp, 2003). Therefore, using IATs might be a fruitful direction to take to examine SDS at an implicit within-subjects level (see, for example, Sakaluk & Milhausen, 2012).

Our findings regarding questionnaire type indicated that questionnaires differ in the extent to which they yield evidence for SDS, which might also explain the nonequivalent findings in studies using these methods. Studies using the DSS (Caron et al., 1993) reported reversed double standards, whereas studies using the SDSS (Muehlenhard & Quackenbush, 1998) reported more traditional double standards, which might be explained by differences in content and scoring of the questionnaires. In the DSS all but one items are formulated in the direction of a traditional double standard (e.g., “It is up to the man to initiate sex.”) and participants answer the items on a scale ranging from strongly agree to strongly disagree. Such a questionnaire design cannot distinguish between people with reversed and egalitarian sexual standards, because both groups of people will (strongly) disagree with the traditional items. Therefore, we cannot be completely sure that the negative combined mean found in studies using the DSS actually reflects reversed double standards, or an egalitarian view about the sexual behavior of men and women instead. In contrast, the SDSS consists of 20 items occurring in pairs, with parallel items about men’s and women’s sexual behavior (e.g., “A [girl/boy] who has sex on the first date is easy”). In addition, six items contrast men’s and women’s sexual behavior, with some items formulated in the direction of traditional SDS (e.g., “A man should be more sexually experienced than his wife.”) and others formulated in an egalitarian way (e.g., “A woman’s having casual sex is just as acceptable to me as a man’s having casual sex.”). Participants answer all items on a scale ranging from disagree strongly to agree strongly. Difference scores are computed between the 10 male and female items and the six individual item-scores are added to these difference scores. The design of the SDSS makes it possible to assess a more complete range of reversed to traditional double standards than with the DSS. However, the SDSS score range is asymmetrical (−30 to 48). Thus, the more traditional double standards appearing in studies using the SDSS might have been an artifact of the possible range of scores.

Limitations and Future Directions

Some limitations of this meta-analytic study need to be addressed. First, the available body of quantitative research on SDS is highly homogeneous in terms of participant age, ethnicity, and educational level. According to biosocial theory, these factors are important in the social construction of gender roles, and more specifically for the social construction of SDS (Wood & Eagly, 2002, 2012). Therefore, future studies should examine SDS in more diverse samples in terms of ethnicity, age, and educational level.

Second, almost all studies included in this meta-analysis measured SDS in a relatively explicit way, by using self-report questionnaires, even though implicit measures, such as IATs or priming tasks, are less prone to social-desirable responding than explicit measures of stereotypes, and are often better predictors of behavior (Gawronski & Bodenhausen, 2006). Thus, researchers should make use of more implicit tasks to assess SDS. Relatedly, previous research has used many different conceptualizations of SDS, sometimes combining attitudinal aspects with stereotypical aspects within one questionnaire. We advise future researchers to be more theory-driven in their conceptualization, operationalization, and predictions regarding SDS. For example, dual-process models (Gawronski & Creighton, 2013) or social cognition frameworks (e.g., Greenwald et al., 2002) could be used to further conceptualize different aspects of people’s SDS-cognitions, that is, implicit, explicit, attitudes, stereotypes, knowledge of stereotypes, prescriptive versus descriptive aspects, and personal versus collective aspects. New measures need to be developed and validated before we can examine the interplay between different double standard components.

Furthermore, studies assessing SDS via questionnaires sometimes used questionnaires that did not distinguish between people with reversed and egalitarian sexual standards. With such questionnaires, it is impossible to study predictors of individual differences in SDS-cognitions. When researchers would like to use a questionnaire in future studies on SDS, they should use questionnaires with symmetrical scales to assess the complete range of SDS from reversed to traditional (e.g., 20 item-pairs of the SDSS; Muehlenhard & Quackenbush, 1998) or develop new questionnaires that can assess the complete range.

