+++ Abstract submission is closed +++ Registration is open +++

Conference Schedule

The titles, authors and abstracts of all paper presentations are listed below the program overview and can be accessed directly by clicking on the sessions. The presenters are marked with an asterisk. Please note that the program is preliminary and may be subject to change.

Session 1: Effects of Policies

Thursday | December 5, 2024 | 11:00 am - 12:15 am (CET)

• Educational disparities in trust changes during the Covid-19 Pandemic (2017-2022)
  Gundula Zoch* (University of Oldenburg, LIfBi Bamberg)
  Steffen Wamsler (LIfBi Bamberg, University of Bamberg)
  ≡ Abstract

  The pronounced and socially stratified increase in political trust at the start of the Covid-19 pandemic is well-documented. However, our understanding of the underlying causes and mechanisms remains incomplete. Drawing on identity theory and previous empirical research on crisis vulnerability, we examine educational disparities in trust changes throughout the pandemic in Germany. Based on the combination of educational disparities in stratified pandemic exposure, resulting from pre-existing and reinforced labour market inequalities, as well as inequalities in skills and perception, we hypothesize that individuals with lower levels of education are more likely to experience a more pronounced initial increase in political trust compared to those with tertiary education (rally-around-the-flag-effect). However, as the pandemic persisted, the unequal social and economic consequences of containment measures became increasingly apparent. Therefore, compared to those with tertiary education, individuals with lower levels of educational attainment are expected to experience a more rapid decline in trust (H2) and a more pronounced overall reduction in political trust (H3).
  We utilize data from the Adult Cohort of the National Educational Panel Study (NEPS, 2017-2022; N=21,021) on trust in state and non-state actors, as well as the media, across four waves before and during the pandemic combined with county-level infection rates and restriction measures. Results from individual-level fixed-effects regression models reveal a sharp initial trust increase, particularly among lower-educated individuals and toward core government institutions. As the pandemic progresses, trust declines, with more pronounced changes among lower-educated individuals and in regard to trust in the media and the police. Conversely, trust in the government remains above pre-pandemic levels. Surprisingly, infection rates and restriction measures on the district level as well as changes in individual objective and subjective pandemic affectedness do not substantively explain educational disparities in trust changes. However, further analyses reveal that lower ICT literacy partly explains the observed educational disparities. From a broader perspective, the results therefore point towards the importance of socially stratified competencies that are likely to shape short-term attitudinal change in times of crises and its socially stratified perception.



• Empty class to empty streets? The effect of permanent school closures on regional economic development and voting behaviour
  Sebastian Vogler* (LIfBi Bamberg, Dresden Technical University)
  Marcel Helbig (LIfBi Bamberg)
  ≡ Abstract

  This paper investigates the effect of permanent school closures in Germany on regional economic development. For this purpose, this paper draws upon several distinct datasets: First, we evaluate administrative data on schools, provided by the Ministries of Education from all federal states in Germany and link these data with precise longitudinal and latitudinal information. This allows us to precisely track any permanent school closure and the respective distance to the next comparable institution. Secondly, datasets from the Federal Institute for Research on Building, Urban Affairs and Spatial Development were utilised, which provided detailed information on small-area economic performances such as total tax income, unemployment rates, net migration, birth statistics and territorial changes for all municipalities in Germany. Finally, data on different voting outcomes, including federal, state and local elections, were used to assess any voting behaviour. This was provided by the respective State Statistical Offices. The combination of all data sets allows for the construction of a balanced panel data set on a yearly basis, spanning the period from 2000 until 2019, encompassing all 10,753 municipalities in Germany.
  Using two-way fixed effects and matched difference-in-differences estimations, we investigate the effect of permanent school closures on economic performance and voting behaviour.
  Our preliminary findings indicate that permanent school closures in Germany do not significantly impact population or gender composition within the affected municipality. However, they do result in a three-percentage point increase in long-term unemployment and a two-percentage point decrease in taxation income, leading to a decrease in economic power of affected municipalities. Furthermore, as a results of demographic changes, there is evidence of a shift in voting behaviour in communal elections from traditional political parties to smaller parties, single candidates and voter associations.



• Large families in Germany: Social profile, child poverty and policy solutions
  Stephan Köppe* (University College Dublin)
  Megan Curran (Columbia University)
  Iñigo Aldama (University College Dublin)
  ≡ Abstract

  Historically, researchers and policymakers alike recognized the risk of poverty among large families, but family size is often neglected in the contemporary literature. This article revives an examination of the connections between family size and poverty risk for children with a focus on Germany. We take a child-centred perspective by analysing a sample of 13-14 year-old children from the National Educational Panel Study (NEPS).
  First, we analyse the social profile of large families and the household and parental characteristics of growing up in a large family. Second, we provide a detailed overview of the welfare and tax policies aimed at large families in Germany. Next, we estimate the poverty risk and prevalence for children in large families (looking at families with 3+ and 4+ children). Finally, we discuss how the policy and socio-economic context interacts with the risk of poverty. We identify that the means-tested social assistance scheme penalizes large families, while the child benefit would only acknowledge higher need of middle-income families with three or more children.
  The article is situated in a wider research agenda on the social profile and disadvantage of large families within advanced welfare states in an upcoming special issue. The presentation will give an outlook on comparative datasets that can be combined with NEPS.

back to top

Session 2: Effects of School Contexts

Thursday | December 5, 2024 | 01:15 pm - 02:55 pm (CET)

• Timing of school entry and personality traits in adulthood
  Anton Barabasch (Friedrich Alexander University Erlangen-Nuremberg, Dresden University of Technology)
  Kamila Cygan-Rehm* (Dresden University of Technology, LIfBi Bamberg)
  Andreas Leibing (Dresden University of Technology, DIW Berlin)
  ≡ Abstract

  This paper examines the long-term consequences of later school entry on personality traits. School enrollment laws are among the most common public policies worldwide. They typically define a cut-off date such that all children born before this date are enrolled in a given school year, and those born after this date are enrolled one year later. The cut-off date largely determines the timing of a child's enrollment, and extensive research has examined its consequences for human capital development, especially cognitive skills. Much less is known about its effects on the development and persistence of personality traits.
  From a theoretical perspective, long-term effects on personality traits would be consistent with the dynamic model of skill formation (Cunha and Heckman, 2007, 2008), i.e., even if some differences initially arise solely because of chronological age advantage, self-productivity and dynamic complementarity of skills create mechanisms that could sustain the effects into the teenage years and beyond. Moreover, starting school can be a stressful life event (Landersø et al., 2020), and it occurs at a critical time for human capital formation (Currie and Almond, 2011). Thus, the unique combination of the importance of school enrollment and its timing within the life cycle is another reason why we might expect persistent effects on personality traits.
  We provide a comprehensive analysis of the long-term effects on the Big Five Inventory of personality traits. To establish causality, we exploit plausibly exogenous, institutionally induced variation in the timing of individuals' school entry. Specifically, we exploit statutory cut-off rules for school enrollment in Germany within a regression discontinuity design (RDD). The German context is particularly suitable for our research design because, unlike the US or the UK, there is no automatic relationship between the timing of school entry and compulsory schooling requirements. We apply this method to survey data from the National Educational Panel Study (NEPS): Starting Cohort 6 - Adults and the German Socio-Economic Panel (SOEP) and to administrative data from school entry examinations.
  We find that older school entrants enjoy a long-term reduction in levels of neuroticism. This trait refers to emotional volatility and is typically associated with poor well-being and worse socioeconomic outcomes (Gensowski et al., 2021). The effect is driven by women and persists into adulthood. This has important implications for gender gaps in the labor market, as women typically score significantly higher on neuroticism at all stages of life, which puts them at a disadvantage.



• Do SES gaps in educational achievement decrease when secondary schools introduce all-day care? Evidence from Germany
  Adrian Greiner* (University of Bamberg)
  ≡ Abstract

  In response to the 2001 ‘PISA shock’, the German government invested in the expansion of all-day care (ADC), leading many schools to adopt it. This initiative was partly motivated by the assumption that introducing ADC would reduce socio-economic disparities in children's educational achievement (KMK 2002). Studies challenging this assumption with causal estimates are scarce and have focused on ADC in primary school (Schmitz 2022; Seidlitz and Zierow 2022). That is why I investigate whether all-day care in secondary school reduces socio-economic disparities in educational achievement. My findings serve not only as an evaluation of a major German educational policy, but also contribute to the discussion whether educational interventions can be effective both in early childhood and adolescence (Heckman 2006; Dahl et al. 2018; Guryan et al. 2023).
  My paper proceeds as follows: By adopting rational choice theory and a theoretical framework developed in early education research, I theorize on the SES gradient in both ADC attendance and ADC effects (Jonsson and Erikson 1996; Cascio 2015). As dimensions of achievement, I consider students’ conscientiousness as well as their grades and competences in math and German. To test my hypotheses, I rely on data from the German National Educational Panel Study (NEPS). I combine subsamples from the NEPS starting cohorts 3 and 4 to create a two-wave panel at the school level that covers two cohorts of 9th graders. The resulting dataset allows me to implement a 2x2 Difference-in-Differences design at the school level. My estimates for the intention-to-treat effect of ADC introduction mostly show very similar effects for all students, regardless of whether their parents have obtained (at least) an Abitur or not. When focusing on German grades and conscientiousness as outcomes, I find slightly exacerbating effects on socio-economic disparities. Based on these results, I argue that for now ADC in secondary education has not fulfilled the disparity-reducing role that German policymakers have ascribed to it. My analyses further illustrate a novel use of the NEPS data that can facilitate future research on (policy) interventions in German secondary school.



• The dramaturgy of grade retentions – Act I: When state and school characteristics become risk factors for grade retentions
  Paul Fabian* (Ludwig Maximilian University Munich)
  Benjamin Gröschl (Ludwig Maximilian University Munich)
  Katja Scharenberg (Ludwig Maximilian University Munich)
  ≡ Abstract

  Nationally and internationally, grade retention is a highly controversial measure to homogenize students with different competencies. In Germany, some federal states (e.g., Hamburg and Berlin) have already abolished grade retention. In Bavaria and Bremen, however, grade retention rates are above the national average (2.3%, Statistisches Bundesamt (Destatis), 2018). Empirical evidence on the effectiveness of grade retention is still insufficient, but the tendency of effects is rather neutral (Fabian, 2020; Goos et al., 2021).
  While retention decisions are generally based on academic performance, grades and absences from school (EURYDICE, 2011), student characteristics (Krohne et al., 2004; Gröhlich & Bos, 2007; Klapproth et al., 2015) and school context factors (Bless et al., 2004; König & Darge, 2010; Hong & Raudenbush, 2005; Klapproth et al., 2015) also influence retention or promotion.
  Our study focuses on identifying school-level factors influencing retention rates using data from NEPS - Starting Cohort 3 (NEPS-Network, 2024). Separate analyses are conducted for waves 1 (n=104) and 3 (n=180). Retention rates are calculated as the z-standardized ratios of retained students to total enrollment. Predictor variables are stepwise integrated into regression models. All analyses are conducted using R 4.3.3 (R Core Team, 2024).
  In Wave 1, retention rates were higher at intermediate tracks (B=.45, p<.10), in urban regions (B=.37, p<.10), in public schools and in states 3 (B=1.07, p<.10) and 12 (B=1.44, p<.10). In contrast, teacher time for consultative meetings (B=-.08, p<.05) was negatively associated with grade retention. Beyond this, grade retention rates were lower in all-day schooling (B=-.43, p<.05).
  In Wave 3, higher grade retention rates were associated with a higher proportion of students with a migration background (B=.21, p<.10) as well as for those at academic (B=.56, p<.10) and intermediate tracks (B=1.05, p<.00). Multi-track schools (B=.73, p<.05), schools in urban areas (B=.54, p<.05), as well as the dependency of funding on numbers of student enrollments (B=.23, p<.05) also showed positive associations. State 4 also had an association (B=.77, p<.10) with the retention rate, while larger schools exhibited lower rates (B=-.18, p<.10).
  In conclusion, some states showed higher grade retention rates even after accounting for control variables. Grade retention rates were higher in schools compared to private ones. Interestingly, schools whose funding is dependent on number of student enrollments had higher grade retention rates, despite this often leading to a decline in enrollment due to students’ (mostly downwards) track changes. In future analyses, we plan to include also socio-spatial data.



• Grade retention: Removing the “bad apple”?
  Samuel Jalalian* (Ludwig Maximilian University Munich)
  Fabian T. Pfeffer (Ludwig Maximilian University Munich)
  ≡ Abstract

  This study examines how the retention of a student (“repeater”) affects the marks and competencies of the promoted students from whose class the repeater was removed (“promoted class”). While much of the prior literature questions the benefits of grade retention for repeaters themselves (Contini & Salza, 2024; Goos et al., 2021; Hattie, 2009, S. 97ff.), some attention has been given to its effects on students in classes receiving the repeater (Bietenbeck, 2020; Hill, 2014; Gottfried, 2013b, 2013a). However, no research has explored the effects of grade retention on the students in the promoted class, which is the focus of this study.
  Building on Lazear's (2001) bad-apple model of peer effects, we hypothesize that retaining a low-performing student might improve their former classmates' academic outcomes by reducing classroom disruptions and enabling the teacher to better focus on the remaining students (H1a). On the other hand, the additional explanations provided for the low-achieving student may have benefited the entire class, so their removal could negatively impact the academic performance of the remaining students (H1b). We also examine how these effects vary by individual performance level (H2a/b).
  To test our hypotheses, we track the performance of non-retained students in the promoted class over at least two years, comparing individual academic outcomes before and after the repeater’s removal. For this, we utilize the longitudinal class identifier developed by Florean et al. (2019), available for starting cohorts 2 (grades 1-4) and 3 (grades 5-9), limiting our analysis to these cohorts and grade levels. We apply fixed-effects regression models with individual fixed effects and clustered standard errors by class.
  Our findings reveal no statistically significant effect of grade retention on the German and math marks (SC2, SC3) or on science (SC2), vocabulary (SC2), math (SC3), and reading (SC3) competencies of a promoted student, on average. However, statistically significant (p<.000) negative effects are observed in elementary students' math competencies, particularly among low achievers. These results challenge the “bad-apple” hypothesis, suggesting that grade retention does not benefit, and may even harm, the remaining promoted students.
  Given the high financial costs of grade retention (around €900 million per year; Klemm, 2009) and the lack of clear benefits for repeaters or non-repeaters, our findings call for reconsidering the practice of grade retention.

back to top

Session 3: Mobility and (Higher) Education

Thursday | December 5, 2024 | 04:45 pm - 06:00 pm (CET)

• The untold story of internal migration in Germany: Life-cycle patterns, developments, and the role of education
  Anton Barabasch* (Friedrich Alexander University Erlangen-Nuremberg, Dresden University of Technology)
  Kamila Cygan-Rehm (Dresden University of Technology, LIfBi Bamberg)
  Guido Heineck (University of Bamberg)
  Sebastian Vogler (LIfBi Bamberg, Dresden Technical University)
  ≡ Abstract

  Research question: Internal migration plays a crucial role in shaping economies and societies, yet little is known about migration patterns in Germany due to data limitations. While research has focused heavily on the U.S., this paper fills a gap by providing a comprehensive analysis of lifetime regional mobility in Germany. Apart from the well-documented East-West and gender gaps, we shed light on the role of education, as it has been long recognized as a key factor for understanding why some individuals move across regions and others do not (known as the “positive skill selection”).
  Data: We utilize data from the National Education Panel Study: Starting Cohort Adults (NEPS-SC6), which offers detailed biographical information on residential moves and educational paths. Our sample includes nearly 13,000 individuals born in Germany between 1944 and 1986, tracking their geographic mobility on a monthly basis from birth until 2020.
  Theory: From an economic perspective migration can be viewed as an investment decision, similar to schooling. Following this concept, education can affect an individual's location choices through several channels. First, education may enhance individuals’ ability to react to disequilibria (Schultz, 1975), such that they migrate in response to regional differences, e.g., in wages or employment opportunities. This assumes that increased education enhances individuals' ability to acquire and interpret information accurately and take actions that result in appropriate relocation. Second, education may impact migration behavior if local labor markets for higher-educated workers become relatively thin. This mechanism requires that education affects educational credentials that are transferable between regions. This may be particularly important in countries like Germany, where secondary school degrees and postsecondary diplomas play a crucial role.
  Methods and results: We describe life-cycle mobility patterns, highlighting differences by space, time, and socio-demographic factors. We find significant mobility around important educational decisions and disparities across groups, especially by educational attainment. We then investigate causality in the substantial education-mobility gradient. For identification, we exploit two policy-induced sources of variation, each shifting towards better education at a different margin of the ability distribution. Using a difference-in-differences and regression discontinuity design, we find no effect of these policies on internal mobility.



• Life satisfacation trajectories at the transition to higher education
  Rafael Warkotsch* (DZHW Hannover)
  Nicolai Netz (DZHW Hannover)
  ≡ Abstract

  Entering higher education (HE) is a major life event that comes with profound changes for individuals. It not only brings about a substantial change in the learning environment and constitutes a major step in the transition to adulthood (Arnett, 2007); additionally, we argue that it is also inherently linked to spatial mobility decisions. Because HE institutions are not evenly dispersed within countries and because leaving school provides an opportunity to also leave the parental home, students are faced with the decision to stay in their home region or to migrate or commute to other regions.
  However, even though entering HE is a major life event and spatial mobility is common among HE students (Quast et al., 2022), only very few studies have investigated changes in life satisfaction following the transition to HE (Dietrich et al., 2022; Reuter et al., 2022). While there is a growing literature on the effects of spatial mobility on life satisfaction in non-student populations (e.g., Erlinghagen et al., 2021; Kratz, 2020), no study has the role of spatial mobility at the transition to HE for individuals‘ life satisfaction trajectories. Given the specificities of the transition to HE (educational setting, age, little prior mobility experience of students), results for non-student populations are not readily transferrable to students entering HE. We aim to narrow this research gap by applying available evidence from other populations (e.g., employees) to the setting of HE. Drawing on social production functions theory (Ormel et al., 1999) as a theoretical framework, we explore the life satisfaction trajectories of students who stay in their home region, migrate within Germany, or commute long distances and test the differences between these trajectories.
  We use data from the German National Educational Panel Study Starting Cohort 4 (NEPS SC4) to estimate fixed-effects models of students’ life satisfaction trajectories. We find substantial increases in life satisfaction upon entering HE for stayers and internal migrants, which are smaller for long-distance commuters.
  Overall, our results highlight the importance of considering the coupling of multiple life events (Buchmann & Kriesi, 2011), such as spatial mobility and the transition to HE, when examining life satisfaction trajectories.



• Disparities in academic risk taking and learning success across gender and socioeconomic groups
  Vanessa Hübner* (University of Bamberg)
  Maximilian Pfost (University of Bamberg)
  ≡ Abstract

  Theoretical background: Students engaging in academic risk-taking (ART) participate in university activities despite uncertainty, potential errors, and possibly being perceived as less competent by instructors or peers (Beghetto, 2009; Clifford, 1991). The positive learning effects of ART in higher education has been discussed since the 1980s (Clifford, 1984; Krochmal & Roth, 2017); however, these benefits may vary across social groups.
  According to multifactorial gender theory, gender is learned, dynamic and shaped by traits, behaviors, and societal expectations (Spence, 1993). Individuals construct their gender identity i.e., how masculine or feminine they feel, by incorporating gender-specific indicators, such as ART, into their everyday life (Vantieghem et al., 2014). Since societal expectations and stereotypes tend to link women with shyness (Meyer et al., 2017) and men with dominance (Bhatia & Bhatia, 2021), structural differences in how ART is employed to construct gender identity are conceivable.
  Additionally, students possess a set of knowledge and skills acquired through socialization and experiences, known as cultural capital (Bourdieu, 2003). Cultural capital may play an important role regarding ART and academic outcome. High-SES students’ cultural capital is generally more valued in academic settings (Edgerton & Roberts, 2014), leading to greater self-confidence (Engstrom et al., 2023) and academic success (Rodríguez-Hernández et al., 2020). Conversely, a mismatch between cultural capital and academic demands can cause uncertainty and withdrawal (Qiu & Ye, 2023), indicating that low-SES students may struggle more with ART.
  Objectives: This study investigates gender and SES differences in ART and their association with learning success. Furthermore, it examines whether ART is moderated by gender and SES, and whether ART directly predicts learning success.
  Method: A sample of N=381 German university students (70.6% female, M age = 23.0 years) was used. ART was measured using two dimensions: (1) ART shown in front of the seminar group (including instructors) and (2) ART displayed only among peers. Results: Data was analyzed using structural equation modeling.
  Men showed more ART on the seminar group dimension, whereas women showed more ART among peers. Being male indirectly predicted higher learning success via the seminar group dimension of ART. Furthermore, SES and gender moderated the relationship between ART and learning success. Both the seminar group dimension and the peer dimension of ART directly predicted learning success.
  Conclusions: Our research enhances the understanding of social disparities in higher education and informs inclusive teaching strategies, highlighting the role of intersectionality in learning.

back to top

Session 4: Competencies and Measurement

Thursday | December 5, 2024 | 04:45 pm - 06:00 pm (CET)

• Bidirectional effects between reading literacy, reading enjoyment and reading interest of students between Grades 5 and 9
  Nora Heyne* (University of Bamberg)
  Maximilian Pfost (University of Bamberg)
  ≡ Abstract

  Building on previous cross-sectional results of positive associations between reading literacy and intrinsic reading motivation (e.g., Bos et al., 2012), longitudinal studies based on expectations of reciprocal effects between both revealed inconsistent results: While only a few results supported assumptions of reciprocal relations (e.g., Schaffner et al., 2016), other studies only showed reading-promoting effects of intrinsic reading motivation (e.g. Miyamoto et al., 2018) or positive effects of reading literacy on later reading motivation (e.g., McElvany et al., 2008). In addition to these different relations that studies revealed with respect to intrinsic reading motivation as a general construct, other studies showed inconsistent findings on the relation between reading literacy and different components of intrinsic reading motivation, especially reading enjoyment and reading interest (e.g., Möller & Bonerad, 2007; Schiefele et al., 2016). Thus, regarding the course of schooling, there are still open questions about the relation between reading literacy and intrinsic reading motivation, in particular their components reading enjoyment and reading interest.
  Against this background, the study investigates reading enjoyment and reading interest in adolescents’ autonomous reading and their relations to reading literacy between Grades 5 and 9. In detail, it examines whether these relations are characterized by reciprocal effects and whether they differ in mutual comparison as well as in the course of adolescence.
  The analyses are based on longitudinal data of students (N=2,854) at three measurement points from the National Educational Panel Study in Germany and are conducted with two approaches of cross-lagged panel models, i.e., the approach of inter-individual differences and changes over time with autoregressive second lags (CLPM-L2), and the approach of intra-individual change that differentiates between-person and within-person effects (RI-CLPM).
  The results indicate that reading literacy is reciprocally related to reading enjoyment, but not to reading interest. The comparison of both components in relation to reading literacy shows that the effects of reading enjoyment on reading literacy exceed the effects of reading interest. Moreover, the relations between reading literacy and the respective components of intrinsic reading motivation showed almost no change over the observed period. Overall, concerning the relation between reading literacy and intrinsic reading motivation, the results suggest that future research and practice should distinguish between reading enjoyment and reading interest. Moreover, promoting reading enjoyment rather than reading interest is a promising way to promote young people's reading literacy.



• Rapid guessing in elementary school students
  Luis Fischer* (University of Bamberg, LIfBi Bamberg)
  Jana Welling (LIfBi Bamberg)
  Kathrin Thums (LIfBi Bamberg)
  ≡ Abstract

  Large-scale assessments provide important data for scientific research and education policy decisions. However, participants may decide to select responses arbitrarily in order to finish quickly. Such rapid guessing (RG) puts the quality of the generated data at risk. So far, research on test-taking motivation primarily focusses on adolescents and adults, while evidence for younger target groups is scarce. Younger participants in an assessment generally tend to have a shorter attention span than adults and may therefore be particularly susceptible to RG. Literature also suggests that test-takers, whose native language is not the test language, demonstrate more RG. Taking a language test in a foreign language likely requires additional cognitive resources that could be a barrier and thereby a threat to test fairness.
  The purpose of this project is to investigate the prevalence and impact of RG in a sample of elementary school students. First, it will be explored how frequently RG occurs in this group (research question 1a). To this end, a 20% normative threshold and a visual threshold based on item response times are determined. Second, a chi-squared test is used to test whether RG differs between children who only speak German at home and those who do not (research question 1b). Finally, the impact of RG on data quality will be assessed by means of Cronbach’s Alpha and an effort-moderated IRT-model (research question 2).
  The study is based on data from NEPS Starting Cohort Newborns (SC1, Wave 9). The sample consists of N = 1,467 (50.9% female) elementary school students (M age = 8.29, SD age = 0.13) who took part in a telephone-proctored and web-based early reading competence test (ELFE II) in their homes in 2020. First results using the normative thresholds show that 2.7% of the responses were flagged as rapid guesses. The median response time effort was 1, indicating high engagement. With the normative threshold, 94.1% of elementary school students were classified as engaged on more than 90% of the items. Response time effort does not differ significantly between children who only speak German at home and those who do not. Results for the visual thresholds as well as research question 2 will be reported in the presentation and discussed with regard to the test motivation of elementary school students and general test fairness.



• Can rapid guessing be anticipated using the hitherto test course, personal dispositions and item characteristics?
  Leo Köllich* (DIPF Frankfurt/Main)
  Lena Engelhardt (DIPF Frankfurt/Main)
  Frank Goldhammer (DIPF Frankfurt/Main, ZIB Frankfurt/Main)
  ≡ Abstract

  Disengaged test-taking behavior, especially rapid guessing without an attempt to thoroughly process and solve the task, is considered a wide-spread problem of low-stakes assessments (e.g. Finn, 2015; Silm et al., 2020; Wise et al., 2021). This issue has gained increasing attention, not least due to rise of online assessments. For example, the number of methods to identify rapid guesses (RGs) has grown significantly in recent years (see e.g., Wise, 2019). So far, relatively little research has been conducted on whether and to what extent RGs can be anticipated and explained by the hitherto test course (see e.g., Lindner et al., 2019), supplemented by personal dispositions and item characteristics.
  This question will be addressed using data from a computer-based, 28-minute multistage mathematics test from the National Educational Panel Study (NEPS; Blossfeld & Roßbach, 2019) (N=4,410 of the adult cohort “SC6”, 9th wave (2016)).
  First, responses were classified into “solution behavior” and “RGs” using the NT20 method (with threshold cap 20s). This resulted in a total of 1,355 RGs (1.9%). The classification will be refined in future iterations. No distinction is made between RGs and rapid omits (quickly skipping questions without inputting an answer).
  First results indicate that a small part of the sample (0.7%) with particularly high disengagement (over 50% RGs) is source of a notable portion of all RGs (27%). 31% of RGs are directly anteceded by another RG, i.e. later parts of an ongoing RG streak. As main analysis we conducted a binomial generalized linear mixed model predicting/explaining the RGs. The preliminary model attained 18.9% variance explanation (marginal R²) using item position (estimate 2.42***), information regarding the persons hitherto RG behavior (latest answer being a RG (0.73***), RG quota (7.40***)), recent unsuccessfulness despite time investments (-0.48), and trends in recent response times (direction (-0.36***), strength of in- (0.69***) or decreasingness (0.75***)) as test-course based predictors of RG. Conscientiousness (0.41), agreeability (-1.36*) and neuroticism (0.78) were used as dispositional predictors, and text length (-0.60), amount of on-screen numbers (0.72*), and whether a formula is shown (0.89**) as item level predictors. Despite trend strengths, all predictors are ranged 0-1. An additional 13.9% of variance were explained by varying baseline RG probabilities for items and persons (random intercepts; conditional R²=.328).
  Research on RG explanation and prediction could help prevention-, identification- and handling of RGs. However, apart from previous RG behavior, predictors showed very low effects – further explanation is needed.

back to top

Session 5: Aspirations and Job-Related Information

Friday | December 6, 2026 | 09:00 am - 10:40 am (CET)

• The impact of an asylum seeker influx on occupational aspirations of German youth
  Eva Lickert* (ETH Zurich)
  Audrey Au Yong Lyn (ETH Zurich)
  ≡ Abstract

  We study how the 2015 arrival of primarily unskilled, male asylum seekers affected occupational aspirations of the young incumbent population in Germany. We leverage data from the NEPS Starting Cohort 3, comprising students in ninth grade during the peak of the influx. These students were at a critical point, just before entering the labor market or pursuing the academic track in high school, making them highly susceptible to external influences on their labor market aspirations. We combine the NEPS data with regional datasets from the German Federal Office for Building and Regional Planning (BBSR) to obtain district-level figures for asylum seekers and other relevant variables.
  The sudden demographic shock caused by the influx likely increased competition for certain jobs. We hypothesize that incumbents aspire to more female-dominated, prestigious, and psychosocially demanding occupations, as asylum seekers –- who are predominately male, less educated, and often traumatized or socially disintegrated –- do not naturally sort into such occupational fields. We anticipate shifts in labor market aspirations, especially among those at risk to end up in low-paid jobs, determined by factors such as educational attainment, socio-economic background, and immigrant background.
  Using district-level allocations of asylum seekers as a source of variation, we employ a difference-in-differences model and address potential bias from selective migration. Specifically, individuals are sorted into treatment and control groups based on the absolute rate of increase in the number of asylum seekers between 2014 and 2015. We sort German districts according to the absolute magnitude of the increase in the asylum seeker rate per 100,000 inhabitants between 2014 and 2015 and designate those districts with an increase above the median as the treatment group, while those with an increase below the median form the control group. To ensure that the estimate is not biased due to selective migration, we fix the individuals to their sample school districts.
  Our estimates reveal significant shifts in aspirations toward female-dominated, highly prestigious, and psychosocially demanding occupations. On average, incumbent students aspire to occupations with a 4.334% higher share of females relative to the mean. They aspire high-prestige occupations 11.563% more, while avoiding low-psychosocial-exposure occupations 34.555% less, compared to the mean. We provide suggestive evidence that these outcomes are caused by a competition effect.



