Data Research Centre

Past Statistics and Data Series Presentations

2013-2014 Presentations

Sept 18, 2013 - Access to Data and Geospatial Layers

Full Title: Access to Data and Geospatial Layers at Western:Western Libraries’ Map and Data Centre (MDC) and Statistics Canada’s Research Data Centre (RDC) Program

Speakers: Vince Gray, Elizabeth Hill, Christine Homuth, Brad Corbett

Nov 20, 2013 - Introduction to Cost Effectiveness Analysis

Presentation Slide Example


Date: April 10, 2013
Time: 12:30-1:30 pm

Sisira Sarma

Sisira Sarma

Assistant Professor
Department of Epidemiology & Biostatistics

Given the steady increase in healthcare costs, there has been a pressure within the healthcare system to provide health services to the population on a cost-effectiveness basis. This presentation will provide an introduction to the cost-effectiveness techniques to evaluate costs and benefits of health interventions. Some practical examples will be used to demonstrate how cost-effectiveness analysis is useful to decision makers in the health sector.

Dec 11, 2013 - Introduction to Structural Equations Modelling

Piotr Wilk

Piotr Wilk, PhD

Assistant Professor, Department of Paediatrics
Assistant Professor, Department of Epidemiology & Biostatistics
Scientist, Division of Children's Health and Therapeutics, Children's Health Research Institute

Structural Equations Modelling is a statistical tool for simultaneously estimating complex relationships between observed (measured) and unobserved (latent) variables. This popular technique offers a more reliable and flexible estimation approach than the traditionally used methods such as multiple regression or ANOVA. This presentation provides a general introduction.

Jan 22, 2014 - Introduction to Program Evaluation

NiravPresentation


Date: January 22, 2014
Time: 12:30-1:30 pm

Nirav Mehta

Nirav Mehta

Assistant Professor
Department of Economics

"Program Evaluation" is the use of statistical inference to quantify the effect of a treatment on outcomes of interest. This talk will provide examples and potential solutions to inferential problems frequently encountered when evaluating social programs. Estimating treatment effects is difficult because researchers only observe outcomes of participants under one scenario: either they receive a particular treatment or did not.  Consider a policymaker who wants to understand how introducing school choice, implemented by providing access to nonsectarian and Catholic public schools, would affect student academic achievement.  To assess how introducing the program has affected student achievement, we need to compare student outcomes under the school choice regime with those when there was no school choice.  This comparison may be complicated by various factors, such as the potential selection of households into schools.

Mar 19, 2014 - Multilevel Models for Social Network Analysis


Date: March 19, 2014
Time: 12:30-1:30 pm

Paul-Philippe Pare

Paul-Philippe Paré

Associate Professor
Department of Sociology

Multilevel analysis and social network analysis are typically taught as two different quantitative methods. In this presentation, we will learn how they can be combined: multilevel models are useful to isolate the effects of network characteristics (Level 2) from individual characteristics (Level 1) on a dependent variable. Variables specifically relevant to social network analysis (Degree-centrality, Closeness, Betweenness) can also be included in the multilevel models. An example based on adolescent delinquency will be presented.

