Assessment and evaluation

What is the Kirkpatrick Model?

Definition

The Kirkpatrick Model is a framework for evaluating training on four levels: how people react, what they learn, how they behave, and what results follow. It groups the data so an organization can see a program's impact.

Key takeaways

  • Four levels, running from how trainees respond up to results for the organization
  • Level 1 data on its own isn’t enough to show whether people will apply what they learned
  • Giving Levels 1 and 2 more weight than Levels 3 and 4 is a known pitfall
  • Planning starts at Level 4, with the program’s purpose, and works back to Level 1

Why the Kirkpatrick Model matters

The four levels help an organization judge a program on evidence rather than instinct. Evaluation built this way is meant to support decisions, performance, and improvement instead of counting activity. A US government field guide built on the model names lopsided use as a pitfall: heavy weight on Levels 1 and 2, with little attention to Levels 3 and 4.

How the Kirkpatrick Model works

The levels are numbered from the reactions of individual participants up to results for the organization.

  1. Level 1 measures reaction
    Level 1 asks how far participants find the experience favorable, engaging, and relevant, and whether it supports the work the targeted outcomes need. On its own, this data isn’t enough to tell whether people will put the learning to use.
  2. Level 2 measures learning
    Level 2 checks how far participants gain the intended knowledge and skills, along with the attitude, confidence, and commitment that the targeted outcomes require. Confidence and commitment are early signals of what may happen back at work.
  3. Level 3 measures behavior
    Level 3 looks at how far people carry out the critical behaviors at work, with support and accountability such as job aids or manager check-ins. Kirkpatrick Partners says measurement can’t wait 90 days after learning.
  4. Level 4 measures results
    Level 4 asks how far the organizational outcomes a program targeted occur because of it and the performance support around it. Barnett and Mattox frame Level 4 as a question about training’s effect on the bottom line.
  5. Plan from Level 4 down
    Planning begins with why the program exists and how it should affect the organization, then works back to Level 1.
  6. Feed findings back into design
    The levels connect, and the model works as a cycle rather than a straight line. Findings from any level can change the program’s design, how stakeholders are aligned, or the strategy behind it.

Kirkpatrick Model example

A sales enablement lead plans product training for account managers, starting at Level 4 by asking sales leaders which outcome the program should move. At Level 3, she lists the customer conversations account managers should hold and asks managers to observe them. Level 2 gets a short quiz plus a confidence question. At Level 1, a short survey asks whether the content feels relevant. Each round of findings goes back into the course design.

Kirkpatrick Model vs Phillips ROI Model

Phillips’ ROI Model follows Kirkpatrick’s levels and adds a fifth one that evaluates the return on investment (ROI) of a training program after it ends. Phillips also recommends evaluating most programs at Level 1 and very few for ROI.

CriterionKirkpatrick ModelPhillips ROI Model
Number of levelsFour, from reaction to resultsFive, with ROI as the added level
Final questionDid the targeted results happen?Was it worth the cost?
Where evaluation startsPlan from Level 4, the program’s purpose, back to Level 1Evaluate most programs at Level 1 and only a few for ROI

How LearnWorlds supports this

Kirkpatrick Model in LearnWorlds

LearnWorlds analytics and reports track survey responses and exam results. Survey responses give you reaction data for Level 1, and exam results give you learning data for Level 2. With the form and survey builder, you can add forms as a course learning activity. You can also schedule automated reports to run daily, weekly, or monthly, and automate report distribution to keep stakeholders up to date.

FAQ

The answer depends partly on use. A field guide based on the Kirkpatrick Model lists overemphasizing Levels 1 and 2 as a pitfall. In a survey Barnett and Mattox report, nearly every course got Level 1 evaluation, and coverage dropped as the levels went up [3]. Kirkpatrick Partners calls the belief that Level 4 is hardest to measure “a common misconception” [1].

To apply the Kirkpatrick Model, identify the leading and lagging indicators for Level 4, then look for the ones the organization often tracks already. Key stakeholders agree in advance on what success means to them, which Kirkpatrick Partners calls return on expectations (ROE).

References

  1. Kirkpatrick Partners. What is The Kirkpatrick Model? Kirkpatrick Partners, LLC.
  2. U.S. Office of Personnel Management (2011). Field guide to training evaluation: Demonstrating the value of training at every level. OPM.
  3. Barnett, K., and Mattox, J. R., II. Measuring success and ROI in corporate training. Journal of Asynchronous Learning Networks, 14(2), via ERIC.
  4. Schroeder-Strong, M. P., Schreiber, B., and Bennett, W (2022). A methodology for projecting the return on investment of training technologies. Military Psychology, via PubMed Central.

Related terms

Instructional design

ADDIE

The ADDIE model is a five-stage process for designing courses and training: analyze, design, develop, implement, and evaluate.

Read definition

Measure how learners react and what they learn

Track progress, exam results, and survey responses, then schedule daily, weekly, or monthly reports for your stakeholders.

Published:

Last Update: