Workshop registration is now closed.
The Workshop consists of three sessions, with two concurrent presentations in each session.
SESSION I (1:00PM - 2:00PM)
- Overcoming Your Fears about Longitudinal Pronunciation Research: Anxiety Reduction
- Practical Steps for Increasing Openness and Reproducibility in Scientific Research
SESSION II (2:30PM - 3:30PM)
- An Introduction to Fitting and Evaluating Mixed-Effects Models in R
- Corpus Linguistics and Pronunciation Analysis
SESSION III (4:00PM - 5:00PM)
- Introduction to Using Mechanical Turk for Linguistics Research
- Visualizing Speech in a Classroom Setting Using Interactive Ultrasound Imaging
You will be taken to a registration page where you can register for the Research Methods in L2 Pronunciation Workshop and/or PSLLT 2017.
Presented by Tracey Derwing & Murray Munro
We will draw on our own experiences with a ten-year, largely quantitative study (Derwing & Munro, 2015) and other recent longitudinal work. Participants will consider a research question regarding an aspect of pronunciation learning, and together we will go through the stages necessary to design a longitudinal study. Target sample size and attrition, developing relationships with programs and participants, options for data analysis, and devising tangential studies will all be discussed. Together we will develop a checklist of strategies to guide prospective researchers toward a methodical and practical approach to conducting longitudinal research.
To conduct effective longitudinal research, it is important to try to anticipate problems that can arise and to mitigate their effects. Of particular concern are such challenges as a closed subject set, attrition, test focus, testing effects, missing data, and demotivation of participants. Without considering these issues in advance, researchers tread in risky waters indeed. In this workshop, participants will gain an understanding of how to implement the core aspects of a longitudinal study through (a) extensive planning prior to initiation of a project; (b) creating designs that allow for incremental publication, rather than waiting until the end; (c) incorporating flexibility to recognize opportune but previously unanticipated studies; (d) teamwork; and (e) organization and management of data, taking into account changing technological platforms.
About the presenters
Tracey Derwing is Professor Emerita at the University of Alberta and an Adjunct Professor in Linguistics at Simon Fraser University. Her primary interests are factors affecting communicative success for L2 speakers.
Murray Munro is a professor of Linguistics at Simon Fraser University. He is an applied phonetician whose interests include L2 pronunciation and forensic linguistics.
Presented by Mallory Kidwell
This practical workshop will review laboratory and personal research practices to improve reproducibility and create more rigorous, open, and impactful research. Topics covered in this workshop will include defining reproducible research, understanding statistical and organizational barriers to reproducibility, and learning how to use the Open Science Framework (https://osf.io/) to manage data and supplementary materials, facilitate collaborations, contain bias, and extend the reach of your research through private or public laboratory operations. This workshop does not require any specialized knowledge to participate but it will be hands-on, so please bring a laptop.
By the end of this workshop, participants will be able to (1) distinguish between different types of reproducibility and their barriers, (2) understand the impact of power and bias on reproducibility, (3) set up a research project using the Open Science Framework (OSF), and (4) use the OSF to keep research more organized, reproducible, and extendable.
About the presenter
Mallory Kidwell is a graduate student in the Clinical Psychology department at the University of Utah. Previously, she held the position of Metascience Project Coordinator at the Center for Open Science, coordinating projects that empirically evaluated reproducibility and the efficacy of open science practices in published scientific literature. Presently, she studies the psychophysiological mechanisms of risky or resilient behavior among adolescents exposed to trauma.
Presented by Charles Nagle
Mixed-effects modeling is a multidimensional statistical analysis that allows the
researcher to partition and explain sources of within- and between-subjects variation
by systematically manipulating the fixed and random effects structures of the model.
Mixed-effects models are advantageous over more widely employed ANOVA because they
are robust in the face of missing data and do not impose the same restrictions related
to independence of observations. This workshop concentrates on fitting mixed-effects
models in RStudio using the lme4 package. Participants will be provided with a brief
overview of mixed-effects models and two methodological review articles before the
By the end of this workshop, participants will be able to (1) describe the advantages
of mixed-effects modeling, (2) read data into RStudio and examine its structure, and
(3) fit and evaluate basic mixed-effects models, including the unconditional and unconditional
linear growth models, using the code provided. Participants will work with a longitudinal
data set and an R script containing annotated code for all of the operations to be
carried out during the workshop. A list of resources related to mixed-effects models
will also be provided.
About the presenter
Charles Nagle is an Assistant Professor of Spanish in the Department of World Languages
and Cultures at Iowa State University. His research examines individual differences
and their relationship to pronunciation development over time.
