Describes a naturalistic paradigm and discusses the value of using a naturalistic, or qualitative, approach to study the effectiveness of computer-based instruction (CBI) and to design effective courseware. Topics discussed include data collection; data analysis; and findings of naturalistic studies of CBI effectiveness. (53 references) (LRW)

Discusses the evaluation of instructional software, describes two versions of a software evaluation model, and reports the results of two studies that compared evaluative decisions based upon subjective software ratings and data collected using both versions of the evaluation model. Teacher use of student data for decision making is also examined.…

Describes a new model for evaluating instructional software that focuses on the extent to which students learn the skills a software package is intended to teach. Results of a field test of fifth grade students are reported, and implications of the initial evaluation of the model are discussed. (14 references) (LRW)

Describes two studies that compared university students' judgments about the readability and studyability of two sets of computer screens: a set of model displays and a set of real screens copied from computer-assisted instruction programs. Relationships among viewers' preferences and field dependence/independence, conceptual style, and gender…