Population Change and Lifecourse

EN/FR

Past Data Development Projects

Past Projects By Year

2013 Projects

QICSS Summer School 2013

The QICSS Summer School is intended for professors, university or public sector researchers, graduate or postdoctoral students, and, in rare cases, for undergraduate students wishing to master the methods that will allow them to take advantage of large and complex surveys such as those of Statistics Canada and the Institut de la statistique du Québec in order to complete the research they are carrying out in the context of their work or studies. Relying on the expertise of both Francophone and Anglophone universities of Québec, only the QICSS Summer School offers a bilingual training programme. Some modules are offered in French and others in English, depending on the expertise of the instructors and training needs.

For 2013, the QICSS plans to offer several modules previously offered and to add new ones in order to respond better to researchers’ needs and to the challenges posed by the use of new data sets available to the QICSS (such as administrative data matched among themselves or merged with survey data) and to raise awareness of potential methods and tools for evaluation and public policy formulation (such as microsimulation models and econometric methods of evaluation of policies and programmes).

Team: Danielle Gauvreau, Professor, Department of Anthropology and Sociology, Concordia University, and QICSS Director; Jean Poirier, University of Montréal and QICSS Deputy Director; Catherine Haeck, Professor, Department of Economic Sciences, Université du Québec à Montréal and Associate Researcher at the QICSS.

Custer support for the project: $9830

2012 Projects

2012 CRDCN National Conference

Data & Methodology Issues Presentations

  1. Functional Databases for Longitudinal Analyses and Tips of the Trade: The Case of the NPHS in Canada
    Amélie Quesnel-Vallée and Emilie Renahy (McGill University)
  2. LISA: Anticipating the Next Generation of Longitudinal Data
    Andrew Heisz (Statistics Canada)
  3. Analyse des trajectoires scolaires: Portée et limites de l’Enquête auprès des jeunes en transition (EJET)
    Pierre Cansisius Kamanzi (Université de Montréal)
  4. Developing Gini and Zinga Indexes for the Analysis of Contextual Income Inequalities Within Canada
    Bradley Corbett (University of Western Ontario)

Developing Gini and Zinga Indexes

Developing Gini and Zinga Indexes for the Analysis of Contextual Income Inequalities within Canada

Data Development and Research Infrastructure

A team, led by Bradley Corbett, Statistics Canada Research Data Centre, University of Western Ontario, proposes to undertake a complex analysis of Canada’s 2006 Census data to produce two indexes containing contextual measures of income inequality. The team will produce estimates of the Gini and Zinga Indexes for various levels of geography as they are defined by Statistics Canada. The Gini Index is a popular measure of income inequality that is well understood and easily interpreted in the social sciences. The Zinga Index is a newer measure of income inequality that is better able to identify the relative nature of ‘poor’ and ‘rich’ because it is based on the ratio of upper and lower group means. These contextual indexes will provide measures of income inequality which can be merged with individual level data when the dataset contains either postal codes or Statistics Canada’s geography codes.

Cluster support for the project: $6250

Requirements Surrounding the Storage and Use of Provincial Data

Legislative Requirements Surrounding the Storage and Use of Provincial Administrative Data by Academics in Academic Institutions

Michael Haan, Department of Sociology, and Ted McDonald, Department of Economics, University of New Brunswick, propose to prepare a report that details the necessary steps for gaining access to administrative data used by provincial government departments. They will review and report on the necessary legislation that must be considered by academics seeking to make use of provincial administrative data for research purposes. There is a wide variety of organizational models across provinces, arising in large part because of differences in legislation governing access to confidential information. The report will build on Harmonizing Research & Privacy: Standards for a Collaborative Future - Privacy Best Practices for Secondary Data Use (SDU) (2006), a comprehensive CIHR funded study where researchers examined privacy legislation across all provincial jurisdictions and produced a detailed toolkit and summary report for both researchers and institutes/centres that act as data repositories for government health data. The proposed report would examine existing privacy legislation across all provincial jurisdictions in an effort to determine what changes, if any, have taken place over the past six years.

Cluster support for the project: $6000
Project withdrawn

2011 Projects

New Uses for Government and Administrative Data

A half-day symposium at the 2011 Congress of the Humanities and Social Sciences, this project aims to promote the use of provincial and other administrative data sources. There has been a growing recognition in recent years of how municipal, provincial and federal administrative databases can be used by academic researchers to conduct policy‐relevant research, helping to provide an evidence‐base for government decisions. The symposium – organized by Martin Cooke of the University of Waterloo and Michael Haan of the University of New Brunswick and jointly sponsored by the Canadian Population Society and the Population Change and Lifecourse Strategic Knowledge Cluster – will attempt to further improve the integration of these data into the Canadian social data system. The experiences of the users of various types of administrative data will be presented and there will be a discussion between current and potential users, data producers, and others with an interest in Canada’s data infrastructure.

