Drawing a picture of the referent of a word produces considerably better recall and recognition of that word than does a baseline condition, such as repeatedly writing the word, a phenomenon referred to as the drawing effect. Although the drawing effect has been the focus of much recent research, it is not yet clear what underlies the beneficial effects of drawing to memory. In 3 experiments, we explored the roles of item and order information following drawing versus silent reading and produced 2 important findings. First, the drawing effect in recall was substantially larger when the 2 conditions were intermixed in a single list compared to appearing in separate lists--in other words, drawing produced a design effect. Second, the studied order was better retained for silent reading compared to drawing in pure lists. These findings are consistent with the item-order account: Memory for the order of drawn lists is poorer because the elaborative act of drawing disrupts the encoding of interitem associations (i.e., relational information) between all list items. Thus, the selection of an encoding task should be informed by how that memory might be used in the future; if one wants to remember individual items, then an elaborative task such as drawing is recommended; if one wants to remember their sequence, it is likely better to read silently.