Statistics by subject – Statistical methods

Filter results by

Help for filters and search
Currently selected filters that can be removed

Keyword(s)

Type of information

2 facets displayed. 0 facets selected.

Year of publication

1 facets displayed. 1 facets selected.

Filter results by

Help for filters and search
Currently selected filters that can be removed

Keyword(s)

Type of information

2 facets displayed. 0 facets selected.

Year of publication

1 facets displayed. 1 facets selected.

Filter results by

Help for filters and search
Currently selected filters that can be removed

Keyword(s)

Year of publication

1 facets displayed. 1 facets selected.

Filter results by

Help for filters and search
Currently selected filters that can be removed

Keyword(s)

Year of publication

1 facets displayed. 1 facets selected.

Other available resources to support your research.

Help for sorting results
Browse our central repository of key standard concepts, definitions, data sources and methods.
Loading
Loading in progress, please wait...
All (28)

All (28) (25 of 28 results)

  • Articles and reports: 81-595-M2003009
    Description:

    This paper examines how the Canadian Adult Education and Training Survey (AETS) can be used to study participation in and impacts of education and training activities for adults.

    Release date: 2003-10-15

  • Technical products: 75F0002M2003006
    Description:

    This document presents the questions, question flows and possible responses for the 2002 Survey of Labour and Income Dynamics (SLID) preliminary questionnaire.

    Release date: 2003-09-09

  • Articles and reports: 11F0019M2003207
    Description:

    The estimation of intergenerational earnings mobility is rife with measurement problems since the research does not observe permanent, lifetime earnings. Nearly all studies make corrections for mean variation in earnings because of the age differences among respondents. Recent works employ average earnings or instrumental variable methods to address the effects of measurement error as a result of transitory earnings shocks and mis-reporting. However, empirical studies of intergenerational mobility have paid no attention to the changes in earnings variance across the life cycle suggested by economic models of human capital investment.

    Using information from the Intergenerational Income Data from Canada and the National Longitudinal Survey and Panel Study of Income Dynamics from the United States, this study finds a strong association between age at observation and estimated earnings persistence. Part of this age-dependence is related to a general increase in transitory earnings variance during the collection of data. An independent effect of life cycle investment is also identified. These findings are then applied to the variation among intergenerational earnings persistence studies. Among studies with similar methodologies, one-third of the variance in published estimates of earnings persistence is attributable to cross-study differences in the age of responding fathers. Finally, these results call into question tests for the importance of credit constraints based on measures of earnings at different points in the life cycle.

    Release date: 2003-08-05

  • Articles and reports: 12-001-X20030016609
    Description:

    To automate the data editing process the so-called error localization problem, i.e., the problem of identifying the erroneous fields in an erroneous record, has to be solved. A paradigm for identifying errors automatically has been proposed by Fellegi and Holt in 1976. Over the years their paradigm has been generalized to: the data of a record should be made to satisfy all edits by changing the values of the variables with the smallest possible sum of reliability weights. A reliability weight of a variable is a non-negative number that expresses how reliable one considers the value of this variable to be. Given this paradigm the resulting mathematical problem has to be solved. In the present paper we examine how vertex generation methods can be used to solve this mathematical problem in mixed data, i.e., a combination of categorical (discrete) and numerical (continuous) data. The main aim of this paper is not to present new results, but rather to combine the ideas of several other papers in order to give a "complete", self-contained description of the use of vertex generation methods to solve the error localization problem in mixed data. In our exposition we will focus on describing how methods for numerical data can be adapted to mixed data.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016612
    Description:

    In the present investigation, we introduce new calibration equations to estimate the population mean in stratified simple random sampling, making use of second order moments of the auxiliary character. We also suggest ways for estimating the variance of the proposed estimator. The resultant new estimator can be more efficient in stratified sampling than the combined regression estimator. Thus, we have extended the idea to double sampling in a stratified population and have studied the simulation results.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016600
    Description:

