This paper studies how corpora can be incorporated into a study in Applied Linguistics field to increase validity. This study aims at testing the validity of a previous study (Atar, 2014) which focused on the reverse transfer from English to Turkish of Turkish bilingual speakers of English. In that study (Atar, 2014) the processing of predictive conditionals (conditionals which require a real possibility) by monolingual and bilingual Turkish speakers are contrasted and it was hypothesized that the bilinguals would use future tenses in their Turkish predictive conditionals (as a result of reverse transfer from English) which is an ungrammatical structure in Turkish. However, one flaw of this study was that the hypothesis took it for granted that English always has future tenses (or modals) in predictive conditionals (please see Atar, 2014 for details). This is an important issue, because if English already uses present tenses in predictive conditionals, the research cannot claim that Turkish bilinguals use future tenses in their Turkish predictive conditionals as a result of the reverse transfer. Consequently, this study sets out to make the description of English predictive conditionals from the British National Corpus (the BNC) to put the claims of this research on a reliable base and this has implications for the use of Corpus Linguistics in Applied Linguistics studies.