The International Association of Survey Statisticians has developed a preliminary program of short courses to be offered just prior to or after the 55th Session of the International Statistical Institute (ISI).

The courses are led by international high-level experts and are addressed to practitioners, researchers and students in statistics and survey methodology. The program includes courses which have already been held in previous ISI Sessions as well as courses which are offered for the first time. The official language of the courses is English. The courses will be held at the Australian Bureau of Statistics at St Andrews House, Sydney Square, Sydney.

Funding assistance for supporting statisticians from developing and transition countries is being sought.

The following courses have been planned.

Course A: Workshop on Survey Sampling

Presented by:
Graham Kalton (Westat)
Steven Heeringa (Survey Research Center, University of Michigan)
Duration: 2.5 days
Dates: 1 and 2 April 2005 (full days); 3 April 2005 (morning)

The workshop will focus on practical aspects of sampling for household surveys. It will start from basic principles and build up to complex stratified multi-stage sample designs. It will cover the main sampling techniques and also such issues as sampling frames, weighting, and imputation. It will end with an introduction to variance estimation with complex sample designs.

Course B: Variance Estimation in Complex Surveys

Presented by:
Wayne Fuller (Iowa State University)
Kirk Wolter (University of Chicago)
F. Jay Breidt (Colorado State University)
Anthony An (SAS Institute)
Duration: 2 days
Dates: 3 April 2005 (afternoon); 4 April 2005 (full day); 5 April 2005 (morning)

The purpose of this course is to provide training in variance estimation in complex surveys for survey statisticians, especially those from developing countries. The course will cover methods of estimating variances for statistics such as means, proportions, ratios, regression coefficients, and statistics arising in the analysis of two-way contingency tables. Both linearization and replication methods will be discussed. The use of computer software for computing variances of statistics from complex sample designs will be demonstrated and instruction will be given in practical applications. About one-half of the course will be devoted to implementation on the computer.

Course C: Workshop on Editing and Imputation of Survey Data

Presented by:
John G. Kovar (Statistics Canada)
Eric Rancourt (Statistics Canada)
Duration: 1.5 days
Dates: 4 April 2005 (full day) and 5 April 2005 (morning)

Surveys and censuses conducted by national statistical agencies, research institutes and other survey organisations suffer from various degrees of nonresponse even under ideal conditions. In order to try to alleviate the problems caused by nonresponse, editing and imputation methods are usually applied. Since the process of editing and imputation is time and resource intensive, care must be exercised in controlling the efficiency as well as the effectiveness of the methods. The aim of this short course is to introduce the students to methods of prevention, detection and treatment of nonresponse. Evaluation of such methods and their impact on the survey outputs will be highlighted. Existing edit and imputation software will be compared. Numerous examples will be provided to illustrate the material presented.

Course D: Introduction to Survey Quality

Presented by:
Paul Biemer (RTI International and University of North Carolina)
Lars Lyberg (Statistics Sweden)
Duration: 2 days
Dates: 3 April 2005 (afternoon), 4 April 2005 (full day) and 5 April 2005 (morning)

The course is designed for a broad audience that includes experienced survey researchers who would benefit from a better understanding of the survey data quality as well as others with little or no prior experience in survey methods. It will provide a brief introduction survey quality using total survey error paradigm. The course begins with a discussion of total survey error and its relationship to survey costs. Then the major sources of survey error are discussed, focusing on four major sources: coverage error, nonresponse, data processing error and measurement error. We also discuss some methods that are most often used in practice for evaluating the effects of the source on total survey error.

Course E: Statistical Disclosure Control

Presented by:
Anco Hundepool (Statistics Netherlands)
Eric Schulte Nordholt (Statistics Netherlands)
Peter-Paul de Wolf (Statistics Netherlands)
Duration: 2 days
Dates: 13 and 14 April 2005 (full days)

The purpose of this course is to provide the participants with an understanding of the methodological aspects of Statistical Disclosure Control, to train them in solving problems on this topic and to demonstrate the ARGUS software. The meaning and impact of Statistical Disclosure Control can only be appreciated in the light of practical problems and policy related issues. Therefore, some attention is also paid to such topics without putting heavy emphasis on them. Topics covered include theory and methods on microdata, exercises on microdata, demonstration of Mu-ARGUS, theory and methods on tabular data, exercises on tabular data, demonstration of Tau-ARGUS, legal issues, on-site facilities and remote access.

Course F: Design and Analysis of Repeated Surveys

Presented by:
David Steel (University of Wollongong)
Craig McLaren (Australian Bureau of Statistics)
Duration: 2 days
Dates: 13 and 14 April 2005 (full days)

This course will consider the interaction between the design of a repeated survey and the methods used for estimation and analysis. The choice of rotation pattern will be considered in terms of the impact on the estimation of levels and changes. Composite and other forms of estimators will be reviewed and the interaction between design and estimation explored. Estimation of seasonally adjusted and trend estimates from repeated surveys will also be considered.