AbstractIn spite of widespread concern over the state of public science knowledge, relatively little empirical research has been done toward improving the communication of science information to laymen. This paper reports the results of an empirical, quasi‐experimental study on textual variables in science writing and their effects on a lay audience. Strong relationships were found between textual variables such as readability and use of example, and audience variables such as information gain and attitude favorability toward the topic (in this case, enzymology). The second phase of this study, a true experimental design in which the results of the first phase are currently being tested, is describe