Analysis

Description:

A finite population of size N is supposed to contain M (unknown) units of a specified category A (say) constituting a domain with mean \mu. A procedure which involves drawing units using simple random sampling without replacement till a preassigned number of members of the domain is reached is proposed. An unbiased estimator of \mu is also derived. This is seen to be superior to the corresponding possibly biased estimator based on a comparable SRSWOR scheme with a fixed number of draws. The proposed scheme is also shown to admit unbiased estimators of M and the domain total T.

Description:

In response to a need for data on disabled persons in Canada, Statistics Canada undertook a program to create a disability database. This includes using supplements to the Canadian Labour Force Survey in the Fall of 1983 and the Spring of 1984, as well as including questions on the 1986 Census of Population. A general discussion of the background and content of the survey is presented. A comparison of screening methodologies conducted by Statistics Canada in November 1982 and January 1983 is presented and the results are compared.

Description:

This presentation describes the important and urgent task of providing useful expressions for analytical statistics for complex sample designs. The following topics are discussed: effects of complex designs, sampling error for analytical statistics, subclasses involved in analytical statistics, comparisons of paired means, computation of analytical statistics and categorical data analysis.

Description:

Univariate statistical models, linear regression models and generalized linear models are briefly reviewed. Examples of a two-way analysis of variance, a three-way analysis of variance and logistic regression for a three way layout are given.

Description:

This paper proposes a modification to the method of Denton (1971) for adjusting sub-annual series to yearly totals. These totals originate from more reliable sources and constitute annual benchmarks. The benchmarked series derived according to the modified method is more parallel to the unbenchmarked series than this is the case with the original method. An additive and a proportional variant of the method are presented. These can easily be adapted for flow, stock and index series. Also presented are a few recommendations about the preliminary benchmarking of current data and the management of “historical” estimates of the series.

Description:

Using data from the Family Expenditures Surveys over time, consumer expenditures on in-home and transportation energy from 1969 to 1982 are being studied. This article briefly summarizes some of the procedures being used to explore the data, summarize it and develop insights into shifts in consumption for policy implications purposes. With such a complex data set and such a complex, multi-faceted subject for analysis some effort must be made to reduce information flows and at the same time increase the information content of each factor of both input and output in the analyses.

Description:

Standard chisquared (X^2) or likelihood ratio (G^2) tests for logistic regression analysis, involving a binary response variable, are adjusted to take account of the survey design. The adjustments are based on certain generalized design effects. The adjusted statistics are utilized to analyse some data from the October 1980 Canadian Labour Force Survey (LFS). The Wald statistic, which also takes the survey design into account, is also examined for goodness-of-fit of the model and for testing hypotheses on the parameters of the assumed model. Logistic regression diagnostics to detect any outlying cell proportions in the table and influential points in the factor space are applied to the LFS data, after making necessary adjustments to account for the survey design.

Description:

Most sample surveys conducted by organizations such as Statistics Canada or the U.S. Bureau of the Census employ complex designs. The design-based approach to statistical inference, typically the institutional standard of inference for simple population statistics such as means and totals, may be extended to parameters of analytic models as well. Most of this paper focuses on application of design-based inferences to such models, but rationales are offered for use of model-based alternatives in some instances, by way of explanation for the author’s observation that both modes of inference are used in practice at his own institution.

Within the design-based approach to inference, the paper briefly describes experience with linear regression analysis. Recently, variance computations for a number of surveys of the Census Bureau have been implemented through “replicate weighting”; the principal application has been for variances of simple statistics, but this technique also facilitates variance computation for virtually any complex analytic model. Finally, approaches and experience with log-linear models are reported.

Description:

The paper shows different estimation methods for complex survey designs. Among others, estimation of mean, ratio and regression coefficient is presented. The standard errors are estimated by different methods: the ordinary least squares procedure, the stratified weighted sample procedure, the stratified unit weight procedure, etc. Theory of large samples and conditions to apply it are also presented.

Description:

Cost models to determine an optimum allocation of the sample among stages in cluster samples are considered. Results from a proposed cost model, which directly considers the implications of follow-up visits to sample clusters as well as other travel to and from the field by data collectors, are compared with results from existing cost models. The proposed model generally calls for fewer clusters with more elements selected per cluster than the existing models.

