Data Relating to Criminal Justice in the United States

Data and Models for Decision Making

This book is concerned with making decisions about the design and operation of criminal justice systems in the United States. Models which are representations of the respective systems are used as an aid in this process. To build these models, one must have appropriate data

O'

What is the sequence of events in the criminal justice system?

Flowchart of process within the criminal justice system (retrieved on April 7, 2020, from the Office of Justice Programs, Bureau of Justice Statistics, at https://www.bjs.gov/content/largechart.cfm)

Figure 1.1 Flowchart of process within the criminal justice system (retrieved on April 7, 2020, from the Office of Justice Programs, Bureau of Justice Statistics, at https://www.bjs.gov/content/largechart.cfm).

Multiple Objective Analyticsor information about the relevant system(s). Data can be of one of two types: qualitative or quantitative in nature. Examples of qualitative data would be things like the processes followed when a person is arrested for an alleged crime. Examples of quantitative (or numerical) data would be the percentage of defendants of a socioeconomic class who appear for their court date after being released on bail.

One important aspect of data used for building a model for decision making is categorization of that data. For example, it has been shown that there is a strong relationship between age and crime. Hence, to build a model which will test the effects of a new' law related to some type(s) of crime, demographic information related to age and crime is important to track.

In addition to using data for the construction of models, data related to performance can be useful in tracking the effectiveness of a new law/policy.

Data collected and used regarding criminal justice can be categorized according to the various functions shown in Figure 1.1. Following from left to right in Figure 1.1, data can be categorized according to that related to (1) crime and arrests; (2) pretrial and trial services; and (3) jail, prison, probation, and parole. Much of the data collection on criminal justice systems in the United States is accomplished through the Bureau of Justice Statistics of the Department of Justice. As discussed below, data is gathered through the many local agencies throughout the country, and then transmitted to the Bureau of Justice Statistics for compilation at the national level.

The Bureau of Justice Statistics web page on “all data collections” (All Data Collections, n.d.) lists the following categories of relevant information: Corrections, Courts, Crime Type, Criminal Justice Data Improvement Program, Employment and Expenditure, Federal, Indian Country Justice Statistics, Law Enforcement, and Victims. Each of these major categories is divided into subcategories. For example, Corrections is divided into the subcategories of total correctional population, local jail inmates, jail facilities, etc. Finally, within each of these subcategories, there is information on (1) data collection and surveys and (2) publications and projects (arising from the data collection and surveys).

The number of publications and projects arising from each of the subcategories is indeed large. In the following paragraphs, we will discuss just a few of the most important of these, within the areas of (1) crime and arrests, (2) courts, and (3) corrections. The reader is referred to the Bureau of Justice Statistics web page for the actual reports.

The United Crime Reporting Program, the Summary Reporting System, and Crime Rates

First, we will consider crime and arrests. The United Crime Reporting (UCR) Program (more recently called the Summary Reporting System) was started by the FBI in 1930. The Program collects data on the numbers of crime, categorizing them under two broad categories: Violent Crimes and Property Crimes. Approximately 17,000 city, county, state, and federal law enforcement agencies (out of about 18,000 total agencies) participate in the program on a voluntary basis, thereby covering the vast majority of the country (pages 35 and 36 in Rennison ad Dodge, 2018). The violent crimes are divided according to murder, forcible rape, robbery, and aggravated assault. The property crimes are divided according to burglary, larceny-theft, and vehicle-theft. These two crime rates are added to give a total crime rate for each year (United States Population and Rate of Crime per 100,000 People 1960-2018, n.d.). Table 1.1 shows the total crime rate, violent crime rate, property crime rate, and other crime rates (per 100,000 population) in the United States from 1960 to 2018.

One of the things to note from Table 1.1 is that crime rates from each category are added without any weighting. That is, the rates for murder, forcible rape, robbery, and aggravated assault are just added together to give an overall rate for violent crime. Similarly, the rates for burglary, larceny-theft, and vehicle theft are added to give the rate for property crime. Finally, the rates for violent crime and property crime are added to give the total crime rate.

