University of Toronto Social Science Methods Week 2019

April 29 - May 3, 2019

The Social Science Methods Week (SSMW) consists of a series of workshops on methodologies with high potential to improve and advance research in the social sciences. Workshops are taught by expert methodologists drawn from departments across the social sciences at the University of Toronto. SSMW is an interdisciplinary forum where researchers can upgrade their methodological toolkit and build cross-disciplinary ties to like-minded scholars.

Registration is free, and open to graduate students and faculty at the University of Toronto. Participants can register for as many workshops as they wish, though enrollment for each workshop is capped at 30 participants. Those wishing to be placed on a waitlist for a workshop should email their request to ssrmweek@utoronto.ca. Space permitting, same-day walk-ins will also be welcomed on a first come, first served basis.

Light refreshments will be provided for registered workshop participants.

Participants should bring a laptop to the workshops with any necessary software pre-installed.

All workshops will be held in Room 36 in the Department of Sociology, St. George Campus, located at 725 Spadina Ave, Toronto, ON M5S.

Questions can be directed to ssrmweek@utoronto.ca.

The University of Toronto Social Science Methods Week is also partnering with PsychHacks 2019, a 24-hour hackathon for students in any discipline who are interested in developing and applying their technical skills. This May 3-4, students will have the opportunity to work together to solve real world data problems and learn from leading scientists! Did we mention food will be provided? Find more details on the event and registration.

Registration

Instructor Biographies

Schedule of Workshops

Workshop Descriptions

Introduction to Python
Length: 2 hours
Instructor: Alicia Eads, Sociology

Python is a programming language. In the hands of social scientists, Python is a powerful tool for collecting, analyzing, and visualizing all kinds of data. This Intro to Python workshop is for both qualitative and quantitative researchers who have little or no experience with Python or any other programming language. We will introduce the basic elements of Python like syntax, data types, objects, and functions. We will learn how to get data into and out of Python by reading and writing files. We will learn how to access and manipulate data. Most importantly, we will introduce the active and exceptionally helpful online Python community, so that once the workshop is over everyone will be able to get good answers on their own to questions about how to do stuff in Python. By the end of this workshop, everyone will get this “joke”: Real programmers count from 0. Please install Python (version 2.7 or 3) on your computer before the workshop. A popular way to do this is to install Anaconda, which includes the spyder application for using Python.

Qualitative Social Science: A Fun Overview
Length: 4 hours
Instructors: Ashley Rubin, Department of Sociology

This workshop offers a crash course in qualitative methods, focusing on (a) designing and conducting good qualitative research, (b) talking about your qualitative research (both with non-qualitative scholars and with other qualitative scholars), and (c) evaluating others’ qualitative research. Because we are focusing on qualitative methods overall—from ethnography to archival/historical to internet-based research—we will not go into the nitty gritty details of any specific method, but instead provide the basic concerns, considerations, and tools all of these approaches have in common. A recurring theme will be inclusivity—recognizing that there are a variety of takes about qualitative research. Scholars using different submethods or coming from different disciplines can offer different advice, some of which can narrowly construe what counts as good or legitimate research—the one right way. In fact, there are many different right ways to do qualitative research, and this workshop will highlight these different takes. Another related theme will be dispelling some common misconceptions about qualitative research (including things like what exactly are qualitative methods, what counts as a research question, how useful is qualitative research, and various research design issues). A final theme is to recognize the (totally normal) anxieties that come with doing all types of research and qualitative research especially.

There are no prerequisites for this course. It will be a good introduction for the novice, a nice refresher for someone with some training, and a different perspective for people trained in one particular tradition or type of qualitative research. There is no software required.

R Basics: Analyzing and Plotting Data
Length: 4 hours
Instructor: Max Barranti, Department of Psychology

Often the most compelling ways to understand our statistical results are through visualizations. In this workshop we will cover how to conduct and plot some of the most common statistical tests for social science research. Analyses covered will include mean comparisons (t-tests, ANOVA) and linear regression (correlation, linear & moderated regression). This workshop will be ideal for those who typically plot their ANOVA and regression results in Excel but want to have greater control over their plotting by using R and the ggplot2 package. Attendees should have the latest version of R installed on their computer. Prior experience with R is helpful, but not required.

Links to download R and RStudio

Download R
Download RStudio

Measuring Theoretical Constructs with Unobserved Variables: An Introduction to Structural Equation Modeling
Length: 3 hours
Instructor: Ulrich Schimmack, Psychology

Empirical sciences require objective measurement, but many constructs in the social sciences are difficult to measure.  The workshop gives a brief introduction to structural equation modeling as a statistical tool for examining construct validity of measures and for testing theoretical predictions at the level of constructs rather than observed measures.  The workshop will also clarify many misconceptions about structural equation modeling and explain why structural equation modeling is superior to simple regression analyses with observed variables.

