Keyword search

Description:

Conceptual arguments and examples are presented suggesting that the Bayesian approach to survey inference can address the many and varied challenges of survey analysis. Bayesian models that incorporate features of the complex design can yield inferences that are relevant for the specific data set obtained, but also have good repeated-sampling properties. Examples focus on the role of auxiliary variables and sampling weights, and methods for handling nonresponse. The article offers ten top reasons for favoring the Bayesian approach to survey inference.

Description:

Periodically, income statistics are updated to reflect the most recent population estimates derived after the census. Accordingly, with the release of the 2020 data from the Canadian Income Survey (CIS), Statistics Canada has revised estimates for 2012 to 2019 using totals from population estimates based on the 2016 Census. This paper presents a comparison between revised and unrevised estimates for key income series, and describes other modifications made to CIS variables.

Description:

We combine weighting and Bayesian prediction in a unified approach to survey inference. The general principles of Bayesian analysis imply that models for survey outcomes should be conditional on all variables that affect the probability of inclusion. We incorporate all the variables that are used in the weighting adjustment under the framework of multilevel regression and poststratification, as a byproduct generating model-based weights after smoothing. We improve small area estimation by dealing with different complex issues caused by real-life applications to obtain robust inference at finer levels for subdomains of interest. We investigate deep interactions and introduce structured prior distributions for smoothing and stability of estimates. The computation is done via Stan and is implemented in the open-source R package rstanarm and available for public use. We evaluate the design-based properties of the Bayesian procedure. Simulation studies illustrate how the model-based prediction and weighting inference can outperform classical weighting. We apply the method to the New York Longitudinal Study of Wellbeing. The new approach generates smoothed weights and increases efficiency for robust finite population inference, especially for subsets of the population.

Description:

The Guide to the Labour Force Survey contains a dictionary of concepts and definitions and covers topics such as survey methodology, data collection, data processing and data quality. It also contains information on products and services, sub-provincial geography descriptions as well as the survey questionnaire.

Description:

This survey was designed to collect details on topics such as: i) the extent to which graduates of postsecondary programs have been successful in obtaining employment since graduation; ii) the relationship between the graduates' program of study and the employment subsequently obtained; iii) the type of employment obtained and qualification requirements; iv) sources of funding for postsecondary education; and v) government-sponsored student loans and other sources of student debt. The survey results are directed towards policy makers, researchers, educators, employers and persons interested in public postsecondary education and graduates' transition from school to work.

Description:

This document brings together guidelines and checklists on many issues that need to be considered in the pursuit of quality objectives in the execution of statistical activities. Its focus is on how to assure quality through effective and appropriate design or redesign of a statistical project or program from inception through to data evaluation, dissemination and documentation. These guidelines draw on the collective knowledge and experience of many Statistics Canada employees. It is expected that Quality Guidelines will be useful to staff engaged in the planning and design of surveys and other statistical projects, as well as to those who evaluate and analyze the outputs of these projects.

Description:

Panel surveys are frequently used to measure the evolution of parameters over time. Panel samples may suffer from different types of unit non-response, which is currently handled by estimating the response probabilities and by reweighting respondents. In this work, we consider estimation and variance estimation under unit non-response for panel surveys. Extending the work by Kim and Kim (2007) for several times, we consider a propensity score adjusted estimator accounting for initial non-response and attrition, and propose a suitable variance estimator. It is then extended to cover most estimators encountered in surveys, including calibrated estimators, complex parameters and longitudinal estimators. The properties of the proposed variance estimator and of a simplified variance estimator are estimated through a simulation study. An illustration of the proposed methods on data from the ELFE survey is also presented.

Description:

Many studies conducted by various electric utilities around the world are based on the analysis of mean electricity consumption curves for various subpopulations, particularly geographic in nature. Those mean curves are estimated from samples of thousands of curves measured at very short intervals over long periods. Estimation for small subpopulations, also called small domains, is a very timely topic in sampling theory.

In this article, we will examine this problem based on functional data and we will try to estimate the mean curves for small domains. For this, we propose four methods: functional linear regression; modelling the scores of a principal component analysis by unit-level linear mixed models; and two non-parametric estimators, with one based on regression trees and the other on random forests, adapted to the curves. All these methods have been tested and compared using real electricity consumption data for households in France.

