Talk:Motor vehicle fatality rate in U.S. by year

2015 Vehicle Miles travelled
Number appears to be off by a factor of ten but I don't know where this data was sourced from. — Preceding unsigned comment added by 192.234.122.28 (talk) 18:18, 10 January 2017 (UTC)
 * It looks in the USA, there are various form for computing VMT:
 * «The reports are based on individual state reports on traffic data counts collected through permanent automatic traffic recorders on public roadways. Data on VMT for urbanized areas are available from the FHWA Highway Statistics Series.» https://www.transportation.gov/mission/health/vmt-capita
 * «The current practices used by the States to prepare these local area estimates vary significantly and often are not thoroughly documented. To gain an understanding of the various practices in use, FHWA conducted a survey of the States in April 2002 through its field division offices;» https://www.fhwa.dot.gov/policyinformation/hpms/statepractices.cfm
 * «TEstimates of the statewide total may involve several factors that account for current economics and social changes such as sales of vehicles by make and model, fuel taxes (gas and diesel), vehicle registrations, vehicle scrappage, population changes, and other inputs from selected use sector activities, to supplement statewide traffic counts.» https://www.fhwa.dot.gov/policyinformation/hpms/statepractices.cfm
 * Hope it answers your question. 81.185.253.94 (talk) 21:43, 23 February 2018 (UTC)
 * Nota: for the factor 10, it should be understood thet VMT migth be computed by 100 million distance traveled or by billion distance traveled (distance in miles or kilometer). 81.185.253.94 (talk) 21:43, 23 February 2018 (UTC)

Color indicators
Shouldn't the numbers that go down be marked with a red arrow ? Also, isn't it possible to compare to the number of actual drivers, rather than the population ? —Preceding unsigned comment added by 24.202.236.203 (talk) 23:04, 9 August 2010 (UTC)

Notable statistic
Of note is that in the two worst years on record, 1979 and 1980, the difference in the total number of deaths was only two.

Why is this of note? It seems completely irrelevant. For example: "Of note, the number of deaths in 1981 and 1993 both had the digit 8 in them." — Preceding unsigned comment added by 98.224.229.100 (talk) 14:21, 13 September 2011 (UTC)

Source
Where is the source of the chart data? None of the links on the page direct you to the origins of that information. — Preceding unsigned comment added by 75.80.222.152 (talk) 06:42, 8 January 2013 (UTC)

Minor differences between NHTSA and CDC
I'll just note that for 2014 the CDC has a slightly different morbidity count than the NHTSA. Presumably this is due to different reporting/counting methods. See for instance: from the CDC which states that 33,954 people died due to vehicular injuries, as opposed to the reported 32,675 people by the NHTSA. It's probably worth figuring out the difference, because the NHTSA only reports on vehicle injuries, whereas the CDC reports on all. On the other hand, the CDC is slower to get their numbers out.--Pigdog234 (talk) 09:12, 8 October 2017 (UTC)

Vandalism in table
I'm not sure what the correct labels are (looks like the vandalism goes back many edits) but there are table labels reading "toad stool" and "mario died" that might be some attempt at a Mario Kart-related joke, perhaps? 108.199.119.137 (talk) 02:49, 8 March 2013 (UTC)

"Pro rated" table entries
I'm tempted to remove the pro-rated data in the table for 2012. It shows a massive 4% increase in fatalities based on only 9 months of data. First off, there is no source for the pro-rated values. Second, in the USA the most dangerous time to drive is during Jan-April when weather is the worst due to winter storms, and during summer when vehicle travel peaks as people vacation. So of course we will see the majority of the fatalities during the first 9 months for both those reasons. Thus extrapolating that rate out for the remainder of the year is very inaccurate. --Dan East (talk) 21:34, 8 August 2013 (UTC)

Source used for population?
The values in the "Fatalities per 100,000 Population" column do not match those reported by the NHTSA. What source is being used for the values in the "Population" column? Is end of year population, beginning of year population, or yearly average population being used? The correct population to use is the yearly average population. --JHP (talk) 22:52, 22 May 2016 (UTC)

Intentional death
Is there anyways to add statistics for the number of intentional deaths (i.e. something like vehicular homicide where a person is trying to kill someone with a car) vs. those cause by a car accident? 67.253.250.154 (talk) 23:23, 10 January 2017 (UTC)


 * But what is a Vehicular_homicide exactly? Does it match the UCR? and where would such data be found? 81.185.253.94 (talk) 21:11, 23 February 2018 (UTC)