    International comparability of Official Statistics is important for domestic uses within any country. But international comparability matters also for the international uses of statistics; in particular the development and monitoring of global policies and assessing economic and social development throughout the world. Additionally statistics are used by international agencies and bilateral technical assistance programmes to monitor the impact of technical assistance.The first part of this paper describes how statistical indicators are used by the United Nations and other agencies. The framework of statistical indicators for these purposes is described ans some issues concerning the choice and quality of these indicators are identified.In the past there has been considerable methodological research in support of Official Statistics particularly by the strongest National Statistical Offices and some academics. This has established the basic methodologies for Official Statistics and has led to considerable developments and quality improvements over time. Much has been achieved. However the focus has, to an extent, been on national uses of Official Statistics. These developments have, of course, benefited the international uses, and some specific developments have also occurred. There is however a need to foster more methodological development on the international requirements. In the second part of this paper a number of examples illustrate this need.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016613
    Description:

    The Illinois Department of Employment Security is using small domain estimation techniques to estimate employment at the county or industry divisional level. The estimator is a standard synthetic estimator, based on the ability to match Current Employment Statistics sample data to ES202 administrative records and an assumed model relationship between the two data sources. This paper is a case study that reviews the steps taken to evaluate the appropriateness of the model and the difficulties encountered in linking the two data sources.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016610
    Description:

    In the presence of item nonreponse, unweighted imputation methods are often used in practice but they generally lead to biased estimators under uniform response within imputation classes. Following Skinner and Rao (2002), we propose a bias-adjusted estimator of a population mean under unweighted ratio imputation and random hot-deck imputation and derive linearization variance estimators. A small simulation study is conducted to study the performance of the methods in terms of bias and mean square error. Relative bias and relative stability of the variance estimators are also studied.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016608
    Description:

    This work deals with the unconditional and conditional properties of some well-known small area estimators: expansion, post-stratified ratio, synthetic, composite, sample size dependent and the empirical best linear unbiased predictor (EBLUP). A two-stage sampling design is considered as it is commonly used in household surveys conducted by the National Statistics Institute of Italy. An evaluation is carried out through a simulation based on 1991 Italian census data. The small areas considered are the local labour market areas, which are unplanned domains that cut across the boundaries of the design strata.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030018802
    Description:

    In this Issue is a column where the Editor biefly presents each paper of the current issue of Survey Methodology. As well, it sometimes contain informations on structure or management changes in the journal.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016607
    Description:

    The Korean Economically Active Population Survey (EAPS) has been conducted in order to produce unemployment statistics for large areas such as metropolitan cities and provincial levels. Large areas have been designated as planned domains in the EAPS and local self-government areas (LSGAs) as unplanned domains. In this study, we suggest small area estimation methods to adjust for the unemployment statistics of LSGAs within large areas estimated directly from current EAPS data. We suggest synthetic and composite estimators under the Korean EAPS system, and for the model-based estimator we put forward the hierarchical Bayes (HB) estimator from the general multi-level model. The HB estimator we use here was introduced by You and Rao (2000). The mean square errors of the synthetic and composite estimates are derived from the EAPS data by the Jackknife method, and are used as a measure of accuracy for the small area estimates. Gibbs sampling is used to obtain the HB estimates and their posterior variances, which we use to measure precision for small area estimates. The total unemployment figures of the 10 LSGAs within the ChoongBuk Province produced by the December 2000 EAPS data have been estimated using the small area estimation methods suggested in this study. The reliability of small area estimates is evaluated by the relative standard errors or the relative root mean square errors of these estimates. Here, under the current Korean EAPS system, we suggest that the composite estimates are more reliable than other small area estimates.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016602
    Description:

    The Canadian Labour Force Survey (LFS) produces monthly direct estimates of the unemployment rate at national and provincial levels. The LFS also releases unemployment estimates for subprovincial areas such as census metropolitan areas (CMAs) and census agglomerations (CAs). However, for some subprovincial areas, the direct estimates are not very reliable since the sample size in some areas is quite small. In this paper, a cross-sectional and time-series model is used to borrow strength across areas and time periods to produce model-based unemployment rate estimates for CMAs and CAs. This model is a generalization of a widely used cross-sectional model in small area estimation and includes a random walk or AR(1) model for the random time component. Monthly Employment Insurance (EI) beneficiary data at the CMA or CA level are used as auxiliary covariates in the model. A hierarchical Bayes (HB) approach is employed and the Gibbs sampler is used to generate samples from the joint posterior distribution. Rao-Blackwellized estimators are obtained for the posterior means and posterior variances of the CMA/CA-level unemployment rates. The HB method smoothes the survey estimates and leads to a substantial reduction in standard errors. Base on posterior distributions, bayesian model fitting is also investigated in this paper.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016601
    Description:

