50+ training survey questions mapped to Kirkpatrick's four levels (2026)

US companies spend roughly $100 billion a year on workplace learning, and most of them measure it with a post-session smile sheet that asks learners to rate the instructor. Training researcher Will Thalheimer has shown those ratings correlate with on-the-job behavior change at roughly r=0.09. Near zero. If your training evaluation stops at "did you enjoy the session?", you are not measuring training. You are measuring entertainment.

This guide gives you 40+ training survey questions mapped to Donald Kirkpatrick's four levels of evaluation, explains how to structure Level 1 questions that actually predict transfer, and shows how to extend the model to ROI with Jack Phillips' methodology. Every question is tagged with its level, timing, and effectiveness rating.

What you will find in this guide:

• Why measuring training effectiveness matters
• Kirkpatrick's four levels explained
• 50+ training survey questions grouped by level
• How to write Level 1 questions that predict Level 3 behavior
• Measuring training transfer (Level 3) with 30, 60, 90-day follow-ups
• Measuring ROI (Level 4) with the Phillips model
• Best practices for timing, length, and anonymity
• Common mistakes in training evaluation
• Free Formbricks training survey template

---

Why measuring training matters

The business case for training evaluation is straightforward: if you do not know whether a program works, you cannot improve it and you cannot justify the spend. And yet most L&D programs never measure past Level 1.

• $100 billion. Estimated annual US spend on workplace learning and development, per ATD's State of the Industry reports.
• Low transfer rates. Classic research by Baldwin and Ford (1988) estimated that only about 10% of training content is actually transferred to job performance. More recent estimates run higher but rarely above 40% for typical corporate training.
• Manager support matters more than content. Multiple meta-analyses show that manager support before, during, and after training is the single largest predictor of transfer, bigger than content quality or delivery format.

The goal of a training survey program is not to rate the instructor. It is to answer three questions: Did learners react positively? Did they learn something? Are they doing it on the job?

---

Kirkpatrick's four levels

Donald Kirkpatrick introduced the four-level model in 1959, and Kirkpatrick Partners has maintained and updated it since. The four levels represent a hierarchy of training outcomes: each level is harder to measure than the last, and each level tells you more about actual effectiveness.

Level	What it measures	Typical timing	Difficulty	Example question
1. Reaction	How learners felt about the training	Immediately after	Easy	"How relevant was this training to your role?"
2. Learning	What learners know or can do after training	Immediately after	Medium	"What are the three key steps for handling a customer escalation?"
3. Behavior	What learners are doing differently on the job	30 to 90 days after	Hard	"How often are you using the technique from the training?"
4. Results	What changed for the business	3 to 12 months after	Hardest	Linked business metric movement

Level 1: Reaction. Measures immediate perception. Did the training feel worth the time? Was the content relevant? Was the instructor prepared? Level 1 is cheap and fast but only weakly predicts behavior change unless the questions are designed carefully.

Level 2: Learning. Measures knowledge and skill gains. Can the learner answer questions, demonstrate the skill, or solve a problem they could not solve before? Level 2 is usually measured with a quiz, scenario, or demonstration task.

Level 3: Behavior. Measures application. Are learners using the new knowledge or skill in their actual work? Level 3 requires a delay (30 to 90 days) to give learners time to apply the training, and usually a survey to the learner and their manager.

Level 4: Results. Measures business impact. Did the intended business metric move as a result of the training? Level 4 is the hardest to isolate because many factors other than training affect business metrics.

Most L&D programs stop at Level 1. The ones that produce measurable value go at least to Level 3.

---

50+ training survey questions by level

Each question is tagged with the Kirkpatrick level, the recommended question type, and a priority (Essential, Recommended, or Nice-to-have).

Level 1: Reaction questions (questions 1-12)

Send immediately after the training. These are the traditional smile sheet questions, rewritten to predict transfer.

1. How relevant was the content to your current role?

• Type: Likert (1-5) | Essential
• Relevance is a stronger transfer predictor than enjoyment.

2. How confident do you feel applying what you learned to your actual work?

• Type: Likert (1-5) | Essential
• Self-efficacy is one of the best Level 1 predictors of Level 3 transfer.

3. How clear was the instructor's explanation of the material?

• Type: Likert (1-5) | Recommended
• Instructional quality.

4. How well did the training reflect real situations you encounter at work?

• Type: Likert (1-5) | Essential
• Fidelity to the job context.

5. How likely are you to apply at least one thing from this training in the next two weeks?

• Type: Likert (1-5) | Essential
• Intent to apply. A Thalheimer-style performance-focused reaction item.

6. What is one thing you plan to start doing differently because of this training?

• Type: Open-ended | Essential
• Forces the learner to name a specific application.

7. What part of the training will you find hardest to apply back on the job?

• Type: Open-ended | Essential
• Surfaces anticipated transfer barriers before they happen.

8. How useful were the examples and exercises?

• Type: Likert (1-5) | Recommended
• Specific feedback on instructional design.