Presented by Shelley Staples and Okim Kang
Corpora can help us to understand actual patterns of language use, including pronunciation, in particular contexts and by particular language learners. However, the use of corpora in the analysis of pronunciation features has been limited. At the same time, speech science has made progress toward identifying various pronunciation features that can show second language (L2) learners’ progress of language learning or predict their proficiency. It has become common for elements of speech production to be detected by instrument and computer-assisted acoustical analysis (e.g., PRAAT), which characterizes different aspects of pronunciation by examining patterns of speech properties.
This workshop will provide an introduction to using corpora for examining pronunciation features, including 1) building corpora of speech samples; 2) using Praat with pronunciation analysis; 3) coding corpus texts for pronunciation features; 4) using a free program to explore the use of coded texts to examine patterns.
About the presenters
Shelley Staples is an Assistant Professor in the English Applied Linguistics Program at University of Arizona. Her research focuses on the use of corpus linguistics for applications to language teaching and language assessment.
Okim Kang is an Associate Professor in the Applied Linguistics Program at Northern Arizona University. Her research concerns aspects of L2 pronunciation, speech perception and production, automated speech scoring, oral language proficiency assessment, and language attitudes.
Presented by Emily Moeng
The goal of this workshop is to introduce researchers to Mechanical Turk, with a focus
on behavioral experiments that may require auditory stimuli. The workshop is primarily
aimed at those with little to no experience with Mechanical Turk. This workshop will
behavior experiments, including those that might require auditory stimuli. JsPsych
is particularly well-suited for those who wish to run behavioral experiments through
MTurk. Although some web coding experience is suggested, neither web coding nor experience
By the end of this workshop, participants will (1) be equipped with the basic tools
and knowledge to run an experiment through Mechanical Turk, (2) have experience with
using JsPsych to create web-based behavioral experiments, (3) have an understanding
of the limitations of an experiment conducted online and how to best adapt procedures
for such an experiment, and (4) have an awareness of some pros and cons on conducting
linguistic experiments through Mechanical Turk.
About the presenter
Emily Moeng is a graduate student in the Linguistics Department at the University of North Carolina at Chapel Hill. She studies phonological acquisition, specifically how acquisition at one level of categorization affects categories at other levels.
Presented by Sonya Bird and Heather Bliss
This workshop will introduce participants to ultrasound imaging as a tool for teaching and researching pronunciation. We will review technical details and practical issues around ultrasound imaging, including affordability and accessibility. The challenges and benefits of using ultrasound across various contexts will be discussed. We will provide concrete examples of paradigms that can be used concurrently to both teach pronunciation using ultrasound and research its effectiveness. The workshop will include an interactive ultrasound session, in which participants will have the chance to practice their articulation of specific sounds and sound sequences that can pose challenges for learners.
Learning outcomes (i) Participants will learn about the benefits and limitations of using ultrasound imaging technology as a tool for researching and teaching pronunciation of challenging sounds and sound sequences. (ii) Participants will acquire practical information on issues such as accessibility and affordability of ultrasound, as well as examples of teaching and research methods for different learning settings (e.g., individual learners, small groups, classrooms). (iii) Participants will gain hands-on and practical experience in how to use ultrasound imaging through an interactive session in which they can practice their own articulations using an ultrasound machine. (iv) Participants will gain an appreciation of the value of visualizing speech in teaching and learning the sounds of unfamiliar languages.
About the presenters
Sonya Bird is an Associate Professor in the Department of Linguistics at the University of Victoria. She studies the phonetic structures of Indigenous languages of the Pacific Northwest. Heather Bliss is a Banting Postdoctoral Fellow in the Department of Linguistics at the University of Victoria and an Adjunct Professor in the Department of Linguistics at the University of British Columbia. Her work focuses on Indigenous language documentation.
We invite proposals for hour-long L2 pronunciation research methods workshops. Topics may include, but are not limited to, the use of corpora, analysis and manipulation of speech, ultrasound, eye tracking, electropalatography, neuro-imaging methods.
Workshop presenters will receive a $500 stipend (one stipend per workshop) to offset workshop and/or travel costs.
Presentations will be given in English. The online abstract submission site will open on January 9, 2017, and the deadline for submissions is April 7, 2017. Six workshops will be selected, and review outcome notifications will be sent in early May.
Proposals should be given a descriptive title, be double-spaced, have no more than 500 words, and should clearly indicate the workshop’s learning outcomes.
Abstracts should be uploaded to http://linguistlist.org/easyabs/pslltworkshop.