User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Are there any studies that show the percent usage of the following techniques for scrolling a long web page?

  • Mouse scroll wheel
  • Up/Down arrow key
  • Page Up / Page Down | Home / End keys
  • Dragging scroll bar
  • Clicking up/down arrows on scroll bar
  • Clicking the negative space of the scroll bar

Some of my Javascript code runs very slow when users drag the scroll bar, but it runs perfectly smooth in all other techniques. If the percentage of scrollbar dragging is small, I might just let this bug slide and work on more important things.

Personally, I rarely use drag the scroll bar. I only use it when I have memorized the position of an element, like a particular photo in Windows Explorer, for example. I will drag the bar to that memorized position. My website's content changes from hour to hour, so memorizing scrollbar position would not help the user in this case.

share|improve this question
I just want to add two more scrolling methods that I use personally - I click the scroll wheel and drag to pan, and when using a Thinkpad, I click the middle button and use the pointer stick to scroll. – Jonathan Jul 14 '11 at 2:00
Also, some people have those fancy Apple mouses that have a touchpad integrated in the surface of the mouse and accept scrolling gestures. – Jonathan Jul 14 '11 at 2:01
I use all of the methods you mention... Depends on the page size, the scroll bar size, whether I am on a notebook or at my desktop machine, etc. – Marjan Venema Jul 14 '11 at 6:14
On a laptop the other method is to use the trackpad vertical scroll area. However this goes horribly wrong on pages with a google map. – PhillipW Jul 14 '11 at 9:34
Have you observed the same slow performance in all browsers? There's a good chance if you tested in Internet Explorer that it is only happening there and that automatically narrows it down to 25% of users (who are used to bad performance anyway right? :P) – Anson Kao Jul 14 '11 at 17:06
up vote 4 down vote accepted

You're asking about the most common situation. It is very much going to depend on the type of mouse and the device the user is using and more importantly, your type of content (see below).

I know there are touch screen, mobile and other device to consider, but really the question is specifically about the most common situation in which the OP's website will be accessed and that sounds like it's going to be a standard desktop.

According to Microsoft's own research, the qualitative results of their analysis of scrolling techniques shows that the ScrollPoint joystick mouse such as the IBM ScrollPoint Pro and accelerated mouse wheels (where the acceleration is a software setting in the OS, Browser, or device settings) that have a 'flick' type of action are good for scrolling long distances - provided it's set up at 3 scroll lines per notch in the case of the accelerated mouse wheel. Microsoft seem to suggest that 3 lines per notch is the default setting, and that's what I'm seeing on my machine right now.

However - ScrollPoint joystick mice and mouse wheels and associated wheel acceleration settings that afford flicking are probably not the most common type of scrolling environment. I'd imagine the commonest, by percentage, to be the standard (cheap) mouse.

Up down arrows, negative space, page keys are more likely used for local examination - moving back and forth from the current page in order to visually relate the previous or next chunk with the current chunk. A kind of progressive step mode.

With very long pages, dragging the scrollbar is commonly used to scan (visual search) or get an overview of the whole page - to determine its usefulness - a kind of 'see before you buy'. The longer the page, the more likely users are to reach out and actually drag the scrollbar up and down.

The more your content affords random access viewing as opposed to linear access, the more the scrollbar will be used. And that depends on your content and whether users are likely to want to examine the content in successive step mode - like reading a long single article (linear access), or to flip about, or get an overview - eg a collection of articles of which some are of interest, and some not (random access).

Also factor in whether your content promotes the beginning and the end pages to be of more interest to the middle bulk (eg examining first and last items in a long ordered list). In which case Home and End keys come back into play a bit more.

I'd recommend you try some user testing. It might be worth getting a few different users to actually use the website and either watch them, or gather statistics about how the user is moving about on your page: This article from superuser SE could help: Application to gather mouse usage statistics.

share|improve this answer
Can you define "pointing stick" and "accelerated mouse wheel"? Is the accelerated wheel on the Logitech MX Revolution? I own one and I turn off hyperscrolling because it doesn't work well. The wheel will rebound slightly at the end of the flick. I don't know how Microsoft concluded that "pointing sticks" and "accelerated wheels" are the norm when I've never heard of "pointing sticks", and the "accelerated wheel" is so poorly implemented. – JoJo Jul 14 '11 at 18:22
My mistake - that was confusing - I did not mean pointing stick (I should have said Scrollpoint joystick) - I will edit that. They're not saying they are 'the norm' - just good for scrolling long distances, thus viable as an alternative to dragging the scrollbar. The devices are described in the linked research doc. An example of the joystick mouse is the ibm scrollpoint pro mouse, and the accelerated wheel is a software implementation which is under analysis in the paper. Acceleration settings are OS/Desktop/Browser/Device dependent - Google: "mouse wheel acceleration", eg Firefox option – Roger Attrill Jul 14 '11 at 18:50

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.