9. How was the pace of the training?

• Type: Multiple choice | Recommended
• Too slow / About right / Too fast.

10. How would you rate the overall quality of this training?

• Type: Rating (0-10) | Essential
• Program-level NPS-style score for benchmarking across programs.

11. Would you recommend this training to a colleague in a similar role?

• Type: Binary (Yes/No) | Essential
• Advocacy proxy.

12. What is one thing we should change about this training?

• Type: Open-ended | Recommended
• Open improvement feedback.

Level 2: Learning questions (questions 13-18)

Paired with a knowledge check or scenario task. Administered immediately after the training.

13. What are the three most important things you learned from this training?

• Type: Open-ended | Essential
• Free-recall is a stronger Level 2 measure than recognition.

14. Before the training, how would you have rated your ability to [skill]?

• Type: Likert (1-5) | Essential
• Retrospective pre-test to avoid response shift bias.

15. After the training, how would you rate your ability to [skill]?

• Type: Likert (1-5) | Essential
• Compare with question 14 to measure self-reported gain.

16. Describe how you would handle [scenario from the training].

• Type: Open-ended | Essential
• Application task, more reliable than multiple choice for complex skills.

17. Which of the following are true about [topic]? (Select all that apply)

• Type: Multiple choice | Recommended
• Knowledge check, appropriate for factual content.

18. How confident are you in your ability to explain [concept] to a colleague?

• Type: Likert (1-5) | Recommended
• Protégé effect: the ability to teach predicts mastery.

Level 3: Behavior questions (questions 19-28)

Send 30, 60, or 90 days after the training. Ask both the learner and the learner's manager.

19. How often are you using the techniques from the training in your day-to-day work?

• Type: Frequency scale (Never / Rarely / Sometimes / Often / Always) | Essential
• Primary Level 3 transfer measure.

20. How effective has the training been in helping you do your job better?

• Type: Likert (1-5) | Essential
• Self-reported effectiveness at a meaningful delay.

21. What has made it easier to apply what you learned?

• Type: Open-ended | Essential
• Surfaces enablers: manager support, tools, documentation, practice opportunities.

22. What has made it harder to apply what you learned?

• Type: Open-ended | Essential
• Surfaces barriers: workload, conflicting priorities, lack of manager support.

23. How supportive has your manager been of applying this training?

• Type: Likert (1-5) | Essential
• Manager support is the largest documented transfer predictor.

24. Have you been given opportunities to practice what you learned?

• Type: Likert (1-5) | Recommended
• Practice opportunity is a structural enabler of transfer.

25. Have you shared what you learned with anyone else on your team?

• Type: Binary (Yes/No) | Recommended
• Knowledge-sharing proxy.

26. Can you give a specific example of a situation where you applied something from the training?

• Type: Open-ended | Essential
• The single highest-value question on the Level 3 survey.

27. How much of the training content do you still remember clearly?

• Type: Likert (1-5) | Recommended
• Decay check.

28. What additional support would help you apply this training more fully?

• Type: Open-ended | Recommended
• Forward-looking, action-oriented.

Manager-side Level 3 questions (questions 29-33)

Ask the manager of each learner 60 to 90 days after training.

29. Have you noticed changes in [learner]'s behavior since the training?

• Type: Likert (1-5) | Essential
• Observer view of transfer.

30. Which specific behaviors have you seen applied from the training?

• Type: Open-ended | Essential
• Concrete observations.

31. How supportive has your team environment been for applying this training?

• Type: Likert (1-5) | Recommended
• Manager assessment of team enablers.

32. What would help [learner] apply more of the training content?

• Type: Open-ended | Recommended
• Action-oriented for the manager and L&D together.

33. How valuable was this training for your team overall?

• Type: Likert (1-5) | Recommended
• Manager-level program rating.

Level 4: Results and business impact (questions 34-42)

Paired with business metric data. Send 90 to 180 days after training.

34. What measurable results have you seen since applying this training?

• Type: Open-ended | Essential
• Qualitative linkage to business outcomes.

35. How has this training affected your team's performance metrics?

• Type: Likert (1-5) | Essential
• Self-reported impact at the manager level.

36. Has any business KPI moved that you attribute, at least in part, to this training?

• Type: Open-ended | Essential
• Specific metric and direction.

37. Would you invest in this training again for other team members?

• Type: Binary (Yes/No) | Essential
• Willingness-to-pay proxy.

38. How valuable was this training compared to the time invested?

• Type: Likert (1-5) | Recommended
• Cost-benefit self-assessment.

39. What other training would complement what was covered?

• Type: Open-ended | Recommended
• Feeds L&D curriculum planning.

40. If you had to rank this training against other training you have done, where would it fall?

• Type: Ranking | Nice-to-have
• Comparative rating.

41. Would you recommend this program continue, change, or be discontinued?

• Type: Multiple choice | Essential
• Program decision input.

