The posters will be exhibited throughout the entire NEPS conference. The presentation and discussion will take place informally during the breaks between the sessions and especially during the lunch break on Thursday. The presenters are marked with an asterisk. Please click on ≡ to open the abstract of the poster.
The role of gender dominance in job satisfaction: An examination of STEM and gender-dominated professions - CANCELED -
Mara Osterburg* (DZHW Hannover)
≡ Abstract
The persistent phenomenon of occupational gender segregation is a global observation, whereby men and women are systematically employed in distinct occupational fields that are perceived to be either male- or female-dominated (Bächmann et al., 2024; OECD, 2023). In particular, women constitute a notable yet still underrepresented demographic in the context of skills shortages in STEM occupations (science, technology, engineering and mathematics) within the labour market. They frequently encounter discrimination, stereotyping and less favourable working conditions (Barone & Assirelli, 2020). Conversely, men are markedly underrepresented in social occupations such as nursing, healthcare and education (Federal Labour Office, 2021; Federal Statistical Office, 2019). Irrespective of the occupational field, women are more likely to occupy lower status positions, to work part-time, to experience slower career progression, to receive lower pay and to leave the company (Leuze & Strauß, 2016; Allen et al., 2016).
Against this background, it becomes pertinent to investigate the determinants of job satisfaction for women and men in different occupational fields. Does the job satisfaction of women generally exhibit a lower level of satisfaction, or does it vary depending on the gender dominance within the occupational field? For example, what role does the birth of a child play in job satisfaction, particularly in STEM professions?
The study presented here expands existing research on gender-specific career choices by adopting a long-term perspective and analysing the career paths of women and men in various occupational fields over an extended period of time. The objective is to examine the job satisfaction of women and men, as well as the underlying causes and consequences of gender inequality in STEM and other occupational fields in a differentiated manner. The analysis is focused on two key questions: firstly, how job satisfaction differs between women and men in STEM professions; and secondly, how job satisfaction in male and female-dominated occupational fields behaves in a gender comparison.
The database is a longitudinal dataset from the National Educational Panel Study (NEPS) from 2010 to 2020, provided by the Research Data Centre of the Leibniz Institute for Educational Trajectories (FDZ-LIfBi). The sample is comprised of individuals who are currently employed and have a minimum of five years' work experience. In order to analyse gender-specific differences in job satisfaction, event study and random effects (RE) models are employed, taking into account various potential influencing factors in order to adequately capture both individual differences and the causality between gender and job satisfaction.
- The impact of educational and employment transitions on health of young adults depending on internal and external health resources
Julia Fritzsche* (University of Rostock)
≡ Abstract
In the field of health research, it is essential to not only examine the risk factors that contribute to the onset of diseases or disorders but also to explore the factors that promote good health, even in the face of various stresses and strains. Theoretical frameworks, such as the Salutogenesis model and the Systematic Demand-Resource (SDR) Model, suggest that an individual's health is contingent upon their ability to effectively cope with a range of demands by utilizing various internal and external resources. These frameworks particularly emphasize psychological factors, such as personality traits (internal resources), and social factors, such as social networks and social support (external resources).
This project focuses on young adults transitioning from education to employment, a critical phase of life that may have long-term consequences for health. Understanding the factors that promote good health during this transition is therefore essential. The primary objective of this project is to investigate the impact of transitioning from education to employment on young adults' self-rated health (SRH). Additionally, the study aims to examine the significance of health resources in successfully navigating this transitional phase.
The analysis draws on data from the German National Educational Panel Study (NEPS) up to the most recent wave, specifically Cohort 5, which tracks university students from the start of their studies in most cases to their entry into the labor market. A difference-in-differences (Diff-in-Diff) regression approach is applied to evaluate the effect of completing a degree or entering the workforce on health outcomes. The observed group includes students aged 25 to 50, resulting in an unbalanced panel with 105,770 observations. The analysis controls for gender, household size, social background, and geographic location (East or West Germany). Furthermore, the aim of the study is to investigate the moderating effect of potential internal (measured by the Big Five personality traits) and external (social trust/social capital) health resources.
