Every survey designer eventually faces the same choice: do you ask a simple yes/no question and get a clean answer, or do you ask a scaled question and capture the nuance?
The dichotomous question, any question with exactly two mutually exclusive options, is the workhorse of survey design. It's fast, easy to analyze, and incredibly reliable when used correctly. It's also one of the most overused formats, applied to questions that don't have a binary answer in real life.
This guide covers what dichotomous questions are, when they genuinely beat the alternatives, 25 real examples across customer research, employee, health, education and political surveys, the pros and cons, and exactly when to reach for a Likert scale or multiple choice instead.
What are dichotomous questions?
A dichotomous question is a closed-ended survey question that offers exactly two answer options. The two options must be mutually exclusive (you can't pick both) and collectively exhaustive (every respondent fits one of them).
The word comes from the Greek dichotomia, literally "cutting in two." Every respondent gets cut into one of two buckets.
The most common forms are:
- Yes / No: "Have you used our product before?"
- True / False: used heavily in quizzes and knowledge tests
- Agree / Disagree: sometimes used in attitude research
- Fair / Unfair: common in policy and ethics research
- Present / Absent: medical or observational data
- This / That: "Coffee or tea?", "iOS or Android?"
✔ The canonical example
"Have you purchased from us in the last 30 days? (Yes / No)"
Clean, binary, and reflects a real-world fact. Either the purchase happened or it didn't. There's no middle ground to lose.
Dichotomous questions sit in a specific spot in the survey design toolkit: simpler than multiple choice, simpler than Likert, more decisive than open-ended. The trick is knowing when "simpler" is the right call.
When to use a dichotomous question
There are four situations where a dichotomous question is clearly the right tool:
- The underlying reality is binary. "Are you currently employed?", "Do you own a car?", "Are you over 18?" These all have a real-world yes-or-no answer. No scale will improve them.
- You're screening or qualifying respondents. A first question like "Have you used our product in the last 90 days?" filters who continues. The branching depends on a binary answer.
- You're branching the survey. "Did you contact support?" routes one group to follow-up questions and skips them for everyone else. Conditional logic in form builders runs on binary signals.
- You want the highest response rate on a sensitive question. Binary options are the lowest-friction format, so people are likeliest to answer them honestly, especially when the topic is uncomfortable.
And three situations where you should not use one:
- The attribute is continuous. Satisfaction, agreement, frequency, intent, trust: these all have natural gradations. A 5- or 7-point Likert scale captures the gradient; a yes/no flattens it.
- The respondent has mixed feelings. If the honest answer is "kind of," forcing yes or no produces noise, not data.
- You actually want gradation in your reporting. If your dashboards or KPIs require averages, distributions, or trend comparisons, you need a scale that produces them. Binary data limits you to "% yes."
25 examples of dichotomous questions
Here are 25 dichotomous questions pulled from real customer research, employee, healthcare, market research, education, and political surveys. Each is paired with the kind of insight it produces, and a note on when a different format would be better.
Customer research surveys
Example 1
"Have you purchased from us in the last 30 days? (Yes / No)"
Screener that cleanly splits buyers from non-buyers so the next questions can branch correctly.
Example 2
"Did you find what you were looking for today? (Yes / No)"
Common post-visit survey. Binary fits because the goal is task completion, not satisfaction.
Example 3
"Would you recommend our product to a friend? (Yes / No)"
A binary recommendation question. If you need the full NPS gradient, use the 0–10 scale instead.
Example 4
"Have you contacted customer support in the last 6 months? (Yes / No)"
Screener for a support-experience follow-up. Branch logic depends on a clean binary.
Example 5
"Are you currently a paying customer? (Yes / No)"
Segmenter. Lets you analyze free vs paid respondents separately without ambiguity.
Employee & HR surveys
Example 6
"Have you spoken with your manager about your career goals in the last quarter? (Yes / No)"
Behavioral fact, not opinion. Binary captures whether the conversation actually happened.
Example 7
"Did you take any vacation days in the last 6 months? (Yes / No)"
Usage signal. Reveals burnout risk without measuring sentiment.