Cluster support for the project: $9,755

Four Training Modules in the 2011 Summer School of the QICSS

Support given to the preparation and the realization of four training modules in the context of the 2011 Summer School of the QICSS

In 2011, the QICSS intends to continue developing training to better meet the needs of the users of large databases, such as the ones created by Statistics Canada and by the Institut de la statistique du Québec. The recent support of the “Population Change and Lifecourse” cluster will be used to support four modules that correspond well to the objectives of the cluster. These modules are: (1) Modèle dynamique de microsimulation; (2) Préparation des données pour l’analyse longitudinale; (3) Mise en valeur et analyse comparative des données de recensement; (4) Introduction to Multilevel (hierarchical) models. These four modules will add to the six modules that had previously been developed and will be offered in the context of the 2011 Summer School, entirely funded by the CIQSS and its partners: (1) Introduction à l’analyse spatiale; (2) Introduction à l’analyse multiniveau; (3) Panel data analysis; (4) Introduction to Structural Equations Modelling; (5) Analyse des données de panel; (6) Analyse des trajectoires développementales.

Cluster support for the project : $13,970

2010 & Prior Projects

YITS for Stata and SPSS: Automating transfer and documentation

YITS for Stata and SPSS: Automating the transfer and documentation of the YITS data files into ready to use Stata and SPSS files

Benoît Laplante, Institut national de la recherche scientifique

Data from Statistics Canada’s Youth in Transition Survey (YITS) are structured as a set of data files, each file containing a different set of variables and most files containing data pertaining to a different cycle of the survey. Statistics Canada provides YITS users with a “data extraction tool” but the software is ill adapted to research projects such as lifecourse analyses that require complex data assembly. Further, the YITS files are provided in SAS format. To meet our own research needs, we have written a series of programs which automate the transfer of the YITS files from the SAS to the Stata and SPSS formats. We have also written a series of programs which automate the preparation of a single one-line-per-person data file that contains all the information from all cycles for each interviewee. The project will complete, review, clean, check and document the existing software and use it to produce a new version of the documented data files. The resulting files and the programs to generate them would be made available to users at the Research Data Centres.

Cluster support for the project: $7,091.70
Workshop on Preparing YITS Data for Longitudinal Analysis by Benoît Laplante

SLID for Stata: Automating the transfer and documentation

SLID for Stata: Automating the transfer and documentation of the SLID data files into ready to use bilingual Stata files.

Benoît Laplante and María Constanza Street, Institut national de la recherche scientifique

There are efficient programs to transfer Survey of Labour and Income Dynamics (SLID) data files into ready to use Stata files, but they are not written in a way that makes the transfers and documentation fully automated. Furthermore, they are undocumented and in need of lot of cleaning and fine check. The proposal is to complete, review, clean, check and document the existing software and use it to produce a new version of the documented data files each time Statistics Canada releases a new version of SLID. Given that SLID is managed with the intent to minimize changes, our software should be operational for several years without major revision. The resulting files and the programs to generate them would be made available to users at the Research Data Centres.

Cluster funding support for the project: $7,041.84

Functional databases for longitudinal analyses:

The case of the National Population Health Survey in Canada

Amélie Quesnel-Vallée and Emilie Renahy, McGill University

The project aims at producing a tool that creates functional databases to facilitate longitudinal analyses of RDC analysts and researchers using the National Population Health Surveys. It will provide a package that allows the renaming of all or a selection of time-varying variables (those asked each year or at least in two different cycles) in a conventional format that facilitates the creation and use of databases either in person-level format or in person-period format depending on the software or the kind of longitudinal model used. The package will be provided to every Research Data Centre and will be composed of:  (a) a SAS macro program that renames either all variables or a list of selected variables, (b) a SAS macro program that transforms a data set in a person-period format, and (c) a user’s guide. The proponents will also be available to provide follow-up assistance to the RDC users.

Cluster funding support for the project:  $5000.

Click here to download the SAS files, the user guide and the background document

Click here to download Stata program and example

Functional Databases for Longitudinal Analyses and Tips of the Trade: The Case of the NPHS in Canada
Amélie Quesnel-Vallée and Emilie Renahy (McGill University)

Using Administrative Data to Analyze the Determinants of Success in University

Martin Dooley, A. Abigail Payne, and Philip DeCicca of the Department of Economics at McMaster University are the first recipients of the grant awarded through the Data Development, Research Infrastructure, and Associated Training initiative of the Cluster.  The specific objective of their project is to improve the administrative data infrastructure available at the Public Economics Data Analysis Laboratory (PEDAL) at McMaster University. PEDAL was established in 2003 in order to improve access by researchers to administrative data. The researchers are working with data on applications to Ontario universities and student academic progress at four such universities.  The $8,000 fund for this project will be used to clean and document these data, and establish typical patterns of student academic progress. 

One research project using data developed at PEDAL aims at improving the understanding of the determinants of the relationship between success in post-secondary education and a series of student background factors such as the characteristics of the student (sex, academic performance, financial aid), the student’s neighbourhood (income, education) and high school (peer composition, stream emphasis). For example, how strongly related are socio-economic background and high school performance to post-secondary performance?  Are there differences in post-secondary academic performance among students from neighbourhoods with varying proportions of immigrants?  Are the real costs of education and access to financial aid linked to rates of academic progress?  Is there a link between changes in high school curriculum and the rate at which students progress towards a diploma or degree?

For further information, including exploring the possibility of accessing PEDAL data, please contact Abigail Payne at paynea@mcmaster.ca.