Workshops

Workshop 1. Record Linkage Methods
Workshop 2. From Traditional Demographic Calculations to Projections by Microsimulations
Workshop 3. Using Administrative/Operating Systems to Strengthen Statistical Survey/Census Systems

Three full-day workshops will take place on Tuesday October 26 from 9:00 to 16:30. Simultaneous translation will be provided.

Please note that the number of seats for each workshop is limited.

Top of Page

Workshop 1. Record Linkage Methods

Karla Fox, Statistics Canada
Lori Stratychuk, Statistics Canada

Record linkage is simply the integration of information from two or more independent sources. Under this framework, records are linked on the basis of common data. Record linkage has become increasingly important in many different domains: maintenance of electronic registry information, health care administration, demographic studies and medical research. Whether following a cohort over time or linking patients to vital statistics in order to calculate survival curves, researchers can benefit from a greater understanding of record linkage.

Using examples from different surveys and programs at Statistics Canada, this workshop will cover the general principles of record linkage, including both statistical matching and exact matching. Related to record linkage, and equally important, other topics to be covered include: data preparation, linkage techniques and linkage evaluation. Practical examples will be used to illustrate different points throughout the workshop. Participants will gain an appreciation for the different concepts involved, an understanding of how to approach a problem requiring record linkage, as well as an awareness of what issues they may face when analyzing a linked file.

Top of Page

Workshop 2. From Traditional Demographic Calculations to Projections by Microsimulations

André Cyr, Statistics Canada
Julien Bérard-Chagnon, Statistics Canada
Éric Caron Malenfant, Statistics Canada
Dominic Grenier, Statistics Canada

« By 2031, (...) between 29% and 32% of the population could belong to a visible minority group (...). » (The Daily, March 9, 2010)

Demography is a statistical study at the heart of social policy and societal challenges. Demographic data not only help to better understand the multiple facets of Canadian reality, but also factor into calculations for various federal-provincial fiscal arrangements, such as the Equalization Program and the Canada Health and Social Transfer (CHST).

The workshop will aim to demystify the estimation and demographic projection methods used at Statistics Canada. It will be divided into two parts:

• The morning will be devoted to the production of demographic estimates. These estimates require the simultaneous use of multiple sources of data with several challenges. They are based on censuses, which are adjusted via reverse record check (RRC). Next, administrative sources are used to estimate the population and its demographic components.
  
  
• In the afternoon, the two main methods for demographic projection will be presented. The first method, by component, uses aggregate data to predict the effective future of the population according to different hypotheses regarding its evolution. The second method, by microsimulation, uses the censuses to predict individuals one by one based on their characteristics and a series of hypotheses. Various surveys and administrative sources are used to calculate the risks of simulated events.
  

Top of Page

Workshop 3. Using Administrative/Operating Systems to Strengthen Statistical Survey/Census Systems

Fritz Scheuren, National Opinion Research Center, scheuren@aol.com
Asaph Young Chun, National Opinion Research Center

The process of “Turning Administrative Data into Statistical Information” is now well along and increasingly welcomed worldwide. There are difficulties, however, both technical and cultural. While in the course we recommend how to handle technical issues, frankly we do not have good answers to some of the “cultural” issues that exist between administrative and statistical organizations—both within a country and/or internationally. Cultural issues may be a topic of class discussion, depending on who attends the workshop.

For well over 50 years, administrative data have been regularly used in a myriad of ways for establishment surveys and business censuses in Canada and the United States. Many European countries now conduct register-based censuses rather than population-based demographic ones. Research is underway—too slow for some—on how to adapt what is done on demographic census taking in Europe to fit the North American context. Our course covers this.

The course involves administrative statistical applications not only connected to censuses but also going beyond them to include administrative data uses in enumerative survey design and survey estimation. Record linkage issues will be featured. Some material is also presented on analytic surveys and contrasts made between them and enumerative efforts.

Traditionally, the seam between administrative data and data collected directly is divided up into sampling and nonsampling opportunities. We cover this as well. Important issues addressed include definitional differences, measuring uncertainties and quantifying biases.

Finally, if we have enough time, the class will be asked to engage in speculations about the future relative importance of information creation between administrative and statistical data. By now, you may be able to guess what our perspective is.