Description:

This article proposes a criterion for calculating the trade-off in so-called “mixed” allocations, which combine two classic allocations in sampling theory. In INSEE (National Institute of Statistics and Economic Studies) business surveys, it is common to use the arithmetic mean of a proportional allocation and a Neyman allocation (corresponding to a trade-off of 0.5). It is possible to obtain a trade-off value resulting in better properties for the estimators. This value belongs to a region that is obtained by solving an optimization program. Different methods for calculating the trade-off will be presented. An application for business surveys is presented, as well as a comparison with other usual trade-off allocations.

Description:

Based on auxiliary information, calibration is often used to improve the precision of estimates. However, calibration weighting may not be appropriate for all variables of interest of the survey, particularly those not related to the auxiliary variables used in calibration. In this paper, we propose a criterion to assess, for any variable of interest, the impact of calibration weighting on the precision of the estimated total. This criterion can be used to decide on the weights associated with each survey variable of interest and determine the variables for which calibration weighting is appropriate.

Description:

This survey was designed to collect details on topics such as: i) the extent to which graduates of postsecondary programs have been successful in obtaining employment since graduation; ii) the relationship between the graduates' program of study and the employment subsequently obtained; iii) the type of employment obtained and qualification requirements; iv) sources of funding for postsecondary education; and v) government-sponsored student loans and other sources of student debt. The survey results are directed towards policy makers, researchers, educators, employers and persons interested in public postsecondary education and graduates' transition from school to work.

Description: The Canadian Tobacco, Alcohol and Drugs Survey (CTADS) is a biennial general population survey of tobacco, alcohol and drug use among Canadians aged 15 years and older, with the primary focus on 15- to 24-year-olds. The CTADS is a telephone survey conducted by Statistics Canada on behalf of Health Canada.

Description:

Statistics Canada was approached by Stentor Resource Centre Incorporated to conduct a survey to monitor the telephone penetration rates across Canada. The survey determines if the respondents have a telephone line in their residence. If they do not have a telephone line, information is collected as to the reasons why. Information is also collected on the income characteristics of the selected households.

The management of the survey was transferred from Stentor to Bell Canada in the Fall of 1998.

The Labour Force Survey (LFS) supplementary capacity is used to conduct this biannual survey. A sample of approximately 44,000 respondents is used for this survey (five out of six rotation groups). The survey data are collected using Computer Assisted Interviewing (CAI). The first data collection procedure took place during November's LFS week in 1996.

This microdata file is prepared biannually and contains the variables from the survey, plus geographical variables from the LFS (province, census metropolitan area, urban/rural breakdown). No other variables from the LFS are added to the file.

Description:

The International Adult Literacy Survey (IALS) was a collaborative effort by twenty governments and three intergovernmental organizations. The countries of Australia, Belgium/Flanders, Canada, Germany, Ireland, Netherlands, New Zealand, Poland, Sweden, Switzerland (French, German and Italian), United Kingdom, United States, Chile, Czech Republic, Denmark, Finland, Hungary, Italy, Norway, and Slovenia participated in the IALS.

The Canadian IALS survey had a number of objectives. These were: a) to provide an updated profile of adult literacy abilities for Canada for comparison to that provided by the 1989 Survey of Literacy Skills Used in Daily Activities (LSUDA); b) to provide sufficiently large numbers of Franco-Ontarians, seniors, social assistance recipients, unemployment insurance recipients and out-of-school youth to profile their skill levels; c) to shed light on the relationship between performance, educational attainment, labour market participation and employment for those at certain literacy levels; and d) to compare Canadian literacy levels with those in other countries.

Description:

The public use microdata file from the 2010 Canada Survey of Giving, Volunteering and Participating is now available. This file contains information collected from nearly 15,000 respondents aged 15 and over residing in private households in the provinces.The public use microdata file provides provincial-level information about the ways in which Canadians donate money and in-kind gifts to charitable and nonprofit organizations; volunteer their time to these organizations; provide help directly to others. Socio-demographic, income and labour force data are also included on the file.

Description:

This publication presents a comprehensive overview of the Canadian trucking industry, both for-hire and private (own account). Principal information includes statistics on revenues and expenses, equipment operated, investment, employment, and commodities transported from point of origin to point of destination. Also included are special studies, a glossary and an explanation of data quality measures and methodology.

