Brett LindenbergEvaluation questions are the questions an evaluation sets out to answer about a program, course, or service. Typical examples are "Did the program reach the people it was built for?", "Was it delivered as planned?", "What changed for participants?", and "Was the result worth the cost?" You answer each one with smaller survey or interview questions, such as "How many times have you used the feedback model since the course?" In the short video below, choose what a training team wants to learn after a supervisor course, and the video shows a matching evaluation question.
Evaluation questions and the questions you ask people
In program evaluation, an evaluation question (often called a key evaluation question) names what the evaluation will judge. The Evaluation Questions Checklist for Program Evaluation by Lori Wingate and Daniela Schroeter (Western Michigan University) says these questions ask about a program's merit, worth, or significance. Unlike survey questions, they are not meant to produce a single data point. The CDC Program Evaluation Framework, 2024 adds that they are broad, open-ended, and limited to what the evaluation's data can answer.
The questions you put on a survey, in an interview, or in a video are data-collection questions. Each one gathers one piece of evidence, and you need several of them, often from more than one source, to answer one evaluation question.
| Source | Question you ask |
|---|---|
| Supervisors who took the course | In the past 30 days, how many one-on-ones did you hold with each direct report? What made them hard to schedule? |
| Their direct reports | How often has your supervisor met with you one-on-one this month? What did you discuss? |
| Calendar or HR records | Do meeting records show a recurring one-on-one for each direct report? |
Evaluation questions examples for programs, by type
The CDC's 2024 framework names five types of evaluation. The sample questions below follow those types and use a supervisor training program as the example. Replace the nouns with your own program or service.
Formative (needs and design) evaluation questions
Formative evaluation checks whether a program is feasible, appropriate, and acceptable before it is fully rolled out. A training needs assessment answers many of these questions.
- What problem does the program address, and what evidence shows the problem is real for this group?
- Which skill or knowledge gaps matter most to the people the program serves?
- Is the planned format realistic for participants' schedules, devices, and locations?
- What did pilot participants find confusing or hard to use?
- Can we run the program with the budget and staff we have?
Process (implementation) evaluation questions
Process evaluation looks at how the program is delivered compared with the plan.
- Was the program delivered as designed? Where did it change, and why?
- How many of the intended participants enrolled, and how many finished? Who dropped out, and why?
- Did facilitators follow the curriculum, and where did they adapt it?
- Which parts of the program did participants use least?
- What got in the way of delivery?
Outcome evaluation questions
Outcome evaluation measures how far the program achieved its intended outcomes.
- To what extent did participants gain the intended knowledge and skills?
- How many participants use the new practice at work after the program?
- Which outcomes did some groups reach and others not?
- What unintended outcomes, good or bad, appeared?
Impact evaluation questions
Impact evaluation compares outcomes with and without the program to find out whether the program caused the change.
- How do results for participants compare with a similar group that did not take part?
- How much of the change comes from the program, and how much from other changes at the same time?
- Did the effect last after the program ended?
- What worked, for whom, and under what conditions?
Economic (cost) evaluation questions
Economic evaluation weighs the program's effects against its costs. For training, see our guide to measuring the ROI on training, or run your own figures in the training ROI calculator.
- What did the program cost per participant, including staff time?
- Do the benefits justify the cost?
- Could a shorter or cheaper format reach the same outcome?
- How does the cost per outcome compare with the alternative we considered?
Simple evaluation questions for a course, workshop, or event
After a single session, keep the form short: a few rating questions and one or two open questions. Use the same scale for every rating item so you can compare sessions. The table gives ten questions you can copy. Pick the ones whose answers you will act on.
| Question | Answer format |
|---|---|
| Overall, how useful was this session for your work? | 1 to 5 rating |
| How relevant was the content to your current role? | 1 to 5 rating |
| How clear were the explanations and examples? | 1 to 5 rating |
| How was the pace? | Too slow, about right, or too fast |
| How confident are you that you can now [do the skill]? | 1 to 5 rating |
| What is one thing you will do differently because of this session? | Open text |
| What was least useful, and what should we change? | Open text |
| Did you expect to learn anything that we did not cover? | Open text |
| How likely are you to recommend this session to a colleague? | 0 to 10 scale |
| Did you have what you needed to take part, such as working audio and the course materials? | Yes or no, with a comment box |
Keep the "one thing you will do differently" question even on the shortest form. It asks for a commitment, and it gives you a specific behavior to check on later.
Training evaluation questions examples by Kirkpatrick level
Training teams often sort evaluation questions by the four levels that Donald Kirkpatrick introduced in the 1950s. The definitions below come from Kirkpatrick Partners. The people you ask, and when you ask them, change from level to level.
Level 1: reaction (ask learners at the end of training)
Reaction is how favorable, engaging, and relevant learners find the training. Kirkpatrick Partners calls relevance the reaction measure that best signals whether learners will use the training on the job.
