Sabine Zinn

Mic-core: a tool for microsimulation

Microsimulation is an increasingly popular tool in the social sciences. Individual behavior is described by a (commonly stochastic) model and subsequently simulated to study outcomes on the aggregate level. Demographic projections are a prominent area of application. Despite numerous available tools often new software is designed and implemented for specific applications. In this paper we describe how a modeling and simulation framework, JAMES II, was used to create a specialized tool for population projections, the Mic-core. Reusing validated and well-tested modeling and simulation functionality significantly reduced development time while keeping performance levels high. We document how the Mic-core was built as plug-ins to JAMES II and illustrate the performance of the resulting tool. We demonstrate how the concept of a modeling and simulation framework enabled successful software reuse of available functionality and briefly report of future work.

Migranten und ihre Nachkommen im deutschen Bildungssystem: Ein aktueller Überblick

In diesem Kapitel wird ein Uberblick uber die aktuelle Situation von Zuwanderern und deren Nachkommen im deutschen Bildungssystem gegeben. Anhand der Daten des Nationalen Bildungspanels (NEPS) kann gezeigt werden, dass sich das vergangene und gegenwartige Migrationsgeschehen in der Zusammensetzung der verschiedenen Bildungsetappen nach Generationenstatus und Herkunftsgruppe widerspiegelt. Mit Blick auf den Bildungserwerb zeichnen sich nach wie vor ausgepragte Unterschiede zwischen Personen mit und ohne Migrationshintergrund ab. So weist die zugewanderte Bevolkerung uber alle Bildungsetappen hinweg ein geringeres sprachliches Niveau als die einheimische Referenzgruppe auf und besucht vergleichsweise haufiger die weniger prestigetrachtigen Bildungszweige der Sekundarstufe I. Die Befunde legen eine differenzierte Betrachtung unter anderem nach Generationenstatus und Herkunftsgruppe nahe. Zusammengefasste Betrachtungen, die lediglich nach Migrationshintergrund oder nach Staatsangehorigkeit unterscheiden, greifen hier zu kurz.

A DEVS model for demographic microsimulation

Microsimulation is increasingly applied in demography to project the development of populations. A stochastic model is being introduced that describes individual life courses on a continuous time base. Life courses are determined by sequences of demographic events. We show how this demographic multi-state projection model can be specified as an atomic model in DEVS. Thereby, strengths but also limitations of this approach are revealed. The limitations are addressed by exploiting a DEVS variant that supports variable structures, i.e., DYNPDEVS, and by modeling individuals as atomic models and the population as a network model. An example projection of a synthetical population based on the population of Italy shows the plausibility and the feasibility of the developed model.

Multistate modelling extended by behavioural rules: An application to migration

We propose to extend demographic multistate models by adding a behavioural element: behavioural rules explain intentions and thus transitions. Our framework is inspired by the Theory of Planned Behaviour. We exemplify our approach with a model of migration from Senegal to France. Model parameters are determined using empirical data where available. Parameters for which no empirical correspondence exists are determined by calibration. Age- and period-specific migration rates are used for model validation. Our approach adds to the toolkit of demographic projection by allowing for shocks and social influence, which alter behaviour in non-linear ways, while sticking to the general framework of multistate modelling. Our simulations yield that higher income growth in Senegal leads to higher emigration rates in the medium term, while a decrease in fertility yields lower emigration rates.

Simulating Synthetic Life Courses of Individuals and Couples, and Mate Matching

We present a novel microsimulation approach which is enriched by elements from agent-based modeling. Concretely, we have designed a simulation model and software which facilitate describing and simulating life courses of individuals and couples and conducting mate matching. To define individual and couple behavior we use a continuous-time multi-state model, that is, we use a continuous-time microsimulation model. For mate matching we apply agent-based modeling: first, to each individual who is seeking for a partner a random value is assigned that captures his or her aspiration level regarding the fit with a potential partner. Then, via an empirical likelihood equation we assess the probability that a given woman and a given man would mate. Thereafter, we simulate a decision making process whether two individuals form a couple applying individual aspiration levels and their mating probability. Our description puts into perspective the differences between microsimulation and agent-based modeling, their relative strength, and limitations. We use the ml-DEVS formalism to specify the novel simulation approach and the modeling and simulation framework JAMES II to implement it. Projecting a hypothesized population based on the population of the Netherlands serves to illustrate its potential.