Last, most studies included in this meta-analysis focused on SDS in behaviors associated with high sexual activity and only a few studies have been conducted specifically on behaviors associated with low sexual activity. However, further study of differences in the strength of traditional SDS between behaviors associated with high sexual activity (more male-typical) and behaviors associated with low sexual activity (more female-typical) is important. Such research can test whether boundaries for male-typical (sexual) behavior are more strict than for female-typical behavior (Hort et al., 1990). Also, research on how people acquire traditional SDS-cognitions now is essential for designing future interventions that foster egalitarian sexual standards and sexual equality for men and women.

Cognitive & affective wellbeing: Commonly cited events had little effect on wellbeing (promotion, being fired, friends passing), others had profound impacts (financial loss, death of partner, childbirth)

The differential impact of major life events on cognitive and affective wellbeing. Nathan Kettlewell et al. SSM - Population Health, Volume 10, April 2020, 100533. https://doi.org/10.1016/j.ssmph.2019.100533

Highlights
• We study the effect of 18 major life events on wellbeing.
• We use a large population-based cohort and fixed-effect regression models.
• Effects on affective and cognitive wellbeing are compared.
• Effects generally smaller when conditioning on other events.
• Events sometimes have different impacts on affective versus cognitive wellbeing.

Abstract: Major life events affect our wellbeing. However the comparative impact of different events, which often co-occur, has not been systematically evaluated, or studies assumed that the impacts are equivalent in both amplitude and duration, that different wellbeing domains are equally affected, and that individuals exhibit hedonic adaptation. We evaluated the individual and conditional impact of eighteen major life-events, and compared their effects on affective and cognitive wellbeing in a large population-based cohort using fixed-effect regression models assessing within person change. Several commonly cited events had little, if any, independent effect on wellbeing (promotion, being fired, friends passing), whilst others had profound impacts regardless of co-occurring events (e.g., financial loss, death of partner, childbirth). No life events had overall positive effects on both types of wellbeing, but separation, injury/illnesses and monetary losses caused negative impacts on both, which did not display hedonic adaptation. Affective hedonic adaptation to all positive events occurred by two years but monetary gains and retirement had ongoing benefits on cognitive wellbeing. Marriage, retirement and childbirth had positive effects on cognitive wellbeing but no overall effect on affective wellbeing, whilst moving home was associated with a negative effect on cognitive wellbeing but no affective wellbeing response. Describing the independent impact of different life events, and, for some, the differential affective and life satisfaction responses, and lack of hedonic adaptation people display, may help clinicians, economists and policy-makers, but individual's hopes for happiness from positive events appears misplaced.

Keywords: Life eventsAffective wellbeingCognitive wellbeingHedonic adaptation

Discussion

The present study confirms what people know; that not all life events are equal and many are concurrent with other events. In some respect, this may seem to be a self-apparent conclusion to anyone who has ever lived but epidemiological research often ignores this by using summed checklists to assess impact, or just evaluates the impact of one event (Dohrenwend, 2006; Gray et al., 2004; Wethington et al., 1997). Our results also quantify the difference and allow us to infer the average effect in the population. Other studies have noted differences between events in the magnitude or duration of effect on wellbeing (Frijters et al., 2011; Luhmann et al., 2012), however we focus on the total impact (both magnitude and duration). Previous longitudinal studies following individuals across time also indicate health shocks (the duration of disability) (Lucas, 2007), and separation (divorce) (Lucas, 2005; Lucas, Clark, Georgellis, & Diener, 2003) have long-term negative effects but unlike Lucas (2005), we found that the impact of the death of a spouse seemed to diminish by 2 years. The evidence for long-term effects of marriage and unemployment is mixed, with some studies showing that they continue to influence wellbeing long after they have occurred (Lucas, Clark, Georgellis, & Diener, 2004), while others report adaption to these same events (Clark et al., 2008; Frijters et al., 2011) as we found. Fig. 5 provides a comparison of the total impact (magnitude and duration) of each event on wellbeing. For instance, on average the impact of a major financial loss on both types of wellbeing was the greatest whilst health shocks, losing a loved one (widowed), separation or divorce tended not to have as much negative impact on both. Conversely, getting married, a major financial gain, retirement and childbirth had positive effects on cognitive wellbeing with little overall positive effect on affective wellbeing. These data demonstrate that the practice of treating life events as comparable is untenable.