Comparisons with other countries: per capita or per VMT?
The argument that the US fatality rate hasn't fallen as fast as the EU is based on the choice to compare deaths per 100,000 people. This makes sense in the context of all causes of death: heart disease, cancer, car crash. Americans drive more than other countries, and so car crashes play a greater role in their mortality. But if you are asking "how safe is it to drive in _____?" it has more to to with deaths per billion km or per VMT. Then you get much different results. France is much worse than the US. The UK and the US are about the same..Automobile safety and Epidemiology of motor vehicle collisions are probably better articles to make these comparisons and argue various points of view and various opinions. --Dennis Bratland (talk) 21:50, 22 February 2018 (UTC)


 * You removed content related to this.
 * I do not consider that one argument should override one other one. In Europe fatalities are counted as integer number of people, while it looks like in the USA they are divided by VMT. I do not say that one is good and the other is bad. According to the wikipedian concept of neutrality point of view (NPOV), I assume both views should be displayed.
 * I am not so convinced by fatalities per VMT because I do not know how (accurately) VMT are computed and/or estimated, I am not convinced there is any relationship between fatalities and VMT (who has any clue?), I am not convinced fatalities and VMT might come from same class of road users (drunk+nigth+pedestrian+cyclists+speed+...) and same class of road condition (city+rural+speed+night+states+...), I am not convinced fatalities per VMT takes into account the speed which would impact the fatalities per driving hours. But anyway, the fact is that it is a computed, published, and famous statistic make it as valuable as other statistics, with the only limit of its meaning... Anyway I would be happy to find in Wikipedia any data which might change my mind.
 * I do not defend the unreliable fake news providers who say that the more VMT, the more cars; the more cars, the more traffic jams; the more traffic jams, the lower speed; the lower speed, the lower accident gravity/fatality.
 * For the comparison of VMT between France and the USA, List_of_countries_by_traffic-related_death_rate OECD/ITF, ed. (12 October 2015) suggests that fatalities by VMT would be 7.1 fatalities by bv/km in the USA, and 5.8 fatalities by bv/km in France. The same source suggests that fatalities by VMT would be 7.1 fatalities by bv/km in the USA, and 3.6 fatalities by bv/km in the UK. A native speaker might had used the word twice to compare the US and the UK . This might make France and the US closer to each other than with the UK, at least that year.
 * Anyway to have a full picture, which does not focus to a specific year rather than another, I believe a graph is better as it might show both year 2005 data and year 2015 data. But many decreasing VMT graph have an issue: they sometimes/often do not reflect a decrease of fatalities, but an increase of VMT, which might lead to misunderstanding and fake/wrong interpretation.


 * I also understood that the title of a page should be the more generic possible and the less specif. For this reason, I assume the title of the page should be Motor vehicle fatality rate in U.S. by year rather than Motor vehicle fatality rate in U.S. by year.
 * Wikipedia is a global topic. So, I suggest the details that pertains to the USA remains in the USA pages, even if the parts of the USA data of global interest might find its place for a more global article. 81.185.253.94 (talk) 20:01, 23 February 2018 (UTC)

And might be fatalities are not linked to VMT
It is sometimes understood that fatalities increase with the increase of VMT, but this is not systematic as fatalities might remain quite stable while the VMT change.


 * The addition you want here is not neutral. It's extremely argumentative and has an agenda. It's fine to have points of view on Wikipedia, but they have to be framed appropriately. You're presenting opinions as facts, and placing them in the wrong context.It appears there is a discrepancy between the Wikipedia article's statistics and the New York Times reporting of data from the OECD and the International Traffic Safety Data and Analysis Group's data. Since it's behind a paywall, I can't verify whether or not that article has a mistake or not. Anyway, we don't cite Wikipedia as a source. From what I can verify, France and Japan had far more dangerous driving conditions from 1970 until 2005, when they caught up with the US. Since 2005 those countries might have done better.We do know that Americans drive at least twice as much as these other countries, and that they drive more in rural conditions. Arbitrarily comparing all 50 US states with a small and homogeneous place like France or Japan or the UK is misleading. The US has many states that are not "rich", and are as large as many countries in Europe. The data in this set isn't even the whole EU: it's only 19 countries. It excludes Greece, and Bulgaria and Romania, Latvia, Estonia. Why are these countries excluded, but we don't exclude the poorest US states, like Mississippi, South Carolina, or Kentucky? Alaska has a huge land area and a minuscule population who drive incredible distances in rural areas. Driving from Anchorage to Fairbanks is comparable to driving the entire length of France.This only goes to show that this is a complex question. The elaborate charts you've posted are evidence of how complex it all is. An analysis on this level of depth is in no way appropriate for an article which is is simply about fatality rates in the US. This article is not a comparison between countries. You found one article where it is: List of countries by traffic-related death rate. I pointed out that Epidemiology of motor vehicle collisions is also an appropriate place for this kind of analysis and comparison. --Dennis Bratland (talk) 22:35, 23 February 2018 (UTC)


