Closed-ended question

A closed-ended question is a survey question with predefined answer options: the respondent picks one or several instead of writing their own text. It is used when the possible answers are known in advance and you need data that can be counted right away, for example "How often do you use the service?" with once a day, once a week, less often.

This is a closed-ended question. In SurveyNinja it is built with the "Radio Button", "Multiple Choice", "Dropdown" and other elements, and the report shows the share of each option. Below: when a closed question beats an open one and what the pitfalls are.

Closed questions give structured data - ready for analysis. Open ones require parsing and grouping. They are often combined: a closed question for the metric, an open one for context. A comparison of the types is in the article on open vs closed questions.

Choosing from a list rather than free text

Closed-ended question - a question with predefined answer options. The respondent chooses one or several, but does not write their own text. The options are set by the researcher - a full list or a scale. The data comes out structured: percentages, shares, you can compute an average (for scales), build cross-tabulations and segment with no extra processing.

In short: "choose from a list" instead of "describe in your own words".

Radio buttons, checkboxes, dropdown list

Single choice. One option from the list - radio buttons, dropdown list. "What is your gender?", "Which of the three plans?". Fits mutually exclusive categories. The result is a single variable with discrete values.

Multiple choice. Several options - checkboxes. "Which channels do you use for news?" - you can mark several. An "Other" option with a text field is often added so you do not lose answers that did not fit the list. Analysis is harder: you have to decide whether to count answers by individual options or by combinations.

Dropdown list. Options are hidden until you click - it saves screen space. Handy for a long list (regions, years). On mobile it can be awkward - an extra tap. The alternative is radio buttons with scrolling.

Options are known: demographics, frequency, plans

When the options are known and can be listed. Demographics (gender, age, region), usage frequency, choosing a product/plan, rating on a scale. For quantitative research closed questions are the backbone: fast, clear, easy to count.

When the options are unknown - it is better to start with an open question (a pilot) or to add "Other" with a text field. Otherwise respondents will pick the "closest" option or skip the question.

Ready-made percentages vs the risk of an incomplete list

Pros: a quick answer for the respondent (a click instead of typing text), a higher response rate, data ready for analysis. Percentages, averages, cross-tabulations - with no coding. Comparing across waves and segments is easy.

Cons: the list may not cover every answer. The respondent will pick the "closest" option or leave. There is a risk of acquiescence and central tendency - people gravitate toward the middle and toward "agree". The options need to be thought through in advance - otherwise the data will be distorted.

A complete list, option order, randomization

The options should cover every possible answer. Otherwise "others" or skips will appear. Mutually exclusive and exhaustive is the classic rule. For non-obvious lists a pilot with an open question helps.

Option order matters: the primacy effect and the recency effect - the first and last options get picked more often. Randomizing the order reduces this influence. In SurveyNinja you can turn on random order of answer options.

Being required is another setting. A closed question can be made required so that everyone answers. For sensitive topics the option to skip is sometimes kept - otherwise people drop out of the survey.

Elements in SurveyNinja: adding, reports, export

Closed questions in the SurveyNinja builder are the "Radio Button" and "Multiple Choice" elements. You add them from the panel, write out the options, and if needed enable "Custom option" (with a text field). The "Dropdown" is a separate element for long lists. More detail is in the element settings.

Answers are shown in the reports - percentages and charts for each option right away. Export to Excel (XLSX) or CSV gives a column with the selected value; for multiple choice it gives separate columns or a combined field, depending on the format. From the export you can build cross-tabulations in a spreadsheet or a statistics package.

Pairing with an open question: NPS and "Why?"

A common pattern: a closed NPS question (0-10), followed by an open "What had the biggest impact on your score?". The closed question gives the metric, the open one gives context. The same with rating scales: "Rate from 1 to 5" - closed, "Why this rating?" - open. The two types complement each other.

Incomplete list, mixed criteria, too many options

Incomplete list. Options are missing - the respondent guesses or skips. The fix: a pilot, "Other" with text, or reworking the options.

Mixed criteria in one question. "Pick a channel: website, social media, friends, advertising" - friends and advertising are not channels in the same sense. Better to split it or clarify the wording. See double-barreled question.

Too long a list. 15-20 options are tiring. A dropdown or grouping by category will make the choice easier.

Illustrative example: a plan survey

A survey for pricing: "Which plan do you use?" - single choice from 5 options. Plus "Other" with a text field - for non-standard cases. 800 responses - a summary in a couple of minutes: 40% plan A, 35% plan B, 15% plan C, 7% other, 3% no answer. A cross-tabulation by gender or age is easy to build from the export. The closed question delivered ready-made figures with no manual parsing.

Asked openly, "Which plan?" would mean reading the answers, catching typos ("basc", "premiumm") and grouping them by hand. A closed list removes this task and makes the data comparable across survey waves.

A closed-ended question - with predefined answer options, the respondent chooses from a list. Single or multiple choice. The data is ready for analysis with no coding. In SurveyNinja - the "Radio Button" and "Multiple Choice" elements, and the "Dropdown".