Search

Background: Subjective survival probabilities are affected by individual-specific judgment and vary by factors known to differentiate actual mortality. Objective: The aim of this study is to evaluate whether sociodemographic characteristics, physical and mental health, and lifestyle are incorporated into subjective survival probabilities of Europeans aged 50 or higher. Methods: We use data from Wave 6 of the Survey of Health, Ageing and Retirement in Europe (SHARE) and period life tables from the Human Mortality Database (HMD). For the statistical analysis we employ multinomial logistic regression models. Results: Our results show that common factors drive the self-reported subjective survival probabilities. Certain factors affecting actual mortality are considered when forming subjective survival probabilities: income, education, poor physical and mental health, activities of daily living (ADLs), smoking, physical activity, diet, quality of life, and number of children. Other factors are not considered in a manner consistent with actual mortality patterns: age, gender, marital status, and body weight. The findings regarding cognitive function are inconclusive; whereas some aspects seem to be integrated in subjective survival probabilities (e.g., memory or self-writing skills), others are not (e.g., numeracy or orientation in time). Contribution: The contribution of this study is the grouping of sociodemographic, health, and lifestyle characteristics according to the subjective survival probabilities' direction and consistency with general population mortality and actual mortality patterns. Hence, we assess which traits are incorporated in the formation of subjective survival probabilities among Europeans aged 50 or higher.

Background: In the second half of the 20th century, remarkable marriage changes were seen: a great proportion of never married population, high average age at first marriage, and large variance in first marriage timing. Although it is theoretically possible to separate these three elements, disentangling them analytically remains a challenge. Objective: This study's goal is to answer the following questions: Which of the three effects, nonmarriage, delayed marriage, or expansion, has the most impact on nuptiality changes? How does the most influential factor differ by time periods, birth cohorts, and countries? Methods: To quantify nuptiality changes over time, we define the measure 'expected years ever married' (EYEM). We illustrate the use of EYEM, looking at time trends in 15 countries (six countries for cohort analysis) and decompose these trends into three components: scale (the changes in the proportion of never married - nonmarriage), location (the changes in timing of first marriage - delayed marriage), and variance (the changes in the standard deviation of first marriage age - expansion). We used population counts by sex, age, and marital status from national statistical offices and the United Nations database. Results: Results show that delayed marriage is the most influential factor on period EYEM's changes, while nonmarriage has recently begun to contribute to the change in North and West Europe and Canada. Period and cohort analysis complement each other. Conclusions: This study introduces a new index of nuptiality and decomposes its change into the contribution of three components: scale, location, and variance. The decomposition steps presented here offer an open possibility for more elaborate parametric marriage models.

1. IntroductionOn average, 15% of European children live in one-parent households, most of them headed by women, and this kind of household has a higher risk of poverty than couple-headed households (OECD 2014). Furthermore, there is evidence that under some conditions, nurturing children in single-mother households has negative consequences for the children's cognitive and emotional development that affect them throughout their entire lives (see Sigle-Rushton and McLanahan 2004 for a review). In Southern European countries, described as family-based welfare regimes (Esping-Andersen 1999), state protection for children is among the lowest in the European Union. However, when comparing poverty ratios between children nurtured in single-parent families and those in two parent households, family-based regimes show smaller differences than other welfare regimes. In the mid-1990s the risk of poverty among children residing in single-parent families was as much as twice as high as the risk of poverty among children residing in two-parent households, while this ratio was much higher in the other Western nations (Luxembourg Income Study, Key Figures. Data from mid-1990's)It seems paradoxical that the relative standard of living of children in single-parent households is better where state social investment is lower. However, this ceases to be such a paradox when we take into consideration the composition effect of single parenthood in Southern Europe. Previous studies with data from the 1990s or early 2000s showed a positive relationship between single motherhood and educational level in Southern Europe (Flaquer, Almeda, and Navarro 2006; Garriga 2010; McLanahan 2004), while this relationship was negative in Northern Europe (McLanahan 2004; Kennedy and Thomson 2010). Furthermore, a high percentage of single mothers in Southern Europe live with their parents in order to balance work and family life or escape from poverty (Trevino 2006; Flaquer and Garriga 2009).The purpose of our study is to examine whether such a composition effect still exists in Southern Europe, given the changing social context characterized by: a) a dramatic change in labor market conditions, b) huge changes in the ethnic composition of the younger population as a result of massive immigration waves, and c) deep-seated changes in family values affecting divorce, cohabitation, and out-of-wedlock patterns. Given that all these changes have not happened with equal intensity among all southern countries, we have selected Spain and Italy as two cases representing two different trends (Moreno and Mari-Klose 2013). Our results confirm that the relationship between educational level and single parenthood has been clearly inverted in Spain, where we have observed a positive association between less-educated women and single motherhood, where the risk of becoming a single mother is now higher among the less-educated. Similar changes, though lower in intensity, are happening in Italy among younger women and in the wealthiest regions. All in all, these changes seem more related to long-term trends than to any effect derived from the economic crisis or migration.2. The Second Demographic Transition and educational variation in single parenthoodSingle parenthood arises from widowhood, out-of-wedlock birth, divorce, and remarriage. Single parenthood is an option for very young girls who get pregnant out of wedlock, most of them from working-class or underclass backgrounds (Kearney and Levine 2012). This source of single parenthood is more frequent in the American continent than in Western Europe, with the exception of the United Kingdom and Ireland, and is very rare in Southern Europe (Part et al. 2013). Widowhood is another source of single parenthood, and for a long time it was the main cause of single parenthood in societies where divorce was forbidden or tightly restricted by law, such as in pre-democratic Spain (Flaquer 1998). Widowhood is more frequent among the lower social strata, given their shorter life expectancy. …