Comments on: A Link Between the Murder Rate and Traffic Fatalities? http://www.howwedrive.com/2009/11/13/a-link-between-the-murder-rate-and-traffic-fatalities/ Just another WordPress weblog Mon, 13 Feb 2012 15:22:48 +0000 http://wordpress.org/?v=2.5.1 By: ogmb http://www.howwedrive.com/2009/11/13/a-link-between-the-murder-rate-and-traffic-fatalities/#comment-9953 ogmb Mon, 30 Nov 2009 12:03:51 +0000 http://www.howwedrive.com/?p=1293#comment-9953 I think there's a bit of a problem with understanding the concept of a proxy for empirical analysis. When an independent variable is designated as a proxy, it means the actual underlying variable ("aggression" or "alcoholism") is either unobservable or the data could not be obtained by the researcher. The proxy variable serves as a stand-in, to be replaced if more accurate measures will become available. So it doesn't really matter that liver diseases can be caused by other conditions than alcoholism, what matters for this study as that the share of liver diseases caused by alcoholism are roughly identical from state to state. Similarly, the urban-suburban distinction in murders vs. traffic fatalities (and similarly, the relatives-strangers distinction) isn't really relevant as long as human aggression seeks the same outlets in all states. I think there’s a bit of a problem with understanding the concept of a proxy for empirical analysis. When an independent variable is designated as a proxy, it means the actual underlying variable (”aggression” or “alcoholism”) is either unobservable or the data could not be obtained by the researcher. The proxy variable serves as a stand-in, to be replaced if more accurate measures will become available. So it doesn’t really matter that liver diseases can be caused by other conditions than alcoholism, what matters for this study as that the share of liver diseases caused by alcoholism are roughly identical from state to state. Similarly, the urban-suburban distinction in murders vs. traffic fatalities (and similarly, the relatives-strangers distinction) isn’t really relevant as long as human aggression seeks the same outlets in all states.

]]> By: salas http://www.howwedrive.com/2009/11/13/a-link-between-the-murder-rate-and-traffic-fatalities/#comment-9817 salas Thu, 19 Nov 2009 13:00:16 +0000 http://www.howwedrive.com/?p=1293#comment-9817 In the results, they combined factors that increase the traffic fatality rate (homicide rate per capita, proportion of male drivers, proportion of drivers over 64 years of age, and deaths caused by alcohol related liver failures per capita) with factors that decrease the traffic fatality rate (physicians per capita, safety-belt usage rate). Not sure which category income per capita falls into, though I would guess it decreases the traffic fatality rate. In the results, they combined factors that increase the traffic fatality rate (homicide rate per capita, proportion of male drivers, proportion of drivers over 64 years of age, and deaths caused by alcohol related liver failures per capita) with factors that decrease the traffic fatality rate (physicians per capita, safety-belt usage rate). Not sure which category income per capita falls into, though I would guess it decreases the traffic fatality rate.

]]> By: Cory http://www.howwedrive.com/2009/11/13/a-link-between-the-murder-rate-and-traffic-fatalities/#comment-9754 Cory Mon, 16 Nov 2009 13:40:01 +0000 http://www.howwedrive.com/?p=1293#comment-9754 Most murders are between people who know one another. Most traffic accidents are not. Seems to me to be some odd form of random correlation that has nothing to link the two. I suspect that population density would be a more likely common factor. Most murders are between people who know one another. Most traffic accidents are not. Seems to me to be some odd form of random correlation that has nothing to link the two. I suspect that population density would be a more likely common factor.

]]> By: Kevin Love http://www.howwedrive.com/2009/11/13/a-link-between-the-murder-rate-and-traffic-fatalities/#comment-9737 Kevin Love Sun, 15 Nov 2009 14:12:41 +0000 http://www.howwedrive.com/?p=1293#comment-9737 I certainly agree that culture is a deciding factor in road safety. Ontario's government brags about having the safest roads in North America. Its a cultural thing. People may make fun of "Toronto the Good" but it is a great place to live. Our murder rate is very low also. Source: http://www.mto.gov.on.ca/english/safety/orsar/orsar06/chp1_06.shtml I certainly agree that culture is a deciding factor in road safety. Ontario’s government brags about having the safest roads in North America. Its a cultural thing. People may make fun of “Toronto the Good” but it is a great place to live.

Our murder rate is very low also.

Source:

http://www.mto.gov.on.ca/english/safety/orsar/orsar06/chp1_06.shtml

]]>