Summer Workshop Series

Workshop Description:

This workshop will highlight and describe meta-analysis and network meta-analysis. Participants will learn how to conduct a meta-analysis to quantitatively answer a review question; interpret and present results; the theory behind different meta-analytic techniques; limitations associated with traditional pairwise meta-analysis; and a brief introduction to network meta-analysis. No prior experience with meta-analysis is required. Having previous knowledge about systematic reviews may be beneficial, but is not required for this workshop.

Presenter:

Myanca Rodrigues, MSc Candidate
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University

Workshop Description:

This workshop will introduce data visualization principles and their applications using the R and STATA statistical programming languages. Participants will be introduce to data visualization practices and concepts developed by Edward Tufte such as data-ink ratio and graphical integrity; learn how to use the ggplot2 and other graphical packages within R to apply Tufte's data principles; learn visualization techniques in STATA; And be introduced to interactive plots using RShiny. Prior experience using R and/or STATA is helpful but not required. Having an understanding of regression analysis and other such statistical concepts are also useful but not required. 

Presenter:

Steve Lee, MA (Masters in Quantitative Methods for Social Sciences), MSc Candidate
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University

Workshop Description:

This workshop will walk through how Bayesian statistics differs from Frequentist statistics, light theory of the Bayesian paradigm, and how Bayesian statistics can be used in applied research.  Participants will learn about Bayesian concepts (e.g. prior distributions, posterior predictive checks, credible interval estimates) and how to implement Bayesian models in the open source language Stan.  No prior experience in Bayesian statistics is assumed.  Experience in R is beneficial, but not required.

Presenter:

Demetri Pananos, PhD Candidate
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University

Workshop Description:

This workshop will provide an applied introduction to the basics of using SAS with your data. During the workshop we will work through all the steps necessary to transform your raw data from excel (and other file formats) into a clean SAS dataset ready to be analyzed. We will go through the process of loading various types of data files into SAS, combining datasets, transforming variables, computing new variables, labeling variables and values, exploring data with simple statistics and graphs, and working with a subset of your data. The workshop will also provide a brief introduction to the basics of various statistical analyses such as t-tests, chi-square analyses, correlations and regressions. This workshop does not teach statistical concepts but teaches how to use these tools in SAS.

No programming experience or SAS knowledge is required. Having completed an undergraduate course in statistics covering p-values, hypothesis testing, analysis of variance and regression is helpful, but not required.

Presenter:

Klajdi Puka, PhD Candidate
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University

Workshop Description:

Observational studies in epidemiology must contend with bias due to selection, measurement and confounding. It is never possible to be certain that study results are unbiased. One can however, explore the potential magnitude and direction of systematic errors offering alternative explanations for the observed results.  This workshop will introduce simple methods for such sensitivity analyses using examples from the literature. Resources for more advanced methods will be provided although this workshop is intended for students with only a basic understanding of epidemiology and biostatistics. Although attention will focus on observational studies the unique challenges of quantitative bias analyses for randomized trials will also be considered.

Presenters:

Dr. Neil Klar, Associate Professor and Undergraduate Chair
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University 

Workshop Description:

This workshop is designed to provide a conceptual understanding and hands-on practice in analyzing complex survey data with missing data. It consists of three parts: 1) introduction to proper design-based analysis of survey sampling data, accounting for sampling weights, clustering, and stratification; 2) introduction to multiple imputation in handling missing data, issues considered include missing data mechanism, building of imputation models, and methods of imputation; and 3) Analysis of complex survey sampling data by application of methods for design-based analysis and multiple imputation. A ‘hand-sized’ dataset will be used to gain insights of the methods, while a large dataset will be used to illustrate proper analysis of complex survey data with SAS, Stata, or R.

Prior knowledge of designed-based analysis of complex survey data and multiple imputation is not required. However, participants with prior knowledge of linear regression analysis and experience with either SAS, or Stata, or R will benefit the most from this workshop.

Presenters:

Dr. Guangyong (GY) Zou, Professor
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University 

Workshop Description:

Two-part series: ‘Health Outcomes, Costs and Cost-effectiveness Analysis’

Health outcomes and costs are the building blocks of cost-effectiveness studies. The aim of this workshop is to provide an introduction to the fundamental concepts and analytical methods used in the analysis of outcomes, costs and cost-effectiveness studies. The workshop will be delivered in two sessions, and participants can attend either or both sessions without losing continuity.

Part 1:

The morning session will start with an introduction to Economic Evaluation methods, with a specific focus on cost-effectiveness analysis. Participants will learn how health economists conceptualize and measure health outcomes for resource allocation decisions. The workshop will provide hands-on experience of using and valuing quality-of-life instruments, analyzing health outcomes data, and interpreting cost-effectiveness results in practice. Participants will also learn about challenges of using health outcomes, including issues such as: who should value outcomes; do different measures produce different results; and how private companies may use and misuse health outcomes data. No prior experience of cost-effectiveness is required. Having some previous experience with Excel and Stata will be useful but not essential.

Presenter:

Shehzad Ali, PhD
Associate Professor
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University

Workshop Description:

Two-part series: ‘Health Outcomes, Costs and Cost-effectiveness Analysis’

Health outcomes and costs are the building blocks of cost-effectiveness studies. The aim of this workshop is to provide an introduction to the fundamental concepts and analytical methods used in the analysis of outcomes, costs and cost-effectiveness studies. The workshop will be delivered in two sessions, and participants can attend either or both sessions without losing continuity.

Part 2:

The afternoon session will cover statistical models to analyse costs or expenditure data (uncensored) commonly available in survey-based, randomized trials and administrative datasets. Participants will learn a variety of statistical models to analyze costs and health expenditure data, how to estimate and interpret the results, understand the strengths and limitations, and when to appropriately use them. It is to be noted that there does not exist a single best statistical model for all situations. Researchers, therefore, must find out the model that is most appropriate for their research question and data at hand. Previous knowledge of linear regression model is required. Having some previous experience with Stata will be useful but not essential. 

Presenter:

Sisira Sarma, PhD
Associate Professor
Department of Epidemiology and Biostatistics
Schulich School of Medicine & Dentistry
Western University