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: This paper looks at how extreme weather claims have impacted the insurance industry in Canada from 2020 to 2023. A specific focus is given to homeowners insurance, which during this time, homeowners insurance consumer price index exceeded all items inflation. Insurer profitability and consumer impacts are analyzed given the significant uptick in catastrophic claims. All four years analyzed ranked in the top 10 most costly claims years in Canadian history, costing billions each year. In addition, other contributing factors are explored to provide a robust overview of trends in profitability of the insurance industry.

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: Using the Employment Insurance Coverage Survey, this paper studies the share of spouses or partners of biological or adoptive mothers who claimed or intended to claim parental benefits from 2006 to 2022.

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: Release published in The Daily – Statistics Canada’s official release bulletin

Description: An alternative to the direct selection of sample is suggested, which while retaining the efficiency at the same level simplifies the selection and variance estimation processes in a wide variety of situations. If n* is the largest feasible pPS sample size that can be drawn from a given population of size N, then the proposed method entails selection of m (=N - n*) units using a pPS scheme and rejecting these units from the population such that the remainder is a pPS sample of n* units; the final sample of n units is then selected as a subsample from the remainder set. This method for selecting the pPS sample can be seen as an analogue of SRS where it is well known that the “unsampled” part of the population as well as any subsample from this part are also SRS from the entire population when SRS is the procedure used. The method is very practical for situations where m is less than the actual sample size n. Moreover, the method has the additional advantage in the context of continuing surveys, e.g. Canadian Labour Force Survey (LFS), where the number of primary sampling units (PSU’s) may have to be increased (or decreased) subsequent to the initial selection of the sample. The method also has advantages in the case of sample rotation. Main features of the proposed scheme and its limitations are given. Efficiency of the method is also evaluated empirically.

Description: The problem considered in this paper is the estimation of various agricultural variables using a multiple frame approach. The list frame is completely contained within the area frame. The stratification for the list and area frames are based on different criteria. Overall, the multiple frame shows some gains in terms of variance over the area frame. However, a more careful analysis reveals problem areas associated with the list frame such as the method of stratification and the degeneration of list strata over time.

Description: This article presents the methodology and analysis of two major pretests undertaken in order to compare the effectiveness of different interviewing methods and to assess the feasibility of collecting information which would meet Victimization Survey information requirements.

Description:

A Hot Deck imputation procedure is defined to be one where an incomplete response is completed by using values from one or more other records on the same file and the choice of these records varies with the record requiring imputation.

General approaches to Hot Deck imputation are outlined, with emphasis on the interaction between the edit constraints and the imputation procedures. Distance functions can be constructed on a mixture of categorical and numeric fields, can be modified to take account of the relative importance of fields and can discriminate against less desirable donors. Matching fields may be correlated with missing fields, may be linked with missing fields by edits or may be natural stratification variables; but increasing the number of matching fields does not necessarily result in a better match. It is important to audit the imputation process and to summarize its performance.

Hot Deck procedures should be evaluated to study the bias and reliability of the estimates, donor usage and frequency of imputation failure in terms of a variety of conditions of the data and variations of the imputation procedure. It appears that the only generally available approach to evaluation is by simulation.

Description:

This paper looks at the current state of development of social statistics in Canada. Some key concepts related to statistics and social information are defined and discussed. The availability and analysis of administrative data is highlighted, along with the need for social surveys. Suggestions are made about the types of data analysis needed for the development of social decision models to meet policy requirements. Finally, an outline of priorities for future work toward the effective use of social statistics is given.

Description: This paper discusses the management of information within the context of the information industry and indicates some likely future trends related thereto. The information industry itself is first briefly described. Then the process used in producing information, the organizational structure required for such production, and the legislation relating to the information industry are discussed in turn. Finally, some approaches to solving the problems of the future are suggested.

Description: This paper includes a description of interviewer techniques and procedures used to minimize non-response, an outline of methods used to monitor and control non-response, and a discussion of how non-respondents are treated in the data processing and estimation stages of the Canadian Labour Force Survey. Recent non-response rates as well as data on the characteristics of non-respondents are also given. It is concluded that a yearly non-response rate of approximately 5 percent is probably the best that can be achieved in the Labour Force Survey.

Description: Two methods for estimating the correlated response variance of a survey estimator are studied by way of both theoretical comparison and empirical investigation. The variance of these estimators is discussed and the effects of outliers examined. Finally, an improved estimator is developed and evaluated.

Description: Let U = {1, 2, …, i, …, N} be a finite population of N identifiable units. A known “size measure” x_i is associated with unit i; i = 1, 2, ..., N. A sampling procedure for selecting a sample of size n (2 < n < N) with probability proportional to size (PPS) and without replacement (WOR) from the population is proposed. With this method, the inclusion probability is proportional to size (IPPS) for each unit in the population.

Description: Approximate cutoff rules for stratifying a population into a take-all and take-some universe have been given by Dalenius (1950) and Glasser (1962). They expressed the cutoff value (that value which delineates the boundary of the take-all and take-some) as a function of the mean, the sampling weight and the population variance. Their cutoff values were derived on the assumption that a single random sample of size n was to be drawn without replacement from the population of size N.

In the present context, exact and approximate cutoff rules have been worked out for a similar situation. Rather than providing the sample size of the sample, the precision (coefficient of variation) is given. Note that in many sampling situations, the sampler is given a set of objectives in terms of reliability and not sample size. The result is particularly useful for determining the take-all - take-some boundary for samples drawn from a known population. The procedure is also extended to ratio estimation.

Description: Papers in this series provide background discussions of the methods used to develop data for economic, health, and social analytical studies at Statistics Canada. They are intended to provide readers with information on the statistical methods, standards and definitions used to develop databases for research purposes. All papers in this series have undergone peer and institutional review to ensure that they conform to Statistics Canada's mandate and adhere to generally accepted standards of good professional practice.

Description: This article presents results from the first Survey of Sexual Misconduct in the Canadian Armed Forces. Namely, the prevalence of general sexualized behaviour in the workplace; discrimination on the basis of sex, sexual orientation, or gender identity; personal experiences of discrimination or sexualized behaviour; the prevalence of sexual assault; and knowledge of policies on sexual misconduct and perceptions of responses to sexual misconduct are examined. Where possible, results are analyzed by sex, environmental command, type of service, age, rank, and number of years of service.

Description: This publication features short, informative articles focusing on specific justice-related issues. For more in-depth articles on justice in Canada, see also Juristat, Catalogue no. 85-002-X.

Description:

Each issue contains a short article highlighting statistical insights on themes relating to agriculture, food and rural issues.

Description: This publication presents energy balance sheets in natural units and heat equivalents in primary and secondary forms, by province. Each balance sheet shows data on production, trade, interprovincial movements, conversion and consumption by sector. Analytical tables and details on non-energy products are also included. It includes explanatory notes, a historical energy summary table and data analysis. The publication also presents data on natural gas liquids, electricity generated from fossil fuels, solid wood waste and spent pulping liquor.

Description: The Departmental Sustainable Development Strategy (DSDS) outlines departmental actions, with measurable performance indicators, that support the implementation strategies of the 2022-2026 Federal Sustainable Development Strategy. The DSDS further outlines Statistics Canada’s sustainable development vision to produce data to help track whether Canada is moving toward a more sustainable future and highlights projects with links to supporting sustainable development goals.

Description: This series provides detailed documentation on the issues, concepts, methodology, data quality and other relevant research related to household expenditures from the Survey of Household Spending, the Homeowner Repair and Renovation Survey and the Food Expenditure Survey.

Description: This report summarizes the annual achievements of the Methodology Research and Development Program (MRDP) sponsored by the Modern Statistical Methods and Data Science Branch at Statistics Canada. This program covers research and development activities in statistical methods with potentially broad application in the agency’s statistical programs; these activities would otherwise be less likely to be carried out during the provision of regular methodology services to those programs. The MRDP also includes activities that provide support in the application of past successful developments in order to promote the use of the results of research and development work. Selected prospective research activities are also presented.

Description: The series covers environment accounts and indicators, environmental surveys, spatial environmental information and other research related to environmental statistics. The technical paper series is intended to stimulate discussion on a range of environmental topics.

Description: This series of analytical articles provides insights on the socio-economic environment in rural communities in Canada. New articles will be released periodically.