The First FBI Crime Report Issued Under Trump Is Missing A Ton Of Info
Clare Malone and Jeff Asher | FiveThirtyEight blog
October 27, 2017
This is a case of missing tables, rather than missing data. But it is a crucial loss because many reporters and state/local government folks just use the tables for reporting purposes. Imagine if the Census Bureau eliminated the tabular data from the Census/ACS. It would be a disaster.
The FBI portal added a new data tool, but it does not have the capability to produce the missing tables because it requires aggregation of many years. And, even for those that use raw microdata, the published tables provide a useful check on coding/weighting assumptions.
It is not unheard of for tables to be re-assessed and deleted, but this review process did not take place.
Changes to the UCR’s yearly report are not unheard of, and the press release that accompanies the 2016 report, which was published in late September, acknowledges the removal of some tables, saying that the UCR program had “streamlined the 2016 edition.” But changes to the report typically go through a body called the Advisory Policy Board (APB), which is responsible for managing and reviewing operational issues for a number of FBI programs. This time they did not.
Another reason to eliminate tables is for efficiency for the data provider. But:
“How much time and savings is there in moving an online table?” Nolan said. “These are canned programs: You create table 71 and table 71 is connected to a link in a blink of an eye.”
UPDATE:
Criminologists Are Asking Jeff Sessions To Release FBI Crime Data
Clare Malone | FiveThirtyEight blog
November 30, 2017
On Tuesday, a research alliance representing two professional associations of criminologists lodged a formal statement of concern with Attorney General Jeff Sessions and acting FBI Director Christopher Wray over a number of data tables that were missing from the FBI’s 2016 Crime in the United States report.
INTERESTING, IF DEPRESSING NUGGET:
The new head of the Bureau of Justice Statistics is a sabermatrician:
. . .Jeffrey Anderson, a former professor of political science with no apparent statistical background besides helping create a system to assess the strength of college football teams, adjusted for their schedule difficulty.