- How relevant was this training to the work you do each week?
- Which part of the course will you use first?
- What, if anything, made the course hard to complete?
Level 2: learning (ask learners during and right after training)
Learning covers knowledge and skills, and also attitude, confidence, and commitment to use what was taught.
- A scenario check, such as: "A direct report misses a deadline for the second time. What is your first step?" with three or four answer choices.
- The same confidence rating before and after: "How confident are you giving corrective feedback?"
- "How committed are you to using the feedback model in your next one-on-one?"
Level 3: behavior (ask learners and their managers after learners have had a chance to apply it)
Behavior is whether people perform the critical behaviors on the job, with the support and accountability to keep doing them. Records such as call logs or audit results add evidence that does not depend on memory.
- "In the past 30 days, how many times did you use the feedback model with a direct report?"
- To managers: "Which behaviors from the course have you seen this person use? Give one example."
- "What has helped or blocked you from using what you learned?"
Level 4: results (ask the program sponsor and use existing business data)
Results is whether the targeted organizational outcomes occur.
- To the sponsor, before the course starts: "Which result would you accept as evidence that the course worked?"
- "Did that measure change after the course, for example turnover on teams led by new supervisors?"
- "What else changed at the same time that could explain the result?"
For methods beyond the four levels, such as Phillips ROI and the Success Case Method, see our guide to training evaluation methods. For a step-by-step plan, read how to measure training effectiveness.
Questions that help you improve and questions that measure and compare
Evaluators separate formative evaluation, which runs while a program can still change, from summative evaluation, which judges a program after it has run. The same split appears in course quizzes that ask which evaluation questions help you improve a product or service and which help you measure and compare.
| Help you improve | Help you measure and compare |
|---|---|
| Which goals are we meeting so far? | What were the results? |
| What works, and what does not? | What are the costs and benefits? |
| What are participants' most common reactions? | Were there unintended results? |
| What should we change before the next group starts? | How do the results compare with last year, or with a group that did not take part? |
If a quiz asks you to pick the three questions that help you improve, choose the ones about which goals you are meeting, what works and what does not, and participants' most common reactions. Improvement questions point to a change you can still make. Measure-and-compare questions support a judgment about whether to continue, expand, or stop.
Sample questions from six evaluation approaches
The evaluation type decides what you ask about. The approach you pick decides how you ask and who takes part. Each of these six approaches comes with its own set of questions.
CIPP model (context, input, process, product)
Daniel Stufflebeam created the CIPP model to help school districts and federal agencies assess War on Poverty projects in the 1960s and 1970s. The Western Michigan University Evaluation Center publishes Daniel Stufflebeam's CIPP Evaluation Model Checklist. While a program runs, CIPP asks four questions: what needs to be done, how it should be done, whether it is being done, and whether it is succeeding. After the program, the checklist turns them around:
- Did the program aim at clear goals based on the needs of the people it serves?
- Did a sound plan address those needs?
- Was the plan carried out well?
- Did the program reach the right people, meet their needs, and produce gains that lasted?
Logic model
A logic model maps a program's inputs, activities, outputs, outcomes, and impact. The W.K. Kellogg Foundation's 2004 Logic Model Development Guide explains how to draw one. Ask one question for each link in the chain:
- Inputs: did we have the staff, budget, and tools the plan assumed?
- Activities: did the sessions happen as planned, for the intended audience?
- Outputs: how many people completed the program?
- Outcomes: what changed in participants' knowledge or practice?
- Impact: did those changes move the long-term goal?
Most Significant Change
Rick Davies and Jess Dart's 2005 guide to the Most Significant Change technique builds an evaluation from participants' stories. You ask one open question about a set period and a set area of change. Groups of stakeholders then pick the most significant stories and record why they chose them. Sample questions:
- Thinking about the last three months, what was the most significant change in how you lead your team?
- Why does that change matter to you?
- What difference has it made so far, and what difference will it make later?
- For the selection panel: why did your group choose this story over the others?
Outcome Harvesting
Ricardo Wilson-Grau and colleagues developed Outcome Harvesting. Rather than measure progress toward set objectives, the evaluator collects evidence of what changed and works backward to whether and how the program contributed. An outcome is a change in what a person, group, or organization does, and it can be intended or unintended, positive or negative.
- What has this team done differently since the program started? When and where did it happen?
- How did the program contribute to that change, if at all?
- Who else could confirm that the change happened?
- To an independent source: is this description of the change accurate, and how much did the program contribute?
Developmental evaluation
Michael Quinn Patton's Developmental Evaluation (2010) is for new programs in complex conditions that keep changing. The evaluator works with the program's decision makers and feeds findings into their choices while the program takes shape.
- What are we seeing in the pilot group that we did not expect?
