as of 6 am  June, 26 2010

OVERVIEW

This document can be downloaded from http://3rdchimp.com/files/June2010Summary.pdf.

Lincoln County (LC) maintains a website.  Among other things it provides a roster of people held in the LC jail (http://lto.locktrack.com/public/reports/lincol_or_ilist.html).  The jail typically houses about 150 people on any given day.

The listing provides descriptions of each inmate, the crime(s) with which they've been charged, and the dates, when available, of when they were lodged in the jail, date of next court appearance and when they will be released.  Court appearance and release dates not always populated.

Each inmate has a unique identifying ID number (SID) as well as, occasionally, an FBI ID number, a “TIP” ID number and a booking number.  The SID is used to connect booking records with personal descriptions and the current list of charges against the inmate.  The particulars of the charges are not static and change over time.

The personal description of each inmate includes a mug shot along with the person's first name, last name, middle initial,  race, sex, height, weight, hair color and eye color.  The races currently used in these profiles are HISPANIC, WHITE, BLACK, ASIAN, UNKNOWN, and AMERIND.There are one or more charges for each inmate.  The information tracked in the charges table are the docket number, a statute ID (often, but not always, derived from Oregon legal codes.  i.e. 811.140 [reckless driving] would appear as 811140]), and a brief description of the charge in English (e.g. DRIVE UNDER INFLUENCE INTOX).  Each charge may appear multiple times for each inmate.  The inmate's bail status is also listed.  That may be a dollar amount, or one of “not set” or “not allowed,” depending on court proceedings.

A simple sampling script that takes a snapshot of the roster is run on a computer four times per day (midnight, 6am, noon, and 6pm).  This simply captures the HTML document offered on the public web page.  An analysis script is run later for each snapshot which parses the HTML and extracts the data described above and places it in a group of data base tables for later analysis.  Monitoring began February 14, 2010.

SUMMARY

As of today, June 26, 2010, there have been 981 different inmates and there are 143 currently present.  The summary is based on 72,587 records from 470 samples.  A total of 185 different charges appear in the records.  By race, the overall record shows 45 UNKNOWNs, 56 HISPANICs, 47 AMERINDs, 14 BLACKs, 10 ASIANs and 809 WHITEs.

The first figure shows the LC jail population distribution by race over the monitoring period.  This shows the variation of raw numbers of overall per-race populations in the jail over the monitoring period.  Compared to the previous month there has been an increase in the non-white population, especially HISPANIC, of the jail since the last summary.  Overall occupation for  June 26, 1020, 143, is on the low side.

The next graph shows the same data normalized to a percentage to better show the per-race distribution of the jail population over time. As mentioned above, the proportion of the non-white jail population is up, near the highest level during the observation period.

As of June 26, the distribution is as follows:

• UNKOWN 6( 2%)
• HISPANIC 10( 7%)
• AMERIND 5 ( 4%)
• BLACK 4( 3%)
• ASIAN 0( 0%)
• WHITE118(84%) 

There are 41 different charges against the HISPANICs. The single most common charge among them is INS HOLD (15.13% of all charges against all HISPANICs1). The only other races (as identified by LC ) showing INS HOLD is UNKNOWN (approx 2.6%), and WHITE (approx 0.06%). The general distribution of charges (normalized to HISPANICs) is shown in the following figure. For non-HISPANIC races the charges not shared with HISPANICS are grouped under “other.”

The numbers shown in the “heat map” graph below are calculated by taking the sum of all occurrences of the charge among all inmates of that race divided by the total number of all charges against inmates of that race multiplied by 100 to yield something like a percent charge load by race.

The HISPANIC charge distribution is distinct in that it represents a subset of charges laid against the population as a whole; the “other” charge category comprises 25% or greater than 30% among all other races. For example, of the 185 different charges brought against 981 different people, CITY ORDINANCE OFFENCE (whatever that is) only appears to have been used against one HISPANIC. It should be kept in mind that data is sparse at this point and unique events don't necessarily represent a trend or tendency and any event will show an unusually high percent charge load as a result of the smallness of the sample. A single charge against a single HISPANIC represents a 0.84% event while a single charge against a single WHITE will be a 0.09% event. A better metric is needed for meaningful evaluation this data.

1Hispanic charge distribution is in spreadsheet http://3rdchimp.com/files/June2010HispanicChargeDistribution.xls