摘要:This article reports on an analysis of assessment items in a Level 1 Earth Science and Climatology module at Liverpool John Moores University (LJMU). It examines the effect of increasing the diversity of assessment methods in the module and increasing the number of summative assessment items from 4 to 8. The effect of doubling the number of assessment items on student performance is examined. Regression analysis highlights differences in how well the marks in the 8 assessment items predict the students’ final module total. The analysis highlights how students ranked overall in the top quartile performed better in the so called ‘deep learning’ assessments (field reports, weather analysis and use of a wiki) whereas students in the lowest quartile performed better in so called ‘shallow learning’ assessments (on-line multiple choice test and the formal written examination). Individual student’s ‘assessment profiles’ are examined and strengths (high class ranking) compared with weaknesses (low class ranking). The reasons for the differences are explored and discussed in the light of students’ motivations, recently introduced ‘graduate skills mapping’ in the University and research by assessment experts.