Example 8
"Have you completed the new hire onboarding modules? (Yes / No)"
Compliance check. The follow-up should be "why not" for anyone who answered no.
Example 9
"Would you apply for a job here again, knowing what you know now? (Yes / No)"
A powerful exit-interview question. The binary forces respondents off the fence.
Building a survey that mixes question types?
Youform has yes/no fields, Likert scales, ranking questions, matrix grids, and conditional branching, all on the free plan with unlimited submissions.
Try Youform freeHealthcare & clinical surveys
Example 10
"Do you currently smoke tobacco products? (Yes / No)"
Genuine binary clinical fact. Used as a risk-factor screener in nearly every intake form.
Example 11
"Have you been diagnosed with diabetes? (Yes / No)"
Medical history. Binary because the diagnosis either exists in the record or it doesn't.
Example 12
"Are you currently taking any prescription medications? (Yes / No)"
Screener that gates a follow-up list. Yes routes to "which medications?"; no skips it.
Example 13
"Did you experience any side effects from the medication? (Yes / No)"
Cleanly captures incidence. Severity would be a separate scaled question.
Market research surveys
Example 14
"Have you heard of our brand before today? (Yes / No)"
Aided awareness measurement. A standard binary in tracking studies.
Example 15
"Would you consider switching from your current provider? (Yes / No)"
Intent screener. The follow-up "why" or "why not" is where the real signal lives.
Example 16
"Do you own a smartphone? (Yes / No)"
Demographic / device segmenter. Branches the survey toward mobile vs desktop questions.
Example 17
"Have you used a subscription meal kit in the last 12 months? (Yes / No)"
Category usage. Used to size addressable markets in concept-testing surveys.
Education & course feedback
Example 18
"Did you complete all the assigned readings for this module? (Yes / No)"
Behavioral check. Lets the instructor compare completion against quiz performance.
Example 19
"Was the homework submitted on time? (Yes / No)"
Operational data, not opinion. Binary because the timestamp either beats the deadline or doesn't.
Example 20
"True or False: The mitochondria is the powerhouse of the cell."
Classic knowledge-test format. Quick to grade, but offers no insight into why a student got it wrong.
Example 21
"Would you recommend this course to a colleague? (Yes / No)"
Binary recommendation. If you need a benchmark you can trend, use NPS instead.
Political & public opinion surveys
Example 22
"Are you registered to vote? (Yes / No)"
Genuine binary administrative fact. Always a screener in election-cycle polling.
Example 23
"Did you vote in the last general election? (Yes / No)"
Self-reported turnout. Useful for weighting samples toward likely voters.
Example 24
"Do you currently identify with any political party? (Yes / No)"
Identity screener that gates "which one?" The binary protects against forcing an answer on independents.
Example 25
"Have you watched any presidential debates this cycle? (Yes / No)"
Media-exposure check. Branches into "which ones" and "how did they affect your views."
Pros and cons of dichotomous questions
Used in the right place, dichotomous questions are a survey designer's best friend. Used in the wrong place, they're the format that quietly destroys the most data.
✔ Strengths
- Fast to answer. Two options, one tap.
- Easy to analyze. Every result reduces to "% yes."
- High response rate. Lowest friction format possible.
- Clean branching. Perfect for conditional logic and skip patterns.
- Reliable for binary facts. No ambiguity about what the answer means.
✖ Weaknesses
- Loses nuance. Mixed feelings collapse into a misleading yes or no.
- Forces a side. No "maybe" option pushes people off the fence.
- Misses gradation. Can't tell strong yes from weak yes.
- Can introduce bias. If neither option fits, respondents pick the less wrong one.
- Hard to trend. Limited statistical depth compared to scaled data.
How to analyze and score dichotomous question data
Binary data is the easiest survey output to work with, but the analysis still needs to be done deliberately. Three common patterns:
1. Calculate the percentage
The default metric for any dichotomous question is the share of respondents who picked each option. Tally the "yes" responses, divide by the total number of valid responses, and report a percentage. If 642 out of 900 respondents said yes, that's a 71.3% yes rate.
