The purpose of this qualitative action research study was to explore improving first semester training practices for graduate teaching assistants (TAs) in the Spanish program at Arizona State University’s (ASU) Tempe Campus. Adding to research on TA training in higher education, a communities of practice (CoP) framework was combined with concrete suggestions on cultivating CoPs to implement a two-part CoP principled training program intervention. Specifically, a goal of the intervention was to address the problem of practice of improving first semester TA lesson planning, communicative language teaching, classroom management, and learning management system usage. Data was collected from interviews, surveys, journals, and training activities from five new TAs who teach Spanish. Data analysis included a multi-cycle qualitative coding process to examine participants’ novice-expert positionalities and the presence of core features of a CoP. Results suggest that regardless of previous experiences, TAs need time to assimilate to the ASU culture, standards, and community. Furthermore, the CoP principled training program showed instances of the necessary core features of a CoP such as joint enterprise and mutual engagement, but also a need for continued community development to address dysfunctions. Implications for these findings point to possible positive effects of continued training through a CoP framework, and a need for reorganization of training practices to allow TAs to legitimately participate in training activities supported by community coordinators as they adjust to the ASU context before beginning their in-service teaching.

The goal of this study was to examine whether there is any effect of phonotactic probability during the early phases of novel word recognition. In order to determine this, I performed two experiments. In Experiment 1, 33 adult monolingual English speakers learned 24 novel word-object pairings, half of which were high English phonotactic probability words and the other half were low English phonotactic probability words. I additionally included three conditions that varied the amount of exposures to each novel word-object pairing (i.e. One Exposure Condition, Two Exposures Conditions, and Five Exposures Condition). Experiment 2 was designed to clarify results found in Experiment 1, with improved randomization and fewer conditions (i.e. One Exposure Condition and Five Exposures Condition). The findings from both experiments were statistically significant in accuracy for Training condition, but not statistically significant for phonotactic probability nor for an interaction between phonotactic probability and Training condition. Although participants demonstrated learning across conditions there is no indication of a relationship between high and low phonotactic probability and novel word recognition. Collectively, these findings suggest that future studies will be necessary to determine if there is indeed an effect of phonotactic probability on early novel word recognition.