when defining buckets for histograms it's often (almost always) useful to keep 0 as its own bucket, since it tends to be a fairly special value; however, there's no easy way to do this in the current interface; for example, I usually want equally varied buckets like in the first image (which gives me the second image), but with zero separated out... I could go to "varied" to create the buckets I want, like in the third image, but notice that the buckets are now "<0, 0-1, 1-2, ..." meaning there's no separate bucket for zero (although I'm guessing the 0-1 bucket is actually just for zero?)
this is what it looks like with the buckets set to None although, obviously, that's not what I want, I want a combination of this and the equally spaced buckets
Yeah, the buckets are "up to but not including" so 0-1 is really "0" but agree it would be nice to just have 0, 1, 2, 3+ as an option for showing it.
I set it up like this so it's more readable but it's kind of a pain to have to remember to do this every time
this is particularly annoying when breaking down by clearly integer quantities (notice the graphs "knows" it's a frequency breakdown and an integer quantity because it says 1-5 times so it should really say 1-4 times, 5-9 times, etc. )
I had to explain this behavior several times to my CEO and then, several months later, he forgets and asks me about it again and I have to look it up again and explain it again 😄
yeah. I had a customer in the UK and they had a hard time with this too. I wonder, just now as I am thinking out loud. What happens if I recast the results to strings . . . .
if you cast it to string then customize buckets has the string interface, so you can manually group the various numbers together, but it's kind of a pain to do
like this, not sure if this is better, especially if your buckets are fairly large
but it's probably a useful workaround in some cases