Race Data incomplete.

[cswroe]

Is there an issue with the dataset? It seems that there is a tremendous change in the data starting in 2021 with race not being reported.

There was a slight trend upward in previous data, but it has nearly made the data unusable for analytics.

[jungle-boogie]

I agree, there is a huge uptick in missing race data.

There's been 368 reported shootings in 2022, and yet, 325 don't have a race.

Perhaps this column should just be dropped from the record, if the Washington Post can't maintain accurate information.

[cswroe]

The dataset should remain as is, but there needs to be some explanation for this. It is far outside the realm of a statistical anomaly and is very consistent since 2021. The Post in their own description of the dataset advertised race first, "In 2015, The Post began tracking more than a dozen details about each killing — including the race of the deceased, the circumstances of the shooting, whether the person was armed and whether the person....."

There are many institutions that utilize this data, and it should be complete and has historically been complete within a reasonable amount of exclusion and/or error.

[jungle-boogie]

I'd also like the data to remain, and be back filled with the missing race info.

But as you point out, downstream organizations depend on this data. I'm certain they would also want the data to be accurate and complete.

[bwolther]

Whereas race was known in 88 percent or more of the cases in years up to 2020, it is now included in only 11 percent of cases so far this year. Race, as a data point in the WaPo shootings database, has gone from being informative to being indistinguishable from noise. (My analysis includes data through 05/22/2022.)

As is common for people to make quantitative claims about shootings, vis a vis race, this data point has been invaluable in helping to characterize and quantify those dynamics. The loss of this data point substantially degrades the informational content of the database, particularly with respect to race. (This change is surprising to me, and unexplained, as far as I know.)

What was the mechanism by which race was previously determined? Why was that mechanism terminated?

[jungle-boogie]

I see there has been a little progress for the 2021 year. I hope this continues and the data becomes filled in.

[MykeMorbius]

There should be a specific count for unarmed people shot since that's what people care about the most, by far. I did my own research using data from 'Mapping Police Violence' via US News and World Report (mainstream media) and came up with an annual average 2013-2019 of 'unarmed blacks' number of just 7. And people think it's hundreds or even thousands, thanx to the MSM.