2. Cross-tabulate against segments
The single percentage is rarely the interesting number; the difference between segments is. Break the "% yes" out by audience cut (new vs returning customers, free vs paid, region, plan tier). A 10-point gap between any two segments is usually where the actual signal lives.
3. Pair binary with a follow-up
A dichotomous question on its own tells you what. The follow-up tells you why. Branch the survey so anyone who answered yes (or no) gets a single open-ended question after. That's where the qualitative depth comes back into binary data.
For trend reporting, plot "% yes" over time the same way you'd plot any other rate metric. Most survey tools (Youform included) export the raw responses as CSV so you can wire them straight into your dashboard of choice.
Dichotomous vs other question types
Picking the right question type comes down to matching the format to the shape of the answer. Here's how dichotomous compares to its closest cousins.
Dichotomous vs multiple choice
Multiple choice has three or more options. Reach for it when the underlying choice isn't binary. "Which device do you use most often?" (iOS / Android / Web) is a three-option fact, not a yes/no opinion.
Dichotomous vs Likert scale
Likert scales (typically 5 or 7 points: strongly disagree → strongly agree) capture intensity. If you want to know how much the respondent agrees, switch to Likert. Yes/no flattens "kind of agree" into the wrong bucket.
Dichotomous vs rating scale
Rating scales (1–5, 1–10, NPS) measure magnitude. Use them when the metric needs to be benchmarked or trended over time. Binary data can't be averaged into a trend with the same precision.
Dichotomous vs open-ended
Open-ended questions capture qualitative depth that binary can't reach. The smart pattern is to combine them: a dichotomous question to filter or measure, followed by an open-ended "why" for the people whose answer interests you most.
Best practices for writing dichotomous questions
A dichotomous question is only as good as its wording. Five rules that separate clean binary questions from broken ones:
- Make the two options mutually exclusive and exhaustive. Every respondent must fit into exactly one. If someone could be "kind of both" or "neither," the format is wrong for this question.
- Stay neutral. Don't bias the wording toward one option. "Did our excellent support resolve your issue? (Yes / No)" is leading; "Did support resolve your issue? (Yes / No)" isn't.
- Ask about one thing at a time. If you find yourself writing "and" or "or" in the stem, you've drifted into a double-barreled question, the worst format mistake in survey design.
- Consider a "not sure" or "prefer not to say" option where it's honest. Forcing a binary answer when respondents genuinely don't know turns guesses into "data." It's not always wrong to add a third option, just know that the question stops being technically dichotomous.
- Pair it with a follow-up. Binary alone tells you what; an open-ended "why" or a Likert follow-up tells you how strongly. Combining the two is almost always better than either alone.
A pre-send checklist before your next survey
- ✔ Each dichotomous question describes a genuinely binary reality, not a continuous one.
- ✔ The two options are mutually exclusive and cover every respondent.
- ✔ The wording is neutral. No adjectives nudge respondents toward one side.
- ✔ Sensitive questions get a "prefer not to say" if honesty depends on it.
- ✔ Every "yes" with downstream value has a follow-up that captures depth.
If you can check all five, your binary questions are doing exactly what they're best at, producing decisive, clean, and trustworthy data.
Frequently asked questions
What is a dichotomous question? +
What is an example of a dichotomous question? +
When should you use a dichotomous question? +
What is the difference between a dichotomous question and a multiple choice question? +
Are dichotomous questions reliable? +
What is the disadvantage of a dichotomous question? +
Is a yes/no question a dichotomous question? +
Build cleaner surveys with Youform
Youform is a free survey and form builder with native support for every question type in this guide: yes/no fields, Likert scales, rating, ranking, matrix grids, NPS, and open-ended. Conditional logic branches the survey on every binary answer, so your screeners and follow-ups stay clean. Unlimited submissions, no respondent limits, and 300+ templates on the free plan.
If you're auditing question wording across a longer survey, also see our guide to double-barreled questions and how to fix them, the format mistake that breaks the most survey data.
No credit card required · Unlimited submissions on the free plan
