Let’s say I’m creating a mobile questionnaire for people in which I ask about their usage patterns of some tool. The user can specify a period of time in which they used a tool, so perhaps between February and March they started to use a hammer, and as they got more confident with it, they ended up using it more and more frequently, then between March and today they used it every day.
I want to represent those choices with a nice label off to the left, and a
SegmentedControlto the right, as there are only three options.
My problem is I can’t come up with descriptions more concise than “increasingly often”, “even throughout” and “decreasingly often”, and so I come here to find out the obvious thing I’ve forgotten.
(I of course can rearrange the layout so the label is on top, or switch to a different input method, but I’d prefer to change the words if possible).
Ah, I think I understand what sense you mean it in now @AdamBarnes.
Not just more often, but an accelerating increase in frequency (“increasingly more often”) – a property of what I believe mathematicians would call the 2nd derivative.
I have to say the prospect of reducing this concept to a single word is implausible.
Your use of “even throughout” is also misleading – does that mean at an evenly increasing frequency, or at an even frequency? Once you introduce a 2nd derivative, there are more than 3 possible qualitative descriptions of the frequency profile. An increase can be accelerating, decelerating, or steady. A decrease can be accelerating, decelerating, or steady. And then there is a steady frequency throughout. 7 in total (if my own reasoning is sound).
You could thus have states like “decreasingly more often” – that is, always more often, but at a slowing rate of increase each time. And “decreasingly less often”, that is always less often over time, but at a slowing rate of decrease over time. Whereas “increasingly less often” means the same as “less and less over time”. I’m giddy at the thought of a user trying to interpret this correctly.
To express this in natural language to an untrained person, I would think paragraphs or even pages with diagrams would be required to explain, never mind one word.
You would not normally present users with such questions – from a statistics point of view you’d almost certainly get junk answers or no response at all.
Instead, ask them to state how often they have used the thing at different points in time. Once you have that, you can infer the derivatives of those figures yourself.