Kieran Healy posts this time series of assault deaths in the USA and eighteen different OECD international locations:
Good graph. I’d simply make three modifications:
1. Label y-axis as deaths per million (with labels at 0, 25, 50, 75, 100) quite than deaths per 100,000. Why? I simply suppose that “per million” is simpler to observe. I can image an space with 1,000,000 folks and say, OK, it could have about 50 or 75 deaths within the U.S., as in comparison with 0 or 15 in one other OECD nation.
2. Put a tough x-axis at y=0. As it’s now, the time sequence kinda float in midair. A zero line would offer a helpful baseline.
3. When itemizing the OECD international locations, use the nation names, not the three-letter abbreviations, and listing them in lowering order of common charges, quite than alphabetically.
I’d additionally wish to see the charges for different international locations on this planet. However might be a large number to cram all of them on the identical graph, so perhaps do a couple of extra: one for Latin America, one for Africa, one for the European and Asian international locations not within the OECD. Or one thing like that. You would show all 4 of those graphs collectively (utilizing a typical scale on the y-axis) to get a worldwide image.
And one other
OK, that was good. Here’s another graph from Healy, who introduces it as follows:
It’s a related scatterplot of whole well being spending in actual phrases and life expectancy of the inhabitants as a complete. The truth that actual spending and expectancy are inclined to steadily improve for many international locations in most years makes the year-to-year connections work regardless that they’re not labeled as such.
And the graph itself:
I’d seen a scatterplot version of this one . . . This time-series model provides a variety of context, particularly displaying how the U.S. used to suit proper in with these different international locations however doesn’t anymore.
And what are my graphics recommendations?
1. Possibly label two or three of these OECD international locations, simply to present a way of the vary? You would choose three of them and coloration them blue, or simply heavy black, and label them instantly on the traces. I’d additionally label the U.S. instantly on the pink line; no want for a legend.
2. To get a way of the time scale, you could possibly put a fats dot alongside every sequence each 10 years. Or, if that’s too crowded, you could possibly do each 20 years: 1980, 2000, 2020. In any other case as a reader I’m in a clumsy place of not having a transparent sense of how the curves line up.
3. Once more, I’d wish to see a couple of extra graphs displaying the opposite international locations of the world.
Kieran Healy posts this time series of assault deaths in the USA and eighteen different OECD international locations:
Good graph. I’d simply make three modifications:
1. Label y-axis as deaths per million (with labels at 0, 25, 50, 75, 100) quite than deaths per 100,000. Why? I simply suppose that “per million” is simpler to observe. I can image an space with 1,000,000 folks and say, OK, it could have about 50 or 75 deaths within the U.S., as in comparison with 0 or 15 in one other OECD nation.
2. Put a tough x-axis at y=0. As it’s now, the time sequence kinda float in midair. A zero line would offer a helpful baseline.
3. When itemizing the OECD international locations, use the nation names, not the three-letter abbreviations, and listing them in lowering order of common charges, quite than alphabetically.
I’d additionally wish to see the charges for different international locations on this planet. However might be a large number to cram all of them on the identical graph, so perhaps do a couple of extra: one for Latin America, one for Africa, one for the European and Asian international locations not within the OECD. Or one thing like that. You would show all 4 of those graphs collectively (utilizing a typical scale on the y-axis) to get a worldwide image.
And one other
OK, that was good. Here’s another graph from Healy, who introduces it as follows:
It’s a related scatterplot of whole well being spending in actual phrases and life expectancy of the inhabitants as a complete. The truth that actual spending and expectancy are inclined to steadily improve for many international locations in most years makes the year-to-year connections work regardless that they’re not labeled as such.
And the graph itself:
I’d seen a scatterplot version of this one . . . This time-series model provides a variety of context, particularly displaying how the U.S. used to suit proper in with these different international locations however doesn’t anymore.
And what are my graphics recommendations?
1. Possibly label two or three of these OECD international locations, simply to present a way of the vary? You would choose three of them and coloration them blue, or simply heavy black, and label them instantly on the traces. I’d additionally label the U.S. instantly on the pink line; no want for a legend.
2. To get a way of the time scale, you could possibly put a fats dot alongside every sequence each 10 years. Or, if that’s too crowded, you could possibly do each 20 years: 1980, 2000, 2020. In any other case as a reader I’m in a clumsy place of not having a transparent sense of how the curves line up.
3. Once more, I’d wish to see a couple of extra graphs displaying the opposite international locations of the world.