2012-13 Presentations

Applications of Time Series Analysis


Date: April 17, 2013
Time: 12:30-1:30 pm

Ivan Medovikov
Brock University

Presentation Slides

Applications of Simulation Methods

Presentation Slide Example


Date: April 10, 2013
Time: 12:30-1:30 pm

Audra Bowlus

Audra Bowlus

Professor
Department of Economics

Presentation Slides

A Social Network Analysis: Examining Community Level Effect


Date: March 20, 2013
Time: 12:30-1:30 pm

Anabel Quan-Haase

Associate Professor
Department of Sociology

Dealing with Missing Data: Strategies for Beginners to Data Analysis

Presentation Slide Example


Date: February 6, 2013
Time: 12:30-1:30 pm

Rachel Margolis

Assistant Professor
Department of Sociology

Presentation Slides

A Practical Introduction to Content Analysis


Date: January 23, 2013
Time: 12:30-1:30 pm

Catherine Corrigall-Brown

Assistant Professor
Department of Sociology

Presentation Slides

Mapping with GIS: The Case of the SW Ontario Children’s Health Atlas

Presentation Slide Example


Date: January 9, 2013
Time: 12:30-1:30 pm

Jason Gilliland and Sandra Kulon

Presentation Slides

Introduction to Panel Data Analysis


Date: November 21, 2012
Time: 12:30-1:30 pm

Youngki Shin

Department of Economics

Presentation Slides

Using the Institute for Clinical Evaluative Sciences (ICES) Western Facility


Date: November 7, 2012
Time: 12:30-1:30 pm

Amit Garg,
Theresa Hands,
and Salimah Shariff

Presentation Slides

Bootstrapping: Example and Counterexamples


Date: October 24, 2012
Time: 12:30-1:30 pm

W.John Braun

University of Western Ontario

Presentation Slides

Access to Data and GIS at Western


Date: October 24, 2012
Time: 12:30-1:30 pm

Vince Gray,
Elizabeth Hill,
and Brad Corbett

University of Western Ontario

Presentation Slides

2011-12 Presentations

Smart Survey Design: From Effective Survey Questions to Layout and Design

April 25, 2012

Iris Gutmanis

Description Questionnaires can be a relatively inexpensive way to gather data from, potentially, a large number of respondents. Questionnaire development is a multi-stage process; a blend of scientific and artistic elements. Some of the issues that will be discussed are: (1) Identifying the objectives of the survey (Are you trying to discriminate, predict or evaluate? Census or sample or secondary data?) (2) Determining the method of administration (In-person/self-administered/by phone/web-based/mixed methods) (3) Writing the questionnaire/cover letter/script (Open or closed questions? Phrasing, layout and flow; Pre-testing) (4) Collating and interpreting the findings.

Presentation Slides

Social Network Analysis: When Social Relationship is the Dependent Variable

March 21, 2012

Anabel Quan-Haase

Description Social network data are unique because of their focus on relationships between people instead of attributes of single individuals. The presentation showed:

  • a general overview of the social network approach
  • the uniqueness of collecting and analyzing social network data
  • how social network data can be re-coded and imported into Stata for further analysis, and
  • an example to illustrate the procedure.

No prior knowledge of social network analysis is required as this presentation only covers basic concepts and procedures.

Presentation Slides

Missing Data Techniques

January 18, 2012

Paul-Philippe Pare

Description Missing data is a common problem in quantitative social research, particularly with large multivariate models and/or sensitive topics. Deleting cases with missing data is often problematic and can lead to smaller and biased samples. We will discuss the merits and limitations of different approaches to deal with missing data in multivariate studies, including traditional methods (e.g. single imputation, dummy variables), modern methods (e.g. multiple imputations, Expectation-Maximization), and treating missing cases as a substantive dependent variable of interest.

Presentation Slides

Simulation Techniques of Analysis

December 14, 2011

Tim Conley

Description The presentation discussed the use of simulation methods in empirical work in the social sciences. The methods involved computer simulations that can be implemented with commonly available software, e.g. Excel, STATA, SAS, MATLAB, and will be useful in estimating statistical models, understanding variability in model estimates, and drawing inferences from observed data.

Presentation Slides

Exploring Data with Statistical Graphics in R

November 23, 2011

Duncan Murdoch

Description There are many ways to explore data through graphical displays. R, the open source statistical package, supports many of them: static graphics such as histograms and scatterplots and multi-panel combinations thereof, as well as interactive graphics. This presentation will give a tour of the capabilities.

Presentation Slides

Research Data Centre: When and How to Access Data

November 9, 2011

Brad Corbett and Mary Ellen Kuenzig

Description: Building on the presentation on October 26 regarding the access to Public Use Files, this presentation will demonstrate how users can gain access to Statistics Canada’s Microdata Files. These files are non-anonymized, and sensitive data such as actual income are not suppressed or categorized. Longitudinal data like the National Longitudinal Survey of Children and Youth (NLSCY) are also available for analysis of trends across time. In Canada, these data are currently accessible through the Research Data Centers in university campuses across the country. The presentation will highlight some of the datasets and explain how researcher can gain access to these data.

Western Libraries Map and Data Centre: Accessing Data at Western using Equinox, <odesi>, and Statistics Canada's website

October 26, 2011

Vince Gray, MLIS & Elizabeth Hill, MLIS

Description: Western faculty and students have access to a wide array of data for research and teaching purposes through agreements such as the Data Liberation Initiative and the Inter-University Consortium for Political and Social Research. This presentation will show how to locate and download data from three sites: (1) Western's Equinox Data Delivery System, (2) Ontario Council of University Libraries' <odesi> database, and (3) the web site of Statistics Canada. The presentation will also describe differences in products offered at the sites, and demonstrate access to both microdata and aggregated data products such as Beyond 20/20 files from the 2006 Census.

