Psudo on politics
US Urbanity vs. Violent Crime Elasticity
Okay, this one takes some explanation, and some narrative. Please bear with me.
Urbanity is the proportion of the population that lives in urban areas (as opposed to rural). Figures for it properly come from the US Census Bureau. I used a summary of the 1990 Census urbanity data that had already been assembled online:
www.allcountries.org/uscensus/37_urban_and_rural_populati...
In a previous chart on US violent crime rates, I noted that there was an obvious national peak in violence in the early 1990s, but that the states contributed to that peak in a variety of extents ranging from not at all to many hundreds of crimes per 100,000 population. I decided to name the extent to which a state's crime rate changed as part of that national trend, and the name I came up with is "Violent Crime Elasticity."
National violence rate peak graph:
flickr.com/photos/91128695@N07/8298646211/
To calculate a numerical value for this elasticity for each state, I chose a low point in the state's crime rates of the 60s, a high point from the early 90s, and a low point in the past 5 or 6 years. If we call these three values A, B, and C respectively, the formula for Violent Crime Elasticity is |B - (A + C) / 2| where |x| means absolute value of X. In simple terms, the number describes how much that state's violent crime rate rose and fell in the national pattern.
It seems to me that some powerful crime-causing force of society was exerted and that the states were susceptible to that force to varying degrees. That is not properly called a hypothesis because it is not obviously testable, but it potentially explains what's going on. (FactCheck.org says it's "a trend criminologists chalked up to 'changes in the crack cocaine market.'")
WIth that paradigm in mind, I examined this map of the peak crime rates during the most violent years of the early 1990s.
flickr.com/photos/91128695@N07/8299176067/
The states that formed the top of the peak seemed to be ones with major urban populations: It occurred to me that perhaps urban populations were more susceptible to that crime-causing force I suspect may exist.
This chart represents a rough test of that properly testable hypothesis. It results in a 30.5% correlation between elasticity and urban proportion of the population, which is not high enough to mark the the hypothesis as confirmed but is high enough to suggest that a better analysis may be able to. Perhaps a county-by-county analysis could confirm or reject the theory more conclusively.
US Urbanity vs. Violent Crime Elasticity
Okay, this one takes some explanation, and some narrative. Please bear with me.
Urbanity is the proportion of the population that lives in urban areas (as opposed to rural). Figures for it properly come from the US Census Bureau. I used a summary of the 1990 Census urbanity data that had already been assembled online:
www.allcountries.org/uscensus/37_urban_and_rural_populati...
In a previous chart on US violent crime rates, I noted that there was an obvious national peak in violence in the early 1990s, but that the states contributed to that peak in a variety of extents ranging from not at all to many hundreds of crimes per 100,000 population. I decided to name the extent to which a state's crime rate changed as part of that national trend, and the name I came up with is "Violent Crime Elasticity."
National violence rate peak graph:
flickr.com/photos/91128695@N07/8298646211/
To calculate a numerical value for this elasticity for each state, I chose a low point in the state's crime rates of the 60s, a high point from the early 90s, and a low point in the past 5 or 6 years. If we call these three values A, B, and C respectively, the formula for Violent Crime Elasticity is |B - (A + C) / 2| where |x| means absolute value of X. In simple terms, the number describes how much that state's violent crime rate rose and fell in the national pattern.
It seems to me that some powerful crime-causing force of society was exerted and that the states were susceptible to that force to varying degrees. That is not properly called a hypothesis because it is not obviously testable, but it potentially explains what's going on. (FactCheck.org says it's "a trend criminologists chalked up to 'changes in the crack cocaine market.'")
WIth that paradigm in mind, I examined this map of the peak crime rates during the most violent years of the early 1990s.
flickr.com/photos/91128695@N07/8299176067/
The states that formed the top of the peak seemed to be ones with major urban populations: It occurred to me that perhaps urban populations were more susceptible to that crime-causing force I suspect may exist.
This chart represents a rough test of that properly testable hypothesis. It results in a 30.5% correlation between elasticity and urban proportion of the population, which is not high enough to mark the the hypothesis as confirmed but is high enough to suggest that a better analysis may be able to. Perhaps a county-by-county analysis could confirm or reject the theory more conclusively.