Keyword search

The U.S. Bureau of the Census implemented major changes to the design of the Survey of Income and Program Participation (SIPP) with the panel begun in 1996. The revised survey design emphasized longitudinal applications and the Census Bureau attempted to understand and resolve the seam bias common to longitudinal surveys. In addition to the substantive and administrative redesign of the survey, the Census Bureau is improving the data processing procedures which yield microdata files for the public to analyse. The wave-by-wave data products are being edited and imputed with a longitudinal element rather than cross-sectionally, carrying forward information from a prior wave that is missing in the current wave. The longitudinal data products will be enhanced, both by the redesigned survey and new processing procedures. Simple methods of imputing data over time are being replaced with more sophisticated methods that do not attenuate seam bias. The longitudinal sample is expanding to include more observations which were nonrespondents in one or more waves. Longitudinal weights will be applied to the file to support person-based longitudinal analysis for calendar years or longer periods of time (up to four years).

This article extends and further develops the method proposed by Pfeffermann, Skinner and Humphreys (1998) for the estimation of gross flows in the presence of classification errors. The main feature of that method is the use of auxiliary information at the individual level which circumvents the need for validation data for estimating the misclassification rates. The new developments in this article are the establishment of conditions for model identification, a study of the properties of a model goodness of fit statistic and modifications to the sample likelihood to account for missing data and informative sampling. The new developments are illustrated by a small Monte-Carlo simulation study.

The study of social mobility, between labour market statuses or between income levels, for example, is often based on the analysis of mobility matrices. When comparing these transition matrices, with a view to evaluating behavioural changes, one often forgets that the data derive from a sample survey and are therefore affected by sampling variances. Similarly, it is assumed that the responses collected correspond to the ' true value.'

A longitudinal study on a cohort of pupils in the secondary school has been conducted in an Italian region since 1986 in order to study the transition from school to working life. The information have been collected at every sweep by a mail questionnaire and, at the final sweep, by a face-to-face interview, where retrospective questions referring back to the whole observation period have been asked. The gross flows between different discrete states - still in the school system, in the labour force without a job, in the labour force with a job - may then be estimated both from prospective and retrospective data, and the recall effect may be evaluated. Moreover, the conditions observed by the two different techniques may be regarded as two indicators of the 'true' unobservable condition, thus leading to the specification and estimation of a latent class model. In this framework, a Markov chain hypothesis may be introduced and evaluated in order to estimate the transition probabilities between the states, once they are corrected or the classification errors. Since the information collected by mail show a given amount of missing data in terms of unit nonresponse, the 'missing' category is also introduced in the model specification.

The log-linear modelling of categorical longitudinal survey data on income is studied. An emphasis is on inference about change. Special attention is paid to modelling of longitudinal data from two waves. A small illustration is based on data from the Canadian Survey of Labour and Income Dynamics.

The purpose of the present study is to utilize panel data from the Current Population Survey (CPS) to examine the effects of unit nonresponse. Because most nonrespondents to the CPS are respondents during at least one month-in-sample, data from other months can be used to compare the characteristics of complete respondents and panel nonrespondents and to evaluate nonresponse adjustment procedures. In the current paper we present analyses utilizing CPS panel data to illustrate the effects of unit nonresponse. After adjusting for nonresponse, additional comparisons are also made to evaluate the effects of nonresponse adjustment. The implications of the findings and suggestions for further research are discussed.

The disseminated results of annual business surveys inevitably contain statistics that are changing. Since the economic sphere is increasingly dynamic, a simple difference of aggregates between n-l and n is no longer sufficient to provide an overall description of what has happened. The change calculation module in the new generation of annual business surveys divides overall change into various components (births, deaths, inter-industry migration) and calculates change on the basis of a constant field, assigning special importance to restructurings. The main difficulties lie in establishing subsamples, reweighting, calibrating according to calculable changes, and taking account of restructuring.

We address the problem of estimation for the income dynamics statistics calculated from complex longitudinal surveys. In addition, we compare two design-based estimators of longitudinal proportions and transition rates in terms of variability under large attrition rates. One estimator is based on the cross-sectional samples for the estimation of the income class boundaries at each time period and on the longitudinal sample for the estimation of the longitudinal counts; the other estimator is entirely based on the longitudinal sample, both for the estimation of the class boundaries and the longitudinal counts. We develop Taylor linearization-type variance estimators for both the longitudinal and the mixed estimator under the assumption of no change in the population, and for the mixed estimator when there is change.