 * Why would the addition be non neutral?
 * I agree we should not mix opinions and facts. Fatalities are facts, the believing of a relationship between fatalities and VMT might be an opinion. VMT are published estimates.
 * I agree we should not exclude countries because this might introduce some bias, but only available data are comparable: In eurostat tables, it looks like VMT are not available for every countries for instance.
 * Some piece of answers for your interrogations are available about [Is-Vehicle-Miles-Traveled-VMT-Even-the-Proper-Metric-to-Determine-Traffic-Fatality-Rates] about [international_road_safety_comparisons] and [International_2015_Tables] and for Alaska [state-by-state-overview].
 * To answer the main point («An analysis on this level of depth is in no way appropriate for an article which is is simply about fatality rates in the US.»)
 * Why not illustrate the increase/decrease, with each rate?
 * raw data are useless while a graphic give a fast and comprehensive (over)view
 * a graphic might be generated for each kind of rate (per capita, per VMT, and so on).
 * Why rates should be expressed in fatalities by VMT while it is so complex? — Preceding unsigned comment added by 81.185.253.94 (talk) 12:33, 24 February 2018 (UTC)  anyway, This same data can be displayed n three different ways as here after:


 * You keep citing opinions in an ongoing debate, and that proves my point. Why would anyone write an article called "Is the VMT Even the Proper Metric to Determine Traffic Fatality Rates?" unless the topic is controversial? It's not settled. Experts. Do. Not. Agree. It makes sense to describe these differing opinions in this controversy in an appropriate article. Not here. --Dennis Bratland (talk) 17:54, 24 February 2018 (UTC)


 * I was wondering why «Americans drive at least twice as much as these other countries». In a first time I imagined that US might have a cheaper gasoline which would make travel inexpensive, but in the same time, there are more people in the EU (EU-27)... The only more or less comparable figure I found was the U.S. Passenger-Miles (Millions) and the Transport of passengers by mode in EU-27 (billion pkm) for the same year, the year 2006: 4,678 billion pkm in EU-27 according to the EC; and 3,235,752 millions miles in the US, according to the DoT. . This makes around 5200 billion pkm for the US that is about a 10% difference. But I am wondering if the fatalities are more dependent of the number of passenger traveling, or of the number of cars...
 * For car occupancy, there are around 1.5 occupant(s) by car in Europe, and 1.7 people per car in the US , again this makes 10%.
 * Thus, I do not understand how car «Americans drive at least twice as much as these other countries»? 81.185.253.94 (talk) 00:25, 27 February 2018 (UTC)
 * Nonetheless, some documents published by Statistique Canada have a rating for the accuracy of the published statistics (ranking from A excelent to E not for publication). Neither passenger-kilometers nor gasoline usages are ranked better than vehicle traveled kilometers (Most of them are ranked around B. A représente un c.v. situé entre 0,00 % et 4,99 % et signifie Excellent, B représente un c.v. situé entre 5,00 % et 9,99 % et signifie Très bon.
 * An English-British document suggest the complexity of the measurement of miles traveled, in the UK:
 * In 2014, in the UK, Cars, vans and taxis represent 83% of all passenger distance traveled; 64% of personal trips were made by car; 78% of personal yearly distance traveled was made by car. 37% of VMT were urban (>10000 people); 43% rural; and 21% on motorways. Cars represent 79% of VMT, while LGVs represent 14% of VMT.
 * Between 1970 and now, traveled distance (all modes) increased while travel time (all modes) did not increase so much... Thus I am wondering why the safety should be compared to the traveled distance rather than the traveled time...
 * Main VMT increases occurred on motorways and A roads... This might question if the increase in VMT occured on the safest roads.
 * There is also a social effect measurable with household income: First quintile travel 7,208 miles per person per year against 2825 for last quintile.
 * In UK, most of travel are short: 55 or 56% of travels are less than 5 miles, between 72% and 100% of the trips are less than 50 miles.
 * In Canada, 89 % of Canadians travel less than 60 kilomètres (go & back); 36 % less than 5 kilomètres.
 * Thus, I might agree that those set of data show it is not as simple as main published data suggest: you might be right to say it is a complex question. Provided data might be added to Transport_in_the_United_Kingdom 77.199.96.191 (talk) 00:04, 2 March 2018 (UTC)