Abstract:
This article begins by reviewing the measurement error model proposed by, with particular emphasis on their concept of simple response variance. More recent developments in the modeling of measurement error are linked to their model and are shown to be extensions and generalizations of their essential concepts. The index of inconsistency, which was first formally described in their paper, is shown to have at least three interpretations, depending upon the statistical framework adopted for describing the gross differences in an interview-reinterview study. Several examples illustrate and compare their classical methods with more modern approaches that employ latent class analysis to estimate the error parameters. It is shown that the use of estimates of simple response variance for survey evaluation may obscure important error structures that are more visible using estimates of classification error probabilities to assess data quality.

Abstract:
We distinguish between information about individuals and ⁢statistical data&xit;, a representation of information that does not identify any individual. Implications of this distinction are: (1) neither privacy nor confidentiality pertains to statistical data but rather to information; (2) individuals may have the right to control their information, but not the right to control statistical data derived from that information; (3) information should be protected by providers from release, but statistical data need to be safeguarded with sanctions on users; (4) zero tolerance can be a standard for the release of information, but, for statistical data, the standard must be to exercise reasonable care so that the risk of an identification is very small. Finally, we discuss how recent legislation in the U.S. can be used to strengthen confidentiality, increase research access to statistical and other data, and make available more useful statistical data.

Abstract:
Although design-based inference is the standard form of inference with large-scale sample surveys, in practice some reliance on model-dependent inference is necessary. This article considers issues of model-assisted inference, the population of inference, conditional inference, the effect of measurement errors, and analytic uses of survey data, under the design-based mode of inference. It then discusses the need for model-dependent inference for small area estimation and for handling missing data (unit nonresponse, item nonresponse, and noncoverage). The article also discusses the use of models in variance estimation.

Abstract:
Seasonal adjustment is applied to economic data from periodic surveys to help analysts detect economically meaningful changes in the time series. Adjustment can dramatically alter the measured values. The time series to which it is applied usually have seasonal movements whose pattern changes over time, and many are subject to unsystematic effects, for example abrupt changes of level or the turbulent, market-driven behavior of small numbers of large companies. There are also systematic effects that require appropriate treatment, such as heteroskedasticity between different calendar months, day of week effects, and effects due to the shifting dates of holidays such as Easter.

Progress in seasonal adjustment depends on the development of models or methods that better account for these effects. Perhaps less obviously, progress also depends on the development of diagnostics that identify the presence of such effects and show whether treatments applied for them result in satisfactory adjustments. Much recent progress has come from the use of specialized time series models and appropriately focused diagnostics, especially graphical diagnostics.

Our talk begins with a review of basic motivations, ideas, and issues of seasonal adjustment. There follows a presentation of some seasonal adjustment, modeling, and model selection diagnostics of the Census Bureau's new X-12-ARIMA and X-12-Graph programs, with an emphasis on graphical diagnostics. The applications we provide of these diagnostics show how some of them address fundamental statistical modeling issues.

Abstract:
Beginning with the pioneering contributions of Neyman, Hansen, Mahalanobis and others, a large part of survey sampling theory has been directly motivated by practical problems encountered in the design and analysis of large scale sample surveys. We have seen major advances in handling both sampling and nonsampling errors as well as data collection and processing. In this lecture, I will present some current trends in sample survey theory and methods. After a brief discussion of developments in survey design, data collection and processing, I will focus on inferential issues, resampling methods for analysis of survey data and small area estimation. I will demonstrate the advantages of a conditional design-based approach to inference that allow us to restrict the set of samples to a relevant' subset. Quasi-score tests of hypotheses based on the jackknife method will be presented. I will also discuss issues related to model-based methods for small area estimation.

Abstract:
Statistics are a major means of knowing about society. In democratic societies official statistics are available not only to politicians and government officials, but also to interest groups and ordinary citizens. They thus have a special role to play in the process of policy formation and implementation. Until recently, the use of official statistics in policy debates has been limited by the relative difficulty in accessing large data bases by ordinary citizens and interest groups. The advent of the Internet and the World Wide Web, however, has dramatically changed the situation and greatly increased availability of statistics to ordinary citizens. Future technological changes are likely to enhance the amount of data that the public can easily access.

Such a vast change in accessibility of statistical data about society presents many challenges to the federal statistical system. Three of the most pressing challenges concern the relevance, validity, and timeliness of official statistics. Relevance refers to the questions to which the statistics gathered are the answers. How do we decide what data to collect and to make publicly available? As more people can access data more easily, the demand for statistical data and controversy over what data to collect will grow. With limited budgets, the statistical agencies will come under increasing public scrutiny, and decisions about what data to collect will take on added importance. Validity refers to the relation of statistical measures to the concepts they are intended to measure. Do the statistical measures published by the statistical agencies have the meaning that is ascribed to them by the broader user community? How does the statistical system refine measures to account for changes in society that make existing statistics no longer reflect the realities they are supposed to measure? Timeliness refers not only to the gap between the time of data collection and their availability but also to the periodicity of data collection operations and to the revisions in the measures to reflect changes in society that affect validity. Technology may provide means for decreasing the amount of time necessary for data collection and processing, but may not be able to decrease it sufficiently to satisfy an audience that has almost instant access to whatever is available. Increased strains on statistical agency budgets because of broader federal budget cutbacks or the allocation of more resources to disseminating data may force a decease in the periodicity of some statistical series. Technology, however, may have little to contribute toward the decision of agencies to change measures to reflect social changes that underlie the construction of the measures.

While there are great strengths in our present system, a decentralized statistical system such as we have in the United States is not well situated to meet these challenges successfully. There needs to be a forum in which responses to the challenges of wider and easier accessibility to official statistics can be discussed, responses formulated, and then carried out. Perhaps the greatest challenge to the federal statistical system today is how to organize itself to meet the challenges brought about by the technological revolution in data accessibility.

Abstract:
Joseph Waksberg will describe the development of sampling theory and related survey methods at the Bureau of the Census from about 1940 to 1970. The research carried out during this period of time and its application to survey methodology have profoundly affected current approaches to survey-taking. The talk will discuss the principal research findings, the milieu and context in which the research was carried out, the main participants, the original motivation for the direction of the research and later influences, and the effect of research on methods used for censuses and sample surveys. There will also be a brief review of the first applications of computers to statistical work carried out at the Census Bureau. Mr. Waksberg will provide some personal reminiscences which illustrate the fact that the mathematical statistician on a project needs to pay attention to broader issues in a survey than sampling methods along, e.g., to cost, other sources of error, the management structure of the organization, expected data uses, and alternative survey methods that are available.

Abstract:
The objective of national statistical systems is to provide relevant, comprehensive, accurate and objective (politically untainted) statistical information. The end purposes to be supported by the information are multiple, but include prominently the monitoring of the evolution of the country's economic and social conditions, the planning and evaluation of government and private sector programs and investments, policy debates and advocacy, and the creation and maintenance of an informed public.

The presentation will discuss ten broad criteria for the assessment of national statistical systems. They are: 1) The broad framework, including the legal one as well as the rank and standing of a Chief Statistician; 2) Professional core values; 3) Non-political objectivity and the means used to assure this; 4) Systems and approaches used to assess and meet federal priorities; 5) Systems and approaches used to assess and meet state (provincial) priorities; 6) Systems and approaches to assess and meet other users' priorities; 7) Mechanisms for the coordination of the statistical system; 8) Balancing priorities against available resources; 9) Serving the needs of different client groups; 10) Generating a supportive environment.

Abstract:
Samplers for less developed countries have been educated and trained in the last 50 years in various settings: in universities, in national statistical offices, on research projects, etc. Also both in their own countries and in "host" countries, such as the USA, India, France, etc. Methods of survey sampling are eminently "transferable" between countries (and disciplines), but are not readily accessible (portable) in printed (etc.) form, because the procedures must be modified to fit local resources. Therefore, travels by "learner" statisticians and/or by the "experts" are needed. The teaching/training process should combine academic subjects, individually selected, with practical training. These are difficult to arrange, and survey sampling also suffers neglect in universities, which teach statistical analysis but not data collection. Recruiting suitable candidates and career placement also pose challenges. But the rewards are great, because survey sampling is spreading all over the world.

Abstract:
Drawing on psychological theories of language comprehension and conversational conduct, I review what respondents learn from various features of questionnaires, including the response alternatives and the context in which a question is asked. This review indicates that many apparent "artifacts" in survey measurement reflect a systematic and thoughtful use of contextual information that is licensed by the tacit rules that govern the conduct of conversation in daily life. Implications for questionnaire construction are discussed.

Abstract:
The importance of measurement error for parameter estimation and for the design of statistical studies, particularly sample surveys, is examined. Beginning with a brief review of Hansen's contributions, the discussion concentrates on estimation problems in which measurement error leads to bias in the usual estimators. Distribution functions and regression equations are discussed, and the implications for the design of surveys are presented.