Likert-scale data in pronunciation research: design, analysis, and visualization

Likert scales are widely used in SLA research to gather a broad range of data, from learners’ individual differences to attitudes toward specific language features. In pronunciation research, they are also commonly used to assess global dimensions of L2 speech, such as accentedness and comprehensibility. In this workshop, we will examine and discuss best practices for designing Likert-scale instruments (e.g., number of response options, wording and anchoring). We will then use R to work through descriptive and inferential analyses of Likert-scale datasets from pronunciation research, considering common decisions that shape results (e.g., treating responses as ordinal, handling missing data). In the final part of the workshop, participants will get hands-on practice with a user-friendly R package to produce publication-ready visualizations (e.g., bar graphs, heatmaps). Basic knowledge of R is recommended but not required; participants will leave with code templates they can adapt to their own projects.

Systematic Approaches to Noticing: Scaffolding Our Visual Representation of Pronunciation Features for Learners 

Schmidt (1990) is frequently cited for the concept of “noticing.”  While the concept of “noticing” is often attributed to grammar instruction (e.g., Batstone, 1996; Ellis, 2006; Larsen-Freeman, 2009; Richards & Reppen, 2014), it has also been applied to pronunciation instruction (e.g., Fouz-González, 2015; Kennedy & Trofimovich, 2010; Mora & Levkina, 2017).  This workshop will start with an examination of what visual representation and input enhancement (Sharwood Smith, 1993) looks like for pronunciation instruction and materials development, pulling from various published and online examples.  To explore systematic approaches that can benefit learners at their particular level, the majority of the workshop will engage participants in comparing different transcription systems and testing out various ways to visually represent suprasegmental features.  The goal is to explore options for scaffolding visual representations for pronunciation learners.  Implications for working with word processors and preparing materials for digital distribution and mass production will also be addressed.  The presenter will pull from examples from his new textbook, Practical Pronunciation: Pronunciation Strategies in Practice, to discuss some lessons learned and some modifications made later in the materials development process.

PSLLT Committee contact: psllt@iastate.edu