EvalCompassEvalCompass
Evaluation frameworksTraining evaluationHuman resource developmentLearning effectsBehavior change

Kirkpatrick Four-Level Training Evaluation Model

Kirkpatrick Four-Level Training Evaluation Model
Also known as: Kirkpatrick model、Four-level evaluation model

A framework distinguishing reaction, learning, behavior and results to examine training and workplace outcomes. Work backward from desired results to design learning and workplace support. [1]

Kirkpatrick 4-level measurement worksheet — openReferences and checked sections
Difficulty
Intermediate
Cost
Medium
Duration
Illustrative estimate (adjust for scale and staffing): Levels 1–2 immediately after training to several weeks; Level 3 at 1–3 months; Level 4 at 3–12 months or longer
Content updated: 2026-09-13Source checked: 2026-09-13 (includes AI cross-checking)No documented human reviewIncludes fictional examples

How it works at a glance

The Kirkpatrick model distinguishes 4 levels: participants’ reactions, what they learned, their actual behavior and organizational or other results. It separates satisfaction from the ability to apply learning at work. [1]

Plan backward from desired results to consider necessary behavior, learning and participant experiences. Examine conditions and support for applying learning at work; do not judge later results from post-training questionnaires alone. [1]

Lv.1 ReactionReactionEngagement, relevance, reactionLv.2 LearningLearningKnowledge and skills gained? (test)Lv.3 BehaviorBehaviorUsed at work? (follow-up)Lv.4 ResultsResultsOrganizational results? (KPIs)Four evaluation perspectives
Four perspectives: reaction, learning, behavior, and results. Start from the desired results and examine the environment that supports practice. The numbering does not establish the value of the data or prove causal links.

When to use it and when it is unsuitable

When to use it

  • Measuring and reporting effects of employee training, management development or skills training
  • Organizing desired results across 4 levels during training design (the New World backward approach)
  • Developing evidence to explain training cost-effectiveness to senior management
  • Diagnosing why learning does not transfer to the workplace (Level 3 barrier analysis)

When it is unsuitable

  • Evaluating policies or social programs other than training; other frameworks such as DAC criteria or CIPP are suitable
  • When Level 4 business results are strongly influenced by other factors such as market changes; identifying contribution is difficult

Participants, roles, and deliverables

Participants and roles

Training officers, participants, workplace managers and departments holding necessary work records. Confirm the cooperation needed for collecting outcome indicators in advance.

Main deliverables

  • 4-level measurement plan
  • Level 1 questionnaire findings
  • Level 2 test findings
  • Level 3 behavioral observation records
  • Level 4 results report

Implementation steps

  1. Set evaluation objectives backward from Level 4
    Start with the organizational or work changes expected if training succeeds (Level 4), then work backward to necessary behavior change (L3), learning (L2) and training design (L1). This approach is recommended by the New World Kirkpatrick Model.
  2. Level 1: measure reaction
    Administer a questionnaire immediately after training. Measure 3 indicators—engagement, relevance and customer satisfaction—rather than simply whether it was enjoyable.
  3. Level 2: measure learning
    Measure changes in knowledge, skills, confidence and commitment through pre/post tests, role-play observation or practical checklists. Comparison with pre-tests clarifies the amount of change.
  4. Level 3: measure behavior change
    When opportunities to apply learning arise, examine behavior through observation, supervisor and colleague information and work records. Also examine enabling conditions, including time and workplace support.
  5. Level 4: measure results
    Measure changes in performance indicators such as sales, quality, customer satisfaction or turnover and show their relationship to training. Fully proving causation is difficult, so building evidence of contribution is a practical approach.
  6. Integrate the 4 levels and recommend improvements
    Integrate findings across levels to identify strengths, weaknesses and improvements. Reflect Level 3 barriers, especially uncooperative managers or workplace culture, directly in the next design.

Strengths and limitations

Strengths

  • A simple, intuitive 4-level structure is easily understood by senior managers, training officers and frontline managers
  • Encourages a shift from retrospective evaluation to evaluation planned during design (New World model)
  • Collects evidence on behavior and results as well as reactions to examine training value
  • Widely applicable, with use across business, government, health care and education

Limitations

  • Levels 3–4 are strongly influenced by factors beyond training, making its contribution difficult to isolate
  • Reaction data alone cannot establish changes in learning, behavior or results
  • Level 4 measurement requires long follow-up and organized data collection and is costly

Common pitfalls

  • Reaction surveys end with impressions alone. Ask about issues useful for improvement, such as ease of participation and job relevance, according to purpose.
  • Level 2 tests measure memorization rather than practical application
  • Planning Level 4 measurement only after training; baseline values need to be obtained beforehand

Conditions of use and evidence

  • Design backward from results and examine workplace conditions supporting behavioral application. [1]
Fictional example; not actual results

Connecting counseling record training to improvements in practice

The following fictional example aims to reduce missing handover information through training for 12 counselors.

Outcome indicatorCurrent value (baseline)Target valueMeasurement timingData source
Proportion of records missing required handover information8 of 20 records using the same definition before training = 40% (fictional)Pilot target: 20% or less among 20 post-training recordsBefore training and 4 weeks afterwardAnonymized record audit

Common errors in judgment

Common mistake

Concluding that work results improved because satisfaction was high

How to revise

Check practical skills, workplace behavior and organizational indicators separately

ReasonEach level observes different things.

Common mistake

Blaming staff who learned but did not apply it for training failure

How to revise

Check enabling conditions such as time, supervisor support and forms

ReasonWorkplace conditions affect application of learning.

Organize your thinking by writing

Kirkpatrick 4-level measurement worksheet

A practical template is available to fill in and print on A4 paper.

Open worksheet

Related practical guides

Training effectiveness measurement

Related methods

Logic ModelCIPP Evaluation ModelSuccess Case MethodIndicator DesignResults-Based Management

References and checked sections

  1. The Kirkpatrick Model
    Kirkpatrick Partners
    Section consulted
    Level 1: Reaction; Level 2: Learning; Level 3: Behavior; Level 4: Results
    How it relates to the text
    Design backward from results and examine workplace conditions supporting behavioral application.
    Source checked: (includes AI cross-checking)

Editorial and checking information

Content updated: Source checked: (includes AI cross-checking)No documented human reviewIncludes fictional examplesEditorial policy and scope of checkingCite this pageReport a correction

How to cite this page

EvalCompass Editorial Team. “Kirkpatrick Four-Level Training Evaluation Model”. EvalCompass, updated 2026-09-13. https://evalcompass.jun-nakatani.com/en/methods/kirkpatrick/
Add the date you accessed the page. Text created by this site is available under CC BY 4.0. Quotations, external materials, and bundled fonts have their own terms of use. When reusing content, credit the source, link to the license, and indicate any changes.

Suggest a correction or improvement to this page

Change history
  • — Updated explanations to correspond to the sources and added method-specific fictional completed examples, judgment examples and a change log.