Kirkpatrick's four levels of assessment is a framework for measuring learning effectiveness. Developed in the late 1950s and updated multiple times since then, the levels provide a way for training to be updated so it can be improved in the future.
Reaction measures the learners' reactions and feelings about the training. This level essentially asks: Did they like the training? The ideal outcome is for learners to feel positive about the training, that it was effective and enjoyable.
The information gained from this level can be used to improve future training.
To measure reaction, IDs often use surveys or questionnaires issued to students. Questions to ask could include:
The next level is level 2, the learning level. This level seeks to answer the question: Did they learn it? The ideal outcome for this level is that learners' knowledge will have increased as a result of the training.
The learning objectives that the instructional designer outlined during the process are what should be measured at this level.
The learning level is often assessed through an end-of-course quiz to determine if the learning objectives were achieved. Sometimes, the questions can be issued before training begins and then after it is completed, to measure achievement.
The next level is behavior. Did they use it? In other words, did learners apply the knowledge that they gained? It's important to keep in mind at this level that there may be barriers to behavior changes. Perhaps learners did gain new knowledge but are unable to apply it on the job because of management's reluctance to change. Conditions have to be favorable for a behavior change to take place.
Measuring behavior is something that should take place over time. One way to accomplish this is to go into the learners' workplace and observe behavior. Other means of gathering this information could include interviews with learners or their supervisors and questionnaires. It is important to note if the behavior change is being supported within the organization.
The final Kirkpatrick level is the results level. The question here is: Was the impact of the instruction felt in the organization?
At this level, the ID analyzes the results of the learning. This may be the most difficult level to measure. Like the behavior level, this level needs to be assessed over time, post-training. Depending on the training, considerations could include things such as:
At this level, like level 3, measuring would likely take place over time. Measurement involves measurable results, such as increased sales figures, higher scores on customer satisfaction surveys, or decrease in late deliveries, for example.
There are some considerations to keep in mind about Kirkpatrick's four levels of assessment. The third and fourth levels, behavior and results, may be impractical to assess. It can be expensive and time consuming to evaluate learners on these two levels.
The model is a scientific model of assessing learning, but in reality, utilizing all four levels is often impractical.