Quantitative research in motivation has usually used measures taken outside the context of students' actual learning activities. This study, however, maintains that students' motivation and learning are situated. The paper reports a study that utilises a diary as a research tool to assess students' learning and motivation in context. A class of high achieving students (N=34) kept a lesson diary for 6 weeks about their mathematics lessons. It was found that high achieving students used learning related self-structures to assess the lessons. These learning self-structures mainly include efficacy belief, mastery, interest and self-regulation. Based on students' comments in the open diaries, students' ratings of various motivational constructs in the structured diaries were classified into motivating, unmotivating and neutral groups. ANOVA Tests with Scheffe Post Hoc Tests confirmed that students would show difference in motivation when they perceived a lesson as motivating versus unmotivating. A subsequent discriminant analysis showed that mastery goal, task value and mastery oriented learning climate were the three most important predictor variables that discriminated motivating lessons from unmotivating lessons. The significance of this study is that it provides a theoretical framing that sees motivation not as a static trait but as a process. It also demonstrates initial evidence for the validity of a situated approach to classroom learning and motivation. Further effort is needed to validate the findings with students having diverse learning characteristics and achievement history, using other data collection design as keeping a diary was proved to be ineffective among low achievers.