![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
This paper analyses the reliability and accuracy of the relationships between migration and employment status when estimated using a linked data set. The analysis will be carried out using a new source, the Labour and Geographical Mobility Statistics, which is provided by the Spanish Statistical Office. This statistic is constructed by an exact matching procedure, linking the Labour Force Survey with the Official Population Register. The findings reveal that in order to accurately study geographic labour mobility, timing and geographical coherence between the two data sets must exist. If this is not the case, causality relations between the labour market and geographical mobility cannot be properly analysed. Although our analysis refers to the microdata quality and internal coherence of this new Spanish statistic, our conclusion can be extended to any linked data set for studying geographic labour mobility, including the population censuses mainly based on administrative data.
Disclosure statement
No potential conflict of interest was reported by the authors.
Correction Statement
This article has been republished with minor changes. These changes do not impact the academic content of the article.
Notes
1. See, for example, Oberski et al. (Citation2017) or Angel et al. (Citation2019).
2. It must be noted that the LFS also contains some retrospective information about the previous place of residence but, in the majority of EU countries, the LFS sample design leads to a systematic underestimation of mobility, in the case of recently arrived immigrants (see Martí & Ródenas, Citation2007, Citation2012). We will not address this here because in the LGMS the administrative register determines the mobility, not the LFS information.
3. The Spanish LFS has been conducted by the INE since 1964 and covers the entire national territory. This quarterly survey is aimed at population living in family dwellings, and intended to collect data on the labour force and its various categories (employed, unemployed) as well as on the population outside the labour market (inactive). The quarterly sample size is taken from about 65,000 dwellings, equivalent to approximately 160,000 people. The Spanish LFS is completely harmonized with the standards established by the EUROSTAT for the EU-LFS.
4. Other examples of unique identifiers relating to people are tax identification numbers or health insurance numbers, for example.
5. See, for example, Zhu et al. (Citation2015), Herzog et al. (Citation2007), and ASPE (Citation2002). From a theoretical point of view, the specific problems that may appear when surveys are linked with administrative records using these techniques can be found in Blom and Korbmacher (Citation2018), Sakshaug (Citation2018a, Citation2018b) or Di Consiglio and Tuoto (Citation2018).
6. See Abowd et.al (Citation2018) or Sakshaug et al. (Citation2012) for a discussion on the biases that the requisite of personal consent can generate, even with exact linkage.
7. The focus of this paper is not to assess which data set measures mobility better. A discussion of this issue can be found in Martí and Ródenas (Citation2004, Citation2007, Citation2012) or Ródenas and Martí (Citation2009). Here our objective is to explain the reasons for the differences in the mobility estimations provided by the PR and the LFS based on an identical set of individuals.
8. Although, on more than one occasion in private conversations, the INE has confirmed a very high percentage of coherences between the municipality of the LFS survey and that of residence according to the PR, they have not published this data or provided any figures.
9. The microdata files can be obtained on request at the INE.
10. A level of significance (α) of 0.05 has been used and the records of the sample have been weighted for all observational units in LGMS. The p-value associated with each estimator is equal to 0 in all the proofs.
Additional information
Notes on contributors
Mónica Martí
Mónica Martí is currently a university lecturer in the Department of Applied Economic Analysis at the University of Alicante (Spain). Her main lines of research have focused on labor market analysis, migration processes and their impact on labor markets, as well as the quality and comparability of statistical sources on migration.
Carmen Ródenas
Carmen Ródenas is a Professor in the Department of Applied Economic Analysis at the University of Alicante (Spain). Her main areas of research are: migration processes and their repercussions on labour markets and, in addition, the analysis of the quality and comparability of statistical information on migrations.