This non-experimental study gauged the English lexical preferences of 545 native Japanese university students. Specifically, this univariate design collated the frequency of English loanword selection in preference to non-borrowed lexical items. A major focus of the study design was instrument validity, addressing the flaws of two previous studies (Brown, 1995; Small, 2002). With keen attention to vocabulary level, the Genius English-Japanese dictionary 5th edition (Minamide, 2014) facilitated the compilation of an initial, sizeable loanword corpus. Subsequent meticulous corpus revisions provided the lexical content for an instrument featuring loanword vocabulary options. The research instrument comprised 50 brief sentences, each with four valid lexical items in a multiple choice format. Thirty sentences included the choice of one English loanword. The combined frequency of loanword selection from an initial study (n = 283) and a replicated study (n = 262) (N = 545) showed an overall relative frequency average of 0.42. Chi-square analyses of the data comparing both studies indicated that differences in lexical selection were significant at p<0.05 for 15 of the sentences with loanwords. Some of the limitations of working with nominal data are highlighted. The use of loanwords by native Japanese speakers can impede effective communication in English because they are not always an appropriate lexical choice. To help minimise perceived and pragmatic failure when communicating, Japanese need to be aware of the pitfalls of using loanwords and improve their English proficiency by increasing their lexical range and flexibility.