Positive Effects of On-The-Job Training

Participation in a one-week on-the-job training course can lead to an improvement in performance of as much as 10% among the call centre employees who receive the training. What''s more, there are positive ''spillover'' effects as their untrained colleagues also improve their performance.

These are among the findings of research by Andries De Grip and Jan Sauermann, which looks at call agents in a multinational telephone company in the Netherlands. The results, published in the May 2012 issue of the Economic Journal, suggest that the positive effects of on-the-job training are as much due to improved motivation as to newly acquired skills.

Analysing an experiment with random participation of employees in a one-week on-the-job training course, the researchers find that participation in the course improves the trainees'' performance by about 10%. At the same time, there are substantial spillover effects: when 50% of the team has been trained, the performance of teammates who have not received the training also increases, by 2.5%.

The results also show that following performance improvements in the first few weeks after the training, the effect decreases over time. This suggests that the training effect is partly caused by an increase in motivation, as opposed to learning. If the training effect were only due to new skills acquired, there would not be a ''wearing-off'' of the training effect within a few weeks after the training.

In addition, the study shows that apart from the main effect (of a 10% performance increase), there are also positive effects on different quality measures, such as the extent to which trained employees understand customers'' questions and show relevant knowledge. This suggests that there is no trade-off in staff performance between quality and quantity.

The advantage of studying the effects of training in a call centre is that performance is easily measurable. For example, the average length of an agent''s calls is a useful measure: shorter calls are in general beneficial for the employer because of lower wage costs per call. The authors also use information from customer surveys, which contain information on customer satisfaction and the quality of calls.

To measure the effects of training, the authors employ a field experiment in which a randomly chosen half of the agents were assigned to a training course. Because the second half did not participate in the training course, this setting makes it possible to attribute any measured performance increase directly to the training course.

This is the first study to use experimental methods to measure the effects of training on staff performance. It is also the first study to provide evidence on spillover effects from training within the workplace.

The research adds to a new strand of work that provides evidence on the effects of social interactions in the workplace. Because agents in a call centre are not engaged in teamwork, these findings suggest that in jobs with more teamwork, spillover effects are likely to be even higher.

''The Effects of Training on Own and Co-worker Productivity: Evidence from a Field Experiment'' by Andries De Grip and Jan Sauermann is published in the May 2012 issue of the Economic Journal.