There are numerous studies about stance markers across registers (e.g. Petch-Tyson, 1998; Bithener et al., 2006; Martinez, 2005). Researchers have also carried out stance markers research in the form of contrastive studies (Biber and Finegan, 1988 and 1989 (written and spoken modes); Berman, 2004 (different levels of proficiency in second language writers); Neff et al, 2003 (native and non-native speakers of English)). Many researchers have combined these two approaches by making contrastive studies on stance markers across registers (e.g. Chafe, 1986; Holmes, 1988). This study makes a comparative analysis of stance markers’ usage between non-native and native speakers of English in spoken discourse, here Macau and Scottish speakers. Then, possible reasons for any differences of usage of stance markers between Macau and Scottish speakers will be speculated on. In order to explore the patterns and the characteristics of usage of stance markers, a corpus-based approach to stance markers used by NSs and NNSs is applied in this study. MOC (The Macau Oral Corpus) and SCOTS (Scottish Corpus of Texts and Speech) were selected as the major source data of this research. The Macau Oral Corpus is composed of 6 interviews with Macau local people of similar background talking about their personal experience and the history about Macau. SCOTS is the Scottish Corpus of Texts Speech which was organized and compiled by Glasgow University. I selected a sub-corpus of SCOTS that also consists of 6 interviews of Scottish people talking about their own history. Based on the classification of stance markers by Biber et al. (1989), this research investigates the frequency of a set of these major types of stance markers (stance modals, stance adverbs and stance complement clauses, vi mainly to and that clauses) across MOC and SCOTS’ usage in spoken discourse. The specific classification of stance makers’ sub-distribution are investigated using the sub-corpora from MOC and SCOTS. The hypothesis of this research is that there would be a statistical difference of using stance markers across MOC and SCOTS from the perspective of frequency and distribution. However, these main findings of the t-test in the present study indicate that there is no statistical difference between MOC and SCOTS’s usage of stance markers in spoken discourse in general. Stance modals, stance adverbs and stance complement clauses were investigated respectively across NSs and NNSs. The results generated from the t-test analysis show that there is a significant statistical difference between MOC and SCOTS’s usage of stance modals including three categories (possibility/permission, necessity/obligation, prediction/volition), while this is not the case for stance adverbs (single adverbs and prepositional phrases ) or stance complement clauses used by MOC and SCOTS. It is expected that this study will give suggestions for learners to use stance markers more accurately and precisely. Therefore, they themselves will express opinions and attitudes more clearly in speech context and strengthen the awareness of using stance markers so as to improve their discourse competence. A multi-dimensional approach to stance markers should be applied in future research. It is also hoped that future large corpora of interviews will be helpful to enlarge the findings in this field of research.