Optimal Sampling for the Population Coverage Survey of the New Italian Register Based Census

Abstract

For the first time in 2018 the Italian Institute of Statistics (Istat) implemented the annual Permanent Population Census which relies on the Population Base Register (PBR) and the Population Coverage Survey (PCS). This article provides a general overview of the PCS sampling design, which makes use of the PBR to correct population counts with the extended dual system estimator (Nirel and Glickman 2009). The sample allocation, proven optimal under a set of precision constraints, is based on preliminary estimates of individual probabilities of over-coverage and under-coverage. It defines the expected sample size in terms of individuals, and it oversamples the sub-populations subject to the risk of under/over coverage. Finally, the article introduces a sample selection method, which to the greatest extent possible satisfies the planned allocation of persons in terms of socio-demographic characteristics. Under acceptable assumptions, the article also shows that the sampling strategy enhances the precision of the estimates.

Characteristics of Respondents to Web-Based or Traditional Interviews in Mixed-Mode Surveys. Evidence from the Italian Permanent Population Census

“…The population census provides the official estimate of the resident population at the municipality level (Righi et al 2021). In Italy, as in most European countries, an assessment of the coverage results of the last population census in 2011 and the high costs of data collection led to a change in the strategy for the population census from the traditional door-to-door enumeration (every ten years for the whole population) to a yearly register-based sample survey, combining the use of administrative sources with annual surveys (Chieppa et al 2018;Citro 2014;Crescenzi 2015;Righi et al 2021).…”

“…The population census provides the official estimate of the resident population at the municipality level (Righi et al 2021). In Italy, as in most European countries, an assessment of the coverage results of the last population census in 2011 and the high costs of data collection led to a change in the strategy for the population census from the traditional door-to-door enumeration (every ten years for the whole population) to a yearly register-based sample survey, combining the use of administrative sources with annual surveys (Chieppa et al 2018;Citro 2014;Crescenzi 2015;Righi et al 2021). A further advantage of these data collection methodologies is the greater containment of the census participation burden: as noted by UNECE ( 2018), the response burden on the population is lower with a combined census, both because the number of questions is reduced (since some information is available from the population register), and because the physical presence of a stranger in people's homes is avoided (which could be a reason for a refusal to participate, especially in a pandemic situation).…”

Measuring the Accuracy of Aggregates Computed from a Statistical Register

“…where ) is the model's unbiased estimate of the conditional probability ) such that unit k included in the register belongs to the population and ) is the model's unbiased estimate of the conditional probability P(k [ Rjk [ U) such that unit k belonging to the population is included in the register. This approach has been adopted for the Italian Population Base Register by integrating the register data with the Census Population Coverage Survey carried out each year as a component of the Italian Permanent Census Survey System (Righi et al 2021). The GMSE of Nunb;ðU d Þ may be expressed as and specified as described in Subsection 4.2 by defining the predictions and model's expected values of the w variables (instead of the y variables).…”

Preface

“…The second group of articles deals specifically with census data, with particular focus on new census forms, and the assessment of census data quality, especially with regard to census coverage. Among this group of articles, Righi et al (2021) report on the annual Permanent Population Census implemented in Italy since 2018, which augments a population register with a Population Coverage Survey. The methodological details of this approach are presented, with the sampling design and the extended version of dual system estimator, proposed by Nirel and Glickman (2009), applied to correct for the possible under-and over-coverage of the register-based census.…”