The purpose of this study is to explore the relationship between a set of item characteristics and the difficulty of TOEFL[R] dialogue items. Identifying characteristics that are related to item difficulty has the potential to improve the efficiency of the item-writing process The study employed 365 TOEFL dialogue items, which were coded on 49 variables, including 5 significant variables reported in Nissan, DeVincenzi, and Tang (1996). Of the 5 significant variables in Nissan et al., 3 correlated significantly with item difficulty in this study. Another 11 met a critical probability criterion. These 11 included representatives from three broad categories of variables: 2 in the category of word-level factors, 1 in the category of discourse-level factors, and 8 in the category of task-processing factors. Multiple regression analyses indicate that the variables in this study account for about 40% of the variance in item difficulty. The following are appended: (1) Coding Instructions; and (2) Instructions for Coding Lexical Overlap. (Contains 3 tables.)