Registered reports in behavioral science: A how-to guide and description of benefits for scientific progress
Length: 3 hours
Instructor: Stéphane Côté, Rotman School of Management

The credibility of the scientific enterprise rests on the accuracy and replicability of its findings. Unfortunately, researchers are under enormous pressure to publish papers with large, statistically significant effects, and to shuffle non-significant findings away in the “file drawer,” a practice which can severely distort the scientific record. To combat this, an increasing number of journals have adopted the registered report format where well-designed studies addressing important questions are accepted for publication before the study is conducted and the results are known (i.e., even if the results are not significant). This workshop will describe the many benefits of this format for scientific progress, including improved methods and analysis (through a review process that largely happens before studies are conducted), a reduction in studies stored in “file drawers,” and assurances that results have high evidential value. Benefits for your research program, including the efficient use of research funds, will also be discussed. In addition, this workshop will provide a “how-to” guide to a) preparing and submitting a proposal and b) conducting and writing up the results, based on the facilitator’s recent experience with this format at Nature Human Behavior.

Integrated Network Science: Social Network Analysis with R
Length: 3 hours
Instructor: Craig Rawlings, Sociology

Structure matters. New methods, data sources, and computational powers are affording a way to see social structures in ways only dreamed of in the past. Social network analysis formalizes the study of social structures, providing a lens through which interactions, relations, roles, and the schemas and institutions connected with these take shape. This workshop will cover core questions in the study of social structure and provide social network tools for rendering social structures into useful information and scientific knowledge. The emphasis will be on thinking structurally and formulating social network approaches to answering fundamental questions about social relationships. Familiarity with R is a plus, but not required. Sample code and exercises in R will be provided.

Multilevel Modelling: An Introduction
Length: 3 hours
Instructor: Amanda Sharples, Psychology

The increasing complexity of research methods being used in social sciences has required the development of more advanced quantitative techniques. One technique that is becoming increasingly common is multilevel modelling. This workshop will introduce participants to the basics of multilevel modelling: when and why to use it, how to organize data, how to calculate effect sizes, and how to report results. We will also cover more advanced topics including running models with more than two levels, cross-classified models, and multivariate models. There will be a focus on practical application, such that throughout the workshop participants will be provided with syntax and data to run each analysis in R and SPSS. As this is an introductory workshop, participants do not need to have any prior knowledge of multilevel modelling, however, they should be familiar with regression analysis and have knowledge of basic statistical concepts.

Introduction to Web Scraping and APIs with Python
Length: 3 hours
Instructor: Fedor Dokshin, Department of Sociology

The Internet has opened new opportunities for social research. As an ever-expanding range of social activity moves online, websites become repositories of fine-grained and time-stamped records of human behaviour and interaction. Besides digital traces of online activity, the Internet also offers access to enormous digitized datasets about the non-digital world. These data include, for example, massive amounts of digitized historical text, public documents, complex administrative datasets, and records of real-world events. This workshop will introduce participants to a few practical tools for collecting data from the Internet. After reviewing some of the possibilities that the Internet affords social scientists, we will walk through hands-on examples using Python to collect data from the Internet. In particular, we will cover the basics of working with application programming interfaces (APIs) and the rudiments of web scraping and web crawling. To follow along (which you’re encouraged to do), you will need to install Python on your computer (examples will use Python 2.7). Python is free to download and comes pre-installed on all Mac computers. Some basic familiarity with Python will help you get the most out of this workshop. If you have no experience with Python, you are encouraged to enroll in the Intro to Python workshop.

Hearing Silence in Qualitative Interviews
Length: 2-2.5 hours
Instructor:Ping-Chun Hsiung, Sociology

Researchers conducting qualitative interviews tend to focus on the words and stories uttered by the informants. Insufficient attention has been directed to the unspoken and/or unspeakable. This workshop fills this gap by conceptualizing “silence” as an interactive encounter co-constructed by the researcher and informant in qualitative interviews. Using interview excerpts, this workshop illustrates how to make silence audible and the implications of hearing silence in qualitative interviews. It also tackles broader issues pertinent to the politics of knowledge production in social science inquiry.

Models of Social Change: Methods and Principles of Age-Period-Cohort Analysis
Length: 3 hours
Instructor: Ethan Fosse, Sociology

Over the past two decades there has been an enormous increase in the number of studies using age-period-cohort (APC) models to understand social change: no discipline in the social sciences has been unaffected by this revolution. However, confusion abounds on the appropriate application, interpretation, and scope of APC models. This workshop provides an overview of how applied researchers can conduct APC analyses without making common mistakes. There are three parts to this workshop. First, we cover the basics of the classical APC model as well as early modeling attempts such as the equality constraints model and proxy variables approach. Second, we review the key assumptions of the hierarchical age-period-cohort (HAPC) model and various Moore-Penrose estimators, including the intrinsic estimator (IE). Finally, we introduce a new set of techniques for conducting APC analyses that incorporate partial identification, mechanism-based models, and sensitivity analyses. This workshop is intended for applied researchers familiar with multiple linear regression. Some familiarity with structural equation modeling and multilevel modeling would be also be helpful, but is not required. Examples will be shown using the R programming language but general principles will be given for other statistical software.