Substantiating post-training behavioral transfer, that fickle and elusive third level of Kirkpatrick’s training evaluation model, often seems like a hard-fought battle. It’s arguably the most difficult to pull off, as identifying which employee behaviors are a result of training can be equally challenging as figuring out which metrics will show anything useful. Not to mention it requires on-the-job measurement to capture evidence of training effectiveness.

However, the ultimate goal is to determine whether a training operationally accomplished its purpose, which is to change behavior. That is, people can react well to a training, they can nail post-training learning measures, but neither of those things necessarily means the knowledge and/or skills obtained in training are being used day in, day out, as hoped. In other words, favorable reactions and verifiable learning are necessary but not sufficient preconditions for training transfer to actually occur. Consult most guides to Kirkpatrick’s model, and the discussion of transfer focuses on the extent to which a training course helped an individual employee improve his or her performance—it’s about improving this learner or that learner, and not so much about the organization. But there’s a missing ingredient inherent in this approach to transfer: work climate. To ignore the context in which transfer is supposed to take place isn’t so much a case of failing to see the forest for the trees, as it is writing off the forest altogether.

This partially stems from good ol’ fashioned field theory, courtesy of the psychologist Kurt Lewin. Field theory states that behavior is a function of a person and their environment, involving both objective and subjective variables as a person interprets, reacts to and influences his or her surroundings. This applies to training transfer when one begins looking beyond the analogues of the training course, beyond the post-training learning indicators, and into the literal world of work.

Earlier this year, research was published in the journal Military Psychology by Kemp et al. (2015) that shed some light on the interaction of on-the-job skills transfer and employee context. This research was based on 919 active soldiers in 133 active-deployment operational teams with measurements spanning over roughly a seven-year period, following completion of a four- to-six-month foreign language training. This extensive research showed that the more learning that took place during the training (a.k.a. Kirkpatrick’s level two), the better able those individuals were to apply those skills on the job, maintain those skills over time, and generalize (i.e., transfer) what they learned to job tasks. Further, while language skills were shown to decay over time, which was expected, the effectiveness with which soldiers were able to apply those skills stayed somewhat consistent. But the cherry on top (for our purposes) was the finding that when these skills were considered in a team context, the collective language skill of the team played an important part in sustaining transfer. This was true for maintaining skills (i.e., low skill teams were worse at skill maintenance and vice versa), though the extent to which an individual’s language skills could be generalized depended on the average skill level of peers on the same team. In other words, although individual skills had a bearing on individual performance, context mattered when it came to using trained skills in real-world applications.

So what does this all mean for training transfer? What’s climate got to do with it? Well, what sort of transfer climate do you have at your company? Besides trying to foster a supportive work environment, are you paying attention to the skill level of peers that have gone through the same training? Are you focusing on individual learners in isolation from their department or work group? Again, behavior is a function of a person and their environment—don’t neglect half of that equation.