Pic taken from roadway.report, an interactive map of all known roadway deaths in the USA.

Fatalities resulting from motor vehicle crashes are the third largest cause of accidental deaths in the United States. On average for 2024 there were over 100 deaths per day on US roads.

you are viewing a single comment's thread
view the rest of the comments
[–] 24 points 3 months ago (2 children)
  • [–] 6 points 3 months ago* (last edited 3 months ago) (1 child)

    We should standardize on printing statistics like this in magenta superimposed on the population density printed in yellow—then per capita differences would show up as different hues (pinkish for high per-capita rates or orangish for low).

  • source
  • parent
  • hideshow 2 child comments
  • [–] 3 points 3 months ago

    I think for this map to be salvaged you need to fix the denominator and not use political boundaries for aggregation. Hexagons or raster because the entire point is to make it arbitrary. The formula would need to be something like (Auto Deaths per capita)/(Road Miles Traveled Per capita), or maybe (Auto Deaths per capita)/(Sum of length of roads per cell).

    Regardless, its still a population map because the denominator gets really small in rural areas, and the map still doesn't tell us anything about how safe or dangerous roads are.

  • source
  • parent
  • [–] [S] 3 points 3 months ago (1 child)
  • [–] 6 points 3 months ago* (1 child)

    #EvenThatMapEndsUpBeingAMapOfPopulation,ButMoreOfABoolean,BecausePeopleDon'tDieWhereThereAren'tPeopleAndItTakesTwoToTango

  • source
  • parent
  • hideshow 2 child comments
  • [–] [S] 5 points 3 months ago* (1 child)

    The red is higher rate of death per capita. You can die in a car by yourself in the middle of no where.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 3 points 3 months ago (1 child)

    #IThoughtWeWereHavingThisConversationInHastagsButTheIssueIsTheDenominator,andThatThereIsPracticallyNoOneInRuralAreas,SoAnyNumberOfDeathsOverAMuchSmallerNumBerAppearsBig.ItsStillAMapOfPopulation

  • source
  • parent
  • hideshow 2 child comments
  • [–] [S] 4 points 3 months ago (1 child)
  • [–] 1 point 3 months ago (1 child)

    #TheyShouldProbablyBeUsingAMoreStandardRasterGridAndThenDividingBySomeNumberThatAccountsForTheDistanceTravelledPerUnitRoadInAGivenCell

  • source
  • parent
  • hideshow 2 child comments
  • [–] [S] 2 points 3 months ago (1 child)
  • [–] 1 point 3 months ago (1 child)

    #ItsNotLikeWeDon'tKNowTheseAreCensusTracks,ItsThatCensusTractsAreNotAgreatWayToAggregateThisParticualarSpatialDataBecauseAgain,YouJustEndUpWithSomeVariantOfPopulationMap

  • source
  • parent
  • hideshow 2 child comments
  • [–] [S] 1 point 3 months ago (1 child)

    The most granular population density data in the USA can be found at the census tract level, which provides detailed demographic information for small geographic areas.

    Data is not magical, however much you wish it to be.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 3 months ago

    Data is not magical, however much you wish it to be.

    I've built maps estimating how much of your lawn you are watering at at 15cm pixel resolution. Techniques to count how many cars per driveway are parked at evening hours. How much water, per hour, being used by every address in a county.

    I have seen dashboards developed by collaborators who have access to near real time data regarding the physical positions of hundreds of millions of devices.

    Just because its the highest resolution data that you know about, doesn't mean its the highest resolution data which exists.

  • source
  • parent