Go back to abstract list
Assessing heritage language knowledge using acceptability judgments: The impact of modality
by Melissa Bowles (University of Illinois at Urbana-Champaign)Despite criticisms about their appropriateness (Polinsky, 2016), Acceptability Judgment Tasks (AJTs) have been used extensively to measure heritage speakers’ knowledge and even to assess the effectiveness of heritage language instruction (Montrul & Bowles, 2010; Potowski, Jegerski, & Morgan-Short, 2009). Most often such AJTs have been written, although some studies have employed aural (Knightly et al., 2003; Sherkina-Lieber et al., 2011) or bimodal (Montrul, Bhatt, & Girju, 2015) AJTs, which have been hypothesized to be a more valid measure of heritage speakers’ knowledge. This study empirically tests that premise by having Spanish heritage speakers complete three parallel 75-item Likert-scale AJTs—one written, one aural, and one bimodal—over the course of two testing sessions, interspersed with other tasks. Data collection is ongoing, but preliminary findings from 10 participants reveal no significant differences in judgments based on modality. If these results hold with the full data set, they would suggest that heritage speakers do not actually perform differently on AJTs depending on their modality, as they do on other, more communicative tasks (c.f., performance on oral vs. written tasks in Montrul, Foote, & Perpiñán, 2008). Implications for heritage language assessment are discussed.