    In small area estimation, one uses data from similar domains to estimate the mean in a particular small area. This borrowing of strength is justified by assuming a model that relates the small area means. Here, we suggest a non-informative or objective Bayesian approach to small area estimation. Using this approach, one can estimate population parameters other than means and find sensible estimates of their precision. AMS 1991 subject classifications Primary 62D05; secondary 62C10.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016611
    Description:

    Optimal and approximately optimal fixed-cost Bayesian sampling designs are considered for simultaneous estimation in independent homogeneous Poisson processes. General allocation formulae are developed for a basic Poisson-Gamma model and these are compared with more traditional allocation methods. Techniques for finding representative gamma priors under more general hierarchical models are also discussed. The techniques show that, in many practical situations, these gamma priors provide reasonable approximations to the hierarchical prior and Bayes risk. The methods developed are general enough to apply to a wide variety of models and are not limited to Poisson processes.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016605
    Description:

    In this paper, we examine the effects of model choice on different types of estimators for totals of domains (including small domains or small areas) for a sampled finite population. The paper asks how different estimator types compare for a common underlying model statement. We argue that estimator type - synthetic, generalized regression (GREG), composite, empirical best linear unbiased predicition (EBLUP), hierarchical Bayes, and so on - is one important aspect of domain estimation, and that the choice of the model, including its parameters and effects, is a second aspect, conceptually different from the first. Earlier work has not always made this distinction clear. For a given estimator type, one can derive different estimators, depending on the choice of model. In recent literature, a number of estimator types have been proposed, but there is relatively little impartial comparisons made among them. In this paper, we discuss three types: synthetic, GREG, and, to a limited extent, composite. We show that model improvement - the transition from a weaker to a stronger model - has very different effects on the different estimator types. We also show that the difference in accuracy between the different estimator types depends on the choice of model. For a well-specified model, the difference in accuracy between synthetic and GREG is negligible, but it can be substantial if the model is mis-specified. The synthetic type then tends to be highly inaccurate. We rely partly on theoretical results (for simple random sampling only) and partly on empirical results. The empirical results are based on simulations with repeated samples drawn from two finite populations, one artificially constructed, the other constructed from the real data of the Finnish Labour Force Survey.

    Release date: 2003-07-31

  • Articles and reports: 81-595-M2003005
    Description:

    This paper develops technical procedures that may enable ministries of education to link provincial tests with national and international tests in order to compare standards and report results on a common scale.

    Release date: 2003-05-29

  • Articles and reports: 11F0019M2003199
    Description:

    Using a nationally representative sample of establishments, we have examined whether selected alternative work practices (AWPs) tend to reduce quit rates. Overall, our analysis provides strong evidence of a negative association between these AWPs and quit rates among establishments of more than 10 employees operating in high-skill services. We also found some evidence of a negative association in low-skill services. However, the magnitude of this negative association was reduced substantially when we added an indicator of whether the workplace has a formal policy of information sharing. There was very little evidence of a negative association in manufacturing. While establishments with self-directed workgroups have lower quit rates than others, none of the bundles of work practices considered yielded a negative and statistically significant effect. We surmise that key AWPs might be more successful in reducing labour turnover in technologically complex environments than in low-skill ones.

    Release date: 2003-03-17

  • Articles and reports: 12-001-X20020029055
    Description:

    In this Issue is a column where the Editor biefly presents each paper of the current issue of Survey Methodology. As well, it sometimes contain informations on structure or management changes in the journal.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026418
    Description:

    National statistical offices are often called upon to produce statistics for small geographic areas, in addition to their primary responsibility for measuring the condition of the country as a whole and its major subdivisions. This task presents challenges that are different from those faced in statistical programs aiming primarily at national or provincial statistics. This paper examines these challenges and identifies strategies and approaches for the development of programs of small area statistics. The important foundation of a census of population, as well as the primary role of a consistent geographic infrastructure, are emphasized. Potential sources and methods for the production of small area data in the social, economic and environmental fields are examined. Some organizational and dissemination issues are also discussed.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026432
    Description:

    This paper suggests stratification algorithms that account for a discrepancy between the stratification variable and the study variable when planning a stratified survey design. Two models are proposed for the change between these two variables. One is a log-linear regression model; the other postulates that the study variable and the stratification variable coincide for most units, and that large discrepancies occur for some units. Then, the Lavallée and Hidiroglou (1988) stratification algorithm is modified to incorporate these models in the determination of the optimal sample sizes and of the optimal stratum boundaries for a stratified sampling design. An example illustrates the performance of the new stratification algorithm. A discussion of the numerical implementation of this algorithm is also presented.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026431
    Description:

