New webinar on 3rd March 2026 | 9:00 am PT
In this webinar, discover how Sopact Sense revolutionizes data collection and analysis.
Closed-ended questions: 6 types, 50+ stakeholder survey examples, pros and cons, and the Answer Architecture framework for better decisions.
A nonprofit runs a post-program survey. The question: "What was the biggest barrier to completing the program?" The answer options are (a) cost, (b) time, (c) distance, or (d) other. Sixty-eight percent pick "time." The team builds next year's program around reducing time commitment. Six months later, qualitative interviews reveal the real top barrier — childcare. That word wasn't on the answer list, so respondents whose actual barrier was childcare picked the closest option instead of writing "other" with no detail. The survey data wasn't technically wrong. The answer list was.
This is The Missing Option Problem — when closed-ended answer lists don't include what respondents actually want to say, respondents pick the closest wrong option rather than skipping the question or selecting "other" with no explanation. The data looks clean. Everyone picked something. It's silently inaccurate because the real answers got buried in the wrong buckets. The fix isn't to abandon closed-ended questions. It's to design answer lists that actually cover respondent reality — and pair them with at least one open-ended follow-up that catches what the list misses.
Last updated: April 2026
This is the closed-ended educational anchor in a five-page cluster on survey question design. For the open-ended counterpart, see open-ended questions. For the direct comparison of the two formats, see open-ended vs closed-ended questions. For survey-specific open-ended templates, see open-ended survey questions. For analyzing qualitative follow-ups at scale, see how to analyze open-ended survey responses.
Closed-ended questions are survey, form, or interview items where respondents choose from a preset list of answer options — they cannot write in their own words. Common formats include yes/no, multiple choice, rating scales, Likert scales, and ranking questions. The defining feature is the fixed answer list: every response falls into one of the predefined options, which makes the data immediately countable and comparable across groups or time.
The preset answer list is both closed-ended's greatest strength and its signature weakness. When the list fully covers respondent reality, the data is fast, clean, and statistical. When the list misses what respondents actually want to say — The Missing Option Problem — respondents pick the closest wrong option, and the data becomes silently inaccurate without anyone noticing.
Closed-ended questions mean questions that can only be answered by selecting from a predetermined list of options — the respondent cannot write their own answer. This is the opposite of open-ended questions, where the respondent answers in their own words. Closed-ended produces countable data (numbers, percentages, averages). Open-ended produces narrative data (reasoning, stories, specific detail).
The word "closed" refers to the answer space being closed off — limited to the options the designer chose in advance. Every closed-ended question commits to an answer list before the first response arrives. That commitment is what makes the data analyzable immediately, and also what creates the core design challenge: anticipating what respondents will actually want to say.
An example of a closed-ended question is "On a scale of 1 to 5, how satisfied were you with the program?" The respondent must pick a number between 1 and 5 — they cannot write anything. Every response falls into one of exactly five predefined buckets, which makes the data immediately countable, averageable, and comparable across groups.
Other common examples of closed-ended questions include:
Each one specifies the answer list up front. The respondent's job is to pick the option that best matches their situation — and hope that one of the options actually does.
Closed-ended questions show up in far more places than surveys. Below is a library of 30+ closed-ended question examples grouped by where they appear in practice. Adapt the wording for your situation.
In surveys (5 examples)
In intake forms and applications (5 examples)
In polls and voting (5 examples)
In quizzes and assessments (5 examples)
In customer feedback and market research (5 examples)
In screening and eligibility (5 examples)
Notice the pattern across all six contexts: every closed-ended question requires a preset answer list that covers respondent reality. When the list is good, the data is fast and comparable. When the list is incomplete — The Missing Option Problem — the data is silently inaccurate.
There are six main types of closed-ended questions, each producing a different kind of data and supporting different analytical operations. Pick the type by what you want to learn, not by what's fastest to build.
1. Yes/No (dichotomous) questions offer exactly two options — yes or no, true or false, completed or didn't. They produce the cleanest possible data and the least nuance. Use them for factual verification ("Did you attend?") or gating logic ("Are you currently employed?"). Don't use yes/no for anything where the honest answer is "it depends."
2. Multiple-choice (single-select) questions present three or more options with one answer chosen. They produce categorical data — counts per category, cross-tabs across groups. The main design risk is The Missing Option Problem: if your list doesn't cover respondent reality, people pick the closest wrong option. Always include "other" with a text box when the answer space is open-ended.
3. Multiple-select (checkbox) questions let respondents pick more than one answer from a list. They reveal co-occurring behaviors or preferences — "Which features do you use regularly?" gets you not just one answer but the combination pattern across respondents. Use sparingly; too many checkbox questions tire respondents out.
