Filter results by

Search Help
Currently selected filters that can be removed

Keyword(s)

Author(s)

183 facets displayed. 1 facets selected.

Content

1 facets displayed. 0 facets selected.
Sort Help
entries

Results

All (342)

All (342) (0 to 10 of 342 results)

  • Articles and reports: 11-522-X202200100001
    Description: Record linkage aims at identifying record pairs related to the same unit and observed in two different data sets, say A and B. Fellegi and Sunter (1969) suggest each record pair is tested whether generated from the set of matched or unmatched pairs. The decision function consists of the ratio between m(y) and u(y),probabilities of observing a comparison y of a set of k>3 key identifying variables in a record pair under the assumptions that the pair is a match or a non-match, respectively. These parameters are usually estimated by means of the EM algorithm using as data the comparisons on all the pairs of the Cartesian product ?=A×B. These observations (on the comparisons and on the pairs status as match or non-match) are assumed as generated independently of other pairs, assumption characterizing most of the literature on record linkage and implemented in software tools (e.g. RELAIS, Cibella et al. 2012). On the contrary, comparisons y and matching status in ? are deterministically dependent. As a result, estimates on m(y) and u(y) based on the EM algorithm are usually bad. This fact jeopardizes the effective application of the Fellegi-Sunter method, as well as automatic computation of quality measures and possibility to apply efficient methods for model estimation on linked data (e.g. regression functions), as in Chambers et al. (2015). We propose to explore ? by a set of samples, each one drawn so to preserve independence of comparisons among the selected record pairs. Simulations are encouraging.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100002
    Description: The authors used the Splink probabilistic linkage package developed by the UK Ministry of Justice, to link census data from England and Wales to itself to find duplicate census responses. A large gold standard of confirmed census duplicates was available meaning that the results of the Splink implementation could be quality assured. This paper describes the implementation and features of Splink, gives details of the settings and parameters that we used to tune Splink for our particular project, and gives the results that we obtained.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100007
    Description: With the availability of larger and more diverse data sources, Statistical Institutes in Europe are inclined to publish statistics on smaller groups than they used to do. Moreover, high impact global events like the Covid crisis and the situation in Ukraine may also ask for statistics on specific subgroups of the population. Publishing on small, targeted groups not only raises questions on statistical quality of the figures, it also raises issues concerning statistical disclosure risk. The principle of statistical disclosure control does not depend on the size of the groups the statistics are based on. However, the risk of disclosure does depend on the group size: the smaller a group, the higher the risk. Traditional ways to deal with statistical disclosure control and small group sizes include suppressing information and coarsening categories. These methods essentially increase the (mean) group sizes. More recent approaches include perturbative methods that have the intention to keep the group sizes small in order to preserve as much information as possible while reducing the disclosure risk sufficiently. In this paper we will mention some European examples of special focus group statistics and discuss the implications on statistical disclosure control. Additionally, we will discuss some issues that the use of perturbative methods brings along: its impact on disclosure risk and utility as well as the challenges in proper communication thereof.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100012
    Description: At Statistics Netherlands (SN) for some economic sectors two partly-independent intra-annual turnover index series are available: a monthly series based on survey data and a quarterly series based on value added tax data for the smaller units and re-used survey data for the other units. SN aims to benchmark the monthly turnover index series to the quarterly census data on a quarterly basis. This cannot currently be done because the tax data has a different quarterly pattern: the turnover is relatively large in the fourth quarter of the year and smaller in the first quarter. With the current study we aim to describe this deviating quarterly pattern at micro level. In the past we developed a mixture model using absolute turnover levels that could explain part of the quarterly patterns. Because the absolute turnover levels differ between the two series, in the current study we use a model based on relative quarterly turnover levels within a year.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100016
    Description: To overcome the traditional drawbacks of chain sampling methods, the sampling method called “network sampling with memory” was developed. Its unique feature is to recreate, gradually in the field, a frame for the target population composed of individuals identified by respondents and to randomly draw future respondents from this frame, thereby minimizing selection bias. Tested for the first time in France between September 2020 and June 2021, for a survey among Chinese immigrants in Île-de-France (ChIPRe), this presentation describes the difficulties encountered during collection—sometimes contextual, due to the pandemic, but mostly inherent to the method.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100018
    Description: The Longitudinal Social Data Development Program (LSDDP) is a social data integration approach aimed at providing longitudinal analytical opportunities without imposing additional burden on respondents. The LSDDP uses a multitude of signals from different data sources for the same individual, which helps to better understand their interactions and track changes over time. This article looks at how the ethnicity status of people in Canada can be estimated at the most detailed disaggregated level possible using the results from a variety of business rules applied to linked data and to the LSDDP denominator. It will then show how improvements were obtained using machine learning methods, such as decision trees and random forest techniques.
    Release date: 2024-03-25

  • Articles and reports: 12-001-X202300200002
    Description: Being able to quantify the accuracy (bias, variance) of published output is crucial in official statistics. Output in official statistics is nearly always divided into subpopulations according to some classification variable, such as mean income by categories of educational level. Such output is also referred to as domain statistics. In the current paper, we limit ourselves to binary classification variables. In practice, misclassifications occur and these contribute to the bias and variance of domain statistics. Existing analytical and numerical methods to estimate this effect have two disadvantages. The first disadvantage is that they require that the misclassification probabilities are known beforehand and the second is that the bias and variance estimates are biased themselves. In the current paper we present a new method, a Gaussian mixture model estimated by an Expectation-Maximisation (EM) algorithm combined with a bootstrap, referred to as the EM bootstrap method. This new method does not require that the misclassification probabilities are known beforehand, although it is more efficient when a small audit sample is used that yields a starting value for the misclassification probabilities in the EM algorithm. We compared the performance of the new method with currently available numerical methods: the bootstrap method and the SIMEX method. Previous research has shown that for non-linear parameters the bootstrap outperforms the analytical expressions. For nearly all conditions tested, the bias and variance estimates that are obtained by the EM bootstrap method are closer to their true values than those obtained by the bootstrap and SIMEX methods. We end this paper by discussing the results and possible future extensions of the method.
    Release date: 2024-01-03

  • Articles and reports: 12-001-X202300200010
    Description: Sample coordination methods aim to increase (in positive coordination) or decrease (in negative coordination) the size of the overlap between samples. The samples considered can be from different occasions of a repeated survey and/or from different surveys covering a common population. Negative coordination is used to control the response burden in a given period, because some units do not respond to survey questionnaires if they are selected in many samples. Usually, methods for sample coordination do not take into account any measure of the response burden that a unit has already expended in responding to previous surveys. We introduce such a measure into a new method by adapting a spatially balanced sampling scheme, based on a generalization of Poisson sampling, together with a negative coordination method. The goal is to create a double control of the burden for these units: once by using a measure of burden during the sampling process and once by using a negative coordination method. We evaluate the approach using Monte-Carlo simulation and investigate its use for controlling for selection “hot-spots” in business surveys in Statistics Netherlands.
    Release date: 2024-01-03

  • Articles and reports: 12-001-X202300200012
    Description: In recent decades, many different uses of auxiliary information have enriched survey sampling theory and practice. Jean-Claude Deville contributed significantly to this progress. My comments trace some of the steps on the way to one important theory for the use of auxiliary information: Estimation by calibration.
    Release date: 2024-01-03

  • Stats in brief: 89-20-00062023001
    Description: This course is intended for Government of Canada employees who would like to learn about evaluating the quality of data for a particular use. Whether you are a new employee interested in learning the basics, or an experienced subject matter expert looking to refresh your skills, this course is here to help.
    Release date: 2023-07-17
Stats in brief (2)

Stats in brief (2) ((2 results))

  • Stats in brief: 89-20-00062023001
    Description: This course is intended for Government of Canada employees who would like to learn about evaluating the quality of data for a particular use. Whether you are a new employee interested in learning the basics, or an experienced subject matter expert looking to refresh your skills, this course is here to help.
    Release date: 2023-07-17

  • Stats in brief: 89-20-00062022003
    Description:

    By the end of this video you will understand what confidence intervals are, why we use them, and what factors have an impact on them.

    Release date: 2022-05-24
Articles and reports (337)

Articles and reports (337) (0 to 10 of 337 results)

  • Articles and reports: 11-522-X202200100001
    Description: Record linkage aims at identifying record pairs related to the same unit and observed in two different data sets, say A and B. Fellegi and Sunter (1969) suggest each record pair is tested whether generated from the set of matched or unmatched pairs. The decision function consists of the ratio between m(y) and u(y),probabilities of observing a comparison y of a set of k>3 key identifying variables in a record pair under the assumptions that the pair is a match or a non-match, respectively. These parameters are usually estimated by means of the EM algorithm using as data the comparisons on all the pairs of the Cartesian product ?=A×B. These observations (on the comparisons and on the pairs status as match or non-match) are assumed as generated independently of other pairs, assumption characterizing most of the literature on record linkage and implemented in software tools (e.g. RELAIS, Cibella et al. 2012). On the contrary, comparisons y and matching status in ? are deterministically dependent. As a result, estimates on m(y) and u(y) based on the EM algorithm are usually bad. This fact jeopardizes the effective application of the Fellegi-Sunter method, as well as automatic computation of quality measures and possibility to apply efficient methods for model estimation on linked data (e.g. regression functions), as in Chambers et al. (2015). We propose to explore ? by a set of samples, each one drawn so to preserve independence of comparisons among the selected record pairs. Simulations are encouraging.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100002
    Description: The authors used the Splink probabilistic linkage package developed by the UK Ministry of Justice, to link census data from England and Wales to itself to find duplicate census responses. A large gold standard of confirmed census duplicates was available meaning that the results of the Splink implementation could be quality assured. This paper describes the implementation and features of Splink, gives details of the settings and parameters that we used to tune Splink for our particular project, and gives the results that we obtained.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100007
    Description: With the availability of larger and more diverse data sources, Statistical Institutes in Europe are inclined to publish statistics on smaller groups than they used to do. Moreover, high impact global events like the Covid crisis and the situation in Ukraine may also ask for statistics on specific subgroups of the population. Publishing on small, targeted groups not only raises questions on statistical quality of the figures, it also raises issues concerning statistical disclosure risk. The principle of statistical disclosure control does not depend on the size of the groups the statistics are based on. However, the risk of disclosure does depend on the group size: the smaller a group, the higher the risk. Traditional ways to deal with statistical disclosure control and small group sizes include suppressing information and coarsening categories. These methods essentially increase the (mean) group sizes. More recent approaches include perturbative methods that have the intention to keep the group sizes small in order to preserve as much information as possible while reducing the disclosure risk sufficiently. In this paper we will mention some European examples of special focus group statistics and discuss the implications on statistical disclosure control. Additionally, we will discuss some issues that the use of perturbative methods brings along: its impact on disclosure risk and utility as well as the challenges in proper communication thereof.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100012
    Description: At Statistics Netherlands (SN) for some economic sectors two partly-independent intra-annual turnover index series are available: a monthly series based on survey data and a quarterly series based on value added tax data for the smaller units and re-used survey data for the other units. SN aims to benchmark the monthly turnover index series to the quarterly census data on a quarterly basis. This cannot currently be done because the tax data has a different quarterly pattern: the turnover is relatively large in the fourth quarter of the year and smaller in the first quarter. With the current study we aim to describe this deviating quarterly pattern at micro level. In the past we developed a mixture model using absolute turnover levels that could explain part of the quarterly patterns. Because the absolute turnover levels differ between the two series, in the current study we use a model based on relative quarterly turnover levels within a year.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100016
    Description: To overcome the traditional drawbacks of chain sampling methods, the sampling method called “network sampling with memory” was developed. Its unique feature is to recreate, gradually in the field, a frame for the target population composed of individuals identified by respondents and to randomly draw future respondents from this frame, thereby minimizing selection bias. Tested for the first time in France between September 2020 and June 2021, for a survey among Chinese immigrants in Île-de-France (ChIPRe), this presentation describes the difficulties encountered during collection—sometimes contextual, due to the pandemic, but mostly inherent to the method.
    Release date: 2024-03-25

  • Articles and reports: 11-522-X202200100018
    Description: The Longitudinal Social Data Development Program (LSDDP) is a social data integration approach aimed at providing longitudinal analytical opportunities without imposing additional burden on respondents. The LSDDP uses a multitude of signals from different data sources for the same individual, which helps to better understand their interactions and track changes over time. This article looks at how the ethnicity status of people in Canada can be estimated at the most detailed disaggregated level possible using the results from a variety of business rules applied to linked data and to the LSDDP denominator. It will then show how improvements were obtained using machine learning methods, such as decision trees and random forest techniques.
    Release date: 2024-03-25

  • Articles and reports: 12-001-X202300200002
    Description: Being able to quantify the accuracy (bias, variance) of published output is crucial in official statistics. Output in official statistics is nearly always divided into subpopulations according to some classification variable, such as mean income by categories of educational level. Such output is also referred to as domain statistics. In the current paper, we limit ourselves to binary classification variables. In practice, misclassifications occur and these contribute to the bias and variance of domain statistics. Existing analytical and numerical methods to estimate this effect have two disadvantages. The first disadvantage is that they require that the misclassification probabilities are known beforehand and the second is that the bias and variance estimates are biased themselves. In the current paper we present a new method, a Gaussian mixture model estimated by an Expectation-Maximisation (EM) algorithm combined with a bootstrap, referred to as the EM bootstrap method. This new method does not require that the misclassification probabilities are known beforehand, although it is more efficient when a small audit sample is used that yields a starting value for the misclassification probabilities in the EM algorithm. We compared the performance of the new method with currently available numerical methods: the bootstrap method and the SIMEX method. Previous research has shown that for non-linear parameters the bootstrap outperforms the analytical expressions. For nearly all conditions tested, the bias and variance estimates that are obtained by the EM bootstrap method are closer to their true values than those obtained by the bootstrap and SIMEX methods. We end this paper by discussing the results and possible future extensions of the method.
    Release date: 2024-01-03

  • Articles and reports: 12-001-X202300200010
    Description: Sample coordination methods aim to increase (in positive coordination) or decrease (in negative coordination) the size of the overlap between samples. The samples considered can be from different occasions of a repeated survey and/or from different surveys covering a common population. Negative coordination is used to control the response burden in a given period, because some units do not respond to survey questionnaires if they are selected in many samples. Usually, methods for sample coordination do not take into account any measure of the response burden that a unit has already expended in responding to previous surveys. We introduce such a measure into a new method by adapting a spatially balanced sampling scheme, based on a generalization of Poisson sampling, together with a negative coordination method. The goal is to create a double control of the burden for these units: once by using a measure of burden during the sampling process and once by using a negative coordination method. We evaluate the approach using Monte-Carlo simulation and investigate its use for controlling for selection “hot-spots” in business surveys in Statistics Netherlands.
    Release date: 2024-01-03

  • Articles and reports: 12-001-X202300200012
    Description: In recent decades, many different uses of auxiliary information have enriched survey sampling theory and practice. Jean-Claude Deville contributed significantly to this progress. My comments trace some of the steps on the way to one important theory for the use of auxiliary information: Estimation by calibration.
    Release date: 2024-01-03

  • Articles and reports: 12-001-X202300100003
    Description: To improve the precision of inferences and reduce costs there is considerable interest in combining data from several sources such as sample surveys and administrative data. Appropriate methodology is required to ensure satisfactory inferences since the target populations and methods for acquiring data may be quite different. To provide improved inferences we use methodology that has a more general structure than the ones in current practice. We start with the case where the analyst has only summary statistics from each of the sources. In our primary method, uncertain pooling, it is assumed that the analyst can regard one source, survey r, as the single best choice for inference. This method starts with the data from survey r and adds data from those other sources that are shown to form clusters that include survey r. We also consider Dirichlet process mixtures, one of the most popular nonparametric Bayesian methods. We use analytical expressions and the results from numerical studies to show properties of the methodology.
    Release date: 2023-06-30
Journals and periodicals (3)

Journals and periodicals (3) ((3 results))

  • Journals and periodicals: 12-605-X
    Description:

    The Record Linkage Project Process Model (RLPPM) was developed by Statistics Canada to identify the processes and activities involved in record linkage. The RLPPM applies to linkage projects conducted at the individual and enterprise level using diverse data sources to create new data sources to meet analytical and operational needs.

    Release date: 2017-06-05

  • Journals and periodicals: 89-639-X
    Geography: Canada
    Description:

    Beginning in late 2006, the Social and Aboriginal Statistics Division of Statistics Canada embarked on the process of review of questions used in the Census and in surveys to produce data about Aboriginal peoples (North American Indian, Métis and Inuit). This process is essential to ensure that Aboriginal identification questions are valid measures of contemporary Aboriginal identification, in all its complexity. Questions reviewed included the following (from the Census 2B questionnaire):- the Ethnic origin / Aboriginal ancestry question;- the Aboriginal identity question;- the Treaty / Registered Indian question; and- the Indian band / First Nation Membership question.

    Additional testing was conducted on Census questions with potential Aboriginal response options: the population group question (also known as visible minorities), and the Religion question. The review process to date has involved two major steps: regional discussions with data users and stakeholders, and qualitative testing. The regional discussions with over 350 users of Aboriginal data across Canada were held in early 2007 to examine the four questions used on the Census and other surveys of Statistics Canada. Data users included National Aboriginal organizations, Aboriginal Provincial and Territorial Organizations, Federal, Provincial and local governments, researchers and Aboriginal service organizations. User feedback showed that main areas of concern were data quality, undercoverage, the wording of questions, and the importance of comparability over time.

    Release date: 2009-04-17

  • Journals and periodicals: 89-629-X
    Geography: Canada
    Description:

    This report summarizes the main issues raised in these meetings. Four questions used to identify Aboriginal people from the Census and surveys were considered in the discussions.Statistics Canada regularly reviews the questions used on the Census and other surveys to ensure that the resulting data are representative of the population. As a first step in the process to review the questions used to produce data about First Nations, Inuit and Métis populations, regional discussions were held with more than 350 users of Aboriginal data in over 40 locations across Canada during the winter, spring and early summer of 2007.

    This report summarizes the main issues raised in these meetings. Four questions used to identify Aboriginal people from the Census and surveys were considered in the discussions.

    Release date: 2008-05-27
Date modified: