The reading data from the 1983-1984 NAEP survey were scaled using a unidimensional item response theory model. To determine whether the responses to the reading items were consistent with unidimensionality, four methods were applied: principal component analysis of phi and tetrachoric correlation matrices; principal component analysis of the image correlation matrix, a method based on the work of Guttman (1953); Bock's full-information factor analysis (Bock, Gibbons, and Muraki, 1985); and Rosenbaum's (1984a) test of unidimensionality, monotonicity, and conditional independence. Results indicated that it was not unreasonable to regard the reading items as measures of a single dimension. (54pp.)