4. Rating scale questions ask respondents to pick a number on a scale, typically 1–5, 1–7, or 0–10. They produce ordinal data that can be averaged and tracked over time. The classic version is NPS ("How likely are you to recommend us, 0–10"). Scale design matters: a 5-point scale compresses signal; a 10-point scale sometimes overwhelms.
5. Likert scale questions ask respondents to agree or disagree with a statement on a symmetrical scale — strongly disagree, disagree, neutral, agree, strongly agree. They measure attitudes and beliefs. Named after psychologist Rensis Likert, who developed the format in 1932. Five and seven points are the standard lengths. (Yes, a Likert scale is a closed-ended question — a specific ordinal subtype.)
6. Ranking questions ask respondents to put items in order — most important to least important, favorite to least favorite. They reveal preferences you can't get from independent ratings. The trade-off: ranking is harder cognitively. Limit to five or fewer items or completion rates drop.
Closed-ended questions in quantitative research are the workhorse of the survey instrument — the part that produces data you can run formal statistics on. Survey studies, clinical trials, market research, program evaluation, and government census work all rely on closed-ended formats for the same reason: structured data enables inferential testing, group comparison, and longitudinal tracking.
In quantitative research contexts, closed-ended questions serve five specific jobs:
Closed-ended questions in quantitative research require more design rigor than casual survey questions because downstream analysis depends on measurement quality. A 5-point Likert scale with a poorly anchored middle option ("neither agree nor disagree") invites satisficing — respondents park there to skip the effort of deciding. For serious quantitative research, a 7-point scale with clear anchors at each position produces better data, and for some study designs an even-numbered scale forces a directional choice. Related reading: baseline survey and survey metrics and KPIs.
Closed-ended questions have six main advantages that explain why they're the default format in most surveys, forms, and research instruments:
1. They're fast to answer. Respondents pick from a list instead of composing written responses. A well-designed 20-item closed-ended survey takes three to five minutes. The same 20 questions as open-ended would take 30+ minutes — and completion rates collapse.
2. They're fast to analyze. Structured data counts itself. The moment responses arrive, counts, percentages, averages, and cross-tabs are available. No coding step, no human review for each response, no weeks of qualitative analysis.
3. They allow researchers to collect and analyze large volumes of data efficiently. Closed-ended items scale to thousands of respondents without added analysis effort. This is why closed-ended dominates large market research studies, national surveys, and longitudinal panels — the structure makes volume tractable.
4. They produce directly comparable data across groups and time. When every respondent picks from the same five options, group differences and pre-post shifts are immediately visible. Open-ended answers require coding before comparison is possible; closed-ended answers are comparable by construction.
5. They're easy to chart and report. Percentages, bar charts, distribution graphs, and heatmaps all come naturally from closed-ended data. The visual language of survey reporting is built on closed-ended output.
6. They have higher completion rates. Respondents who would abandon a long open-ended question will pick from a list. For surveys sent to large or distracted populations, closed-ended questions often double or triple completion rates versus open-ended equivalents.
Closed-ended questions also have real disadvantages that justify pairing them with at least one open-ended follow-up per survey:
1. They limit respondents from elaborating on their thoughts. This is the most common disadvantage of closed-ended questions. A respondent who has a specific reason, a complicated situation, or a nuanced opinion has no way to express it — they pick the closest option and the nuance disappears.
2. They can't capture reasoning. A 4-out-of-5 rating tells you the level of satisfaction, not what drove it. Without an open-ended follow-up, the "why" behind every number is invisible.
3. They risk The Missing Option Problem. When answer lists don't include what respondents actually want to say, respondents pick the closest wrong option. The data looks clean but is silently inaccurate. This is the specific failure mode closed-ended questions are uniquely prone to.
4. Poorly designed scales compress signal. A 5-point scale where 4 and 5 mean practically the same thing collapses real differences into the top two categories. Scale design choices shape the data as much as the question stem does.
5. They produce no quotes for reports. Funders, boards, and stakeholders respond to real participant voices. Closed-ended questions produce numbers, not quotes. Reports built only on closed-ended data read like spreadsheets — accurate but unmemorable.
6. Leading options bias results. Options written with subtle positive or negative framing — "Excellent / Great / Good / OK / Not bad" — inflate scores toward the intended direction. Every option level must be symmetrical and neutrally worded; asymmetry creates systematic bias.
7. Changing scale formats breaks comparability. Swapping a 1–5 scale at baseline for a 1–10 scale at follow-up makes the two waves non-comparable. Once a scale is chosen for a study, it's locked until the study ends.
