Were homicides decreasing before control

Could the changes be caused by noise? This calls for a t test, to see
if the average homicide rate changed.

Andy Freeman said:

There's an interesting thing about averages. If you take the average
of a declining series and compare it to the end point, you find that
the average is, not surprisingly, higher.

Lambert's stats show that the pre-control rates were decreasing.

Were the pre-control rates decreasing?

NSW average homicide rate
1901-1910  2.3
1911-1920  2.3

There certainly is no long term trend. Perhaps there is a short term
trend:

NSW average homicide rate
1912-1914 2.3
1915-1917 2.2
1918-1920 2.3

No short-term trend either.

I did a least squares line of best fit on the rates for the years
1911-1920 and the slope of the line turned out to be -0.02, so I
suppose you could claim that the trend was a decrease of 0.02 per
year. (Although a 95% confidence interval on that slope includes 0.)
If that "trend" continued, the average rate for 21-27 would have been
2.14. In fact it was 1.4. A t test shows these two numbers to be
significantly different.

The "trend" over the ten pre-control years in Qld was an increase of
0.14 per year. The "trend" over the pre-control years in England was
a decrease of 0.002 per year. You cannot claim a pre-existing trend
as a cause for the post-control reduction.

The post control rates were rather steady, but right at the point
reached by the previously declining trend, thus the post-control
average is necessarily lower than that of the declining trend.

No matter how many times you say "declining trend" you won't be able
to create one that isn't there.

The significant events were:
1920: NSW controls all guns
1927: Queensland controls handguns and NSW drops controls on long guns

The only significant events, or the only one that Lambert chooses to
talk about? Considering that Lambert previously claimed that an
adjacent state was necessarily a good control ....

Goodness me. I forgot to type in a complete history of Australia in
the period 1911-1937. Those were the significant changes in gun laws.
Other events you may not have heard about: World War I 1914-1918, the
Depression 1929+.

The war seems to have caused a dip in the NSW homicide rate from
1915-1918 (this also happened in England, Germany, Italy..). This is
probably because a significant percentage of the young males (who
commit most homicides) were in the armed forces. The Depression
doesn't seem to have had any effect.

The means:

           NSW  Qld
1911-1920  2.3  4.1

Hold it. Lambert also assured us that NSW and Qld were "all other
factors are the same" except for when they introduced gun control, but
now we find that they weren't identical by his own measure before gun
control.

If you're going to follow up to my postings, PLEASE read them first.
Here is a quote from the original posting:

I couldn't find any data for the homicide rate in Queensland prior to
72. I was able to find the numbers of people charged with
manslaughter and murder from 1900-1977. Adding these and dividing by
the population each year gives the homicide charge rate per 100 000

4.1 is the homicide charge rate. This should be closely correlated
to the homicide rate. Comparing the homicide rate and the homicide
charge rate in Queensland after 72, I find the homicide rate is about
2/3 of the charge rate. This suggests that the homicide rate in Qld
might have been about 2.7, which is comparable to the 2.3 figure in
NSW.

Tags

More like this

Frank Crary said (referring to changes in homicide rate in NSW and Qld): The sharp drop you claim to see, is (in my opinion) not significant compared to the background variations. While this may have been real effect of gun control laws, in is at least as likely that it is a coincidence. Could the…
Andy Freeman said: Why doesn't Lambert tell us about pre-control crime and murder rates and trends in Oz and compare them to post-control rates and trends? If gun control actually worked in Oz, the introduction of controls was associated with a good change in the rate trend. Could it be that there…
Your claim that I have not shown that the situations were stable is false. The homicide rate was roughly constant in the period before gun control and in the period after gun control. Andy Freeman said: The graphs have shown that it was roughly constant AFTER, but before.... there was a dip in…
What on earth do you mean by 'the "nothing else happened" parameter"? Andy Freeman said: Lambert's model is for a transition between two stable situations with some "noise". He uses it to argue that gun control explains the transition. Yet, he doesn't bother to show whether or not anything else…