Solving the many-variables problem in MICE with principal component regression

Multiple Imputation (MI) is one of the most popular approaches to addressing missing values in questionnaires and surveys. MI with multivariate imputation by chained equations (MICE) allows flexible imputation of many types of data. In MICE, for each variable under imputation, the imputer needs to specify which variables should act as predictors in the imputation model. The selection of these predictors is a difficult, but fundamental, step in the MI procedure, especially when there are many variables in a data set. In this project, we explore the use of principal component regression (PCR) as a univariate imputation method in the MICE algorithm to automatically address the many-variables problem that arises when imputing large social science data. We compare different implementations of PCR-based MICE with a correlation-thresholding strategy through two Monte Carlo simulation studies and a case study. We find the use of PCR on a variable-by-variable basis to perform best and that it can perform closely to expertly designed imputation procedures.

Changing attitudes about the impact of women's employment on families: The COVID-19 pandemic effect

We use representative longitudinal panel data from the Dutch European Values Survey (EVS) to study whether the COVID-19 pandemic shifted opinions about how a woman's full-time employment impacts family life. The data was collected before the COVID-19 pandemic in 2017 and in May 2020. The analysis focuses on groups whose unpaid and paid work situation changed abruptly with the COVID-19 pandemic: parents with coresident children, and those who experienced a change in paid workload that clashes with traditional gender role expectations, namely women whose workload increased and men whose workload decreased or who stopped working. We found that groups that faced an abrupt change in their paid and unpaid work routines that clashed with their previously held gender attitude changed their gender attitude in alignment with the new paid or unpaid work situation. For women in couple households with children, this meant that they saw a halt in their progression toward gender egalitarian attitudes. For those who experienced a change in paid workload that clashes with traditional gender role norms, it meant stronger progression toward gender egalitarian attitudes. The results are interpreted on the basis of cognitive dissonance theory and exposure theory and placed in the context of previous findings.

Are we in this together? Changes in anti-immigrant sentiments during the COVID-19 pandemic

The COVID-19 pandemic is posing a threat to people all across the globe. According to traditional literature, threat perceptions induce anti-immigrant sentiments, as ingroup identity and self-interest are strengthened at the expense of the outgroup. In this study, we investigate whether the COVID-19 pandemic indeed increases anti-immigrant sentiments, or that this type of threat elicits other or no group related responses. We also look at whether such responses are expressed more strongly among specific subgroups in Dutch society. To do so, we use unique longitudinal panel data based on the European Values Study 2017, with a repeated measure in May 2020, during the national 'intelligent lockdown' in the Netherlands. Based on structural equation modeling, we demonstrate that anti-immigrant sentiments have not increased due to (perceived threat of) the COVID-19 pandemic. In fact, negative opinions towards immigrants decreased between 2017 and 2020 in the Netherlands, for which we provide alternative explanations. Although some subgroups do experience more threat than others due to the coronavirus, such as women, first generation immigrants, and the elderly, this does not lead to more negative feelings towards outgroups. Whether this is due to the fact that individuals feel threatened by everyone, regardless of group membership, should be explored in future research.

The perceived deservingness of COVID-19 healthcare in the Netherlands: a conjoint experiment on priority access to intensive care and vaccination

BACKGROUND

Amidst the COVID-19 pandemic, governments, health experts, and ethicists have proposed guidelines about ICU triage and priority access to a vaccine. To increase political legitimacy and accountability, public support is important. This study examines what criteria beyond medical need are deemed important to be perceived of priority COVID-19 healthcare access.

METHOD

Two conjoint experiments about priority over ICU treatment and early COVID-19 vaccination were implemented in a probability-based sample of 1461 respondents representative of the Netherlands. Respondents were asked who should receive treatment out of two fictitious healthcare claimants that differed in in age, weight, complying with corona policy measures, and occupation, all randomly assigned. Average marginal coefficient effects are estimated to assess the relative importance of the attributes; attributes were interacted with relevant respondent characteristics to find whether consensus exists in this relative ranking.

RESULTS

The Dutch penalize those not complying with coronavirus policy measures, and the obese, but prioritize those employed in 'crucial' sectors. For these conditions, there is consensus among the population. For age, young people are prioritized for ICU treatment, while the middle-aged are given priority over a vaccine, with younger respondents favoring healthcare for elderly claimants, while older respondents favor support for young cohorts.

