Evan Hensleigh sens me this redesign of the cross-national time use graph:
Here was my version:
And here was the original:
Compared to my graph, Evan’s has better fonts, and that’s important–good fonts can make a display look professional. But I’m not sure about his other innovations. To me, the different colors for the different time-use categories are more of a distraction than a visual aid, and I also don’t like how he made the bars fatter. As I noted in my earlier entry, to me this draws unwanted attention to the negative space between the bars. His country labels are slightly misaligned (particularly Japan and USA), and I really don’t like his horizontal axis at all! He removed the units of hours and put + and – on the edges so that the axes run into each other. What was the point of that? It’s bad news. Also I don’t see any advantage at all to the prehensile tick marks. On the other hand, if Evgn and I were working together on such a graph, we would probably come up with something better than either of us would make alone.
Thanks.
I had looked at the original and said, "Eh, too much work to figure out what it says." Your versions are much easier to grasp.
My only additional suggestions would be:
1. Put faint vertical gridlines every 15 minutes so you can come up with cocktail party nuggets like "The average French person spends X minutes per day more on grooming than the average American."
2. Put an abbreviated version version of the horizontal titles on the right so it's easier to remember what Turkey's bars mean.
Tick marks are pretty useless but faint (e.g., 10% gray or lighter) vertical gridlines are helpful.
Also, I wouldn't mind using opposite colors for the bar fills, such as red and green, depending upon whether the bars are above or below zero. It's not necessary, but it makes multiple comparisons register faster.
IMHO, Evan's graph looks more "magaziney" with the thicker and differently colored bars – more eye candy than chart junk. But I agree that the horizontal axis at the bottom is not great.
With respect to the doughnut chart, if you have represented numbers graphically, and have then added the number values themselves so as to help make the relevant comparisons, then you have likely made a bad graph. An exception may be adding numbers that represent extreme values (e.g. to distinguish between 1% and 0.1%).
I like the square end bar plot better, which I assume was made using the wonderful command called barplot. I knew you were going to point out the unnecessary use of color without even reading.
Thanks for your commentary, Andrew!
I agree with most of your critiques, but I do prefer the thicker bars, because they don't have a minimum width. I find the haloed lines make every bar look significant, even if the actual difference is tiny. (The colors, then, are to keep the rows straight when the bars disappear for a few columns, though there are probably better ways of accomplishing that. Coding positive and negative is definitely a better use of color.)
You're absolutely right about the horizontal axis, though.
Steve:
I assume you're kidding about the 15-minute differences, but just in case you're not . . . I wouldn't trust any differences that small.
Actually, I don't know that I trust any of these numbers. If they confirm one's stereotypes and experiences, they seem believable; otherwise who knows how much to trust them. But it was fun to make the graphs.
Evan:
As I wrote in my entry with the R code, I tried different colors for positive and negative but I found it too distracting from the larger patterns.
colors are very important to keep rows straight. if you want to compare an activity across countries it's much easier if you have colors, one per country. that's the main improvement over your version, imho.
I would add a column of numbers to the right of the activity labels and the left of France, giving the unweighted average across countries of time put into each activity.
R code for this one, too?