Comparison of Training Transfer and Effectiveness Models

Abstract

The costs of training devices and simulators have induced the military to formulate models for predicting the training effectiveness, including transfer effectiveness, of the devices during their design and development. Analysis of existing models compared them on the following dimensions: objectives, components, units of analysis, metrics, and development. Development included level of completion, validation, and automation. The models analyzed were those for predicting effectiveness or prescribing device characteristics rather than models for empirical evaluation of existing devices. Baseline models included the military Instructional Systems Development (ISD) model and the Navy's Training Effectiveness and Cost Effectiveness Prediction model. Other models were the Training Device Effectiveness model (TRAINVICE); Training Efficiency Estimation Model; training resource estimators; manpower, personnel, and training estimation models; multi-attribute utility estimation; and methods for specifying training device features. Most of the models were prescriptive, rather than predictive, of effectiveness. Few (including TRAINVICE) estimated transfer to the operational setting.

Human Performance Models for Computer-Aided Engineering

“…An example of the beginning of such an attempt is the Knerr, Nadler, Dowell, and Trifano (1983) army project. The modeling might be either in the factor analytic tradition or in nonlinear factor analysis (e.g., connectionist modeling) techniques.…”

Issues in Developing and Implementing Computer-Based Instruction for Military Training

“…Training design for these tasks followed the Training Effectiveness and% Cost Effectiveness Prediction (TECEP) model (Knerr, Nadler, & Dowell, 1984). TECEP translates task descriptions and learning principles into prescriptions for training.…”

Applying Training System Estimation Models to Army Training. Volume 1. Analysis of the Literature

“…Third, a few review articles and reports have appeared, trying to describe and evaluate some of the models: Adams and Rayhawk (1987a), Braby (1973), Carter (1982, Goldberg and Khattri (1987), Knerr, Nadler, and Dowell (1984), and Rosen, Berger, and Matleck (1985).…”