The impact of some events is negligible after accounting for the impact of concurrent events. In general, the conditional effects of life events were a little closer to zero than the unconditional effects, but in almost all cases this was minimal, reflecting how uncommon co-occurrence actually was. However the unconditional positive effect of pregnancy on cognitive wellbeing was all but reversed once concurrent events (childbirth) were accounted for.

These results also challenge the notion of many of the identified life events as being intrinsically “stressful”, the implication of which is that they should have some negative effect on wellbeing. Holmes and Rahe's Social Readjustment Scale (Holmes & Rahe, 1967) weights marriage as the sixth most stressful event yet we found no negative impact on affective wellbeing and a profound anticipatory and subsequent positive effect on life satisfaction. Conversely people's wellbeing in the lead up to some positive events was impaired, the most notable being reconciliation which most likely demonstrates the effect of relationship difficulties just prior to the event.

The differential impact of events on the components of affective and cognitive wellbeing supports their distinction as separate constructs, although both show hedonic adaptation. A novel aspect of the present study is the comparative differences of the affective and cognitive wellbeing response to certain events. For instance, some positive events had a substantial impact on cognitive wellbeing while eliciting relatively little impact on affective wellbeing or “happiness” (e.g., Married, Retired, Childbirth, Pregnant). In contrast, negative events tended to have comparable and untoward effects on both cognitive and affective wellbeing, with the exception of Separated which again elicited a greater (negative) impact on cognitive wellbeing, and Moving which had no affective response but reduced life satisfaction. The differential impact of events on the components of affective and cognitive wellbeing supports the distinction between wellbeing components and their treatment as separate constructs. It also implies that, on average, hoping for happiness from positive events appears misplaced.

Limitations

A few general issues are worth discussing in large, longitudinal models and studies of this kind. Such studies preclude the use of the experience sampling method of assessing affective wellbeing which many consider the best method for assessing short term intra-individual variation in affective wellbeing. The fixed effects models exclude anyone who did not experience the event in the time window of interest. This means that in any particular event, such as marriage, average differences in subjective wellbeing between married people and unmarried people may be present, however these between-group differences will not be revealed by the fixed effects model which estimates within-subject changes in the sample of interest. As a result, these population estimates can reveal what to expect once an event has occurred, but cannot be used to predict whether an event such as marriage will increase or decrease wellbeing in any particular case. That is, the effects of marriage may be specific to the kinds of people who get married and should not be offered as evidence or a reason to get married.

We used an unbalanced panel, which means a slightly different set of individuals may contribute to the pre- and post-event coefficients (although there is considerable overlap). A balanced approach (Clark et al., 2008) only includes people with measurements before and after the event, which ensures the same cohort is followed over time. However, balancing reduces efficiency and risks inducing potential selection effects, so other researchers have taken a more liberal approach and included anyone with more than one consecutive observation, regardless of when those observations occurred (Frijters et al., 2011), which we follow in this study. In a sensitivity analysis we restricted the sample to a balanced panel observed pre- and post-event (see Balanced Models in Supplementary Materials) which did not materially change the overall results or inferences.

We also note some causes of potentially non-random measurement error inherent in any dynamic model of this sort. First, due to censoring issues we do not know at time $t=1$ if a life event occurred before that first year (e.g., 2002). Similarly, at time $t=T$ we do not know whether an event occurred after the final year (e.g., 2016). We do not expect this to significantly bias our estimates since many events occur infrequently and this only affects years close to the endpoints of our data. A similar issue arises in the case of missing life event information‚ either because the respondent did not complete that part of the questionnaire or because they are missing from the sample in a particular year. In both cases, we assumed no life event occurred in the missing year when constructing pre- and post-indicators. Again, we expect any bias to be small given that most life events are infrequent and more than 65 percent of people are responding year-to-year (see Table S2 in Supplementary Material). In a follow-up analysis (Uncontaminated Models in Supplementary Materials), we excluded from the sample any observations within three years of missing life event data to estimate an uncontaminated (as well as balanced) model. This means we only estimated effects for the years 2005–2012, and so after balancing and de-contamination this was our most restricted sample. As a result, our estimates became less precise and, while generally qualitatively similar to the main results, some effects became statistically insignificant (particularly for the positive events).