CONCLUSION

People who have no control over their social risk and are able to reciprocate to society are considered as more deserving of priority of COVID-19 healthcare. Our findings provide fair support for the implemented ethical guidelines about ICU-treatment and COVID-19 vaccines.

Mind over matter. The impact of subjective social status on health outcomes and health behaviors

Recent insights have shown subjective status to impact health and health behavior. It is however unclear how this exactly happens. In this study we explore two mechanisms: this of a direct, mediating effect of subjective status explaining the impact of material class on health outcomes and behavior and an indirect, moderating impact on the relationship between material class and health outcomes and behavior. To test this empirically we conduct two studies, focusing on Great-Britain, using survey-data from the English Longitudinal Study of Ageing (N: 2709-3448) and the Whitehall II-study (N: 6275-6467). Our linear and logistic regression analyses show subjective status has a mainly direct impact on health outcomes and has both a direct, mediating and indirect, moderating impact on health behavior. In the conclusion of our article we reflect on the theoretical reasons why subjective status has a direct impact in certain cases, while playing an indirect role in other cases.

Migration to European Countries: A Structural Explanation of Patterns, 1980–2004

Various theoretical approaches have provided us with insights to explain the pattern of migration flows. Economic theory considers migration to be a reaction to labor market and economic incentives. Cultural theories predict that migration flows will occur according to a center-periphery pattern, while social network analysis assumes that migrants follow already established migration networks. We test these three approaches simultaneously using OECD and Eurostat data on the migrant inflow into the European countries between 1980 and 2004. The analysis demonstrates that migration flows react to economic incentives, mainly with regard to the labor market, but also to cultural and colonial linkages. There is no indication that the importance of the colonial past is declining over time. The response of migration patterns to shortages in the labor market is shown to be highly efficient, while the analysis shows that immigrants are not attracted by high levels of social expenditure.

Imagining class: A study into material social class position, subjective identification, and voting behavior across Europe

The traditional approach to class voting has largely ignored the question whether material class positions coincide with subjective class identification. Following Sosnaud et al. (2013), this study evaluates party preferences when Europeans' material and subjective social class do not coincide. Seminal studies on voting behavior have suggested that members of lower classes are more likely to vote for the economic left and cultural right and that higher classes demonstrate the opposite pattern. Yet, these studies have on the one hand overlooked the possibility that there is a mismatch between the material class people can be classified in and the class they think they are part of, and on the other hand the consequences of this discordant class identification on voting behavior. Analyzing the 2009 wave of the European Elections Study, we find that the majority of the Europeans discordantly identify with the middle class, whereas only a minority of the lower and higher classes concordantly identify with their material social class. Further, material class only seems to predict economic voting behavior when it coincides with subjective class; for instance, individuals who have an inflated class identification are more likely to vote for the economic left, even when they materially can be classified as middle or high class. We conclude this paper with a discussion on scholarly debates concerning class and politics.

Hard times and European youth. The effect of economic insecurity on human values, social attitudes and well-being

While economic downturns have adverse effects on young people's life chances, empirical studies examining whether and to what extent human values, social attitudes and well‐being indicators respond to sudden economic shocks are scarce. To assess the claim that human values are less affected by economic shocks than social attitudes and well‐being, two distinct yet related studies based on the European Social Survey (ESS) are conducted. The first employs a fixed effects pseudo‐panel analysis of the 2008–2014 ESS‐waves to detect whether changes over time in the socio‐demographic group's unemployment risk and national youth unemployment affect individual dispositions to varying degrees. The second study captures micro‐ and cross‐national effects in the 2010 ESS cross‐section. Unique for this set‐up is that we can test whether the findings hold for over‐time changes in youth unemployment within countries (pseudo‐panel), as well as for cross‐country differences in youth unemployment (multilevel). Both studies indicate that political trust, satisfaction with the economy and subjective well‐being are lowered by economic risk and hardship, while social trust and self‐rated health are less affected by changes in youth unemployment. Secondly, human values are immune to economic risk, underscoring that values transcend specific situations and are therefore resistant against sudden economic shocks.