Learning Program Evaluation methods - the old and the new!

As someone working in the learning and development function I have asked and been asked questions around “how to measure the effectiveness of a training program?” many times. Each time the answer comes short of satisfactory. Mostly because it is a difficult question to answer and a difficult problem to solve, the variables are enough to make you feel dizzy. 

Let’s look at the considerations - 

• The People function does not look at an investment not only into development, but also engagement and in some cases a branding tool to attract talent
• Most L&D people are in the profession because they want to help others improve their performance. They didn't get into it to collect and analyze data
• The Kirk Patrick model,developed in the 1950s, is the most used model for evaluation - I am sure this is the “go to” when we think of the topic (see below)
• The next most used model is the one by Jack Philips with the added Level 5 - ROI 

These models work best when the training outcomes are fairly isolated and not affected by other factors (this almost never happens in life!) 

There are many other evaluation methods that can help by themselves or in addition to others to create maybe a more robust design based on the needs of our program. 

Some other ways - 

1. By comparing programs to the established gold standards in the industry
2. Based solely on the performance of the employee before, right after and in the next year or so.
3. In any training, some participants are going to do better than others. The Success case method for evaluation was developed by Robert Brinkerhoff to address this. The model essentially studies what the participants who did the best did differently and uses this insight to improve the program

The experts are doing some really interesting work in the fields of forward looking ways for evaluation, i.e. predictive analysis. 

What ways do you use to evaluate the success of your training?