• Vocational aspirations and the cognitive requirement level of training occupations
  Anett Friedrich* (BIBB Bonn)
  Annalisa Schnitzler* (BIBB Bonn)
  ≡ Abstract

  In Germany training occupations are heterogeneous in many respects, including the cognitive requirement level (Friedrich et al. 2023). Occupations with higher cognitive requirements should offer more opportunities to learn additional and more transferable skills, which should be remunerated accordingly (Meyer/Sacchi 2020). Access to these occupations is strongly linked to the school-leaving certificate (Friedrich et al. under review). Young people with maximum a lower secondary school-leaving certificate rarely succeed in making the transition to a more cognitively demanding training occupation (ibid.), even though some of them have cognitive skills that are comparable to those of graduates with an intermediate secondary school-leaving certificate or Abitur (Holtmann et al. 2018). So far, it remains unclear whether young people with no more than a lower secondary school-leaving certificate face firm-induced and institutional barriers or whether they do not aspire to more demanding occupations. Therefore, we investigate who aspires to occupations with higher cognitive requirements and which individual characteristics are related to those aspirations.
  Our theoretical framework is based on the Expectancy-Value-Model (Eccles/Wigfield 2002), according to which each occupational option is associated with a subjective expectancy of success and a subjective task value, which consists of four elements: (1) the interest-enjoyment value (interest and enjoyment of activity); (2) the attainment value (relevance for confirming one's own identity); (3) the utility value (usefulness for one's goals); and (4) the relative costs (e.g. required effort, fear of failure, opportunity costs).
  We use the NEPS-SC 3 (NEPS Network 2024) to estimate linear probabilistic models (Wooldridge 2010) for young people who have ever expressed an interest in vocational education and training. Our dependent variable is the cognitive requirement level of the realistically aspired training occupation. We use an indicator from the Vocational Psychological Service (BPS) of the Federal Employment Agency (Friedrich et al. 2023).
  The preliminary results indicate that young people with better mathematics grades and those who perceive that attaining an intermediate school-leaving certificate would entail minimal learning stress are more likely to aspire to occupations with higher rather than lower cognitive requirements. This might suggest that these people enjoy cognitive challenges and attribute relevance to them. Young people who pursue a lower secondary school leaving-certificate are less likely to aspire to occupations with higher rather than lower cognitive requirements, which could indicate that they anticipate their poorer chances and possible failure. For risk aversion and performance-related learning motivation we find no significant coefficients.



• Has my employer noticed my promotion? How employers and individuals report the requirement level in the German Classification of Occupations 2010
  Nadine Bachbauer* (IAB Nuremberg)
  Ann-Christin Bächmann (IAB Nuremberg)
  Basha Vicari (IAB Nuremberg)
  ≡ Abstract

  The German Classification of Occupations 2010 (KldB 2010) does not only provide information on occupational specialization, but also differentiates occupational units according to their requirement level into four groups ranging from (1) unskilled or semi-skilled activities to (4) highly complex activities (Paulus & Matthes 2013). The requirement level (5th digit of the KldB 2010 code) thus offers a new analytical potential to observe individual career advancement and hence address crucial issues of social inequality, e.g., by analyzing possible gender differences in career trajectories. To date, however, it is unclear whether the measure of the requirement level and possible changes over time are reliably reported. Depending on how occupational information is collected, different types of measurement errors may occur. For instance, survey data can be inaccurate and prone to coding errors. Administrative data can also be subject to errors, e.g., employer notifications might under-report changes in the annual occupational coding routine, including changes in the requirement level.
  In our paper, we analyze possible differences in occupational information and, in particular, the requirement level, on the basis of respondent and employer reports. For this purpose, we use the NEPS-SC6-ADIAB data that links survey data of the Adult Cohort of the German National Educational Panel Study with administrative data from the Integrated Employment Biographies (IEB) of the Institute for Employment Research.
  First comparative analyses, suggest only little differences in the reported requirement levels of the two data sources, with requirement levels tending to be higher in survey data. However, these differences show up only for men. The number of reported promotions differs only marginally between the data sources, although the distribution differs between the genders. In the administrative data, the number of promotions is higher for men, while it is balanced in the survey data. These findings raise a puzzle: How can these differences be explained? Do women tend to be promoted to an intermediate level, while men are mainly promoted to the highest level? To resolve this puzzle, we next examine the reporting differences in more detail, including relevant control variables, such as employer change. In a final step, we demonstrate how the differences in reporting affect substantive analyses of gender differences in promotion processes.



• Social contacts at the transition from apprenticeship training to work in Germany
  Markus Weißmann* (GESIS Mannheim, University of Mannheim)
  Tobias Roth (GESIS Mannheim)
  ≡ Abstract

  For many young adults, apprenticeship training in Germany marks the first step into the labour market, as around three in four graduates remain in the company after graduation (Bundesinstitut für Berufsbildung, 2024). However, graduates might also decide to look for and eventually move on to jobs in other companies. Social contacts can be helpful here, as they might give information about job positions or even work as referrers, thereby increasing the set of employment alternatives (Burt, 1992; Granovetter, 1995; Krug and Rebien, 2012; Lin, 2001; Weiss and Klein, 2011). In this regard, it has been argued that especially weak ties such as colleagues are helpful (Granovetter, 1995), while strong ties such as family contacts might be less helpful (Oesch and Von Ow, 2017).
  In our paper, we ask how assumed help from different types of social contacts influence the decision to search and apply for jobs and how actual help from these contacts influences the likelihood of being employed within or outside the training company.
  For our analyses, we rely on data from Starting Cohort 4 of the National Educational Panel Study (NEPS SC4; Blossfeld and Roßbach, 2019). We are using both, prospective and retrospective measures of information from several strong (e.g., family, friends) and weak (e.g., colleagues, teachers) ties about open job positions as well as their efforts to get training graduates such positions (i.e., referrals). Prospective information was measured in the last training year before graduation and represents the assumed availability of help from social contacts, while retrospective information was measured in the first interview after having graduated and represents the help that was actually mobilized.
  Results show that graduates that in their last year before training graduation stated that teachers would give them information about job positions or helped to get a job have a higher likelihood to have searched and applied for jobs once they graduated. In contrast, available help from workplace contacts decreases these likelihoods. The assumed availability of help from strong ties, like family or friends, did not influence job search or application behaviour. A different picture emerges when looking at the consequences of having mobilized social contacts for employment outcomes. While referrals from colleagues also increase the likelihood of remaining in the company, referrals from mothers, partners, and relatives outside the nuclear family decrease this likelihood. Information or referrals from teachers, however, are not significantly related to the likelihood of any employment outcome.

back to top

Session 6: (Early) Child Development

Friday | December 6, 2024 | 09:00 am - 10:40 am (CET)

• Maternal work trajectories and childcare histories: An in-depth exploration of the interplay and its consequences for child development
  Susanne Schmid* (University of Oldenburg, University of Bamberg)
  Hannah Sinja Steinberg (DJI Munich)
  ≡ Abstract

  In most industrialised countries, maternal employment has increased in recent decades, thereby transforming family life and children’s early childhood education environments affecting their development. To facilitate the reconciliation of work and family, mothers have been shown to adjust their careers to more flexible working conditions (Felfe, 2012). Simultaneously, the availability of alternative caregivers is a prerequisite for maternal employment but it is often limited. This impacts the choice of childcare arrangements. Despite a broad literature about the beneficial effects of institutional childcare supply on maternal employment and child development, little is known about (1) how maternal employment conditions are linked to the usage of different childcare types and combinations and (2) its consequences for child development.
  Drawing on the life-course perspective, we focus on the linked lives paradigm and explore the interplay between mother’s and children’s life domains (Elder et al., 2003). In accordance with Cunha and Heckman (2010), we expect that mother’s employment decisions form children’s learning environments and affect available resources, e.g., finances and time. The chosen childcare arrangement involves various learning opportunities, which differ between institutional childcare and individual care at home. We contribute by providing (1) novel evidence on the interplay between maternal employment histories and certain childcare arrangements from a trajectory-based perspective, rather than isolated events; and by (2) highlighting the impact of maternal employment on family dynamics which leads to disparities in child development.
  The study addresses this gap by examining the relation between childcare and maternal employment trajectories using multichannel sequence analysis based on the Newborn Starting Cohort of the German Educational Panel Study (NEPS-Network [LIfBi] et al., 2023; NEPS-SC1-ADIAB7521, 2012-2021, N=~1990). This survey provides monthly information on childcare usage from birth to school entry, along with annual measures of social-emotional and cognitive competences (math, vocabulary). Additionally, this dataset has been enriched with mother’s administrative employment records.
  Descriptive results indicate highest usage of parental childcare for children under one year of age, while the use of institutional childcare and its combination with grandparental care increases steadily as children grow older. At age one, approximately one-third of mothers are employed with this increasing proportions of part-time work, while full-time employment remains stable. Future stages of this study will include a joint sequence analysis of childcare and employment trajectories to emphasize their interdependence. Additionally, we will identify potential risk groups characterized by patterns that may have adverse consequences for child development.



