The active and passive voice are equally comprehensible in scientific writing

Thesis (Ph. D.)--University of Washington, 1997The passive voice is typically thought to be wordy, impersonal, and difficult to read. Although early studies examining the passive voice appeared to show that passive voice sentences are harder to process and less preferred than active voice sentences, later studies pointed out a number of communication situations when passive sentences were not more difficult to process and were indeed preferred to their active counterparts.The present investigation focused on the use of passive voice in scientific reports. Three studies were performed. Study 1 focused on the incidence of the use of passive voice verbs in empirical reports in fields of varying object-orientation (psychology, botany, and chemistry/physics) and across article report sections (Introduction, Method, Results, and Discussion) of varying objectivity. The incidence of passive voice verbs was greatest in the most object-oriented field (chemistry/physics), second highest in a somewhat less object-oriented field (botany), and lowest in the least object-oriented field (psychology). Similarly, the objective article sections (Methods and Results) had a higher percent of passive voice verbs than the interpretive article sections (Introduction and Discussion).Study 2 was an empirical investigation designed to assess reading speed and comprehension for "high-active" vs. "high-passive" versions of scientific reports. Two experiments were conducted, each of which included two articles that varied solely in voice, the "high-active" version having 100% active voice verbs and the "high-passive" version having 42% passive voice verbs. Subjects read one high-active article and one high-passive article. After reading each article, they answered 16 multiple-choice questions designed to test their understanding of the article. Large samples were used to ensure highly sensitive experiments. For both experiments, there were no significant differences in either reading speed or reading comprehension due to differences in voice.Study 3 compared student lab papers rated by instructors as weak versus strong to determine which factors, including voice, might account for the differences in instructor ratings. The analysis revealed differences in content development, paper length, incidence of violations of American Psychological Association guidelines, organization, and degree of sentence cohesion, all in favor of strong papers. However, it did not reveal any significant differences in percentage of passive voice verbs used. Additionally, the pattern of usage in the lab reports was similar to that observed for the published articles examined in Study 1.These results have several implications: (a) passive voice verbs are widely used in scientific writing (Study 1); (b) passive voice verbs are generally used appropriately, primarily for presenting objective information (Studies 1 and 3); (c) rhetorically-appropriate passive voice verbs do not elicit lower reading comprehension scores nor longer reading times than active voice verbs (Study 2); and (d) the incidence of passive voice verbs in student lab reports did not affect instructor ratings of paper quality (Study 3). This study demonstrates that passive voice need not create the kind of problems with which they are commonly associated. It also demonstrated the usefulness of evaluating linguistic devices such as the passive voice in stand-alone documents (as opposed to isolated sentences or short passages) and in specific rhetorical contexts (scientific reports)