Abstract: It is not always possible or practical to gather data over a long period of time, and researchers have used retrospective data from questionnaire and interview to gain insights into change over time. There has been little discussion within SLA as to the validity of this data, and it is seemingly accepted at face value. This paper examines both qualitative and quantitative retrospective accounts of changes in self-efficacy, by collecting quantitative questionnaire data over a one-year period, and then asking students to quantitatively recollect their feelings of self-efficacy over that time frame, followed by qualitative description. Results showed that students were able to accurately recollect quantitative data, although there was a noticeable degradation in accuracy with time. Interview data suggested that students were reluctant to admit lack of ability to recall events, and sometimes produced conflicting accounts of events.  This may in part be due to the dual role of teacher and researcher adopted in this study. The implications for researchers and also for teachers are considered.