Presentation Slides

Time Series Analysis

October 19, 2011

Ivan Medovikov

Description: The presentation will introduce basic time-series models, such as the Auto-Regressive Moving Average (ARMA) model and the Generalized Auto-Regressive Conditional Heteroscedasticity (GARCH) model. It will focus on some of the issues that arise when working with time-series data, and include a brief demonstration of the popular software for time-series analysis.

Presentation Slides

2010-11 Presentations

Survival Analysis: The Study of Lifetimes and their Distributions

April 20, 2011

Bruce Jones
Department of Statistical and Actuarial Sciences

Description This will be an introductory talk highlighting the main concepts and statistical methods used in survival analysis. Specifically, the following topics will be discussed:

  • Lifetimes and the nature of lifetime data;
  • Functions that characterize lifetime distributions;
  • Parametric and nonparametric models for lifetimes; and
  • Models involving fixed or time-varying covariates.

Presentation Slides

Logit, Probit and Tobit: Models for Categorical and Limited Dependent Variables

March 23, 2011

Rajulton Fernando
Department of Sociology

Description Responses to survey questions are often measured as categorical, either nominal or ordinal. Likewise, response variables are often measured in a limited way; for example, a response variable could take a value of less than 10 but we do not know by how much less. This presentation will deal with modeling such response variables, with greater emphasis on the Probit and Tobit than the Logit model that is more commonly used. Applications with the familiar statistical software and correct interpretation of results will be demonstrated.

Presentation Slides

Moving Towards a More Relevant, Flexible and Timely Data System

March 9 , 2011

Christian Dea
Director General, Knowledge and Data Management,
Human Resources and Skills Development Canada

Description Christian Dea will provide an overview of the modernization of the data portfolio at the Human Resources and Skills Development Canada (HRSDC), and discuss the engagement with the academic community in the development and implementation of the modernized data portfolio.

Presentation Slides

Multi-level Models of Analysis

February 23, 2011

Piotr Wilk
Department of Epidemiology and Biostatistics

Description Characteristics of environments or contexts are often included in the analysis of individual behaviour. Thus statistical analysis, in addition to individual level variables, requires the handling of information on number of levels such as neighbourhood, communities, cities, etc. What are techniques of multi-level analysis? How are models developed, and parameters estimated? These and other basic steps in multi-level modeling will be presented with illustrations using data gathered through a national survey.

Presentation Slides

Bootstrapping: When and how to use this weighting technique

January 19, 2011

David Bellhouse
Department of Statistical and Actuarial Sciences

Description The basics of bootstrap techniques and some applications are described for the case of independent and identically distributed observations. For large survey data sets with observations that are no longer independent and identically distributed, survey files come with a set of bootstrap weights, for which bootstrap techniques have been developed. The talk explores when bootstrapping is appropriate and how bootstrapping can be carried out, both generally and in terms of specific statistical package programs.

Presentation Slides

WES and PALS: Uses in Business Research

November 24, 2010

Alison Konrad
Professor of Organizational Behavior, Richard Ivey School of Business

Brad Corbett
Analyst, Research Data Centre, on Access to RDC Data

Description: What data are available from the Workplace & Employment Survey and the Participation & Activity Limitation Survey? How useful are they for study of business and organizations? What are their strengths and limitations? How can they be accessed?

Mapping Data: Research with Geographic Information Systems

October 20, 2010

Dr. Jason Gilliland, Richard Sadler & Don Lafreniere
Human Environments Analysis Laboratory, Department of Geography

Description: What is a Geographic Information System (GIS)? How can you use a GIS for research? What data are amenable for mapping and analysis with GIS? What GIS software is available at Western? This presentation will offer answers to these questions, and a demonstration on how GIS is being used at Western.

Western Libraries Map and Data Centre: Accessing Data at Western using Equinox, <odesi>, and Statistics Canada's website

September 15, 2010

Vince Gray, MLIS & Elizabeth Hill, MLIS

Description: Western faculty and students have access to a wide array of data for research and teaching purposes through agreements such as the Data Liberation Initiative and the Inter-University Consortium for Political and Social Research. This presentation will show how to locate and download data from three sites: (1) Western's Equinox Data Delivery System, (2) Ontario Council of University Libraries' <odesi> database, and (3) the web site of Statistics Canada. The presentation will also describe differences in products offered at the sites, and demonstrate access to both microdata and aggregated data products such as Beyond 20/20 files from the 2006 Census.

