Abstract
The aim of the study was to map the influence of reading task and text format on reading results with an online text. To this purpose, an experiment performed by S. Gordon et al. (1988) was replicated and enhanced. In four conditions, subjects were given a reading task (summarize or answer specific questions) and an online text (linear or hypertext format). In all conditions, both text and task were administered through the World Wide Web, After the subjects had completed their reading, all were given the same assignment: make a summary and answer specific questions. No significant main effects of the independent variables (format and task) were found on the performance of the subjects. There proved to be a significant interaction effect, however, on the completeness of the summaries. The most thorough summaries were written by subjects who were told before the experiment that they would have to summarize the text, and who were presented with the text in a linear version. As far as reading time was concerned, there was a significant difference between the format conditions: reading the text in linear format took more time than reading the text in hypertext format.