    When stand-alone sampling frames that list all establishments and their measures of size are available, establishment surveys typically use the Hansen-Hurwitz (HH) PPS (probability proportional to size) estimator to estimate the volume of transactions that establishments have with populations. This paper proposes the network sampling (NS) version of the HH estimator as a potential competitor of the PPS estimator. The NS estimator depends on the population survey-generated establishment frame that lists households and their selection probabilities in a population sample survey, and the number of transactions, if any, of each household with each establishment. A statistical model is developed in this paper to compare the efficiencies of the HH and NS estimators in single-stage and two-stage establishment sample surveys assuming the stand-alone sampling frame and the population survey-generated frame are flawless in coverage and size measures.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026423
    Description:

    The reputation of a national statistical office (NSO) depends very much on the quality of the service it provides. Quality has to be a core value: providing a high quality service has to be the natural way of doing business. It has to be embedded in the culture of the NSO.

    The paper will outline what is meant by a high quality statistical service. It will also explore those factors that are important to ensuring a quality culture in an NSO. In particular, it will outline the activities and experiences of the Australian Bureau of Statistics in maintaining a quality culture.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026434
    Description:

    In theory, it is customary to define general regression estimators in terms of full-rank weighting models (i.e., the design matrix that corresponds to the weighting model is of full rank). For such weighting models, it is well known that the general regression weights reproduce the (known) population totals of the auxiliary variables involved. In practice, however, the weighting model often is not of full rank, especially when the weighting model is for incomplete post-stratification. By means of the theory of generalized inverse matrices, it is shown under which circumstances this consistency property remains valid. In this paper,, we discuss the non-trivial example of consistent weighting between persons and households as proposed by Lemaître and Dufour (1987). We then show how the theory is implemented in Bascula.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026428
    Description:

    The analysis of survey data from different geographical areas where the data from each area are polychotomous can be easily performed using hierarchical Bayesian models, even if there are small cell counts in some of these areas. However, there are difficulties when the survey data have missing information in the form of non-response, especially when the characteristics of the respondents differ from the non-respondents. We use the selection approach for estimation when there are non-respondents because it permits inference for all the parameters. Specifically, we describe a hierarchical Bayesian model to analyse multinomial non-ignorable non-response data from different geographical areas; some of them can be small. For the model, we use a Dirichlet prior density for the multinomial probabilities and a beta prior density for the response probabilities. This permits a 'borrowing of strength' of the data from larger areas to improve the reliability in the estimates of the model parameters corresponding to the smaller areas. Because the joint posterior density of all the parameters is complex, inference is sampling-based and Markov chain Monte Carlo methods are used. We apply our method to provide an analysis of body mass index (BMI) data from the third National Health and Nutrition Examination Survey (NHANES III). For simplicity, the BMI is categorized into 3 natural levels, and this is done for each of 8 age-race-sex domains and 34 counties. We assess the performance of our model using the NHANES III data and simulated examples, which show our model works reasonably well.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026429
    Description:

    In most telephone time-use surveys, respondents are called on one day and asked to report on their activities during the previous day. Given that most respondents are not available on their initial calling day, this feature of telephone time-use surveys introduces the possibility that the probability of interviewing the respondent about a given reference day is correlated with the activities on that reference day. Furthermore, non-contact bias is a more important consideration for time-use surveys than for other surveys, because time-use surveys cannot accept proxy responses. Therefore, it is essential that telephone time-use surveys have a strategy for making subsequent attempts to contact respondents. A contact strategy specifies the contact schedule and the field period. Previous literature has identified two schedules for making subsequent attempts: a convenient-day schedule and a designated-day schedule. Most of these articles recommend the designated-day schedule, but there is little evidence to support this viewpoint. In this paper, we use computer simulations to examine the bias associated with the convenient-day schedule and three variations of the designated-day schedule. The results support using a designated-day schedule, and validate the recommendations of the previous literature. The convenient-day schedule introduces systematic bias: time spent in activities done away from home tends to be overestimated. More importantly, estimates generated using the convenient-day schedule are sensitive to the variance of the contact probability. In contrast, a designated-day-with-postponement schedule generates very little bias, and is robust to a wide range of assumptions about the pattern of activities across days of the week.

