Evaluating Training Effectiveness
Measuring the effectiveness of training programs consumes valuable time and resources. As we know all too well, these things are in short supply in organizations today. Why should we bother?
Many training programs fail to deliver the expected organizational benefits. Having a well-structured measuring system in place can help you determine where the problem lies. On a positive note, being able to demonstrate a real and significant benefit to your organization from the training you provide can help you gain more resources from important decision-makers.
Consider also that the business environment is not standing still. Your competitors, technology, legislation and regulations are constantly changing. What was a successful training program yesterday may not be a cost-effective program tomorrow. Being able to measure results will help you adapt to such changing circumstances.
The Kirkpatrick Model
The most well-known and used model for measuring the effectiveness of training programs was developed by Donald Kirkpatrick in the late 1950s. It has since been adapted and modified by a number of writers, however, the basic structure has well stood the test of time. The basic structure of Kirkpatrick’s four-level model is shown here.
Figure 1 – Kirkpatrick Model for Evaluating Effectiveness of Training Programs
An evaluation at each level answers whether a fundamental requirement of the training program was met. It’s not that conducting an evaluation at one level is more important that another. All levels of evaluation are important. In fact, the Kirkpatrick model explains the usefulness of performing training evaluations at each level. Each level provides a diagnostic checkpoint for problems at the succeeding level. So, if participants did not learn (Level 2), participant reactions gathered at Level 1 (Reaction) will reveal the barriers to learning. Now moving up to the next level, if participants did not use the skills once back in the workplace (Level 3), perhaps they did not learn the required skills in the first place (Level 2).
The difficulty and cost of conducting an evaluation increases as you move up the levels. So, you will need to consider carefully what levels of evaluation you will conduct for which programs. You may decide to conduct Level 1 evaluations (Reaction) for all programs, Level 2 evaluations (Learning) for “hard-skills” programs only, Level 3 evaluations (Behavior) for strategic programs only and Level 4 evaluations (Results) for programs costing over $50,000. Above all else, before starting an evaluation, be crystal clear about your purpose in conducting the evaluation.
Using the Kirkpatrick Model
How do you conduct a training evaluation? Here is a quick guide on some appropriate information sources for each level.
Level 1 (Reaction)
- completed participant feedback questionnaire
- informal comments from participants
- focus group sessions with participants
Level 2 (Learning)
- pre- and post-test scores
- on-the-job assessments
- supervisor reports
Level 3 (Behavior)
- completed self-assessment questionnaire
- on-the-job observation
- reports from customers, peers and participant’s manager
Level 4 (Results)
- financial reports
- quality inspections
- interview with sales manager
When considering what sources of data you will use for your evaluation, think about the cost and time involved in collecting the data. Balance this against the accuracy of the source and the accuracy you actually need. Will existing sources suffice or will you need to collect new information?
Think broadly about where you can get information. Sources include:
- hardcopy and online quantitative reports
- production and job records
- interviews with participants, managers, peers, customers, suppliers and regulators
- checklists and tests
- direct observation
- questionnaires, self-rating and multi-rating
- Focus Group sessions