Description:

This guide is for the Household Internet Use Survey microdata file. The Household Internet Use Survey is being conducted by Statistics Canada on behalf of Industry Canada. The information from this survey will assist the Science and Technology Redesign Project at Statistics Canada to fulfil a three-year contractual agreement between them and the Telecommunications and Policy Branch of Industry Canada. The Household Internet Use Survey is a voluntary survey. It will provide information on the use of computers for communication purposes, and households' access and use of the Internet from home.

The objective of this survey is to measure the demand for telecommunications services by Canadian households. To assess the demand, we measure the frequency and intensity of use of what is commonly referred to as "the information highway" among other things. This was done by asking questions relating to the accessibility of the Internet to Canadian households both at home, the workplace and a number of other locations. The information collected will be used to update and expand upon previous studies done by Statistics Canada on the topic of the Information Highway.

Description:

The main objective of the 2002 Youth Smoking Survey (YSS) is to provide current information on the smoking behaviour of students in grades 5 to 9 (in Quebec primary school grades 5 and 6 and secondary school grades 1 to 3), and to measure changes that occurred since the last time the survey was conducted in 1994. Additionally, the 2002 survey collected basic data on alcohol and drug use by students in grades 7 to 9 (in Quebec secondary 1 to 3). Results of the Youth Smoking Survey will help with the evaluation of anti-smoking and anti-drug use programs, as well as with the development of new programs.

Description:

The Adult Education and Training Survey (AETS) is Canada's most comprehensive source of data on individual participation in formal adult education and training. It is the only Canadian survey to collect detailed information about the skill development efforts of the entire adult Canadian population. The AETS provides information about the main subject of training activities, their provider, duration and the sources and types of support for training. Furthermore, the AETS allows for the examination of the socio-economic and demographic profiles of both training participants and non-participants. This survey also identifies barriers faced by individuals who wish to take some form of training but cannot. The AETS was administered three times during the 1990s, in 1992, 1994 and 1998, as a supplement to the Labour Force Survey (LFS).

The content of the AETS was revised to take into account recommendations coming from consultation exercises. As a result, more than half of the 2003 survey is made up of new questions and the target population has been modified.

The main objectives are:1) To measure the incidence and intensity of adults' participation in job-related formal training.2) To profile employer support to job-related formal training.3) To analyze the aspects of job-related training activities such as: training provider, expenses, financial support, motivations, outcomes and difficulties experienced while training.4) To identify the barriers preventing individuals from participating in the job-related formal training they want or need to take.5) To identify reasons explaining adults' lack of participation and of interest in job-related formal training.6) To relate adults' current participation patterns to their past involvement in and plans about future participation in job-related training.7) To measure the incidence and frequency of adults' participation in job-related informal training.8) To examine the interactions between participation in formal and informal job-related training.

The population covered by the AETS consists of Canadians 25 years of age and older. This is a change from the population previously targeted by the AETS, which consisted of Canadians aged 17 years of age and older. A primary consideration for this change was the practical difficulties in applying the definition of adult education to individuals in the 17 to 24 years of age group. By definition, adult education excludes students who are still involved in their first or initial stage of schooling. As previous AETS did not precisely identify students still in their initial stage of schooling, analyses using these data had to rely on an ad hoc definition of adult learners. According to this definition, individuals aged 17 to 24 who were not in one of the following situations were excluded from the analysis: full-time students subsidized by an employer and full-time students over 19 enrolled in elementary or secondary programs.

Description:

Section 1 described results for small for-hire carriers whose operating revenues were between $30,000 and $1,000,000. Section 2 contains data for all owner operators included in the Annual Motor Carriers of Freight Survey of Small For-hire Carriers and Owner Operators including some firms whose operating revenues exceeded $1,000,000. Section 3 provides a general discussion of the Annual Motor Carriers of Freight Survey of Small For-hire Carriers and Owner Operators methodology and data quality.

Description:

Conceptual arguments and examples are presented suggesting that the Bayesian approach to survey inference can address the many and varied challenges of survey analysis. Bayesian models that incorporate features of the complex design can yield inferences that are relevant for the specific data set obtained, but also have good repeated-sampling properties. Examples focus on the role of auxiliary variables and sampling weights, and methods for handling nonresponse. The article offers ten top reasons for favoring the Bayesian approach to survey inference.

Description:

Periodically, income statistics are updated to reflect the most recent population estimates derived after the census. Accordingly, with the release of the 2020 data from the Canadian Income Survey (CIS), Statistics Canada has revised estimates for 2012 to 2019 using totals from population estimates based on the 2016 Census. This paper presents a comparison between revised and unrevised estimates for key income series, and describes other modifications made to CIS variables.

Description:

We combine weighting and Bayesian prediction in a unified approach to survey inference. The general principles of Bayesian analysis imply that models for survey outcomes should be conditional on all variables that affect the probability of inclusion. We incorporate all the variables that are used in the weighting adjustment under the framework of multilevel regression and poststratification, as a byproduct generating model-based weights after smoothing. We improve small area estimation by dealing with different complex issues caused by real-life applications to obtain robust inference at finer levels for subdomains of interest. We investigate deep interactions and introduce structured prior distributions for smoothing and stability of estimates. The computation is done via Stan and is implemented in the open-source R package rstanarm and available for public use. We evaluate the design-based properties of the Bayesian procedure. Simulation studies illustrate how the model-based prediction and weighting inference can outperform classical weighting. We apply the method to the New York Longitudinal Study of Wellbeing. The new approach generates smoothed weights and increases efficiency for robust finite population inference, especially for subsets of the population.

Description:

Panel surveys are frequently used to measure the evolution of parameters over time. Panel samples may suffer from different types of unit non-response, which is currently handled by estimating the response probabilities and by reweighting respondents. In this work, we consider estimation and variance estimation under unit non-response for panel surveys. Extending the work by Kim and Kim (2007) for several times, we consider a propensity score adjusted estimator accounting for initial non-response and attrition, and propose a suitable variance estimator. It is then extended to cover most estimators encountered in surveys, including calibrated estimators, complex parameters and longitudinal estimators. The properties of the proposed variance estimator and of a simplified variance estimator are estimated through a simulation study. An illustration of the proposed methods on data from the ELFE survey is also presented.

Description:

Many studies conducted by various electric utilities around the world are based on the analysis of mean electricity consumption curves for various subpopulations, particularly geographic in nature. Those mean curves are estimated from samples of thousands of curves measured at very short intervals over long periods. Estimation for small subpopulations, also called small domains, is a very timely topic in sampling theory.

In this article, we will examine this problem based on functional data and we will try to estimate the mean curves for small domains. For this, we propose four methods: functional linear regression; modelling the scores of a principal component analysis by unit-level linear mixed models; and two non-parametric estimators, with one based on regression trees and the other on random forests, adapted to the curves. All these methods have been tested and compared using real electricity consumption data for households in France.

Description:

This article proposes a criterion for calculating the trade-off in so-called “mixed” allocations, which combine two classic allocations in sampling theory. In INSEE (National Institute of Statistics and Economic Studies) business surveys, it is common to use the arithmetic mean of a proportional allocation and a Neyman allocation (corresponding to a trade-off of 0.5). It is possible to obtain a trade-off value resulting in better properties for the estimators. This value belongs to a region that is obtained by solving an optimization program. Different methods for calculating the trade-off will be presented. An application for business surveys is presented, as well as a comparison with other usual trade-off allocations.

Description:

Based on auxiliary information, calibration is often used to improve the precision of estimates. However, calibration weighting may not be appropriate for all variables of interest of the survey, particularly those not related to the auxiliary variables used in calibration. In this paper, we propose a criterion to assess, for any variable of interest, the impact of calibration weighting on the precision of the estimated total. This criterion can be used to decide on the weights associated with each survey variable of interest and determine the variables for which calibration weighting is appropriate.

Description:

This paper develops statistical inference based on super population model in a finite population setting using ranked set samples (RSS). The samples are constructed without replacement. It is shown that the sample mean of RSS is model unbiased and has smaller mean square prediction error (MSPE) than the MSPE of a simple random sample mean. Using an unbiased estimator of MSPE, the paper also constructs a prediction confidence interval for the population mean. A small scale simulation study shows that estimator is as good as a simple random sample (SRS) estimator for poor ranking information. On the other hand it has higher efficiency than SRS estimator when the quality of ranking information is good, and the cost ratio of obtaining a single unit in RSS and SRS is not very high. Simulation study also indicates that coverage probabilities of prediction intervals are very close to the nominal coverage probabilities. Proposed inferential procedure is applied to a real data set.