    Release date: 2003-01-29

Data (0)

Data (0) (0 results)

Your search for "" found no results in this section of the site.

You may try:

Analysis (27)

Analysis (27) (25 of 27 results)

  • Articles and reports: 81-595-M2003009
    Description:

    This paper examines how the Canadian Adult Education and Training Survey (AETS) can be used to study participation in and impacts of education and training activities for adults.

    Release date: 2003-10-15

  • Articles and reports: 11F0019M2003207
    Description:

    The estimation of intergenerational earnings mobility is rife with measurement problems since the research does not observe permanent, lifetime earnings. Nearly all studies make corrections for mean variation in earnings because of the age differences among respondents. Recent works employ average earnings or instrumental variable methods to address the effects of measurement error as a result of transitory earnings shocks and mis-reporting. However, empirical studies of intergenerational mobility have paid no attention to the changes in earnings variance across the life cycle suggested by economic models of human capital investment.

    Using information from the Intergenerational Income Data from Canada and the National Longitudinal Survey and Panel Study of Income Dynamics from the United States, this study finds a strong association between age at observation and estimated earnings persistence. Part of this age-dependence is related to a general increase in transitory earnings variance during the collection of data. An independent effect of life cycle investment is also identified. These findings are then applied to the variation among intergenerational earnings persistence studies. Among studies with similar methodologies, one-third of the variance in published estimates of earnings persistence is attributable to cross-study differences in the age of responding fathers. Finally, these results call into question tests for the importance of credit constraints based on measures of earnings at different points in the life cycle.

    Release date: 2003-08-05

  • Articles and reports: 12-001-X20030016609
    Description:

    To automate the data editing process the so-called error localization problem, i.e., the problem of identifying the erroneous fields in an erroneous record, has to be solved. A paradigm for identifying errors automatically has been proposed by Fellegi and Holt in 1976. Over the years their paradigm has been generalized to: the data of a record should be made to satisfy all edits by changing the values of the variables with the smallest possible sum of reliability weights. A reliability weight of a variable is a non-negative number that expresses how reliable one considers the value of this variable to be. Given this paradigm the resulting mathematical problem has to be solved. In the present paper we examine how vertex generation methods can be used to solve this mathematical problem in mixed data, i.e., a combination of categorical (discrete) and numerical (continuous) data. The main aim of this paper is not to present new results, but rather to combine the ideas of several other papers in order to give a "complete", self-contained description of the use of vertex generation methods to solve the error localization problem in mixed data. In our exposition we will focus on describing how methods for numerical data can be adapted to mixed data.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016612
    Description:

    In the present investigation, we introduce new calibration equations to estimate the population mean in stratified simple random sampling, making use of second order moments of the auxiliary character. We also suggest ways for estimating the variance of the proposed estimator. The resultant new estimator can be more efficient in stratified sampling than the combined regression estimator. Thus, we have extended the idea to double sampling in a stratified population and have studied the simulation results.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016600
    Description:

    International comparability of Official Statistics is important for domestic uses within any country. But international comparability matters also for the international uses of statistics; in particular the development and monitoring of global policies and assessing economic and social development throughout the world. Additionally statistics are used by international agencies and bilateral technical assistance programmes to monitor the impact of technical assistance.The first part of this paper describes how statistical indicators are used by the United Nations and other agencies. The framework of statistical indicators for these purposes is described ans some issues concerning the choice and quality of these indicators are identified.In the past there has been considerable methodological research in support of Official Statistics particularly by the strongest National Statistical Offices and some academics. This has established the basic methodologies for Official Statistics and has led to considerable developments and quality improvements over time. Much has been achieved. However the focus has, to an extent, been on national uses of Official Statistics. These developments have, of course, benefited the international uses, and some specific developments have also occurred. There is however a need to foster more methodological development on the international requirements. In the second part of this paper a number of examples illustrate this need.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016613
    Description:

    The Illinois Department of Employment Security is using small domain estimation techniques to estimate employment at the county or industry divisional level. The estimator is a standard synthetic estimator, based on the ability to match Current Employment Statistics sample data to ES202 administrative records and an assumed model relationship between the two data sources. This paper is a case study that reviews the steps taken to evaluate the appropriateness of the model and the difficulties encountered in linking the two data sources.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016610
    Description:

