Developing mental orthographic representations in refugee spellers with low literacy: how much input is too much?

Abstract

Background: We used two principles of implicit learning, variability and complexity, to train mental orthographic representations in refugee English learners to improve spelling. Methods: Twenty-eight refugees enrolled in a 10-week English class were trained on classroom words using stimuli designed to encourage implicit learning. We contrasted high-variability visual input combined with either high-linguistic or low-linguistic complexity, using a short (<5 minute) PowerPoint-based training. Participants were regularly tested on their spelling and were compared with themselves using single subject design. Individual effect sizes were aggregated across participants, and we used dependent measures t-tests to compare conditions. Results: Participants learned significantly more treated words than control words in the high-variability/low-complexity condition, but not in the high-variability/highcomplexity condition. Conclusions: Refugees can benefit from interventions designed to promote implicit learning but can be overwhelmed by too much input.

Promoting Literacy in Adult Second Language Learners: A Systematic Review of Effective Practices

“…We selected four experimental or quasi-experimental studies (Blackmer and Hayes-Harb 2016;Condelli et al 2010;Kotik-Friedgut et al 2014;Smyser and Alt 2018), in which the efficacy of a specific pedagogical intervention was tested by comparing a treatment group and a control group (exposed to alternate treatment or no treatment), and two studies (Fanta-Vagenshtein 2011; Trupke-Bastidas and Poulos 2007) where, despite the absence of a control group, literacy gains were measured at the end of the intervention.…”

“…The only notable exception is the study by Condelli et al (2010), which was conducted on a larger number of learners (n = 1344). One common aspect of many studies is the downsizing of participants during the intervention, mainly due to dropouts (e.g., Blackmer and Hayes-Harb 2016;Kotik-Friedgut et al 2014;Smyser and Alt 2018) or unstable performance on test measures (e.g., Smyser and Alt 2018).…”

“…Participants' first language backgrounds are not always reported, but appear to be heterogeneous (spanning from Burmese, Arabic, Nepali, Somali, Kunama, and Kirundi to Swahili, etc.) in the majority of the studies (Blackmer and Hayes-Harb 2016;Condelli et al 2010;Smyser and Alt 2018). Two studies (Fanta-Vagenshtein 2011;Kotik-Friedgut et al 2014) were specifically focused on Ethiopian migrants whose primary language is Amharic.…”

“…Program durations in the studies considered spanned from 10-12 weeks (Condelli et al 2010;Smyser and Alt 2018;Trupke-Bastidas and Poulos 2007) to 30-32 weeks (Blackmer and Hayes-Harb 2016;Fanta-Vagenshtein 2011). One study reports longer duration, up to one year (Kotik-Friedgut et al 2014) Half of the considered studies describe steps taken to guarantee the fidelity of implementation (Blackmer and Hayes-Harb 2016;Condelli et al 2010;Kotik-Friedgut et al 2014).…”

“…Finally, the study by Smyser and Alt (2018) does not provide information about the fidelity of implementation. However, it is legitimate to assume that the proposed treatment (PowerPoint presentation of target word stimuli, likely regulated by settings imposed by the researchers) was independent from the teacher's style of instruction.…”