Preliminary multivariate analysis results indicate that the effect of completing a degree has a generally positive impact on respondents' self-rated health (SRH) (Diff-in-Diff Coef. 0.037 | p<0.001), controlling for all aforementioned covariates. Moreover, both personality traits and social trust are significantly associated with SRH. While the project is still in progress, these early results suggest that additional research is needed to fully understand the interplay between educational transitions and health resources. Further analyses will deepen the understanding of how internal and external health resources influence health during this pivotal life stage.
- How do students perceive STEM lessons? A study based on the DIAMONDS taxonomy
Lukas Ketscher* (Friedrich Alexander University Erlangen-Nuremberg)
Heidrun Stoeger (University of Regensburg)
Albert Ziegler (Friedrich Alexander University Erlangen-Nuremberg)
≡ Abstract
Following Rauthmann et al. (2014) situations can be classified in different ways. A fundamental assumption of their approach is that situations can be defined in terms of cues, classes, and characteristics (cf. Saucier et al., 2007; Van Mechelen & De Raad, 1999). Their DIAMONDS taxonomy (Duty, Intellect, Adversity, Mating, pOsitivity, Negativity, Deception, Sociality) focuses on the characteristics of everyday situations. Its main contribution is that situations are ascribed to attributes in much the same way that personality traits are ascribed to people (Edwards & Templeton, 2005; Kola et al., 2020, 2023).
Why is the perception of situation characteristics so critical in educational settings? Prior research has primarily concentrated on the analysis of cues in learning environments (e.g., Cheryan et al., 2014; Guardino & Fullerton, 2010; Jennings & Greenberg, 2009), particularly within the context of STEM education (e.g., Cheryan et al., 2009, 2011, 2015). However, long-term increases in student achievement and further outcome variables have only been documented in select instances (OECD, 2015, 2023). Therefore, further investigation of characteristics is an exciting avenue for future research that has the potential to shed light on the dark (Fairhurst et al., 2023; Funder, 2009, 2016; Ketscher et al., in preparation).
In order to establish a foundation for this approach, it is first necessary to present descriptive data. A DIAMONDS-based questionnaire was conducted concerning STEM education as part of a longitudinal survey of students at the secondary level in several federal states in Germany. Two fundamental research questions were addressed as a descriptive basis for further inquiry: (1) Does the student’s behavior exhibit a connection with situational characteristics? (2) Do different groups differentially perceive STEM lessons?
Parametric methods were employed to analyze data from a sample of 1,132 students. The results revealed (1) notable correlations between specific situation characteristics and students' behavior. Moreover, (2) a link was observed between the perception of STEM education and the somewhat negative associations female students held compared to their male counterparts. Additionally, distinct perceptions concerning students' achievement levels were identified. Accordingly, examining situational characteristics may also prove instrumental in elucidating behavioral processes in the long run, particularly in educational settings. Nevertheless, further research is necessary to achieve comprehensive enlightenment.
- Parental job loss and child development
Lucas Bütje* (University of Konstanz)
≡ Abstract
I study the effects of parental job loss on child development and the mechanisms through which these effects may operate. Much of the economic literature on this topic relies on large administrative datasets to maximize the credibility of their causal identification strategy (e.g. Britto, Melo, and Sampaio, 2022; Hilger, 2016; and Rege, Telle, and Votruba, 2011). Such administrative data sources, however, lack crucial information to understand the mechanisms behind the effect of parental unemployment on children. This is important because income loss, the only channel frequently available in administrative data, has often been found to be insufficient to explain the entire effect (Ruiz-Valenzuela, 2021). These findings highlight the importance of studying other channels such as parental mental well-being and parenting activities (Fox, Han, Ruhm, & Waldfogel, 2013; Ruiz-Valenzuela, 2021).
To bring information on all these channels together, I use longitudinal survey data from the Millennium Cohort Study (MCS). The MCS follows children born in the UK around the year 2000, starting at the age of nine months. Data is collected every two to four years and includes extensive information on the children and their parents as well as standardized tests of child development. In the most recent wave, children are 17 years old.
