Evaluating the learning in learning objects

A comprehensive review of the literature on the evaluation of learning objects revealed a number of problem areas, including emphasizing technology ahead of learning, an absence of reliability and validity estimates, over‐reliance on informal descriptive data, a tendency to embrace general impressions of learning objects rather than focusing on specific design features, the use of formative or summative evaluation, but not both, and testing on small, vaguely described sample populations using a limited number of learning objects. This study explored a learning‐based approach for evaluating learning objects using a large, diverse, sample of secondary school students. The soundness of this approach was supported by estimates of reliability and validity, using formal statistics where applicable, incorporating both formative and summative evaluations, examining specific learning objects features based on instructional design research, and testing of a range of learning objects. The learning‐based evaluation tool produced useful and detailed information for educators, designers and researchers about the impact of learning objects in the classroom.