42. Is there anything else you would like to share about the training experience?

• Type: Open-ended | Recommended
• Catch-all.

---

Beyond smile sheets: writing Level 1 questions that predict Level 3

Will Thalheimer's performance-focused smile sheets research found that traditional Level 1 questions ("How would you rate the instructor?") correlate weakly with on-the-job behavior change. The fix is to rewrite Level 1 questions so they probe transfer-relevant variables.

Performance-focused Level 1 questions:

• Ask about relevance to the learner's actual work, not general interest.
• Ask about self-efficacy ("How confident do you feel applying this?"), not enjoyment.
• Ask about anticipated barriers ("What will be hardest to apply?"), not perceived quality.
• Ask about intent to apply in a specific timeframe.
• Ask about specific learning moments, not overall satisfaction.

A well-written Level 1 survey can serve as an early signal for transfer before the delayed Level 3 survey is available. A poorly written one is just a rating of the room.

---

Measuring training transfer (Level 3)

Level 3 is where training programs prove their value, and also where most programs quietly fail. Transfer measurement requires patience (30 to 90 days) and effort (surveying both learners and managers), which is why so many programs skip it.

The 30-60-90 cadence for Level 3:

• 30 days: quick pulse on early application. Are learners starting to use what they learned? What barriers have shown up?
• 60 days: deeper transfer check. Are specific behaviors present on the job? Is the manager observing change?
• 90 days: final Level 3 reading. Is the behavior embedded? Has it produced early Level 4 signals?

Paired learner-manager surveys. Ask the learner and their manager the same core questions independently. Compare answers. Gaps between learner self-assessment and manager observation are some of the richest signals in training evaluation.

Focus on enablers and barriers. The Level 3 survey is not just a pass/fail check. Its biggest value is surfacing what is blocking transfer so you can adjust the program, the manager support system, or the on-the-job enablement.

---

ROI and the Phillips model

Jack Phillips extended the Kirkpatrick model with a fifth level, Return on Investment, and a methodology for isolating training-driven business impact from other factors.

The Phillips ROI process:

1. Collect Level 1-4 data.
2. Convert Level 4 results to monetary value.
3. Isolate the training's contribution using one of several methods (control groups, estimation by experts, learner estimation with confidence).
4. Calculate costs, including salaries, materials, facilities, and learner time.
5. Compute ROI = (Program benefits minus costs) / costs.

Phillips ROI is expensive to run well and only justified for high-stakes programs. Most L&D programs should aim for reliable Level 3 measurement first, and only advance to Level 4 and ROI for programs where the investment is large enough to require a financial case.

---

Best practices

Start with Level 3. If you are only going to measure one level, measure behavior change, not reaction. Level 3 tells you whether the training actually worked.

Pair learner and manager surveys. Two perspectives always beat one. Gaps between self-report and observation are signals.

Respect time. Keep every survey under 10 minutes. A 10-minute Level 3 survey that gets a 60% response rate beats a 30-minute survey at 20%.

Benchmark against yourself. Track the same program across cohorts. A well-measured program improves year over year because you can see what is working.

Close the loop. Share aggregate findings with learners, managers, and program owners within 30 days of each survey window. See our guide on closing the feedback loop.

---

Common mistakes

Stopping at Level 1. The most common mistake. Level 1 data alone cannot tell you whether training worked.

Running Level 1 immediately, then never following up. If you measure reaction and forget to measure transfer, you are measuring satisfaction, not effectiveness.

Asking generic questions. "How would you rate this training?" returns noise. Use specific, performance-focused items.

Ignoring manager support. Manager support is the largest transfer predictor. If your survey program does not measure manager behavior, it is missing the biggest variable.

Skipping anonymity on Level 3. Learners may underreport transfer barriers that involve their manager. Anonymity on Level 3 surveys produces more candid data.

Not acting on results. If Level 3 data surfaces a transfer barrier and nothing changes, the program owner has wasted the survey. Data without action is noise.

---

Free training survey template

Skip the blank page. Formbricks is an open-source experience management platform with free training survey templates you can deploy in minutes.

Why Formbricks for training surveys:

• Open source and self-hostable. Learner and manager feedback stays on your infrastructure.
• Anonymous by design. Anonymity is a first-class feature, which matters for honest Level 3 feedback about managers and on-the-job barriers.
• Flexible distribution. Deploy via email, Slack, link, or in-LMS widget.
• Automated cadence. Schedule the 30/60/90 transfer surveys once and they run automatically after each cohort.
• Free tier. Launch without a credit card.

How to get started:

1. Sign up at formbricks.com
2. Pick the training evaluation template or start from scratch
3. Map your questions to Kirkpatrick levels
4. Automate delivery at the end of session, 30 days, 60 days, and 90 days
5. Review cohort data and ship improvements

Start measuring training transfer with Formbricks →

For more survey design guidance, see our employee survey questions guide, our survey questions examples library, and the content quality evaluation template for a ready-to-use starting point.

---

Frequently asked questions