My empirical strategy is a difference-in-differences event study. Thereby, I compare children in households that experienced a job loss during a given period to those whose parents stayed employed. The difference-in-differences design accounts for any potential confounders that may be different between these groups of households, but constant over time. I analyze the following outcomes: child development (cognitive and non-cognitive), income, parental mental well-being and parenting activities. To address the issue of varying outcome measures over time and make the event study feasible, most outcomes are converted into within-wave ranks.
In preliminary results, parental job loss appears to have a slightly positive effect on cognitive skills and not much of an effect on non-cognitive skills. This could be explained by the finding that, while job loss decreases income and mental well-being of parents, maternal activities with the child increase (regardless of which parent experiences the job loss). It should be noted that these results have been computed with children up to age 11, while data on the more recent waves is still being processed.
- One or both parents work in STEM: Is there a social inheritance effect?
Johanna Pfeuffer* (Friedrich Alexander University Erlangen-Nuremberg)
Heidrun Stoeger (University of Regensburg)
Albert Ziegler (Friedrich Alexander University Erlangen-Nuremberg)
≡ Abstract
The article deals with parents' occupation and their significance for their children's individual identification and choices in STEM fields. Bourdieu's habitus is the theoretical point of departure and was applied to families and the STEM field by Archer et al. (2012). Accordingly, the family habitus characterizes the daily practices, resources, values, and identity concepts of families (Archer et al., 2012). It can, therefore, be assumed that families differ in terms of mindset and capital levels depending on their STEM background (Moote et al., 2020) and differently shape the aspirations of children (Archer et al., 2012).
We analyzed family constellations and compared children's STEM choice intentions. In the first step, it was investigated whether the choice intentions differ depending on which parent works in the STEM sector (RQ1: The more, the better?).
Due to the disagreement regarding the gender-specific effect of (STEM) parents (Lee et al., 2020; Hoferichter & Raufeder, 2019; Sonnert, 2009), we tested whether the group differences apply equally to girls and boys. It was assumed that a parental STEM background would be more meaningful for girls than for boys (RQ2: Booster for girls?).
Lastly, we examined the explanatory factors and analyzed how individual STEM identification can explain the association between parents' STEM background and STEM choice intentions (RQ3: Inherited identities).
For the analysis, survey data of students from middle school level from secondary schools with a STEM focus from the Germany-wide FösaMINT study (N>1,000) were used. Analyses of variance and post-hoc tests were applied to answer the questions, and mediation models were estimated.
RQ1: Based on the data, there were significant group differences depending on the parental STEM background about pupils' STEM choice intentions.
RQ2: There was no interaction effect between family STEM background and gender about the dependent variable STEM choice intentions. Significant mean differences were found for boys for the group comparison of two STEM parents and no STEM parent. Significant differences were found between girls and those with no STEM parents and those with a STEM mother, a STEM father, and two STEM parents.
RQ3: The mediation analysis showed that a stronger individual identification with STEM largely mediated the association between the parental STEM background and the STEM choice intentions. With regard to the explained effect, there were differences in how many people have a STEM background in the family.
- Disciplinary divides: Exploring the development of political interest in higher education
Yuxin Zhang* (University of Trento)
≡ Abstract
Higher education is commonly regarded as a key mechanism for preparing individuals for active democratic participation in public and political spheres. Much research has documented a positive correlation between higher education attainment and political engagement across various countries. As higher education has expanded globally, increasing attention has been directed toward horizontal inequalities within institutions, particularly the division between students in STEM (Science, Technology, Engineering, and Mathematics) and SHAPE (Social Sciences, Humanities, Arts for People, and Economy) disciplines. STEM fields are typically associated with technical and scientific skills, whereas SHAPE disciplines are more focused on moral and social objectives. Previous studies suggest that STEM students may exhibit lower levels of civic and political engagement compared to their SHAPE counterparts. However, the limitations inherent in cross-sectional data and non-representative samples have hindered efforts to draw convincing conclusions.