- How has our understanding of the problem changed since we started?
- Based on this round, what one change should we try next?
- Which early signs would tell us to stop, or to scale up?
Utilization-focused evaluation
Patton's Utilization-Focused Evaluation (fourth edition, 2008) judges an evaluation by its intended use by its intended users. You start with the people who will act on the findings.
- Who will use this evaluation, and which decision will they make with it?
- What evidence would lead them to expand, change, or stop the program?
- When do they need the answer?
- What will they do if the answer is yes, and what if it is no?
If you hire or train an evaluator, the American Evaluation Association publishes evaluator competencies and its Guiding Principles for Evaluators, last updated in 2025.
How to write good evaluation questions
Wingate and Schroeter's checklist names five qualities of a good evaluation question: evaluative, pertinent, reasonable, specific, and answerable. The table shows a weak question for each quality and a stronger version.
| Quality | Weak question | Stronger question |
|---|---|---|
| Evaluative: asks for a judgment, not a single fact | How many people attended? | Did the course reach the supervisors who most needed it? If not, why not? |
| Pertinent: tied to the program's purpose and users' needs | Did learners like the catering? | Which parts of the course did learners find least relevant to their work, and why? |
| Reasonable: within what the program can influence | Did the course reduce turnover across the company? | Did turnover change on teams led by course graduates, compared with similar teams? |
| Specific: names the part of the program, the measure, and who is affected | Was the program effective? | To what extent do graduates hold weekly one-on-ones and use the course feedback model? |
| Answerable: possible with the data, time, and staff you have | Did the course change how employees feel about the company? | How did engagement survey scores for graduates' teams change between the last two survey rounds? |
The checklist also asks that the full set of questions cover the evaluation's purpose, with a reason for each topic you leave out. The CDC framework adds that you should limit the set to the questions you can answer in time with the resources you have.
For the participant questions that collect the evidence:
- Ask about one thing per question. "Was the session clear and useful?" is two questions.
- Ask about specific, recent behavior ("in the past 30 days") rather than general habits.
- Use balanced scales with the same number of positive and negative options.
- Keep the wording neutral. "How helpful was the excellent workbook?" leads the answer.
- End with one open question, so people can tell you what your other questions missed.
How to ask evaluation questions inside a training video
An evaluation form emailed after training reaches learners once they are back at work. When the training is a video, you can ask the questions inside it, at the moment each one applies. In Mindstamp, a question pauses the video, and playback continues after the learner answers. The question types that suit evaluation are:
- Multiple choice for scored knowledge checks and scenarios (Level 2).
- Rating (1 to 5 stars) for reaction and confidence.
- Likert scale with five or seven points and your own labels, so an existing survey keeps its wording.
- Free response for open feedback and commitments.
- Voice response for short spoken stories, which suits Most Significant Change prompts.
Put each knowledge check right after the concept it tests. Our guide to creating an interactive video quiz shows the build step by step. Ask the same confidence rating near the start and again near the end, so you can compare before and after. Do not leave every evaluation question for the final frame. Across 8.07 million views of 18,315 Mindstamp videos (2018 to October 2026), the average view played 46.5% of the video. Those are mostly training, marketing, and sales videos made by our customers, so treat the figure as a pattern and not a benchmark for all video. A question placed only at the end will miss many of those viewers. On supported plans, the "Prevent passing required interactions" setting stops viewers from skipping past a question they must answer.
Mindstamp records each answer against the learner who gave it and lists it under Questions in reporting. To analyze answers in a spreadsheet, filter the Views list to one video, group, or playlist and export a CSV with a column for each question. If your LMS tracks completion, you can export the video as a SCORM 1.2 or 2004 package. To build a stand-alone feedback form in a video, see how to turn any video into a survey.
In-video questions work best for Levels 1 and 2, because the learner is present to answer. For behavior and results you need evidence from later and from other people. Ask managers and check records for Level 3, and pull Level 4 numbers from your own business systems.
Common questions
What is an example of an evaluation question?
"To what extent did the program achieve its intended outcomes for the people it served?" is a typical evaluation question. A participant question that helps answer it is "What have you done differently at work since the program?"
What are typical evaluation questions?
Typical evaluation questions ask whether a program is needed and feasible, whether it was delivered as planned, whether it achieved its outcomes, whether it caused the change, and whether the benefits justified the cost. Those five areas match the formative, process, outcome, impact, and economic evaluation types in the CDC's 2024 framework.
How many evaluation questions should an evaluation have?
There is no fixed number. Keep only the questions your evaluation can answer in time with your data, staff, and budget, and make sure the set covers what the people using the evaluation need to decide.
To start, write down the one decision your evaluation must inform. Turn it into one evaluation question, list who or what can answer it, and draft the participant questions for each source. Test the form with a few people before you send it. To try the questions inside a video, start from the evaluation question selector example.