2009-10 Presentations

Introduction to MPlus

May 12, 2010

Piotr Wilk
Department of Epidemiology and Biostatistics

Description: Mplus is a statistical modeling program for estimating a wide range of models containing continuous and categorical observed variables as well as unobserved (latent) variables. The primary objective of this workshop is to introduce participants to the commands, options, and settings of Mplus. The workshop will demonstrate how Mplus can handle a wide range of statistical analysis such as structural equation models, growth curve models, multilevel models, and mixture models.

Presentation Slides

Introduction to Stata

April 15, 2010

Rajulton Fernando
Department of Sociology

Description: The session will deal with handling files in Stata, such as using Log Files, various Windows, Data Editor, Data Browser, and Do-File Editor. Some basic Stata commands will be shown with examples, including manipulating data, descriptive statistics, using system variables like _n and _N, and statistical models like Multiple and Logistic Regression.

Presentation Slides

Survival Analysis using R

March 24, 2010

Bruce L. Jones
Department of Statistical and Actuarial Sciences

Description: After a brief discussion of the R computing environment, basic techniques for the analysis of lifetime data will be demonstrated. These include non-parametric estimation of the survival function, fitting parametric and semi-parametric models to lifetime data involving covariates, and testing for differences in lifetime distributions. A basic understanding of introductory statistics will be assumed, but no background in survival analysis is required.

Presentation Slides

Introduction to HLM

February 24, 2010

Paul-Philippe Paré
Department of Sociology

Description: HLM is powerful and easy to use for hierarchical linear and non-linear analysis. The first part of the presentation will focus on the theory and logic of hierarchical analysis, with some specific examples from social research. The second part will introduce HLM software. We will learn (1) how to load data from a general statistical software (SPSS, STATA) into HLM; (2) estimate a random intercept linear model; and (3) estimate a random intercept Bernoulli model (i.e. hierarchical logistic regression).

Presentation Slides

Analysis of Complex Survey Data

January 20, 2010

David Bellhouse
Department of Statistical and Actuarial Sciences

Description: The nature of complex surveys is described and why standard analyses do not give the correct results. Typical analyses for survey data include regression analysis for quantitative data and logistic regression for categorical data. Needed for data analysis are either cluster identifiers or a file with a set of bootstrap weights. Using a U.S. survey on youth at risk, several examples of data analyses will be given when the cluster identifiers are available. A description of the bootstrap approach will also be given

Presentation Slides

Survey Methodology at Statistics Canada

December 16, 2009

Brad Corbett
Analyst, Research Data Centre

Access and Use of RDC Data

October 21, 2009

Rod Beaujot, Brad Corbett, and Zenaida Ravanera
UWO Research Data Centre

Western Libraries Map and Data Centre: Accessing Data at Western using Equinox, <odesi>, and Statistics Canada's website

September 16, 2009

Elizabeth Hill, MLIS

Description: Western faculty and students have access to a wide array of data for research and teaching purposes through agreements such as the Data Liberation Initiative and the Inter-University Consortium for Political and Social Research. This presentation will show how to locate and download data from three sites: (1) Western's Equinox Data Delivery System, (2) Ontario Council of University Libraries' <odesi> database, and (3) the web site of Statistics Canada. The presentation will also describe differences in products offered at the sites, and demonstrate access to both microdata and aggregated data products such as Beyond 20/20 files from the 2006 Census.

2008-09 Presentations

Interested in Tracking Social Change in Canada? Try the General Social Surveys

April 3, 2009

Heather Dryburgh
Chief, General Social Surveys, Statistics Canada

Description: How has the General Social Survey improved our knowledge about Canadian society over the past 20 years? On the occasion of the General Social Surveys’ 20th anniversary, Statistics Canada developed a historical database to facilitate research on trends in Health; Time use; Victimization; Education, work and retirement; Family; Access and use of information communication technology; Social support and aging; and Social engagement. The presentation will provide information on what the database can offer and how it can be accessed.

From SPSS to STATA

February 2, 2009

Brad Corbett
Analyst, Research Data Centre

Description: The procedure on how to use STATA after having done statistical analysis with SPSS will be illustrated with various examples focusing on Bootstrap weights. This will also serve as a preparation towards the Western Summer Institute on Longitudinal Data Analysis (see below).

Data Access at the Research Data Centre

January 15, 2009

Roderic Beaujot and Brad Corbett
UWO Research Data Centre

Description: This inaugural session covers the data available at the Research Data Centre and the procedures on how to access the data. The last half hour will be a planning forum inviting suggestions on data and statistical techniques trainings to be pursued in the Series.