Record Nr.

UNINA9910155440103321

Titolo

Patterns and impact of longitudinal measurement error for welfare receipt [[electronic resource]] / Institut für Arbeitsmarkt- und Berufsforschung (IAB) der Bundesagentur für Arbeit, Johannes Eggs

Pubbl/distr/stampa


Bielefeld, : wbv Publikation, 2016

ISBN

3-7639-4112-6

Edizione

[1st ed.]

Descrizione fisica

1 online resource (138 pages) : illustrations, tables

Collana

IAB-Bibliothek (Dissertationen) ; 362

Disciplina

511.43

Soggetti

Arbeitslosengeld II

Dissertation

Längsschnittuntersuchung

Messfehler

Messverfahren

Panel

Persönlichkeitsmerkmale

Statistik

Arbeits- und Industriesoziologie

Lingua di pubblicazione

Inglese

Formato

Materiale a stampa

Livello bibliografico

Monografia

Nota di bibliografia

Includes bibliographical references.

Nota di contenuto

Vorwort 1 Introduction 1.1. The Total Survey Error framework 1.2. The longitudinal perspective 1.3. Measurement error 1.4. Welfare receipt 1.5. Measurement error for welfare receipt 1.6. Data 1.7. The studies 2 Will respondents eventually get it right? Changes in measurement error in a panel survey using dependent interviewing: Results from a five-wave validation study (Johannes Eggs, Annette Jäckle and Mark Trappmann) 2.1. Introduction 2.2. The panel survey and validation data 2.3. Results 2.4. Discussion 3 Measurement error for welfare receipt and its impact on panel models 3.1. Introduction 3.2. Data 3.3. Measurement error 3.4. Measurement error and fixed-effects models 3.5. Discussion & conclusion 4 Errors in retrospective welfare reports and their effect on event history analysis (Johannes Eggs and Rainer Schnell) 4.1. Introduction 4.2. Previous research 4.3. Data 4.4. Hypotheses 4.5. Methods 4.6.

Results 4.7. Discussion 5 Dependent interviewing and sub-optimal responding (Johannes Eggs and Annette Jäckle) 5.1. Introduction 5.2. Theoretical background on false confirmation 5.3. Previous studies on false confirmation and misreporting of benefits 5.4. The panel survey and validation data 5.5. Predictors of sub-optimal responding 5.6. Results 5.7. Discussion 6 Concluding remarks 6.1. Further research Bibliography Appendix Abstract Kurzfassung

Sommario/riassunto

Diese Arbeit beschäftigt sich mit Messfehlern in Längsschnittdaten. Messfehler können in besonderem Maße die Messung von Übergängen und Veränderungen über die Zeit beeinflussen. Die Messung von Veränderungen ist jedoch einer der Hauptgründe für das Erheben von Längsschnittdaten. Allerdings werden Messfehler in Längsschnittdaten selten analysiert. Durch die Verknüpfung von Paneldaten mit Registerdaten auf der individuellen Ebene werden in dieser Arbeit Messfehler für den Bezug von Arbeitslosengeld II für fünf aufeinanderfolgende Panelwellen untersucht. Dabei zeigt sich, dass die Messfehler für den Bezug nicht zufällig verteilt sind, sondern mit der Zeit und persönlichen Charakteristiken korrelieren. Measurement error is a common phenomenon in the empirical sciences. Longitudinal data can especially be affected by it, as measurement error can influence measures of change, which is one of the primary reasons for collecting longitudinal data in panel surveys. However, measurement error in longitudinal data is rarely analysed. In this series of papers, the measurement error for welfare receipt is analysed for up to five consecutive panel waves by linking panel survey data with administrative data on the individual level. Results from all four studies suggest that measurement error for welfare receipt is highly differential. The measurement error for welfare receipt is not based on a random process, but based on personal characteristics and the welfare histories of the respondents.