This question already has an answer here:

In a small sample (n = 3) we found that one user almost always scrolled their tablet/phone by stroking along the margin/gutter where the scrollbar exists. This caused a problem, as we have small fold-out button hanging on this edge of the screen which this user would trigger.

However, the other two users boldly stroked up and down the centre of the screen, avoiding the issue.

I'm pretty sure that most of our development and testing team also use the centre, but only because we've only had that one user report the problem with the fold-out.

I tried looking for "scrolling heatmaps" - but this is the wrong kind of information to answer my question.

So, before looking at how to record this information for myself, I was wondering if there was already data available on where people are running their fingers to scroll in browser windows. Of course, there are some confounding factors like page design and drag/drop functionality, to consider, but some data is better than no data.

Where would be a good place to find this data if it does indeed exist?

With regards to the possible duplicate - the other question is similar, but the answer there would not satisfy my question.

marked as duplicate by Vitaly Mijiritsky, JonW Jul 26 '16 at 10:23

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • 1
    I don't habe any public data, so this goes as a comment: we did a study about this for a tablet version of an app, and we found that the big bold strokes you mention were common on older people, while younger audiences usually grabbed the tablet with both hands, but kept scrolling just as with a mobile. That is: with their thumb. – Devin Jul 25 '16 at 15:06
  • @VitalyMijiritsky, that answer certanly doesn't answer this question. Furthermore it's based on an article that literally says you can translate a phone to a tablet, just like that. The author is ignoring tablets are handled with TWO hands, and that's just the beginning – Devin Jul 25 '16 at 15:11
  • @Devin - interesting - I hadn't considered potential differences by age. But then again, I haven't really observed the behaviour you describe. Small sample set again, but my nieces and nephews may hold the tablets with both hands, but they still let go with one hand and scroll with big bold strokes up the centre when they need to. – HorusKol Jul 25 '16 at 23:21
  • @HorusKol, sorry, I should have clarified first: research included a range from 18 to 70 years old, in our specific case no kid would use this app. Since this was for a financial app, even 18 years old was quite a reach but we played on the safe side as well , which helped in noticing this behavior since younger audiences modified the perception. On a side note, I'm not very young, yet I generally do the same on my iPad (scrolling with thumb) unless I need to do long scrolls – Devin Jul 25 '16 at 23:38

I think no matter what the data heat map looks like (assume we can find one), your user(s) still has trouble to accidently triggers the fold-out function. Jakob Nielsen suggests that using only 5 users to do usability testing is sufficient to find problems (https://www.nngroup.com/articles/why-you-only-need-to-test-with-5-users/).

Although you only have three users, this IS your data. Why bother to have more users coming in, or try to find someone else's data, instead of redesign the position of the fold-out button? Maybe you can have it on the left and eliminate all the potential false triggering? Just an idea, good luck :)

  • his sample test was for 3 users, so 1 means 33.333% and 2 means 66.666%. This can't be seriously considered as representative. What if he does the study with 100 people and he founds that 1 is the only person with the problem? what if the original 2 become 2/100? what if there are additional UI concerns that prevent moving the button? should the OP work on moving something he shouldn't move based on a possibly false premise such as a single user having a problem? Test and research is exactly what he needs – Devin Jul 25 '16 at 17:47
  • Well, we don't want to move the fold-out to somewhere else that will cause a problem, and since this is one thing in a list of about 100 things we want to fix/improve/implement, we need to determine the ROI. Like Devin says, if this is only a 1 in a 100 problem, we will put our energies in what improves things for 99 or 100 users instead. – HorusKol Jul 25 '16 at 23:16
  • Also - 5 testers may show you all problems, but it won't give you a scale of a problem - especially if only one of those 5 reports the problem. – HorusKol Jul 25 '16 at 23:24

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