• Transition from employment to un- and nonemployment – Effects on children‘s competences
  Laura Helbig* (IAB Nuremberg)
  Malte Reichelt (Friedrich Alexander University Erlangen-Nuremberg, IAB Nuremberg, LIfBi Bamberg)
  ≡ Abstract

  In a globalized and digitalized world individual competencies become increasingly important for one’s position in society. Research shows that these competencies are influenced by external factors such as schooling and peer relationships. A key factor, however, is provided by the quality and quantity of parental interaction. While in most industrialized countries, maternal employment has rapidly increased over the past decades, there is considerable heterogeneity in maternal work decisions, affecting childcare organization and time mothers spend with their children. Despite this, little is known about how maternal employment, unemployment, or nonemployment impacts children's early competencies, due to limited data.
  While unemployment and nonemployment increase time with children, unemployment is often linked to lower income and worsened mental health. Based on investment- and social capital-theory, as well as research on employment and mental health, we hypothesize that transitions to unemployment negatively affect children’s competencies, whereas transitions to nonemployment have an unclear impact.
  We use the Newborn-Cohort (SC1) of the National Educational Panel Study, linked with administrative records of parents (NEPS-ADIAB). We use waves 1 to 10 (children aged 7 months – 9 years) and restrict our sample to cases in which mothers have responded to the survey (n=1,898). Following Skopek and Passaretta (2021), we construct a composite index for children’s competencies, combining all available test scores across all waves. We identify all transitions from employment to unemployment (actively seeking employment) and to nonemployment.
  The findings, based on OLS models, indicate that maternal unemployment is associated with lower competencies in children, while nonemployment correlates with higher competencies. However, when controlling for unobserved individual factors using Fixed-Effects models, these differences disappear. This suggests that the relationship between employment status and competencies is not causal but driven by selection effects, where mothers of children with lower competencies are more often unemployed.
  Further analysis shows that transitions to nonemployment benefit children with below-average competencies, while transitions to unemployment harm high-skilled children. The results suggest that increasing parental interaction can be beneficial for those children, who can potentially gain the most: those with below-average competency-development. Maternal transitions to unemployment on the other hand have negative consequences only for high-skilled children, even after controlling for changes in income and hours in public childcare, supporting theories that assume negative effects on mental health and quality of parent-child interaction.



• Cognitive and non-cognitive skill formation in early childhood
  Philipp Kugler (IAW at University of Tübingen)
  Jan Simon Wiemann* (IAW at University of Tübingen)
  ≡ Abstract

  We analyse the development of children’s cognitive and non-cognitive skills from early childhood through primary school. Our research investigates how cognitive, i.e. literacy and numeracy, and non-cognitive skills, i.e. regulatory and social skills, as well as other factors in one period affect the skills in the subsequent period. This mechanism is relevant for the design of programmes aimed at supporting the skill formation of, for example, disadvantaged children. In addition, this research can inform efforts to improve kindergarden environments and school curricula by identifying critical and sensitive periods in child development, when interventions may have the most effective impact on skill acquisition.
  In our econometric approach, we assume that the skill generation process follows a dynamic CES production function (Cunha, Heckman & Schennach, 2010). The production functions links skills at t+1 to skills at t, parental investments, and initial parental skills. This allows to estimate input effects and input substitutability. To address potential endogeneity of parental investments, we use an instrumental variable approach. The model is stratified by gender to account for potentially heterogeneous skill trajectories between males and females. This feature is motivated by an extensive literature documenting gender-specific patterns in the acquisition of cognitive and non-cognitive skills during childhood.
  We use data from the National Educational Panel Study (SC1), which provides various competence tests and information on skills of the children to estimate the skill development in early childhood. To identify the true skill endowments, we assume that test scores are the result of underlying latent skill factors as in Attanasio et al. (2020).
  The estimated latent distributions of cognitive and non-cognitive skills for young boys and girls at period t=0 (between 6 and 17 months) and period t=4 (8 to 9 years) show that there are small differences between males and females for both cognitive and non-cognitive skills at period t=0. These differences appear to decrease by t=4. While this suggests a skill convergence over time our research interest lies in the movement of individuals on the distributions over time and the factors influencing the magnitude and direction of the movements. By applying the CES production function framework, we aim to identify the determinants of individual skill trajectories across developmental periods and provide insights into the dynamic process of skill formation in early childhood.



• Generation 2011 and Lectores – a qualitative longitudinal cohort for a national multidisciplinary project
  Sandra Fontanaud* (Université de Picardie Jules Verne Amiens)
  ≡ Abstract

  How can we get the most out of a longitudinal qualitative corpus without limiting its potential? How can successive teams of researchers use the data in the best possible conditions? How can we ensure that social scientists (sociologists, psychologists, linguists) can all draw on the same material for the data they need to support their research? I'd like to share with you the exciting methodological challenge that has been driving me for several years now: setting up a protocol for transcribing the video data collected as part of the 'Generation 2011' longitudinal cohort.
  Generation 2011 is a qualitative longitudinal sociological study of the different ways in which children grow up and find their place in society. From birth onwards, regular interviews are conducted with mothers and fathers. From the age of 2, the children themselves are involved in the study through games, whose sequences were filmed. The study focuses on 50 children born in 2011, the year in which the ELFE cohort was established. The ELFE cohort follows 18,320 children through family surveys (educational practices, games, care, schooling, psychomotor and language development) and school surveys in kindergarten and first grade (language and reading tests, mathematics tests, teaching materials, characteristics of teachers, class and school). The Generation 2011 team therefore has access to both statistics and much more informal information.
  This work proved to be much more complex than initially planned and rich in lessons. The video transcription protocol set up as part of Generation 2011 had to make it possible to transcribe the verbatim of children aged 2 and over during different interview sequences, to describe the actions taking place simultaneously on the screen, and to meet the needs of researchers and engineers from different disciplines, while taking into account the cost of transcriptions and the team's ability to recruit transcribers whose work met all these requirements. Since 2020, a multidisciplinary team of researchers and engineers has been testing the transcriptions resulting from this protocol as part of the Lectores project research, and we think it would be interesting to share this experience with the scientific community as a whole.

back to top

Session 7: Education and Work

Friday | December 6, 2024 | 11:00 am - 12:40 am (CET)

• Aligning aspirations: Exploring the relationship between work values and occupational fields
  Annalisa Schnitzler* (BIBB Bonn)
  Silvia Annen* (University of Bamberg)
  ≡ Abstract

  Besides factors like social expectations, requirements in the VET system and market conditions (Ahrens et al., 2021), adolescents’ occupational choice is determined by their interests and values (Holland, 1997). Mortimer and Lorence (1979) differentiate two central value dimensions: extrinsic values, which concern the rewards derived from the job but external to the work itself (e.g., income, prestige, security) and intrinsic values, which involve rewards obtained directly from the work experience (e.g., interest, responsibility, autonomy). Research also shows that work values undergo changes due to transitioning to work (Lechner et al., 2017) and in response to value-related rewards that occupations offer (Johnson, 2001).
  Various empirical research studies reveal differences in reward potential for these work values between different occupations. In terms of income, the area of technical research is at the forefront (Tiedemann & Werner, 2024), regarding occupational prestige occupations from, for example, technical research, development, design and production control (Ebner & Rohrbach-Schmidt, 2021). When, in times of digitalisation, operationalising job security through occupations’ substitution potential, skilled workers in chemical and pharmaceutical engineering and mechatronics are particularly at risk (Grienberger et al., 2020). Concerning autonomy, knowledge-related activities, in contrast to production and service activities, seem to go hand in hand with digital self-determination (Kirchner, Meyer & Tisch, 2020).
  In this paper, we seek to investigate whether adolescents’ work values mirror these differences in the reward structure of different occupational fields by focussing on different work values simultaneously and taking a more differentiated look at their significance for the choice of certain occupational areas. Specifically, we look at the question of which work values are relevant in which occupational fields. We resort to data from wave 3 and 9 of NEPS Starting Cohort 4 and analyse adolescents‘ ratings of the personal importance they attribute to a number of work values. For pupils in grade 10 we link their work values to their idealistic occupational aspiration grouped into 12 occupational fields (see Tiemann et al., 2008). In wave 9, we focus on those participants who have or will have successfully completed a dual or school-based apprenticeship. We use the re-assessment of the importance of work values and link the differences in work values between wave 3 and 9 to the occupational field of the successfully completed apprenticeship.
  Chi-square-tests show differences according to occupational field for both initial work values and the rise or fall in importance after having entered an occupational field.



• Dynamics in adults’ digital skills over the life course: The impact of further training and skill use on ICT skills
  Martin Ehlert* (WZB Berlin, FU Berlin)
  ≡ Abstract

  Against the background of technological innovations, digital skills are becoming increasingly important for workers’ employability as well as citizens overall inclusion in society. Yet, many adults did not learn the skills to master information and communication technology (ICT) sufficiently during initial education. To keep up with changes, adults need to learn and update these skills later in life. The aim of this study is to assess the dynamics of ICT skills over the life course.
  This study focusses on two research questions: (1) How do adults’ ICT skills develop over the life course? (2) What are the effects of ICT skill use and further education on ICT skill acquisition during adulthood?
  I examine two explanations for ICT skill formation during adulthood. First, based on human capital theory, participation in further education and training on ICT tropics may increase ICT skills. Second, practice engagement theory by Reder (1994) claims that ICT skill growth due to ICT practices. The basic idea is that engagement with activities that make use of skills in different domains leads to higher skills. The opposite of the coin is the “use-it-or-lose-it” hypothesis which states that workers lose skills if they do not apply them at work.
  I use data from the German National Educational Panel Study (NEPS), Starting Cohort (SC) 6 (adults). ICT skills (“computer literacy”) were measured in wave 5 and 14 using competence tests. In addition to ICT skills, the data provides information about participation in a wide range of adult learning activities and the topics covered. This allows for the identification of learning activities focused on ICT topics. Also, ICT use on the job as well as a wide range of socio-demographic variables are surveyed.
  Preliminary analyses based on the linked WLE scores reveal that ICT skills decreased on average among individuals over the nine years covered by the data. This decline in skills increases with age, with particularly high losses among people aged 65 and above. Among those who took at least one ICT-related non-formal course between the measurement points, the decline is less pronounced. This finding is corroborated by multivariate models indicating a sizeable positive association between ICT training and ICT skills, even after controlling for initial ICT skills and a set of socio-demographic controls. ICT use at work also proved to be significantly correlated with increases in ICT skills.



• Gendered intergenerational educational mobility patterns converge in the cohort sequence: Evidence from Switzerland using administrative data
  Rolf Becker* (University of Bern)
  Richard Nennstiel (University of Bern)
  ≡ Abstract

  In many societies, educational attainment determines social inequality in terms of life chances, and at the same time there is a strong link between social origin and educational success. Therefore, analysis of educational mobility patterns is a central concern for sociologists which seeks to describe and explain the pattern of educational expansion which takes place across historical periods and birth cohorts. In the context of societal changes, such as trend of modernization, educational expansion and significantly increased female participation in education, we use administrative data from different sources (N=556,112) – such as census data and register-based administrative surveys – to examine the extent to which absolute and relative intergenerational educational mobility has changed in Switzerland for women and men from the 1951–1990 birth cohorts.
  We show that there is significantly more upward than downward mobility, while a large proportion of individuals are laterally mobile. By looking at absolute mobility patterns by cohort and gender separately, we extend previous research and show that the decreasing absolute mobility rates are due to the changing educational composition of the parental generations. Following on from previous studies, we reveal that the observed trend toward less relative social mobility continues in the youngest cohorts. It is also worth noting that, while the father’s educational attainment has a higher predictive power for children’s education in all cohorts, the impact of the mother’s education approaches that of the father.
  Overall, the mobility patterns of men and women converge very strongly over the cohort sequence. Beyond these substantive points, our study demonstrates the potential of using administrative data for social stratification research.



• Persistent gender inequalities in the Swiss labor market: A study of educational credentials and their returns
  Richard Nennstiel* (University of Bern)
  Rolf Becker (University of Bern)
  ≡ Abstract

  Switzerland is a country that has experienced a comparatively slow educational expansion compared to other industrialized countries and has a comparatively high level of gender inequality within the labor market. Nevertheless, it is also apparent in Switzerland that women have recently overtaken men in achieving higher educational credentials and that the occupational structure has changed due to the increasing tertiarization of the labor market.
  Given that there is also a very strong link between educational credentials and labor market success in Switzerland, we examine the extent to which returns to education – measured as labor market participation (full-time, part-time, or inactive) and attained social class (using the OESCH class scheme) – have changed within and between genders from 1970 to 2020 in Switzerland. To trace changes over this historical period, we use administrative full censuses (Volkszählung 1970, 1980, 1990, and 2000) and administrative large-scale surveys (Strukturerhebungen 2011-2020).
  We can show that there has been a significant equalization of the gender gap in terms of returns to education in labor market participation. This is mainly because women have been able to significantly increase their likelihood of working part-time across all educational levels. Nevertheless, when comparing women and men with the same educational credentials, women still work significantly less full-time than men. Gender inequalities in returns to education in terms of the social class attained have also decreased. This is mainly due to the increased labor force participation of women and the structural transformation of the labor market. Women with the same educational background as men are more likely to work in the lower-grade service class, while persistent inequalities continue to exist to the disadvantage of women in terms of access to higher-grade service class jobs.

back to top

Session 8: School-Related (Mis)Perceptions

Friday | December 6, 2024 | 11:00 am - 12:40 am (CET)

back to top

• Special needs and disability pathways of children since birth to 10 years old in France: Data from the Elfe Cohort
  Mélissa Arneton* (INED-INSERM, INSEI-Grhapes Paris)
  Laurence Germany (INED-INSERM Paris)
  ≡ Abstract

  Inclusive approach in education can be approach with longitudinal data collected in a life-span perspective. However, several definitions of situations of disability co-exist since adoptions of the International Convention on the Rights of Persons with Disabilities and national laws. Functional disability based on medical diagnoses but also social definition focus on barriers and limitations to the development of people. Even if special educational needs are not exactly the same as deficiencies (Shurr, Minuk and Chahine, 2024), it is important to document and identify such variables during data collection in order to contribute to a better comprehension of development of all children. In the cohort Elfe (French longitudinal study since childhood), several variables are constructed based on main approaches of disability: special needs and their evaluation by parents and teachers in a psychosocial approach, deficiencies related to a medical approach and accommodations based on an administrative approach. Methodological limits and interests of each approach are presented also as their empirical links.
  ELFE investigates individual competences and educational trajectories in a longitudinal study design, in continental France. It is the first nationwide scientific study of children, tracking them from birth to adulthood, in a multidisciplinary perspective: medical, environmental, social and psychological (Charles and al., 2019). Initially sampled in 2011 on a representative base (N=18,329), children and their families received follow-up interviews and competence tests. Important context persons (i.e., teachers, physicians) were included in the assessment design. As of yet, a total of 10 measurement waves is available.
  Comparatively to the NEPS cohort, which makes a focus on SEN-learning in order to create a specific panel (Heydrich and al., 2013), the French cohort provides incidental data in general population like in Millennium Cohort Study (Parsons and Platt, 2017). Each conceptualization of SEN gives different information about the child and his or her environment. At 1 year, 0.79% of parents introduce an administrative procedure for disability with MDPH, they are 2.18% at 10. Even if, during the first year of obligatory schooling (age 6), 27.20% students are considered by their teacher to present special needs, school accommodations concern 18.76%. At age 9, teachers consider that SEN concern 15.93% of children whereas parents perceive limitations for their child in 2.10% (age 10). The longitudinal nature of the data provides rich opportunities for research on the development children, particularly to study interrelations of disability situations, perceptions of their needs and development of their competencies.



