3 ms·

1. I am not an expert here, but unless you are too I wouldn't refute claims made by peer-reviewed publications without at least offering an argument against the

1. I am not an expert here, but unless you are too I wouldn't refute claims made by peer-reviewed publications without at least offering an argument against the analysis made. http://iopscience.iop.org/article/10.1088/1748-9326/10/11/114005/meta;jsessionid=CBB0DEF280F6107CE4771D4142C8BA0D.c5.iopscience.cld.iop.org http://iopscience.iop.org/article/10.1088/1748-9326/10/11/11... If you agree that the environment is worse off and that the environment being worse off causes more deaths, then you agree that they caused deaths. Then it becomes a question of magnitude, which they tried as best as possible to estimate. 2. There is a confidence interval around 59 in the study which doesn't include 0, so it is statistically significant according to the analysis they did. Not sure what sort of analysis would say that 59 deaths is statistically insignificant unless it was a difference between two groups, but that's not the case here. 3. There's a problem here with news articles boiling down an estimate with a wide confidence interval to a single number. There needs to be a way for the public to start understanding statistical conclusions and confidence intervals.

Unklejoe 10y ago
> [If you agree that the environment is worse off and that the environment being worse off causes more deaths, then you agree that they caused deaths.] ...I should have said “worse off than if VW had not cheated and all other variables remained the same”. I don't question that more emissions results in more deaths. I question their ability to reliably calculate the number of deaths caused when dealing with numbers that small (~0.0003%) and leaving out some very important variables (which they do reference but don't incorporate). Of course, the publication was meant to compare the claimed vs actual emissions, but I'm talking in the more general case of overall emissions. The publication claims that: >> [Finally, we note that there may have been environmental benefits of the defeat device that we have not computed. For example, the reduced use of diesel exhaust fluid in selective catalytic reduction may potentially have also reduced ammonia slip, and therefore associated PM2.5 exposure and health impacts. In particular, Dedoussi and Barrett (2014) find that total PM2.5 exposure attributable to road transportation is caused approximately equally by NOx and NH3 emissions.] I can think of a few more inadvertent environmental/health benefits of the defeat device: For example, suppose that VW hadn't cheated and instead spent more time and money on R&D to come up with a legitimate solution. They would then have to pass this cost on to the consumers in the form of a more expensive car. Would all of the people who purchased diesel VW's still have purchased them if they were more money? If not, would they have purchased a car that produces worse emissions instead? The VW Jetta has a 5 star crash rating. Would all of the people who purchased other cars purchased cars with the same crash rating, or perhaps a car with a worse crash rating? There are a bunch of things that you could list that could all have an impact that rivals the 59 cited in the publication. We're talking about 59 out of 17,500,000 (number of annual deaths integrated over 2008-2015). It wouldn't take much to actually find out that the VW emissions scandal actually resulted in more lives saved. EDIT: That said, the publication isn't really trying to measure the amount of deaths caused by the scandal itself, but rather the increased emissions portion of it. Perhaps my original comment was out of line now that I think about it.
- dontreact 10y ago
  Yes, it's very tricky to take an action as complex as the decision to cheat on emissions test and boil down numbers describing the damage done to society. Having numbers like "they killed 59 people" certainly makes it easier to stir enough outrage, but it is an oversimplification. Would you agree with something like "the emissions they caused by cheating killed people, and our current best estimate of that is 59 (+- the confidence interval in the paper)". It's true that the decision to cheat may have had the benefit of getting safer cars to more people, but that is even harder to measure than deaths caused by emissions. To me, the decision to cheat wouldn't be justifiable and I would still be outraged because they knowingly did harm to the environment and society with the hopes that there would be a nearly un-measurable side benefit: increasing the average safety of cars relative to a world that doesn't exist. To me, it seems despicably irresponsible to gamble the well being of many human lives against something so hard to measure.

Unklejoe 10y ago

> [If you agree that the environment is worse off and that the environment being worse off causes more deaths, then you agree that they caused deaths.] ...I should have said “worse off than if VW had not cheated and all other variables remained the same”. I don't question that more emissions results in more deaths. I question their ability to reliably calculate the number of deaths caused when dealing with numbers that small (~0.0003%) and leaving out some very important variables (which they do reference but don't incorporate). Of course, the publication was meant to compare the claimed vs actual emissions, but I'm talking in the more general case of overall emissions. The publication claims that: >> [Finally, we note that there may have been environmental benefits of the defeat device that we have not computed. For example, the reduced use of diesel exhaust fluid in selective catalytic reduction may potentially have also reduced ammonia slip, and therefore associated PM2.5 exposure and health impacts. In particular, Dedoussi and Barrett (2014) find that total PM2.5 exposure attributable to road transportation is caused approximately equally by NOx and NH3 emissions.] I can think of a few more inadvertent environmental/health benefits of the defeat device: For example, suppose that VW hadn't cheated and instead spent more time and money on R&D to come up with a legitimate solution. They would then have to pass this cost on to the consumers in the form of a more expensive car. Would all of the people who purchased diesel VW's still have purchased them if they were more money? If not, would they have purchased a car that produces worse emissions instead? The VW Jetta has a 5 star crash rating. Would all of the people who purchased other cars purchased cars with the same crash rating, or perhaps a car with a worse crash rating? There are a bunch of things that you could list that could all have an impact that rivals the 59 cited in the publication. We're talking about 59 out of 17,500,000 (number of annual deaths integrated over 2008-2015). It wouldn't take much to actually find out that the VW emissions scandal actually resulted in more lives saved. EDIT: That said, the publication isn't really trying to measure the amount of deaths caused by the scandal itself, but rather the increased emissions portion of it. Perhaps my original comment was out of line now that I think about it.

dontreact 10y ago
Yes, it's very tricky to take an action as complex as the decision to cheat on emissions test and boil down numbers describing the damage done to society. Having numbers like "they killed 59 people" certainly makes it easier to stir enough outrage, but it is an oversimplification. Would you agree with something like "the emissions they caused by cheating killed people, and our current best estimate of that is 59 (+- the confidence interval in the paper)". It's true that the decision to cheat may have had the benefit of getting safer cars to more people, but that is even harder to measure than deaths caused by emissions. To me, the decision to cheat wouldn't be justifiable and I would still be outraged because they knowingly did harm to the environment and society with the hopes that there would be a nearly un-measurable side benefit: increasing the average safety of cars relative to a world that doesn't exist. To me, it seems despicably irresponsible to gamble the well being of many human lives against something so hard to measure.