Abstract
Teacher-education students were given the task of drawing two graphs, one to represent
categorical data, the other to represent numerical data. A statistical-thinking framework was
used to evaluate the graphs produced by assigning levels of thinking to the graphs and
comparing the levels associated with the two types of data.. The levels of thinking
associated with the categorical data were higher than those associated with the numerical
data. More students were able to reorganise the categorical data compared to the numerical
data. The levels of thinking were not related to the order of presentation of the data.
Steven Nisbet