TRAINING DESIGN AND EVALUATION: AN EXAMPLE FROM A SATELLITE BASED DISTANCE LEARNING PROGRAM

Abstract

INTRODUCTION Training programs are intended to improve employee workplace performance. If training delivers the requisite learning effects, it certainly fulfills its intended purpose. With those effects in place, training can rightfully be considered an investment in human resource management. In order for training to be the investment so many managers desire, a linkage must exist between the causal sequence or sequences present in the training delivery and learning effects. According to Kraiger, Ford, and Salas (1993), whose work is based on taxonomies constructed by Bloom (1956), there are three areas or domains, as they are called, into which training effects fall. They are the cognitive domain, the skill domain, and the affective domain. These areas of learning outcomes are generally accepted in the literature (Ammons and Niedzielski-Eichner, 1985). Thus, by consensus from the field, the investment in human resources sought by trainers and managers is unlikely to occur if the cognitive domain, the skill domain, and the affective domain are not demonstratively improved among trainees. In other words, unless trainees leave the training sessions with more or different knowledge, more or different skills, and the desire to use the new knowledge and skills, no improved workplace performance is likely to occur. Without workplace performance improvement, human resource training will not be the investment it is supposed to be. Increasing knowledge, skill, and motivation is no mean production task. It should be addressed with complete recognition of the complexity of the effects and their linkage to the delivery process or the causal sequences present in the training. Training should be designed to the knowledge, skill, and motivation effects or with the same care engineers design manufactured goods for the marketplace. Indeed, since the products of training are intangible, greater care may be called for in their design than manufactured goods. To design training properly, one should adhere to systems principles according to Moore and Kearsley (1996). They assert that systems based training is the best way to ensure quality training; since one of the defining ingredients of systems is that they must be purposeful, it follows that training can only be systems based if it delivers improvement in the three domains developed by Kraiger, Ford, and Salas (1993). No purpose is achieved if learners leave the training in the same cognitive, skill, and affective states they were in when they arrived. The systems approach to training design and delivery consists of several steps or stages (Moore and Kearsley, 1996). The first step is to conduct thoroughgoing needs analysis. This step is crucial if learners are to receive relevant, appropriate, and overall sound training. The second step is to align properly adult training production processes with the three learning domains. Simply put, a curriculum delivery process for the appropriate learning effect must be designed. For example, lectures are frequently assumed to align with cognitive outcomes; exercises or some other form of practice logically aligns with skill; and coaching or encouraging in some consistent and systematic fashion is thought to produce motivation. Following steps one and two should be the evaluation stage or, as engineers might call it, quality control. It is in this area that the analysis to follow will report findings from a satellite based distance learning project conducted during the last half of 1996. Too often it is reported that training is not designed on systems principles and, in such instances, evaluation is not allowing redesign to occur using systems principles (Rice, 1988). Training is seen as perquisite or perk not as an investment in human resources or as a benefit of being employed by the organization. There are also instances of training being performed because some problem has occurred in the workplace. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.