Training Evaluation

Training evaluation is the systematic collection of data to better manage training programs and training systems. To be effective, evaluation should answer two questions: How did I do? How can I do better? Note that the form of the question and the necessary data vary by stakeholder role (e.g., learner, trainer, training manager, field supervisor). The evaluation literature recognizes two broad forms of evaluation that overlap with these questions: summative and formative. Summative evaluation addresses the question of whether training “worked”—did trainees master the knowledge and skills covered in training, did training result in improved performance back on the job? Summative evaluation tends to be primarily quantitative and conducted after training. Formative evaluation is typically done during training and is more qualitative. Formative evaluation will often focus on the design and delivery of training and is oriented to improving the quality of the learning experience and the learning outcomes for participants. However, rather than thinking of two “types” of evaluation, it is more instructive that all evaluation should address each question from the perspective of all stakeholders. It is also interesting to note that training evaluation was rarely discussed and likely rarely done prior to about 1950. Most industrial training was done on the job by the trainer, and training continued until the trainee could perform effectively; hence the need to evaluate that training was nil. However, as organizations grew in complexity and became more hierarchical, immediate supervisors became less knowledgeable about work tasks, and thus less capable of training subordinates to proficiency; with the emergence of training professionals in the 1940s and 1950s came the recognition of new methods. Through much of the 1960s and 1970s, the development of Training Evaluation Models and methods was guided by training practitioners. However, since the late 1980s, training researchers have advanced new taxonomies of training outcomes, new models for training evaluation, and new ways of linking training evaluation results to organizational planning and decision making. Throughout this article, “taxonomy” will be used to refer to the classification of a set of variables (i.e., different ways in which learning in training can be measured). “Model” will be used to refer to an approach for measuring training outcomes (i.e., a method for deciding which outcomes are the most important to measure for a particular training program). Note that a particular source can include both a taxonomy (a classification of learning outcomes) and a model (an approach to choosing and implementing learning outcomes).