In longitudinal surveys, sample subjects are observed over several time points. This feature typically leads to dependent observations on the same subject, in addition to the customary correlations across subjects induced by the sample design. Much research in the literature has focussed on modeling the marginal mean of a response as a function of covariates. Liang and Zeger (1986) used generalized estimating equations (GEE), requiring only correct specification of the marginal mean, and obtained standard errors of regression parameter estimates and associated Wald tests, assuming a "working" correlation structure for the repeated measurements on a sample subject. Rotnitzky and Jewell (1990) developed quasi-score tests and Rao-Scott adjustments to "working" quasi-score tests under marginal models. These methods are asymptotically robust to misspecification of the within-subject correlation structure, but assume independence of sample subjects which is not satisfied for complex longitudinal survey data based on stratified multi-stage sampling. We proposed asymptotically valid Wald and quasi-score tests for longitudinal survey data, using the Taylor Linearization and jackknife methods. Alternative tests, based on Rao-Scott adjustments to naive tests that ignore survey design features and on Bonferroni-t, are also developed. These tests are particularly useful when the effective degrees of freedom, usually taken as the total number of sample primary units (clusters) minus the number of strata, is small.

Two-phase sampling designs have been conducted in waves to estimate the incidence of a rare disease such as dementia. Estimation of disease incidence from longitudinal dementia study has to appropriately adjust for data missing by death as well as the sampling design used at each study wave. In this paper we adopt a selection model approach to model the missing data by death and use a likelihood approach to derive incidence estimates. A modified EM algorithm is used to deal with data missing by sampling selection. The non-paramedic jackknife variance estimator is used to derive variance estimates for the model parameters and the incidence estimates. The proposed approaches are applied to data from the Indianapolis-Ibadan Dementia Study.

Statistics of Canadian production and trade in fish are presented in four parts: primary operations (series N1-48), processing (series N49-82), value of exports and imports (series N83-113), and miscellaneous statistics (series N114-142).

This section includes production, exports and imports of metallic and non-metallic minerals, the latter category including structural materials but excluding fuels, which are reported in the Energy chapter, Section Q. The section contains three parts: metallic minerals, series P1-81; non-metallic minerals, series P82-150; and principal statistics, series P151-162. The first two parts contain quantities and value of production, exports and imports; the third part contains number of employees, salaries and wages, cost of fuel and electricity, cost of process supplies and containers, gross value of production and net value added by processing.

This section contains the major statistics pertaining to the energy-producing and energy-distributing industries in Canada. These statistics are collected and maintained by Statistics Canada using a classification based on nature of the resource base and level of activity, rather than the end-use classification that has been adopted here. In the previous edition, portions of the material which appears here appeared in Section N, Minerals and Fuel, others appeared in Section P, Electric Power.

This section updates the official statistics on manufactures for 1870 to 1959 presented in the first edition of Historical Statistics of Canada. Apart from minor revision to some series for the years 1952 to 1959, no revisions have been made to the statistics from 1870 to 1959. The descriptions of the statistics for this period have been reproduced without change from the description in the first edition written by Arthur J.R. Smith.

The statistics in this section are arranged in five subsections. The first subsection (series S1-64) contains summary statistics for construction by type and by region. It also contains general statistics of the construction industry. The second subsection (series S65-166) contains statistics for non-residential construction by the sector 'purchasing' the construction. The third subsection (series S167-219) covers residential construction, the fourth (series S220-246) the housing stock and the fifth (series S247-335) mortgage funding. A major difference between this section and the similar section in the first edition of Historical Statistics of Canada is the elimination of most of the series on repair construction detail. On the other hand, a large number of series have been added. These include the series on mortgage funding and housing unit costs.

The data of this section are in nine subsections as follows: rail transport, (series T1-82); water transport and canal statistics, (series T83-141); roads and road transport, (series T142-194); civil aviation, (series T195-246); oil and gas pipelines, (series T247-250); transportation accident victims (series T251-292); post office statistics, (series T293-314); telecommunications carrier industry, (series T315-352) and radio and television, (series T353-359).

The data of this section are in three parts. Series Vl-331 contain statistics on retail trade, series V332-409 on wholesale trade and series V410-448 on service establishments.

This section is in three main parts as follows: Summary Statistics (series W1-66); Elementary and Secondary Education (series W67-306) and Post-secondary Education (series W307-532). Within each part there are series on enrolment, number and qualifications of teachers and financial data. Additional information on degrees awarded and libraries is given in the third part.

