also, you have presented three different – and conflicting – definitions for the abortion ratio. per the CDC, the abortion ratio is the number of abortions per 1000 live births, *not* the number of abortions per 1000 pregnancies (a number which would be difficult to calculate, as many pregnancies end in spontaneous abortion before the woman knows she’s pregnant), or as you alternatively present it, the “number of abortions per 1000 abortions and live births,” which does not account for known pregnancies that end in miscarriage or stillbirth.

the abortion rate is the number of abortions per 1000 women of reproductive age (15-44).

your charts are pretty, but because you have ignored standard definitions, they are meaningless from a general usage standpoint.

Your small multiples chart displays a more complete image but one may argue that proportions in the last years are relatively stable and we may not need to see the whole time series.

This is where chart design meets information assessment and my fundamental argument over the last years: you can learn some tricks, you can memorize some best practices, but if you don’t know what to do with your data that will be seen in each of your charts.

I believe that small exercises like this (Nathan, over FlowingData, had a similar one last week), can raise awareness for better data management and data reporting skills. Let’s do it more often!