Teaching statistics and research methods to psychology undergraduates is a major pedagogic challenge. Knowledge of students' conceptual problems in mathematics is important in the current climate of widening access, a burgeoning interest in psychology, and fears about declining standards of numeracy and other quantitative skills. This study compared the mathematical knowledge of two cohorts of undergraduates who entered psychology a decade apart--one in 1992, the other in 2002. Six broadly defined components of mathematical thinking relevant to the teaching of statistics in psychology were examined--calculation, algebraic reasoning, graphical interpretation, proportionality and ratio, probability and sampling, and estimation. Both cohorts were also compared with a 1984 cohort on a subset of items reported in a study by Greer and Semrau (1984). Results revealed highly significant differences between the two cohorts on all six components, with 1992 students outperforming their 2002 counterparts. Males were also found to perform significantly better than females on a majority of components. Level of qualification in mathematics was found to predict overall performance. Comparison with Greer and Semrau's (1984) sample revealed an alarming decline in performance across the two decades on a selection of test items.