The data in the tables of Section Y fall into two major divisions. Series Yl-296 concern the federal scene, series Y297-387 the 10 provincial scenes. Within the first major division, series Yl-40 contain data on executive and legislative posts and occupants; series Y41-210 on elections and parliamentary sessions; series Y211-263 on employees and representation abroad; and series Y264-296 on disallowance of provincial acts, reservation of provincial bills and on federal Royal Commissions. Within the second division, series Y297-801 present provincial data on Lieutenant-Governors and premiers; series Y302-387 the results of provincial elections.

This section on crime and justice contains 328 series, arranged in four main sub-sections: crime and law enforcement, (series Z1-65); court proceedings, (series Z66-172); penal institutions, (series Z173-226); and juvenile delinquency, (series Z227-291). Population statistics are appended to facilitate use of the series, (series Z292-328). These series have been selected and arranged not only to provide usable quantitative information covering the field of criminal justice but also to indicate something of the historical development of justice statistics in Canada over the past century.

The Higher Education sector is composed of all universities, colleges of technology and other institutes of post-secondary education, whatever their source of finance or legal status.

This report describes the basic characteristics of the firms using biotechnologies, the type of use of biotechnologies, the stage of use and future use, the obstacles to acquisition and implementation of biotechnologies, the advantages obtained, and internal and external sources of information leading to biotechnology use.

The Juristat on impaired driving, released every two years, presents data on the declining trend in impaired driving at the national, provincial and census metropolitan area (CMA) levels, as well as the characteristics of persons charged with this offence. The analysis is based on police-reported statistics as well as data from the courts and corrections sectors. Other data sources include data on fatally-injured drivers, on trends in alcohol consumption, as well as information on pro-active police measures such as trends in check-stops and roadside license suspensions.

This article provides estimates of the percentage of seniors who are licensed to operate a motor vehicle. It describes the health characteristics of these license holders and reviews research relating to older drivers.

This article examines changes in hospital separations of children aged 1 to 14 between 1986/87 and 1996/97. It focuses on four common causes of childhood hospitalization: asthma, chronic disease of tonsils and adenoids, fractures, and acute appendicitis.

This article focuses on differences in the health status and health care utilization patterns of mothers in two-parent families, women who recently became lone parents, and women who had been lone parents for a longer period. Changes in the health of these women and their health care use over time are also explored.

This article examines associations between long working hours, depression and changes in selected health behaviours. Based on an analysis of people followed over a two-year period, the relationship between changes in work hours and changes in health behaviours is explored.

This article provides estimates of the incidence of arthritis between 1994/95 and 1996/97 among women aged 38 or older. It also examines the association between hormone replacement therapy (HRT) and a new diagnosis of arthritis by 1996/97.

The estimate for research and development in the higher education sector has been revised.

As the Internation Year of the Senior Persons winds down, attention on this growing group of consumers will continue well into the next millenium. This event marked the first year that seniors have been recognized by a worldwide designation.

Series description

This series includes five general reference products - the Preview of Products and Services; the Catalogue; the Dictionary; the Handbook and the Technical Reports - as well as geography reference products - GeoSuite and Reference Maps.

Product description

Technical Reports examine the quality of data from the 1996 Census, a large and complex undertaking. While considerable effort was taken to ensure high quality standards throughout each step, the results are subject to a certain degree of error. Each report looks at the collection and processing operations and presents results from data evaluation, as well as notes on historical comparability.

Technical Reports are aimed at moderate and sophisticated users but are written in a manner which could make them useful to all census data users. Most of the technical reports have been cancelled, with the exception of Age, Sex, Marital Status and Common-law Status, Coverage and Sampling and Weighting. These reports will be available as bilingual publications as well as being available in both official languages on the Internet as free products.

This report deals with coverage errors, which occured when persons, households, dwellings or families were missed by the 1996 Census or enumerated in error. Coverage errors are one of the most important types of error since they affect not only the accuracy of the counts of the various census universes but also the accuracy of all of the census data describing the characteristics of these universes. With this information, users can determine the risks involved in basing conclusions or decisions on census data.

This report deals with sampling and weighting, a process whereby certain characteristics are collected and processed for a random sample of dwellings and persons identified in the complete census enumeration. Data for the whole population are then obtained by scaling up the results for the sample to the full population level. The use of sampling may lead to substantial reductions in costs and respondent burden, or alternatively, can allow the scope of a census to be broadened at the same cost.

Series Description:

This series includes five general reference products - the Preview of Products and Services; the Catalogue; the Dictionary; the Handbook and Technical Reports - as well as two geography reference products - GeoSuite and Reference Maps.

Product Description:

The 1996 Census Dictionary provides detailed information on all of the concepts, variables and geographic elements of the 1996 Census. Information provided for each variable includes a definition, the associated census questions, applicable response categories or classifications and special remarks, namely on historical aspects. Users should make use of this edition of the 1996 Census Dictionary for the most up-to-date information. This final edition is also available on our web site as a free downloadable product.

The National Population Health Survey (NPHS) is a family of surveys with multiple objectives, one of which is to provide information on a panel of people who will be followed over time to reflect the dynamic process of health and illness. Data for the first cycle of the NPHS - Households Survey were collected from June 1994 to June 1995, and were released in September 1995. Data for the second cycle were collected from June 1996 to August 1997. One of the primary outputs for the second cycle is a longitudinal master file. This paper will describe six major strategies that were developed to process the longitudinal master file.

In 1994, Statistics Canada introduced a new longitudinal social survey that collects information from about 23,000 children spread over 13,500 households. The objective of the National Longitudinal Survey of Children and Youth is to measure the development and well being of children until they reach adulthood. To this end, the survey gathers together information about the child, parents, neighbourhood as well as family and school environment. As a consequence, the data collected for each child, is provided by several respondents, from parents to teachers, a situation which contributes to an increased disclosure risk. In order to reach a balance between confidentiality and the analytical value of released data, the survey produces three different microdata files with more or less information. The master file that contains all the information is only available by means of remote access. Hence, researchers do not have direct access to the data, but send their request in the form of software programs that are submitted by Statistics Canada staff. The results are then vetted for confidentiality and sent back to the researchers. The presentation will be devoted to the various disclosure risks of such a survey and to the tools used to reduce those risks.

In epidemiology analysis of longitudinal data is commonly accepted as providing the most robust measures of association between putative risk and selected outcomes such as death or cancer. SMARTIE is a SAS application for efficient analysis of longitudinal data. Based on person days at risk, it can handle multiple exits from and re-entries to risk, and derives outcome measures such as survival rates. Standardised Mortality Ratios (SMRs) and Cancer Incidence Ratios (SIRs). Summary data can be produced in a format easily ported to any modelling package such as Stats 5.0. We discuss the background to its development, the overall program structure, its command language, and finally we say something about the organization of outputs. Findings from survival studies using the Longitudinal Study of the Office for National Statistics (ONS) are used to demonstrate features of SMARTIE. This study is based on one per cent of the population of England and Wales. It is continually updated with the addition of new members and with information from birth, death and cancer records, and from the census.

Models for fitting longitudinal binary responses are explored using a panel study of voting intentions. A standard repeated measures multilevel logistic model is shown inadequate due to the presence of a substantial proportion of respondents who maintain a constant response over time. A multivariate binary response model is shown a better fit to the data.

Longitudinal studies with repeated observations on individuals permit better characterizations of change and assessment of possible risk factors, but there has been little experience applying sophisticated models for longitudinal data to the complex survey setting. We present results from a comparison of different variance estimation methods for random effects models of change in cognitive function among older adults. The sample design is a stratified sample of people 65 and older, drawn as part of a community-based study designed to examine risk factors for dementia. The model summarizes the population heterogeneity in overall level and rate of change in cognitive function using random effects for intercept and slope. We discuss an unweighted regression including covariates for the stratification variables, a weighted regression, and bootstrapping; we also did preliminary work into using balanced repeated replication and jackknife repeated replication.

This paper presents a method for handling longitudinal data in which individuals belong to more than one unit at a higher level, and also where there is missing information on the identification of the units to which they belong. In education, for example, a student might be classified as belonging sequentially to a particular combination of primary and secondary school, but for some students, the identity of either the primary or secondary school may be unknown. Likewise, in a longitudinal study, students may change school or class from one period to the next, so 'belonging' to more than one higher level unit. The procedures used to model these stuctures are extensions of a random effects cross-classified multilevel model.

The British Labour Force Survey (LFS) is a quarterly household survey with a rotating sample design that can potentially be used to produce longitudinal data, including estimates of labour force gross flows. However, these estimates may be biased due to the effect of non-response. Weighting adjustments are a commonly used method to account for non-response bias. We find that weighting may not fully account for the effect of non-response bias because non-response may depend on the unobserved labour force flows, i.e., the non-response is non-ignorable. To adjust for the effects of non-ignorable non-response, we propose a model for the complex non-response patterns in the LFS which controls for the correlated within-household non-response behaviour found in the survey. The results of modelling suggest that non-response may be non-ignorable in the LFS, causing the weighting estimates to be biased.