This study investigates whether higher education genuinely fosters political engagement or if individuals self-select into higher education institutions based on pre-existing dispositions. By examining self-selection and political socialization theories, this research explores the sociopolitical returns to higher education within democratic contexts. We analyze both general trends in engagement throughout the university and variations across different academic disciplines. Utilizing various proxies for political engagement, we expect to answer the following questions: (1) Does higher education generally enhance the development of political engagement? (2) Do individuals exhibit varying levels of political engagement based on their disciplines? (3) Do SHAPE and STEM fields affect individuals' political engagement, and whether the effects are heterogeneous?
Using linear growth curve modeling with NEPS Starting Cohort 5 data, we observed that individuals enter university with already varying levels of political engagement. Although there is a general, slight increase in engagement over time, the rate of development (slopes) shows little difference and often runs parallel across disciplines. Our findings support a modest general socialization effect of the university experience but highlight that self-selection within the university plays a more significant role, with minimal socialization effects associated with specific academic disciplines. These results align with recent contributions emphasizing the importance of the self-selection mechanism and unobserved individual characteristics over environmental factors in shaping individual political engagement during the "impressionable years".
- Beyond the degree: Detangling the effect of family background on overqualification in the UK and Germany
Melanie Hochmuth* (University of Bamberg)
Julia Hufnagl (University of Bamberg)
Richard Wolff (University of Bamberg)
Silvia Annen (University of Bamberg)
Xavier St. Denies (Institut national de la recherche scientifique, Centre Urbanisation Culture Société Montreal)
≡ Abstract
Overqualification is defined as a vertical mismatch in which the level of education attained by a worker exceeds the requirements of his or her current occupation (Erdsiek 2016) and increases costs for all stakeholders, including the individual, the employer, and society as a whole. Although research has identified numerous factors that contribute to the risk of overqualification, there is a lack of understanding of the potential role of overqualification in perpetuating intergenerational inequalities. Moreover, some research has focused on the differences in socioeconomic status attainment between graduates from affluent social backgrounds (Torche 2011; Witteveen and Attewell 2017). However, this research has contributed little to our understanding of the underlying causes of these differences, particularly with regard to the differences in the characteristics of career trajectories by social origin.
To address this, we examine the extent to which parental class background influences an important career dynamic in the UK and Germany: the risk of overqualification and the probability of exit from overqualification. In both cases, family background is likely to affect the risk of overqualification and the probability of exit from overqualification in several ways. First, the family's financial capital could influence overqualification because graduates from wealthy families may have the opportunity to search longer for an adequate job. Second, the job search process may be influenced by parents' social capital and networks (Jonsson et al. 2009). Third, the transmission of cultural capital from parents to children may contribute to higher chances of accessing the most elite jobs available to university graduates from higher social backgrounds (Laurison and Friedman 2020; Rivera 2015). Finally, social differences in aspirations may also affect individuals' occupational choices due to a desire to prevent downward mobility (Jacob and Klein 2013).
To test these hypotheses, we compare the incidence and persistence of overqualification in the UK and Germany between 2010 and 2020. The results are derived from analyses of data from the UKHLS and the NEPS SC6. As first results show, individuals with a father with academic occupation, women, older workers, and those living in West Germany are more likely to be overqualified, while self-employment reduces the risk of overqualification.
- Differences in factors relating to perceived personal and group discrimination
Florian Weber* (LIfBi Bamberg)
≡ Abstract
Perceived (ethnical) Discrimination can be broadly categorized into perceived personal discrimination (i.e. discrimination the individual reports being subjected to because they belong to a certain ethnic group) and perceived group discrimination (i.e. discrimination the individual perceives migrants in general, or the own ethnic group, are subjected to). While differences in levels between those two dimensions have been reported and analysed, differences in the factors associated with perceived discrimination have not been in the focus of research.
