A study tested the hypothesis that essay examination topics specifying full rhetorical contexts are superior to less information-laden versions of the same topics. Subjects were 360 college students who were given 45 minutes to write essays on 6 topics phrased at 3 information (or specificity of rhetorical context) levels--high, low, and moderate. The essays were holistically scored and were analyzed for the effects of subject matter, information load, and length on score. Results showed that full rhetorical context topics yielded the lowest mean score, while moderate information topics yielded the highest. A critical reading revealed that moderate information load essays were apt to be more sharply focused and better organized. There was also a significant correlation between essay length and score, with longer papers receiving higher scores than shorter papers. The findings suggest that the information level of examination topics can affect writing elicited by a timed test. They also call into question the preference for full rhetorical contexts in writing assignments. (The three versions of the six topics are included in the paper.) (FL)