Writing a closed-ended question that actually works comes down to three moves. Every strong closed-ended question across every context uses them.
Move 1 — Design the answer list before you write the question. Most closed-ended design goes in the wrong order. The question gets written, then options get brainstormed last-minute. Better: start with what respondent answers might look like, build the answer list that covers that range, then write the question stem to match. This single reordering prevents most of The Missing Option Problem before it starts.
Move 2 — Test the list against reality, not a conference room. The answer options that feel comprehensive in a design meeting often miss the top signal in respondents' actual lives. Pilot-test your closed-ended questions with 10–15 real respondents before fielding at scale. Look for "other" picks over 10% — that's the signal your list is incomplete.
Move 3 — Add "other" with a text field, and read what comes in. Closed-ended questions should always offer an "other" escape valve when the answer space is open-ended (like barriers, reasons, or categories the designer can't fully anticipate). Then actually read the "other" responses — that's where the Missing Option Problem shows itself. Promote frequent "other" write-ins to named options in the next wave.
For a deeper comparison of when to use closed-ended versus open-ended formats, and the 80:20 mix that strong surveys settle into, see open-ended vs closed-ended questions. For writing open-ended questions that produce usable answers, see open-ended survey questions. For making sense of open-ended responses at scale, see how to analyze open-ended survey responses.
Closed-ended questions are survey, form, or interview items that limit the respondent's answer to a preset list of options. Common formats include yes/no, multiple choice, rating scales, Likert scales, and ranking questions. They produce structured data — counts, percentages, averages — that can be analyzed immediately. The preset answer list is both their strength and their weakness.
An example of a closed-ended question is "On a scale of 1 to 5, how satisfied were you with the program?" The respondent must pick a number between 1 and 5 — they can't write their own answer. Other examples include "Did you complete the program? Yes or No" and "Which barrier affected you most: cost, time, distance, or other?"
The six main types of closed-ended questions are yes/no (dichotomous), multiple choice with one answer, multiple select with multiple answers, rating scale (numeric like 1–5 or 0–10), Likert scale (agreement from strongly disagree to strongly agree), and ranking. Each type produces a different data structure and supports different analytical operations.
Closed-ended questions mean questions that can only be answered by selecting from a predetermined list of options. The respondent cannot write their own answer. This is the opposite of open-ended questions, where the respondent answers in their own words. Closed-ended produces countable data; open-ended produces words and reasoning.
One key advantage of closed-ended questions in market research is they allow researchers to collect and analyze large volumes of data efficiently. Structured responses count themselves — no coding step, no per-response human review. This is why closed-ended dominates large market research studies, national panels, and tracking surveys where hundreds of thousands of responses need quick analysis.
A common disadvantage of closed-ended questions is that they limit respondents from elaborating on their thoughts. The preset options constrain what can be said — respondents with specific reasons or complicated situations have no way to express them. The fix is to pair closed-ended questions with at least one open-ended follow-up per survey section.
Closed-ended questions in quantitative research are the structured items in a survey or assessment that produce data you can run statistics on. They handle demographics, outcome measurement with standardized scales, pre-post comparison, group comparison, and eligibility screening. Quantitative research requires careful scale design because downstream analysis depends directly on measurement quality.
A Likert scale question is a closed-ended question that asks the respondent to agree or disagree with a statement on a symmetrical scale — typically strongly disagree, disagree, neutral, agree, strongly agree. Likert scales measure attitudes and beliefs. Named after psychologist Rensis Likert, who developed the format in 1932. Five and seven points are the standard scale lengths.
Closed-ended questions are fast to answer, fast to analyze, directly comparable across groups and time, easy to chart, and friendly to high-volume surveys. They produce countable data that generates averages, percentages, and cross-tabs automatically. They also enable large-sample analysis that would be impossible with open-ended data. Completion rates run higher than open-ended equivalents.
The main disadvantages of closed-ended questions are that they limit respondents from elaborating, can't capture reasoning behind answers, risk The Missing Option Problem when lists don't cover respondent reality, and produce no quotes for reports. Poorly designed scales compress signal; leading options inflate bias. The fix is to pair closed-ended with one open-ended follow-up per section.
Write a good closed-ended question with three moves: design the answer list before writing the question stem, pilot-test the list with real respondents to check it covers their reality, and always include "Other" with a text field. Then actually read the "Other" responses — that's where incomplete lists show themselves. Promote frequent write-ins to named options next wave.
The Missing Option Problem is when closed-ended answer lists don't include what respondents actually want to say, so respondents pick the closest wrong option rather than skipping or selecting "other" with no detail. The data looks clean — everyone picked something — but it's silently inaccurate because real answers got buried in wrong buckets.
