When an organization implements a training intervention to improve the performance of their employees, they need to make sure that the training is actually effective. In order to assess the efficacy of training, the organization needs to be able to evaluate the intervention at multiple levels. Unfortunately, the methods typically used in an evaluation program limit the kinds of claims that one can make about the efficacy of training.

In designing a training intervention, organizations begin with a picture of their current situation, and establish a set of objectives to improve the situation. Organizations then evaluate the efficacy of the intervention based on whether those objectives were successfully met. And with recent innovations in HR and KPI measurement systems, it has become easier to gather data on whether such interventions are having the desired effects. However, it is still unclear what conclusions can be made from all that data. For example, can the data rule out the possibility that the intervention would have been as effective as an alternative? or even no intervention? Indeed, recent summits  and workgroups  have been organized to discuss how data can be better used to contribute towards the organization’s objectives.

The biggest limitation in current evaluation standards is the lack of a comparison group. In clinical studies, interventions of interest are always compared to a control group that receives an alternative intervention. Otherwise, the two groups are designed to be a similar as possible. In this way, any post-intervention differences in performance between groups can be attributed to the difference in interventions. That is, evaluators will be able to determine whether the intervention of interest actually had a causal effect on producing the desired change.

Of course, the use of a comparison group, or control group, in an organizational setting is not always practical. It can be difficult to convince an organization to deploy a placebo intervention on their employees, as it can be perceived to be a waste of their resources. However, it is important to express that it would be a greater waste of their resources if they deployed an intervention that was only equally effective as a placebo.