crime-data-explorer icon indicating copy to clipboard operation
crime-data-explorer copied to clipboard

Say when a state started reporting NIBRS data

Open cde-feedback-bot opened this issue 7 years ago • 3 comments

What brought you to the site and how can we improve your experience here?

General feedback about the site?

when reviewing Town of Madison PD https://crime-data-explorer.fr.cloud.gov/explorer/agency/WI0131000/aggravated-assault, I noticed the following sentence is very inaccurate and could lead to confusion/misuse of data without proper notes/disclaimers/context

The Town of Madison PD began submitting NIBRS data in 2013. The data description on this page suggests that data displayed in the tables below are from 2004-2014, when in fact the data is only from 2013-2014. This is something that is very concerning, considering the data is open to the public and could result in misuse/misunderstanding by media, etc.

Using this data for something cool? Tell us about it!

cde-feedback-bot avatar Aug 16 '17 13:08 cde-feedback-bot

@cacraig, @jeremiak: This is a good point. We could be clearer about the methodology we use for calculating NIBRS dimensions, like "victim's relationship to offender", and what time range that data is based on.

In regards to the latter, we currently say that the totals for our various dimensions are drawn from the time-range selected for the search, typically a 10-year range. But this isn't always accurate, as a state or agency could have reported NIBRS data for only a portion of the timeframe, leading to situations where the numbers represent only a year or two. Right now we have no way of warning users that the data may not be representative of a full 10 years.

Thoughts on how we can better handle this? Could we add a clarifying statement that explains what years the NIBRS data is based on (based on when the state or agency started reporting NIBRS), if it didn't report NIBRS for the full 10 years?

If there are challenges with "knowing" when an agency started reporting NIBRS, could we mitigate confusion in other ways? How about just adding a generic statement that warns users that NIBRS data may not be representative of the full 10 year timeframe?

cc: @harrisj, @RyanSibley

LarryBafundo avatar Aug 18 '17 20:08 LarryBafundo

I think one of the last things I added to cde_agencies was a field to represent when the "current" NIBRS run had started. This allows us to handle cases where agencies report NIBRS in 1991-1992 and then start up again in 2007. It doesn't handle cases where an agency is reporting off and on for the last 10 years (but not sure how often that happens or not). Maybe that would be helpful.

harrisj avatar Aug 21 '17 12:08 harrisj

I responded to this individual and created separate issues based on their feedback

LarryBafundo avatar Aug 29 '17 21:08 LarryBafundo