• Parental misperceptions and investments: Does information provided by teachers play a role?
  Elena Ziege* (BiB Wiesbaden, Johannes Gutenberg University Mainz)
  Ariel Kalil* (University of Chicago)
  ≡ Abstract

  Parents’ knowledge and beliefs about their child’s academic achievement shapes parents’ decisions about investing in their child’s education (e.g., Francesconi & Heckman, 2016). The standard economic model assumes that parents have perfect information about their children’s true performance when they make decisions to invest in children’s learning. In fact, many parents misperceive their child’s academic progress. This can be seen in the large share of parents who overestimate their children’s performance (e.g., Dizon-Ross, 2019). This is important because parents who overestimate their children’s performance spend less time in learning activities with their children (Kinsler & Pavan, 2021). Parents’ imperfect information can thus increase educational inequalities.
  A small number of experimental and quasi-experimental research in different countries shows that resolving parental misbeliefs about students’ test scores, grades and school-related activities such as absences and completed homework can increase parental investment, change child behavior, and improve child skill (see Barrera-Osorio et al., 2020; Bergman, 2021; Bergman & Chan, 2021; Cobb-Clark et al., 2021; De Walque & Valente, 2023; Dizon-Ross, 2019; Gan, 2021; Kraft & Rogers, 2015; Robinson et al., 2018).
  Our study contributes to this literature by examining how information provided by teachers on a child's skills influences parental misperceptions and investment decisions in Germany. To conduct our analysis, we leverage the variation in the timing of report card distribution across different German federal states and the timing of surveys conducted with parents of first-grade students around the same time they receive their child's first-grade report cards. This quasi-experimental variation in the timing of information allows us to compare parents' perceptions of their child's skills before and after receiving the teacher's evaluation.
  Using a sample of 1,300 parents surveyed in the third wave of the National Educational Panel Study (NEPS) Starting Cohort 2, we confirm previous findings on parental overestimation. 40 to 46 percent of parents overestimate their child’s vocabulary, science and math skills. We show that underestimation also exists but to a smaller degree: one quarter of parents underestimate their child’s science skills one fifth underestimates their child’s language and math skills. This tendency is particularly pronounced among migrant parents and parents without a bachelor's degree. Our preliminary results indicate that providing parents with information about their child's academic performance reduces their misconceptions. The percentage of parents who overestimate their child’s abilities is reduced by half when parents receive information from the child’s school. There is no reduction in parental underestimation.



• Bridging the gap: The role of parent-teacher perception in child cognitive outcomes
  McKayla Jensen* (Brigham Young University)
  Mikaela Dufur (Brigham Young University)
  Jonathan Jarvis (Brigham Young University)
  Shana L. Pribesh (Old Dominion University)
  ≡ Abstract

  How can a young child’s behavior impact their cognitive development? As young children learn and grow, they are exposed to new ideas, skills, and personalities. Among the earliest settings for child cognitive development outside the home are pre-schooling/daycare facilities. A child’s early social and academic development is impacted by the quality of encouragement and examples put forth by parents and educators. Both authority figures employ different methods to aid young children in facilitating growth.
  Although home and daycare settings are vastly different, what happens when parents and educators perceive children’s behavior differently? Do those differing views translate into diminished cognitive development and abilities? This research seeks to understand how differences in parent and teacher opinions on a child’s different behavioral traits affect various realms of a child’s cognitive abilities outside the home.
  To study this question, we draw upon data from waves 1 and 2 of Starting Cohort 2 (Kindergarten) of the National Educational Panel Study (NEPS). Our analysis incorporates survey responses from both the target children’s parents and daycare educators, offering insights into the children’s behaviors and social tendencies from both adult perspectives.
  Past research has shown that for teachers, a more positive relationship with a child’s parents often yields more favorable reports on behavior. For parents, the desire to be productively involved in a child’s development and positive behavior may be affected by parental self-efficacy and mental health struggles. Research has concluded that when a child’s behavioral issues are not acknowledged and addressed, they can cause negative repercussions throughout the child’s life.
  Our study took the perceptions of teachers and parents on ten behavioral traits they feel the child displayed in daycare and home settings respectively. We found that the more those perceptions differed, the greater the negative impact was on the child’s social, concentration, linguistic, mathematical, and knowledge cognitive abilities compared to their peers. This was especially true for differences in the ratings for knowledge and understanding as perceived by the parents and teachers, where four or all five cognitive abilities measured respectively were negatively affected. Additionally, differing opinions on the neatness of a child, surprisingly, lead to greater ratings of cognitive abilities in all five categories.
  Understanding the impacts of parent and teacher congruence or discordance in recognizing their children’s behaviors can aid cognitive development as those children continue to grow and progress to adolescence and adulthood.



• Beliefs about teaching, learning and self-efficacy: Impact on perceived teaching characteristics in heterogeneous classrooms
  Andrea Vorderobermeier* (University of Passau)
  Maximilian Sailer (University of Passau)
  ≡ Abstract

  It has been extensively demonstrated that teachers´ beliefs of teaching and learning as well as motivational beliefs have a significant impact on a range of factors, including student reported teaching characteristics. The objective of this article is to examine whether these findings are applicable to heterogeneous class structures. The study was conducted over the course of a two-weeks summer school in Austria (N = 532 teachers and N = 3,517 students). The students were taught in small groups across different classes and school levels. Given that beliefs can be activated by specific contextual conditions, the objective is to elucidate which beliefs can be identified in the teachers. Furthermore, the main goal is to ascertain whether these beliefs impact the perception of the teaching characteristics instructional strategies, classroom management, student motivation, clarity and structure, challenging practice, individualization and differentiation.
  Our research questions are (1) What beliefs about teaching and learning in the dimensions of transmission, construction, and student orientation, as well as motivational beliefs (self-efficacy), do teachers at the summer school hold? and (2) Do the teachers´ beliefs affect how students perceive the characteristics of the classroom in heterogeneous group settings?
  The data was collected from teachers (teachers, N = 377; student teachers, N = 155) and their pupils (M age = 12.75; SD = 1.58; M class size = 10.39) via an online questionnaire as part of an evaluation of the summer school program. A multi-level analysis was considered based on the data structure and theoretical considerations. However, it was decided that this is not necessary due to weak hierarchization and minor differences between the groups. In response to FF1, a series of t-tests and ANOVAs were conducted. In order to represent the latent constructs of interest at both the student and teacher level, and to analyze their relationship (FF2), structural equation modelling is employed on a separate basis for student teachers and teachers.
  The results indicate that self-efficacy has a weak positive effect on some teaching characteristics (between ß = .12 and .22), as well as constructional beliefs (between ß = .14 and .17). Transmissive beliefs have a modest negative influence (between ß=-.14 and ß = -.28), as well as student orientation (between ß = -.12 and -.17). With regards to the students, only transmissive beliefs are significant, with weak positive effects (between ß = .20 and ß = .25).
  The study demonstrated that even in a diverse setting, teachers´ beliefs have a small but significant impact on the perception of specific teaching characteristics.