Japanese English Learner Self-assessments on the CEFR-J’s A-level Can-do Statements Using Four and Five-point Response Scales

Abstract

Just as the CEFR (Common European Framework of Reference) did in Europe, likewise has the CEFR-J (CEFR-Japan) impacted the foreign language education industry of Japan. Both systems operate using illustrative descriptors which describe a learner’s communicative competences in five language skills across six levels. The current study reports on Japanese English learner self-assessment on the CEFR-J’s 50 A-level descriptors using either a four-point or a five-point scale to determine if a neutral response option (Neither agree nor disagree) influenced participants’ responses. Self-assessment by Japanese language learners has been shown to be subject to cultural factors related to social desirability phenonmena, resulting in high selection rates of mid-scale response options no matter the content of the item or the size of the scale. Overall, no significant differences between mean responses on a four-point (no neutral category) and a five-point (contains an inherent mid-point) rating scale were found following controls for scale size. Conversely, significant interactions were found for rating scale, skill (reading and spoken production) and descriptor difficulty level (A1.1 and A2.2). When the distance between responses and scale mid-point was measured and compared across rating scale to determine whether the inclusion of a neutral option appeared to influence selection rates, no significant differences were found for 68% of all descriptors. While inclusion of a middle response option had a far lesser impact on responses than has been previously shown, further research is required to determine the impact of differing scale types on Japanese English learner self-assessments. The influence on responses from socio-cultural factors, response styles, task-familiarity, language skill, the number of response scale categories and language proficiency are discussed.