    In the presence of item nonreponse, unweighted imputation methods are often used in practice but they generally lead to biased estimators under uniform response within imputation classes. Following Skinner and Rao (2002), we propose a bias-adjusted estimator of a population mean under unweighted ratio imputation and random hot-deck imputation and derive linearization variance estimators. A small simulation study is conducted to study the performance of the methods in terms of bias and mean square error. Relative bias and relative stability of the variance estimators are also studied.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016608
    Description:

    This work deals with the unconditional and conditional properties of some well-known small area estimators: expansion, post-stratified ratio, synthetic, composite, sample size dependent and the empirical best linear unbiased predictor (EBLUP). A two-stage sampling design is considered as it is commonly used in household surveys conducted by the National Statistics Institute of Italy. An evaluation is carried out through a simulation based on 1991 Italian census data. The small areas considered are the local labour market areas, which are unplanned domains that cut across the boundaries of the design strata.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030018802
    Description:

    In this Issue is a column where the Editor biefly presents each paper of the current issue of Survey Methodology. As well, it sometimes contain informations on structure or management changes in the journal.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016607
    Description:

    The Korean Economically Active Population Survey (EAPS) has been conducted in order to produce unemployment statistics for large areas such as metropolitan cities and provincial levels. Large areas have been designated as planned domains in the EAPS and local self-government areas (LSGAs) as unplanned domains. In this study, we suggest small area estimation methods to adjust for the unemployment statistics of LSGAs within large areas estimated directly from current EAPS data. We suggest synthetic and composite estimators under the Korean EAPS system, and for the model-based estimator we put forward the hierarchical Bayes (HB) estimator from the general multi-level model. The HB estimator we use here was introduced by You and Rao (2000). The mean square errors of the synthetic and composite estimates are derived from the EAPS data by the Jackknife method, and are used as a measure of accuracy for the small area estimates. Gibbs sampling is used to obtain the HB estimates and their posterior variances, which we use to measure precision for small area estimates. The total unemployment figures of the 10 LSGAs within the ChoongBuk Province produced by the December 2000 EAPS data have been estimated using the small area estimation methods suggested in this study. The reliability of small area estimates is evaluated by the relative standard errors or the relative root mean square errors of these estimates. Here, under the current Korean EAPS system, we suggest that the composite estimates are more reliable than other small area estimates.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016602
    Description:

    The Canadian Labour Force Survey (LFS) produces monthly direct estimates of the unemployment rate at national and provincial levels. The LFS also releases unemployment estimates for subprovincial areas such as census metropolitan areas (CMAs) and census agglomerations (CAs). However, for some subprovincial areas, the direct estimates are not very reliable since the sample size in some areas is quite small. In this paper, a cross-sectional and time-series model is used to borrow strength across areas and time periods to produce model-based unemployment rate estimates for CMAs and CAs. This model is a generalization of a widely used cross-sectional model in small area estimation and includes a random walk or AR(1) model for the random time component. Monthly Employment Insurance (EI) beneficiary data at the CMA or CA level are used as auxiliary covariates in the model. A hierarchical Bayes (HB) approach is employed and the Gibbs sampler is used to generate samples from the joint posterior distribution. Rao-Blackwellized estimators are obtained for the posterior means and posterior variances of the CMA/CA-level unemployment rates. The HB method smoothes the survey estimates and leads to a substantial reduction in standard errors. Base on posterior distributions, bayesian model fitting is also investigated in this paper.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016601
    Description:

    In small area estimation, one uses data from similar domains to estimate the mean in a particular small area. This borrowing of strength is justified by assuming a model that relates the small area means. Here, we suggest a non-informative or objective Bayesian approach to small area estimation. Using this approach, one can estimate population parameters other than means and find sensible estimates of their precision. AMS 1991 subject classifications Primary 62D05; secondary 62C10.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016611
    Description:

    Optimal and approximately optimal fixed-cost Bayesian sampling designs are considered for simultaneous estimation in independent homogeneous Poisson processes. General allocation formulae are developed for a basic Poisson-Gamma model and these are compared with more traditional allocation methods. Techniques for finding representative gamma priors under more general hierarchical models are also discussed. The techniques show that, in many practical situations, these gamma priors provide reasonable approximations to the hierarchical prior and Bayes risk. The methods developed are general enough to apply to a wide variety of models and are not limited to Poisson processes.

