Abstract

This study aims to prove the influence of a mother tongue upon university students’ performance of a foreign language, in particular that of Thai on English language construction usage. The exploration combines a) a discussion about the English-Thai structural difference of certain cases of measurement expressions, b) corpus evidence and c) an empirical validation of the problematic point through a test. This measure was developed in the form of a specially designed multiple-choice test with its content validity (via the testing index of item-objective congruency) and its internal consistency (via the Cronbach’s alpha coefficient method) verified. The testing was carried out on first-and second-year English major students (two groups for each level) who were required to arrange sentence constituents containing expressions of measurement into complete sentences. Their results were then analyzed statistically (with a Levene’s test and a one-way ANOVA) and qualitatively in order to support that the English usage problem of Thai university students really exists, most plausibly owing to the cross-linguistic distinction. The analysis produced a positive result, and the stability of the measure (through a calculation of Pearson’s product moment correlation coefficient) was substantiated insofar as this measure can be promisingly applied to testing a similar grammar point in other comparable languages.