Small Area Methods and Administrative Data Integration
In: Analysis of Poverty Data by Small Area Estimation, S. 61-82
9 Ergebnisse
Sortierung:
In: Analysis of Poverty Data by Small Area Estimation, S. 61-82
Welfare systems can be observed according to two different perspectives. The former deals with the supply of social protection, i.e. with the funding and provision of social benefits and the production of social services and goods. The latter focuses on the demand of social protection, and particularly on the characteristics of people benefiting from social protection or asking for it. Typically, data on the supply of social benefits have an administrative nature (registers and budgets data) whereas data on beneficiaries come from sample surveys. In theory, administrative data, being census data, can be detailed by territory. On the contrary, sample surveys are usually planned to provide accurate estimates at the national level or for large sub-national areas. This chapter provides an example on the use of different data sets for the Old age and Family/children functions at the province level (LAU 1 in the EU nomenclature). Data on the supply of benefits derive from the SISSIM (Istat Survey on Interventions and Social Services of Individual and associated Municipalities) and from municipalities' budgets. Data on the demand of social protection come from EU-SILC (European Union - Statistics on Income and Living Conditions), a survey that is annually conducted by Istat in a comparable European framework. Earned benefits are estimated applying small area estimation methods, given that the sample size of the EU-SILC survey at the province level is small, so the traditional design-based estimators usually are unreliable. Results are analysed to understand whether administrative and sample survey data can be used to to compose a coherent picture of social protection delivered at the provincial level.
BASE
In: Human Development in Times of Crisis, S. 79-106
In: Socio-economic planning sciences: the international journal of public sector decision-making, Band 91, S. 101794
ISSN: 0038-0121
In: Socio-economic planning sciences: the international journal of public sector decision-making, Band 82, S. 101327
ISSN: 0038-0121
There has been rising interest in research on poverty mapping over the last decade, with the European Union proposing a core of statistical indicators on poverty commonly known as Laeken Indicators. They include the incidence and the intensity of poverty for a set of domains (e.g. young people, unemployed people). The EU-SILC (European Union - Statistics on Income and Living Conditions) survey represents the most important source of information to estimate these poverty indicators at national or regional level (NUTS 1-2 level). However, local policy makers also require statistics on poverty and living conditions at lower geographical/domain levels, but estimating poverty indicators directly from EU-SILC for these domains often leads to inaccurate estimates. To overcome this problem there are two main strategies: i. increasing the sample size of EU-SILC so that direct estimates become reliable and ii. resort to small area estimation techniques. In this paper we compare these two alternatives: with the availability of an oversampling of the EU-SILC survey for the province of Pisa, obtained as a side result of the SAMPLE project (Small Area Methods for Poverty and Living Conditions, http://www.sample-project.eu/), we can compute reliable direct estimates that can be compared to small area estimates computed under the M-quantile approach. Results show that the M-quantile small area estimates are comparable in terms of efficiency and precision to direct estimates using oversample data. Moreover, considering the oversample estimates as a benchmark, we show how direct estimates computed without the oversample have larger errors as well as larger estimated mean squared errors than corresponding M-quantile estimates.
BASE
In: Survey research methods: SRM, Band 6, Heft 3, S. 155-163
ISSN: 1864-3361
"There has been rising interest in research on poverty mapping over the last decade, with the European Union proposing a core of statistical indicators on poverty commonly known as Laeken Indicators. They include the incidence and the intensity of poverty for a set of domains (e.g. young people, unemployed people). The EU-SILC (European Union - Statistics on Income and Living Conditions) survey represents the most important source of information to estimate these poverty indicators at national or regional level (NUTS 1-2 level). However, local policy makers also require statistics on poverty and living conditions at lower geographical/domain levels, but estimating poverty indicators directly from EU-SILC for these domains often leads to inaccurate estimates. To overcome this problem there are two main strategies: i. increasing the sample size of EU-SILC so that direct estimates become reliable and ii. resort to small area estimation techniques. In this paper the authors compare these two alternatives: with the availability of an oversampling of the EU-SILC survey for the province of Pisa, obtained as a side result of the SAMPLE project (Small Area Methods for Poverty and Living Conditions, http://www.sample-project.eu/ ), they can compute reliable direct estimates that can be compared to small area estimates computed under the M-quantile approach. Results show that the M-quantile small area estimates are comparable in terms of efficiency and precision to direct estimates using oversample data. Moreover, considering the oversample estimates as a benchmark, they show how direct estimates computed without the oversample have larger errors as well as larger estimated mean squared errors than corresponding M-quantile estimates." (author's abstract)
The aim of this paper is twofold: first it shows how the identification of seven vulnerable labour market groups in the 2018 European Union Labour Force Survey (EU-LFS) is possible. These groups include age, gender identity, sexual orientation, single parenthood, migration (ethnicity, nationality, and migration status), religion, and disability. Second, it presents a study on how statistically reliable indicators can be obtained for a selection of those identified vulnerable groups. For the first identification part, the exercise showed that the sample size of the vulnerable group is largely dependent on the operationalisation utilised. Several of the vulnerable groups included straightforward definitions that are widely agreed on and correspond well with items in the EU-LFS. For other groups, such as disability and migrant status, the sample size varied widely based on operationalisation used to identify respondents. Age and gender identity all provide for straightforward identification with sizeable sample sizes. Sexual orientation was limited to same-sex couples living in the same household, which faced additional restrictions like anonymisation and limited detailed household data. Identification of single parenthood depended on the age cut off for children in the household, as the EU-LFS defines a dependent child in a way that deviates from research norms. Identifying disability and migrant status also provided difficult as there is not a single operationalisation for either and identification has to be done indirectly. There was no measurement for religion. Additionally, we find issues with removing duplicate responses from the EU-LFS as repeat sampling varies at the country level and there are no consistent identifiers for both household and individuals in all countries In the second part, the use of Small Area Estimation methods was proved to be useful for obtaining reliable estimates of some selected vulnerable groups indicators based on the EU-LFS 2018 data.
BASE
Il presente lavoro, ultimato a inizio luglio 2020, e' il frutto della collaborazione di un gruppo di docenti e ricercatori che operano nei Dipartimenti di Economia e Management, Giurisprudenza e Scienze Politiche dell'Universita' di Pisa. Nato sulla base di una richiesta della Prefettura di Pisa all'Ateneo pisano in piena emergenza sanitaria, esso si propone lo scopo di fornire un'analisi degli effetti che l'emergenza COVID-19 ha avuto sul tessuto economico e sociale della provincia di Pisa e, alla luce dei risultati ottenuti, effettuare alcune "proposte per la ripartenza" per i prossimi mesi. Il convincimento degli autori che questo contributo di analisi e proposte, ancorche' riguardante la realta' territoriale della provincia di Pisa, possa avere una qualche utilita' anche per altre realta' provinciali e regionali, nonche' per quella nazionale. La ragione di tale convinzione e' duplice. Da un lato, l'approccio utilizzato, basato sulla multidisciplinarieta' e sul coinvolgimento delle realtà socio-economiche e istituzionali del territorio, rappresenta un metodo essenziale e generale per la piena comprensione di una realta' nuova e assai complessa quale quella derivante dall'emergenza COVID-19. Gli autori, provenienti da settori scientifici diversi quali l'ambito aziendale, economico, statistico, giuridico e sociolopsicologico, sono stati i primi a rendersi conto di quanto tale metodo di "messa a sistema" delle informazioni e degli attori economici e istituzionali della provincia fosse cruciale, ancorche' inusuale rispetto al carattere tipicamente specialistico delle ricerche in ambito accademico. Dall'altro lato, le proposte contenute nel lavoro, e che sono riportate in modo sintetico val termine di questa introduzione, sono il frutto dell'analisi quantitativa e qualitativa contenuta nei primi capitoli e rappresentano un esempio di come le scienze sociali possano fornire una base informativa essenziale per processi decisionali basati sui fatti (quelli che in ambito scientifico vengono definiti "evidenze empiriche"). In altre parole, le proposte hanno valenza generale, in quanto mettono in evidenza problemi e ipotizzano soluzioni che sono comuni a tutto il territorio nazionale. Il lavoro, organizzato come segue. Il primo capitolo presenta un'analisi strutturalee dinamica dell'economia della provincia di Pisa nel periodo precedente alla crisi sanitaria. Il secondo capitolo contiene una lettura dell'impatto economico, sociale e sanitario dell'emergenza COVID-19 e delle misure di contrasto messe in campo dal governo nei mesi iniziali della crisi (marzo-giugno 2020). Il terzo capitolo contiene un approfondimento dell'analisi economico-aziendale svolta. L'impatto del COVID-19 sull'economia alcuni settori emersi come rilevanti per l'economia provinciale. Il quarto capitolo svolge riflessioni e proposte in ambito giuridico, il quinto capitolo chiude il lavoro presentando alcune proposte di policy. Gli autori desiderano ringraziare il Prefetto di Pisa, per l'attivita' di supporto istituzionale, il Rettore dell'Universita' di Pisa, e tutti gli attori istituzionali e socio-economici che hanno collaborato direttamente – mediante incontri e interviste ‒ o indirettamente – attraverso la messa a disposizione dei dati e informazioni ‒ alla stesura del lavoro.
BASE