| nanarchy said: I don't get why they bother to use 2 seperate scales on that graph, yes the US is 10 fold higher but you could still easily draw it on the same scale and avoid the first sight confusion that makes them look similiar |
Actually, that makes perfect sense once you realize that this is the whole point behind the chart - to suggest to naive people that the US experienced a similar decline in gun crime over the same time period. 
And btw, that chart is flawed in multiple ways. A severe, but not instantly obvious problem with this chart is that they mix data from different sources: from gunpolicy.org and CDC.
Here's the raw data from gunpolicy.org for Australia (Source: http://www.gunpolicy.org/firearms/region/australia):
2012: 1.03 / 2011: 0.86 / 2010: 1.08 / 2009: 1.06 / 2008: 1.10 / 2007: 1.11 / 2006: 1.18 / 2005: 1.09 / 2004: 1.20 / 2003: 1.46 / 2002: 1.49 / 2001: 1.69 / 2000: 1.70 / 1999: 1.84 / 1998: 1.68 / 1997: 2.32 / 1996: 2.84 / 1995: 2.61 / 1994: 2.90 / 1993: 2.91 / 1992: 3.49 / 1991: 3.59 / 1990: 3.51
Now if you try to get the corresponding raw data for the same time period for the US (Source: http://www.gunpolicy.org/firearms/region/united-states), you get aware of a huge problem:
2012: 10.69 / 2011: 10.38 / 2010: 10.26 / 2009: 10.22 / 2008: 10.39 / 2007: 10.37 / 2006: 10.35 / 2005: 10.39 / 2004: 10.10 / 2003: 10.39 / 2002: 10.51 / 2001: 10.38 / 2000: 10.19 / 1999: 10.35
There is simply no data for the time before 1999 for the US!
"Ah!" one might say, "so that's the reason they used data from two different sources; for the US for the time before 1999, gunpolicy.org has no data, so for the US data before 1999 they had no choice to use the available data from CDC instead!"
Which sounds somewhat plausible at first - but apart from the general problem of mixing data from different sources, there's a much bigger problem that one gets aware of when looking for that missing data on the CDC website
Money quote (http://www.cdc.gov/injury/wisqars/fatal.html):
Note: The coding of mortality data changed significantly in 1999, so you may not be able to compare number of deaths and death rates from 1998 and before with data from 1999 and after. Also, cause-of-injury groupings were updated in 2003, which may affect some WISQARS mortality reports.
So even the CDC admits that the US data from before 1999 simply cannot be compared to other data, not even CDC data!
And now have a look at the bar for the US in the chart: All decline in the US was BEFORE that year 1999!
Now when one only uses the available (and comparable) data from the single source gunpolicy.org, which spans the time period from 1999 to 2012, the chart looks somewhat different; suddenly the charts don't look look so comparable after all. And how should they, given that in that time period the numbers for the US rose by about 3%, while the numbers for Australia dropped by 44%?