    Release date: 2003-07-31

  • Articles and reports: 12-001-X20030016605
    Description:

    In this paper, we examine the effects of model choice on different types of estimators for totals of domains (including small domains or small areas) for a sampled finite population. The paper asks how different estimator types compare for a common underlying model statement. We argue that estimator type - synthetic, generalized regression (GREG), composite, empirical best linear unbiased predicition (EBLUP), hierarchical Bayes, and so on - is one important aspect of domain estimation, and that the choice of the model, including its parameters and effects, is a second aspect, conceptually different from the first. Earlier work has not always made this distinction clear. For a given estimator type, one can derive different estimators, depending on the choice of model. In recent literature, a number of estimator types have been proposed, but there is relatively little impartial comparisons made among them. In this paper, we discuss three types: synthetic, GREG, and, to a limited extent, composite. We show that model improvement - the transition from a weaker to a stronger model - has very different effects on the different estimator types. We also show that the difference in accuracy between the different estimator types depends on the choice of model. For a well-specified model, the difference in accuracy between synthetic and GREG is negligible, but it can be substantial if the model is mis-specified. The synthetic type then tends to be highly inaccurate. We rely partly on theoretical results (for simple random sampling only) and partly on empirical results. The empirical results are based on simulations with repeated samples drawn from two finite populations, one artificially constructed, the other constructed from the real data of the Finnish Labour Force Survey.

    Release date: 2003-07-31

  • Articles and reports: 81-595-M2003005
    Description:

    This paper develops technical procedures that may enable ministries of education to link provincial tests with national and international tests in order to compare standards and report results on a common scale.

    Release date: 2003-05-29

  • Articles and reports: 11F0019M2003199
    Description:

    Using a nationally representative sample of establishments, we have examined whether selected alternative work practices (AWPs) tend to reduce quit rates. Overall, our analysis provides strong evidence of a negative association between these AWPs and quit rates among establishments of more than 10 employees operating in high-skill services. We also found some evidence of a negative association in low-skill services. However, the magnitude of this negative association was reduced substantially when we added an indicator of whether the workplace has a formal policy of information sharing. There was very little evidence of a negative association in manufacturing. While establishments with self-directed workgroups have lower quit rates than others, none of the bundles of work practices considered yielded a negative and statistically significant effect. We surmise that key AWPs might be more successful in reducing labour turnover in technologically complex environments than in low-skill ones.

    Release date: 2003-03-17

  • Articles and reports: 12-001-X20020029055
    Description:

    In this Issue is a column where the Editor biefly presents each paper of the current issue of Survey Methodology. As well, it sometimes contain informations on structure or management changes in the journal.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026418
    Description:

    National statistical offices are often called upon to produce statistics for small geographic areas, in addition to their primary responsibility for measuring the condition of the country as a whole and its major subdivisions. This task presents challenges that are different from those faced in statistical programs aiming primarily at national or provincial statistics. This paper examines these challenges and identifies strategies and approaches for the development of programs of small area statistics. The important foundation of a census of population, as well as the primary role of a consistent geographic infrastructure, are emphasized. Potential sources and methods for the production of small area data in the social, economic and environmental fields are examined. Some organizational and dissemination issues are also discussed.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026432
    Description:

    This paper suggests stratification algorithms that account for a discrepancy between the stratification variable and the study variable when planning a stratified survey design. Two models are proposed for the change between these two variables. One is a log-linear regression model; the other postulates that the study variable and the stratification variable coincide for most units, and that large discrepancies occur for some units. Then, the Lavallée and Hidiroglou (1988) stratification algorithm is modified to incorporate these models in the determination of the optimal sample sizes and of the optimal stratum boundaries for a stratified sampling design. An example illustrates the performance of the new stratification algorithm. A discussion of the numerical implementation of this algorithm is also presented.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026431
    Description:

    When stand-alone sampling frames that list all establishments and their measures of size are available, establishment surveys typically use the Hansen-Hurwitz (HH) PPS (probability proportional to size) estimator to estimate the volume of transactions that establishments have with populations. This paper proposes the network sampling (NS) version of the HH estimator as a potential competitor of the PPS estimator. The NS estimator depends on the population survey-generated establishment frame that lists households and their selection probabilities in a population sample survey, and the number of transactions, if any, of each household with each establishment. A statistical model is developed in this paper to compare the efficiencies of the HH and NS estimators in single-stage and two-stage establishment sample surveys assuming the stand-alone sampling frame and the population survey-generated frame are flawless in coverage and size measures.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026423
    Description:

    The reputation of a national statistical office (NSO) depends very much on the quality of the service it provides. Quality has to be a core value: providing a high quality service has to be the natural way of doing business. It has to be embedded in the culture of the NSO.