Description:

This paper investigates the linearization and bootstrap variance estimation for the Gini coefficient and the change between Gini indexes at two periods of time. For the one-sample case, we use the influence function linearization approach suggested by Deville (1999), the without-replacement bootstrap suggested by Gross (1980) for simple random sampling without replacement and the with-replacement of primary sampling units described in Rao and Wu (1988) for multistage sampling. To obtain a two-sample variance estimator, we use the linearization technique by means of partial influence functions (Goga, Deville and Ruiz-Gazen, 2009). We also develop an extension of the studied bootstrap procedures for two-dimensional sampling. The two approaches are compared on simulated data.

Description:

The U.S. Census Bureau is investigating nonrespondent subsampling strategies for usage in the 2017 Economic Census. Design constraints include a mandated lower bound on the unit response rate, along with targeted industry-specific response rates. This paper presents research on allocation procedures for subsampling nonrespondents, conditional on the subsampling being systematic. We consider two approaches: (1) equal-probability sampling and (2) optimized allocation with constraints on unit response rates and sample size with the objective of selecting larger samples in industries that have initially lower response rates. We present a simulation study that examines the relative bias and mean squared error for the proposed allocations, assessing each procedure’s sensitivity to the size of the subsample, the response propensities, and the estimation procedure.

Description:

The Guide to the Labour Force Survey contains a dictionary of concepts and definitions and covers topics such as survey methodology, data collection, data processing and data quality. It also contains information on products and services, sub-provincial geography descriptions as well as the survey questionnaire.

Description:

This document brings together guidelines and checklists on many issues that need to be considered in the pursuit of quality objectives in the execution of statistical activities. Its focus is on how to assure quality through effective and appropriate design or redesign of a statistical project or program from inception through to data evaluation, dissemination and documentation. These guidelines draw on the collective knowledge and experience of many Statistics Canada employees. It is expected that Quality Guidelines will be useful to staff engaged in the planning and design of surveys and other statistical projects, as well as to those who evaluate and analyze the outputs of these projects.

Description:

This report describes sampling, weighting and estimation procedures used in the Census of Population. It provides operational and theoretical justifications for them, and presents the results of the evaluations of these procedures.

Description:

The Canadian Labour Force Survey (LFS) is the official source of monthly estimates of total employment and unemployment. Following the 2011 census, the LFS underwent a sample redesign to account for the evolution of the population and labour market characteristics, to adjust to changes in the information needs and to update the geographical information used to carry out the survey. The redesign program following the 2011 census culminated with the introduction of a new sample at the beginning of 2015. This report is a reference on the methodological aspects of the LFS, covering stratification, sampling, collection, processing, weighting, estimation, variance estimation and data quality.

Description:

In over 70 years, the methodology and questionnaire, as well as the collection and processing techniques of the Canadian Labour Force Survey have undergone major changes. This document summarizes these changes chronologically and provides references to more detailed information sources. Among the most significant changes were two questionnaire redesigns, which occurred approximately 20 years apart, in 1976 and 1997.

Description:

This report describes sampling and weighting procedures used in the 2011 National Household Survey. It provides operational and theoretical justifications for them, and presents the results of the evaluation studies of these procedures.

Description:

This report describes the quality indicators produced for the 2004 Survey of Household Spending. These quality indicators, such as coefficients of variation, nonresponse rates, slippage rates and imputation rates, help users interpret the survey data.

Description:

This report describes the quality indicators produced for the 2005 Survey of Household Spending. These quality indicators, such as coefficients of variation, nonresponse rates, slippage rates and imputation rates, help users interpret the survey data.

Description:

This report describes the quality indicators produced for the 2006 Survey of Household Spending. These quality indicators, such as coefficients of variation, nonresponse rates, slippage rates and imputation rates, help users interpret the survey data.

Description:

This report describes sampling and weighting procedures used in the 2006 Census. It reviews the history of these procedures in Canadian censuses, provides operational and theoretical justifications for them, and presents the results of the evaluation studies of these procedures.