Search results

It has become fashionable to say that ‘training is an investment in human capital’. Unfortunately, many people who like to use this phrase dislike justifying the investment in terms of actual contributions to the company's profit and loss statement. This inconsistency is not surprising because, if this cold, hard criterion of demonstrating useful results were used, many of today's training courses would be thrown lock, stock, and barrel out of the corporate window, along with the men responsible for them. Managers who reject such testimonials as, “I'm sure it's helping,” are doing their companies a service. They are correctly resisting the pressure to continue spending money without some reasonable evidence that such funds will produce at least an average return. The message is clear, therefore. If the personnel and training staff really feel that a particular course should be retained, it is their responsibility to develop a method of measuring the dollar value of this course. The defence that there are just some things that cannot be measured is really an admission that either these individuals have not tried very hard or that this particular course in its present form is probably not achieving the desired result. When an objective criterion for measuring contribution to profitability can be presented, this is the time for management to consider the proposal seriously.

A relatively recent advance in analyzing longitudinal data, structural equation modeling with structured means, for examining the impact of organizational change and development interventions, is presented. Some of the limitations of current approaches to analyzing data collected from “experimental” and “control” groups are discussed, along with why structural modeling is particularly useful for real‐world experiments and quasi‐experiments. An illustration is then given, applying this approach to data collected from a team‐building intervention which involved 2,331 employees in 16 plants of a large garment manufacturer. Implications of the research are briefly considered.