Summery Events on Quantitative Methods

Network Meta-Analysis for Decision-Making

Time: 9:00 AM – 10:00 AM PDT (12:00 PM – 1:00 PM EST), Friday, June 7th, 2019
Speaker: Joseph C. Cappelleri, PhD, Executive Director of Biostatistics, Pfizer Inc., President-Elect of the New England Statistical Society (NESS)
Registration: RSVP by 11:59 PM, June 5th, 2019.
https://www.eventbrite.com/e/network-meta-analysis-for-decision-making-tickets-61604258997

Abstract
Comparative effectiveness research is a rigorous evaluation of the impact of different treatment options – some of which may not have been compared directly – that are available for treating a given medical condition for a particular set of patients. As a key part of comparative effectiveness research, network meta-analysis enables us to combine trials involving different sets of treatments, using a network of evidence, within a single analysis. This integrated and unified analysis incorporates all direct and indirect comparative evidence about treatments. This presentation highlights and illustrates the concepts and assumptions of network meta-analysis.

2019 UBC Research Computing Summer School

Date: Monday, June 24, 2019 to Thursday, June 27, 2019
Location: University of British Columbia – Point Grey Campus (Rooms TBA)
REGISTER HERE 
This June, WestGrid and the University of British Columbia Advanced Research Computing (ARC) team are hosting their third annual Research Computing Summer School. Courses will explore topics in:

  • high-performance computing and parallel programming
  • machine learning
  • research computing with Python, MATLAB and Chapel
  • next-generation sequencing
  • databases on Compute Canada clusters
  • scientific visualization

Bayesian Network Meta-analysis Workshop
Date: July 3-4, 2019
Speaker: Dr. Audrey Beliveau
Host Organizations: Therapeutics Initiative, the Cochrane Hypertension Review Group, and the Department of Statistics
Fee: faculty / staff = $200; student = $100

Abstract:
Dr. Beliveau will cover best-practice guidelines for NMA – Presentation of the recommendations from NICE, PRISMA and ISPOR for the statistical components of an NMA. The workshop will include discussion of model assessment (fixed vs random-effects models, leverage plots, DIC), assessment of homogeneity and consistency, outputs (SUCRA plots, league tables, forest plot), mathematical description of NMA models (various distribution families and link functions). Bayesian paradigm, trace plots, choice of priors will also be covered. In addition, some hands-on exercises will be delivered in the second day.

Research Presentation: Uncovering Text Features of Hyper-partisan News using NLP with Deep Learning

Time: 11am-12pm, Thursday, March 21, 2019
Location: Educational Library Block, Room 278
Moderator: Xuyan Tang (MA student in MERM)
Presenter: Chiyu Zhang (PhD student in iSchool), Xuejun Ryan Ji (PhD student in MERM), & Yadong Liu (PhD student in Linguistics)

Abstract:
This study aims to identify typical text features of hyperpartisan news via a set of text analytic approaches, as well as detect hyperpartisan news by employing various learning architectures. We will demonstrate how to use natural language processing techniques to obtain information from text data.

Research Talk: What the heck are Learning Analytics anyway?

Time: 11am-12pm, Thursday, Feb 28, 2019
Location: Educational Library Block, Room 278
Moderator: Ryan Ji (PhD student, MERM, ECPS)
Presenter: Dr. Leah Macfadyen, Associate Director, MET, Faculty of Education

Abstract:
Developments in education and learning technologies in recent decades mean that universities are now awash in data about learners and learning. Online teaching tools such as Learning Management Systems (e.g. Canvas), discussion forums, messaging and homework systems, simulations, peer feedback environments and audio/video tools used in flipped or blended courses all collect rich sets of data about learner activity, behaviour, course choices, and performance. As a result, we now have a wealth of e-traces about learners, courses, and programs. In this session, I will showcase some concrete examples of recent learning analytics projects at UBC and other institutions.

Here is Dr. Macfadyen’s PowerPoint slides, https://cp.sync.com/dl/3bf01c170/7r9y5qdq-vgdkwy67-sdm84f2w-4eveaage. Another reference about learning analytics:
Slade, S., & Prinsloo, P. (2013). Learning Analytics: Ethical Issues and Dilemmas. American Behavioral Scientist, 57(10), 1510-1529.

Research Talk: Conceptual Challenges for Methodologies Used to Ensure Fairness and Equity in Assessments, Measures, and Surveys

Time: 11am-12pm, Thursday, Jan. 31, 2019
Location: Scarfe Building Room 2415

Moderator: Dr. Anita Hubley (Professor, MERM, ECPS)
Presenter: Dr. Bruno Zumbo (Professor, MERM, ECPS)

Abstract:
I wish to share with you a bit of my thinking, and some of my intellectual journey, as I have come to conceptualize and develop methods used to ensure fairness and equity in assessments, measures, and surveys — in particular differential item functioning (DIF) and measurement invariance. I address two conceptual challenges that have lingered with me and have informed my theoretical developments. I speculate at the end about the ways in which the subject may continue to develop and in which it may connect with other areas of policy, methodology, and philosophy.

Here is a good background reference for you, https://goo.gl/qtZ7fz.

Power Analysis Workshop

Time: 10am-12pm, Thursday, November 22, 2018
Location: Scarfe Library Block Room 278
Presenters:
Ryan Ji (Ph. D. student, MERM program, UBC)
Ben Hives (M.Sc. student, School of Kinesiology)

Outline of the workshop:
Power Analysis is used to determine an adequate sample size needed to detect a given effect size for a planned study. It can also be used to determine the power, given an effect size and a sample size.

In this workshop, we will approach power analysis conceptually other than mathematically, then we will provide demonstrations via GPower for t-test (independent samples and paired samples), ANOVA, Repeated-Measures ANOVA, Mixed-Design ANOVA, and Multiple Linear Regression. Finally, we will discuss the difference between prospective power analysis and retrospective power analysis, and why the retrospective power analysis should be avoided.

Research Talk: Response Process as a Source of Validity Evidence

Time: 11am-12pm, October 25, 2018 (Thursday)
Location: Scarfe Library Block Room 278
Moderator: Vicki Knight (Assistant Professor, Special Education, ECPS)
Speaker:
Anita Hubley (Professor, MERM, ECPS)

Abstract:
What are people doing, thinking, or feeling when completing measures or tasks? Response processes are an opportunity for us to better understand the meaning of scores. Dr. Hubley will discuss what response processes are (and are not), their importance, the current state of response processes research, and how to examine this source of validity evidence. Hubley and Zumbo (2017) is a good chapter to set the tone for this talk. You can download the chapter from here.

Panel Discussion: Validation Framework in Mixed Methods Research

Time: 11am-12:30pm, September 27, 2018 (Thursday)
Location: Scarfe Library Block Room 278
Moderator: Yan Liu (Assistant Professor, MERM, ECPS)
Discussants:
Amery Wu (Associate Professor, MERM, ECPS)
Richard Young (Professor, Counselling Psychology, ECPS)

Abstract:
Mixed methods have been widely used by researchers in all kinds of fields. However, how do we understand validity and use validation in mixed methods research has not been widely discussed. Drs. Wu and Richard will discuss issues of validity and validation in  mixed methods research. We will also invite our audience to join our conversations. Dellinger and Leech (2007) is a good background reading for this discussion. You can download the paper from here.