Additionally, the integration (or discrimination) paradox has attracted interest in the last few years. It describes the phenomenon that perceived discrimination (as an aspect of emotional integration) is positively associated with education and labour market attainment (as aspects of structural integration), while classical theory of integration would suspect a negative association. Several mechanisms have been proposed and analysed in literature, however without focus on possible differences between the two dimensions of discrimination mentioned above. Therefore, I want to answer the following research question: Which proposed mechanisms possibly explaining the integration paradox relate to perceived personal and which to perceived group discrimination?
One central question of the explanation of perceived discrimination in general is whether exposure or a higher susceptibility (e.g. due to more knowledge about discrimination) is the central mechanism of the discrimination paradox. I suspect that higher exposure is related to perceived personal and higher susceptibility to both perceived personal and group discrimination.
One central question of the explanation of perceived discrimination in general is whether exposure or a higher susceptibility (e.g. due to more knowledge about discrimination) is the central mechanism of the discrimination paradox. I suspect that higher exposure is related to perceived personal and higher susceptibility to both perceived personal and group discrimination.
I use various data sources to answer my research question. Firstly, I re-analyse data from a current meta-analysis on factors related to perceived discrimination by Schaeffer & Kas (2023). While they use the dimension (personal vs. group) of the studies included as a binary variable in their meta-regressions, this only allows for differences in levels between those dimensions. I modify this to estimate separate models for studies using perceived personal and group discrimination, respectively. Secondly, I use NEPS SC6 data (with Mikrozensus data additionally used for the generation of certain measures) which contains both dimensions when perceived discrimination was first recorded in the 2018 panel wave.
For both types of analyses, I find that the predictors differ between perceived personal and group discrimination. In the analysis using SC6, several predictors relate significantly to perceived personal discrimination, while there no substantial relations can be found for perceived group discrimination.
- I am good enough: Constructing a short measure of global self-esteem
Diana Steger* (LIfBi Bamberg)
Elena Wittmann (LIfBi Bamberg)
Astrid Schütz (University of Bamberg)
≡ Abstract
Self-esteem is a fundamental aspect of the self. It affects numerous outcomes in mental health, interpersonal relationships, and educational and work-place settings. Despite its significance, existing measures of self-esteem often lack brevity or do not meet psychometric standards. To address this gap, the present study aimed to develop a concise four-item scale that is a) unidimensional, b) reliable, and c) invariant across age to map the inter- and intra-individual differences in self-esteem over the life course.
For the short scale development, we draw on data from NEPS starting cohorts 3, 4, 5 and 6 (N=24,285) covering ages 12 to 31 who had completed the original 10-item version of the Rosenberg Self-Esteem Scale. To allow for an independent validation, in a first step data was randomly split into two separate subsamples: Using 80% of the original sample, we computed unidimensional factor models for all possible item sets, comparing model fit (CFI, RMSEA, SRMR), factor saturation (McDonald’s ω), and factor loading patterns. Furthermore, we computed locally weighted structural equation models for each item set, which allow for the inclusion of age as a continuous covariate to examine differential item functioning (DIF) effects across age. Across all item sets, the most promising solution that included both positively and negatively worded items resulted in sufficient model fit (CFI=.98, RMSEA=.09, SRMR=.02), excellent factor saturation (ω=.72) and sufficient loading patterns (minλ=.39). Three of the four indicators were invariant across age.
In a second step, we used the remaining 20% of the sample to replicate the psychometric properties of the scale and examine the relationship of the newly developed short scale with relevant variables. The replication revealed overall satisfactory psychometric properties (CFI=.96, RMSEA=.12, SRMR=.03, ω=.72, minλ=.36) and no DIF effects across age. Furthermore, the proposed short scale shows correlation patterns that are very similar to those of the long form, it is positively correlated positively to global well-being and health, aspects of personality (i.e, extraversion and conscientiousness), and aspects of social competencies (i.e., prosocial behavior, assertiveness, and conflict ability) and negatively correlated to neuroticism. Consistent with existing literature, men had significantly higher scores than women.
This new scale offers a brief yet reliable tool for assessing self-esteem across the life course, making it highly suitable for contexts requiring efficient psychological assessments, such as large-scale assessments.
back to top