3 ms·

Sources are listed after the article: > Sources: James Trussell, Office of Population Research, Princeton University; Brookings Institution

• raquo 12y ago
  That's not saying much. The guy wrote hundreds of articles on this topic since 1972: http://www.princeton.edu/~trussell/publications.htm http://www.princeton.edu/~trussell/publications.htm The charts also ignore the fact that risk is more correlated with number of usages rather than number of years of usage, so the numbers themselves are quite useless. Overall, I am disappointed by this... um... style of journalism.
  • reubenmorais 12y ago
    > The charts also ignore the fact that risk is more correlated with number of usages rather than number of years of usage, so the numbers themselves are quite useless. I don't know how you're expecting that to be measured/reported. They're just building on top of how effectiveness of contraceptive methods is already reported, which comes from polling actual people and extrapolating from that, not some theoretical calculation of the chance of failure per individual sexual relation.
    • raquo 12y ago
      I'd say it's not unreasonable to expect a question of "how many times did you use said contraception last year" in the poll/questionnaire. If the "perfect" scenario can be self-reported, this is fair game as well I think.