English language education in formal and cram school contexts: an analysis of listening strategy and learning style

ABSTRACT

Formal English language education in Taiwan now starts at Year 3 in primary school, with an emphasis on communicative proficiency. In addition to formal education, attending English cram schools after regular school has become a common phenomenon for Taiwanese students. The main purpose of gaining additional reinforcement in English cram schools is to help pupils enhance their academic performance. However, how pupils use strategies to approach their learning, how they prefer to develop their listening skills, and their relation to different learning environments, say, formal and cram schools, are still unclear. The aim of the present study is to investigate how far formal and cram school English education and the degree of English listening comprehension influences different aspects of listening strategy use and learning styles. The research involved a questionnaire survey of 526 Year 6 primary school pupils. The analysis of the questionnaire was conducted through multivariate analysis of variance and chi-square tests of independence. Results showed that there was a close link between English listening comprehension, learning in English cram schools, certain types of strategy use, and learning styles. Pupils who preferred to learn English in cram schools reported better English listening comprehension and developed better cognitive strategies.

KEYWORDS:

• Listening strategy
• learning style
• primary school
• cram school
• EFL

Notes

1. Primary school pupils are aged from 7 to 12 years (Years 1–6), junior high school students are aged from 13 to 15 years (Years 6–9), and senior high schools include students aged from 16 to 18 years (Years 10–12).

2. The reason why Q3 and Q4 were not used as independent variables in the MANOVA was because of the unequal sample size between those who reported attending cram school (N = 358) and not attending cram school (N = 168). Even though the homogeneities of the variances–covariances of the four factors were not violated, the sample sizes in the two groups have exceeded the recommended ratio: largest/smallest = 1.5 (Howell 2012; Pallant 2013). As a result, the MANOVA test results of these two variables were not reported.