    The paper will outline what is meant by a high quality statistical service. It will also explore those factors that are important to ensuring a quality culture in an NSO. In particular, it will outline the activities and experiences of the Australian Bureau of Statistics in maintaining a quality culture.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026434
    Description:

    In theory, it is customary to define general regression estimators in terms of full-rank weighting models (i.e., the design matrix that corresponds to the weighting model is of full rank). For such weighting models, it is well known that the general regression weights reproduce the (known) population totals of the auxiliary variables involved. In practice, however, the weighting model often is not of full rank, especially when the weighting model is for incomplete post-stratification. By means of the theory of generalized inverse matrices, it is shown under which circumstances this consistency property remains valid. In this paper,, we discuss the non-trivial example of consistent weighting between persons and households as proposed by Lemaître and Dufour (1987). We then show how the theory is implemented in Bascula.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026428
    Description:

    The analysis of survey data from different geographical areas where the data from each area are polychotomous can be easily performed using hierarchical Bayesian models, even if there are small cell counts in some of these areas. However, there are difficulties when the survey data have missing information in the form of non-response, especially when the characteristics of the respondents differ from the non-respondents. We use the selection approach for estimation when there are non-respondents because it permits inference for all the parameters. Specifically, we describe a hierarchical Bayesian model to analyse multinomial non-ignorable non-response data from different geographical areas; some of them can be small. For the model, we use a Dirichlet prior density for the multinomial probabilities and a beta prior density for the response probabilities. This permits a 'borrowing of strength' of the data from larger areas to improve the reliability in the estimates of the model parameters corresponding to the smaller areas. Because the joint posterior density of all the parameters is complex, inference is sampling-based and Markov chain Monte Carlo methods are used. We apply our method to provide an analysis of body mass index (BMI) data from the third National Health and Nutrition Examination Survey (NHANES III). For simplicity, the BMI is categorized into 3 natural levels, and this is done for each of 8 age-race-sex domains and 34 counties. We assess the performance of our model using the NHANES III data and simulated examples, which show our model works reasonably well.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026429
    Description:

    In most telephone time-use surveys, respondents are called on one day and asked to report on their activities during the previous day. Given that most respondents are not available on their initial calling day, this feature of telephone time-use surveys introduces the possibility that the probability of interviewing the respondent about a given reference day is correlated with the activities on that reference day. Furthermore, non-contact bias is a more important consideration for time-use surveys than for other surveys, because time-use surveys cannot accept proxy responses. Therefore, it is essential that telephone time-use surveys have a strategy for making subsequent attempts to contact respondents. A contact strategy specifies the contact schedule and the field period. Previous literature has identified two schedules for making subsequent attempts: a convenient-day schedule and a designated-day schedule. Most of these articles recommend the designated-day schedule, but there is little evidence to support this viewpoint. In this paper, we use computer simulations to examine the bias associated with the convenient-day schedule and three variations of the designated-day schedule. The results support using a designated-day schedule, and validate the recommendations of the previous literature. The convenient-day schedule introduces systematic bias: time spent in activities done away from home tends to be overestimated. More importantly, estimates generated using the convenient-day schedule are sensitive to the variance of the contact probability. In contrast, a designated-day-with-postponement schedule generates very little bias, and is robust to a wide range of assumptions about the pattern of activities across days of the week.

    Release date: 2003-01-29

  • Articles and reports: 12-001-X20020026433
    Description:

    Sitter and Skinner (1994) present a method which applies linear programming to designing surveys with multi-way stratification, primarily in situations where the desired sample size is less than or only slightly larger than the total number of stratification cells. The idea in their approach is simple, easily understood and easy to apply. However, the main practical constraint of their approach is that it rapidly becomes expensive in terms of magnitude of computation as the number of cells in the multi-way stratification increases, to the extent that it cannot be used in most realistic situations. In this article, we extend this linear programming approach and develop methods to reduce the amount of computation so that very large problems become feasible.

    Release date: 2003-01-29

Reference (1)

Reference (1) (1 result)

Date modified: