Slider (slider question)

"Drag the slider from 0 to 100" - instead of buttons, the respondent drags a handle along a scale. This is a slider: a visual element for choosing a value.

It looks modern and saves space. But on mobile, accuracy is lower - it's easy to miss, and a finger covers the scale. For 5-7 gradations, buttons are often more reliable. In SurveyNinja, a slider can be a display option for a scale. Below - when a slider is appropriate and what the pitfalls are.

A slider produces a numeric value - you can calculate the mean and median. But if there are few gradations (1-5), the difference from buttons is mainly visual. The respondent still picks a discrete value. A slider pays off with a continuous scale or a wide range (0-100, 1-10 with decimals).

What a slider is in a survey

Slider - an interface element where the respondent chooses a value by dragging a handle along a horizontal (or vertical) scale. The range is defined - for example, 0-100 or 1-10. The result is a number. The difference from a button-based scale is the way you interact and the visual look. A slider is more compact for a wide range; on mobile and with a small number of gradations, buttons are usually more precise and clearer.

In short: "drag to the right spot" instead of "tap an option."

Slider vs buttons: when to use which

Buttons (1, 2, 3, 4, 5) are an explicit choice - all options are visible. The respondent clicks and immediately sees the result. On mobile, that's convenient. With a slider, you need to figure out where to drag; on a touchscreen accuracy is lower, and a finger covers the scale. For 5-7 gradations, buttons are more reliable.

A slider pays off for a 0-100 or 1-10 range with steps, or when a "continuous" look matters - for example, estimating a percentage or a probability. Visually, a slider takes up one line; 10 buttons are wider. But UX research shows that for surveys, buttons more often produce more stable data.

Problems on mobile and with accuracy

On a smartphone the slider handle slips - it's hard to land on exactly 7 out of 10. People drag "roughly," and clusters appear on round numbers (50, 75, 100) or on gradations. On desktop with a mouse it's more accurate, but still not everyone likes dragging.

No gradations - a slider without labels ("minimum" and "maximum" at the ends, but no markings) increases the approximation. Adding labels (0, 25, 50, 75, 100) helps, but then it's closer to a discrete scale.

In SurveyNinja: the "Scale" element with a slider type

In the SurveyNinja builder, the slider is a display option for the "Scale" element. You choose the "Slider" type instead of buttons or stars. You set the range (min-max) and the labels for the poles. Answers are recorded as a number. More details - in element settings.

Results in reports - mean, distribution. The export to Excel is a numeric column. This is handy for cross-tabs and segmentation. With discrete gradations (1-5), the mean for the slider and for buttons is usually close; at 0-100, the slider produces more variety in values.

Initial position: center or empty

A slider can start in the center (50 out of 100) or be empty (with no selected value). The center carries a risk of an anchor: the respondent leaves it as is or shifts it slightly. Empty forces a choice but can be annoying. A neutral option is to start on the left (minimum) with a "drag" hint; the respondent deliberately drags.

In SurveyNinja, the initial position is configurable - check that you're not creating an anchor by default. For research where objectivity matters, it's better not to set a "hint" through the starting position.

When a slider is appropriate

Estimating a percentage, a probability, "how much" on a 0-100 scale. Visual surveys where a modern look matters. A desktop audience - a mouse is more precise than a finger. When the range is large (20-30 gradations), buttons are awkward and a slider is more compact.

When buttons are better: 5-7 gradations, a mobile audience, rigorous research with an emphasis on data stability. A rating scale with buttons is a proven option.

Common mistakes

A slider without labels. Only "minimum" and "maximum" - the respondent doesn't understand what the middle means. Labels (0, 50, 100) or markings on the scale help.

Using it on mobile without testing. Open the survey on a phone in preview - is it convenient to drag, is the scale visible? If not - switch to buttons.

A discrete slider without a step. "From 1 to 5" with a slider - the result can be 3.7 or 2.9. Do you need that? If the gradations are whole numbers, set the step to 1. Or use buttons.

Required field. A slider with no value is a skip. If the question is required, the respondent must move the handle. Make sure that "empty" is not accepted - otherwise you get false skips from people who didn't realize they had to drag.

Vertical vs horizontal

More often the slider is horizontal - it's more familiar. A vertical one is sometimes used for "higher is better" (for example, volume), but it's rare in surveys. Important: on a narrow screen a horizontal slider gets compressed - the labels overlap each other. Check it in preview on different devices.

Case: an NPS-like 0-10 scale as a slider

Instead of 11 buttons (0-10) - a slider. It saves space and looks tidy. 300 responses on desktop - a mean of 7.2, with a distribution close to the buttons. On the mobile version of the same survey - more clustering at 5, 7, 10; a mean of 6.8. A difference of 0.4 may be an artifact of the slider on mobile. Conclusion: for important metrics and a mixed audience, buttons are safer.

Areas of application

Estimating a probability (0-100%), importance, "how much do you agree" on a continuous scale. Creative and visual surveys where a modern interface matters. A/B tests: slider vs buttons - you can compare how the type of element affects the answers. For standard quantitative surveys with 5-7 gradations, buttons remain the standard.

A slider is choosing a value by dragging a handle along a scale. Compact for a wide range. Accuracy is lower on mobile. For 5-7 gradations, buttons are more reliable. In SurveyNinja - a display option for the "Scale" element.