Until recently, it was normal to design the teaching of spoken English based on the written standards of the language. After the advent of spoken corpora and advancements in sound recording technologies, however, the results of large body of research works began to indicate that spoken language, especially, conversation, has its own grammatical features – spoken grammar, that makes it different from writing. This, in turn, forced the spoken language researchers to question the appropriateness of the standard English model for the teaching of speaking skills based on the argument that people in real-life oral interaction mostly, if not wholly, do not speak the way they write. However, this is not to relegate the written grammar model and throw it to the dusty corner of the room, rather, it is to mean that the grammatical structures of both writing and speaking should be introduced in synergy whenever necessary. To this end, the present study examines the effects of corpus-informed spoken grammar instruction on EFL learners’ oral language production. Moreover, the study also intends to investigate the perceptions of the students and the experimenter teacher towards the instruction. To achieve this, a single group interrupted time series quasi-experimental research design was employed. Purposively selected intact group of Wolkite University undergraduate EFL students and their teacher were taken as participants of the study. The students were exposed to the corpus-informed spoken grammar instruction which lasted for six weeks based on the ‘Three I’s Approach’. Alongside the intervention, six different tests were administered to the students on a weekly basis with the intention of measuring the effects of the instruction on their oral language proficiency. What is more, at the end of the instruction, the students were also asked to fill in an exit questionnaire which was intended to investigate their perceptions towards the instruction. Likewise, teacher-logs were used to measure the perceptions of the experimenter teacher. The results of students’ tests and questionnaire were analyzed by using one-way repeated measures ANOVA, one-sample t-test, and descriptive statistics. Whereas, the results of the teacher-logs were analyzed through thematic coding technique. Accordingly, the findings revealed that the students demonstrated significant improvements throughout the tests indicating that, given enough time and practice, the corpus-informed spoken English grammar instruction can enhance EFL learners’ oral language production. Moreover, the findings also indicated that most of the students and the experimenter teacher had positive perceptions towards the instruction. Key Words: Spoken Grammar, Corpus-informed materials, Oral Language Production viii Chapter One: Introduction