A statistical approach to address the problem of heaping in self-reported income data

Self-reported income information particularly suffers from an intentional coarsening of the data, which is called heaping or rounding. If it does not occur completely at random – which is usually the case – heaping and rounding have detrimental effects on the results of statistical analysis. Conventional statistical methods do not consider this kind of reporting bias, and thus might produce invalid inference. We describe a novel statistical modeling approach that allows us to deal with self-reported heaped income data in an adequate and flexible way. We suggest modeling heaping mechanisms and the true underlying model in combination. To describe the true net income distribution, we use the zero-inflated log-normal distribution. Heaping points are identified from the data by applying a heuristic procedure comparing a hypothetical income distribution and the empirical one. To determine heaping behavior, we employ two distinct models: either we assume piecewise constant heaping probabilities, or heaping probabilities are considered to increase steadily with proximity to a heaping point. We validate our approach by some examples. To illustrate the capacity of the proposed method, we conduct a case study using income data from the German National Educational Panel Study.

The role of languages for modeling and simulating continuous-time multi-level models in demography

Demographic microsimulation often focuses on effects of stable macro constraints on isolated individual life course decisions rather than on effects of inter-individual interaction or macro-micro links. To change this, modeling and simulation have to face various challenges. A modeling language is required allowing a compact, succinct, and declarative description of demographic multi-level models. To clarify how such a modeling language could look like and to reveal essential features, an existing demographic multi-level model, i.e., the linked life model, will be realized in three different modeling approaches, i.e., ML-DEVS, ML-Rules, and attributed pi. The pros and cons of these approaches will be discussed and further requirements for the envisioned language identified. Not only for modeling but also for experimenting languages can play an important role in facilitating the specification, generation, and reproduction of experiments, which will be illuminated by defining experiments in the experiment specification language SESSL.

Modeling competence development in the presence of selection bias

A major challenge for representative longitudinal studies is panel attrition, because some respondents refuse to continue participating across all measurement waves. Depending on the nature of this selection process, statistical inferences based on the observed sample can be biased. Therefore, statistical analyses need to consider a missing-data mechanism. Because each missing-data model hinges on frequently untestable assumptions, sensitivity analyses are indispensable to gauging the robustness of statistical inferences. This article highlights contemporary approaches for applied researchers to acknowledge missing data in longitudinal, multilevel modeling and shows how sensitivity analyses can guide their interpretation. Using a representative sample of N = 13,417 German students, the development of mathematical competence across three years was examined by contrasting seven missing-data models, including listwise deletion, full-information maximum likelihood estimation, inverse probability weighting, multiple imputation, selection models, and pattern mixture models. These analyses identified strong selection effects related to various individual and context factors. Comparative analyses revealed that inverse probability weighting performed rather poorly in growth curve modeling. Moreover, school-specific effects should be acknowledged in missing-data models for educational data. Finally, we demonstrated how sensitivity analyses can be used to gauge the robustness of the identified effects.

Emigration Rates From Sample Surveys: An Application to Senegal

What is the emigration rate of a country, and how reliable is that figure? Answering these questions is not at all straightforward. Most data on international migration are census data on foreign-born population. These migrant stock data describe the immigrant population in destination countries but offer limited information on the rate at which people leave their country of origin. The emigration rate depends on the number leaving in a given period and the population at risk of leaving, weighted by the duration at risk. Emigration surveys provide a useful data source for estimating emigration rates, provided that the estimation method accounts for sample design. In this study, emigration rates and confidence intervals are estimated from a sample survey of households in the Dakar region in Senegal, which was part of the Migration between Africa and Europe survey. The sample was a stratified two-stage sample with oversampling of households with members abroad or return migrants. A combination of methods of survival analysis (time-to-event data) and replication variance estimation (bootstrapping) yields emigration rates and design-consistent confidence intervals that are representative for the study population.

Weighting Panel Cohorts in Institutional Contexts

The National Educational Panel Study (NEPS) surveys and tests, next to adults, undergraduates, and newborns, Kindergarten children and students within their institutional contexts. Individuals who decided to participate in the panel study can refuse participation in specific waves or drop out completely. Weighting adjustments are usually applied to account for nonparticipation. Within the institutional cohorts of the NEPS, these adjustments take clustering at the institutional level into account. In NEPS, information on children is enriched by interviews with their parents. Thus, dealing with two distinct but possibly interdependent participation decisions has to be regarded by a joint modeling approach. The results of models analyzing the participation propensity provide insights concerning factors influencing the participation probability. In general, few potential determinants affect participation decisions. These include place of residence, language spoken at home, age, and having missing values in personal or migration-related characteristics. For later waves the participation status of the previous wave has proved to be a good predictor. Moreover, being surveyed and tested within the institutional context positively influences participation decisions.