A more accurate picture of crime would be given if the overall rates were weighted by the type of crime. For example, clearly, a murder is more serious than a larceny-theft. One way that these rates could be weighted would be by their costs. The RAND Corporation developed a cost of crime calculator that can be used for this purpose (Cost of Crime Calculator, n.d.). This calculator provides an estimated cost for each type of crime as shown in Table 1.1, with appropriate discounting for the year of the crime. For example, the costs associated with each type of crime for the year 2000 are given by:

TABLE 1.1

Crime Rates (Number of Crimes per 100,000 Population in the United States) from 1960 to 2018

Year

Population

Total

Crime

Rate

Violent Crime Rate

Property Crime Rate

Murder

Forcible-Rape

Robbery

Aggravated Assault

Burglary

Larceny-

Theft

Vehicle-

Theft

1960

179,323,175

1,887.2

160.9

1,726.3

5.1

9.6

60.1

86.1

508.6

1,034.7

183.0

1961

182,992,000

1,906.1

158.1

1,747.9

4.8

9.4

58.3

85.7

518.9

1,045.4

183.6

1962

185,771,000

2,019.8

162.3

1,857.5

4.6

9.4

59.7

88.6

535.2

1,124.8

197.4

1963

188,483,000

2,180.3

168.2

2,012.1

4.6

9.4

61.8

92.4

576.4

1,219.1

216.6

1964

191,141,000

2,388.1

190.6

2,197.5

4.9

11.2

68.2

106.2

634.7

1,315.5

247.4

1965

193,526,000

2,449.0

200.2

2,248.8

5.1

12.1

71.7

111.3

662.7

1,329.30

256.8

1966

195,576,000

2,670.8

220.0

2,450.9

5.6

13.2

80.8

120.3

721.0

1,442.9

286.9

1967

197,457,000

2,989.7

253.2

2,736.5

6.2

14.0

102.8

130.2

826.6

1,575.8

334.1

1968

199,399,000

3,370.2

298.4

3,071.8

6.9

15.9

131.8

143.8

932.3

1,746.6

393.0

1969

201,385,000

3,680.0

328.7

3,351.3

7.3

18.5

148.4

154.5

984.1

1,930.9

436.2

1970

203,235,298

3,984.5

363.5

3,621.0

7.9

18.7

172.1

164.8

1,084.9

2,079.3

456.8

1971

206,212,000

4,164.7

396.0

3,768.8

8.6

20.5

188.0

178.8

1,163.5

2,145.5

459.8

1972

208,230,000

3,961.4

401.0

3,560.4

9.0

22.5

180.7

188.8

1,140.8

1,993.6

426.1

1973

209,851,000

4,154.4

417.4

3,737.0

9.4

24.5

183.1

200.5

1,222.5

2,071.9

442.6

1974

211,392,000

4,850.4

461.1

4,389.3

9.8

26.2

209.3

215.8

1,437.7

2,489.5

462.2

1975

213,124,000

5,298.5

487.8

4,810.7

9.6

26.3

220.8

231.1

1,532.1

2,804.8

473.7

(Continued)

Criminal Justice Systems in the United States

M3

TABLE 1.1 (CONTINUED)

Crime Rates (Number of Crimes per 100,000 Population in the United States) from 1960 to 2018

Total

Crime

Violent Crime

Property Crime

Forcible-

Aggravated

Larceny-

Vehicle-

Year

Population

Rate

Rate

Rate

Murder

Rape

Robbery

Assault

Burglary

Theft

Theft

1976

214,659,000

5,287.3

467.8

4,819.5

8.7

26.6

199.3

233.2

1,448.2

2,921.3

450.0

1977

216,332,000

5,077.6

475.9

4,601.7

8.8

29.4

190.7

247.0

1,419.8

2,729.9

451.9

1978

218,059,000

5,140.4

497.8

4,642.5

9.0

31.0

195.8

262.1

1,434.6

2,747.4

460.5

1979

220,099,000

5,565.5

548.9

5,016.6

9.8

34.7

218.4

286.0

1,511.9

2,999.1

505.6

1980

225,349,264

5,950.0

596.6

5,353.3

10.2

36.8

251.1

298.5

1,684.1

3,167.0

502.2

1981

229,146,000

5,858.2

594.3

5,263.8

9.8

36.0

258.7

289.7

1,649.5

3,139.7

474.7

1982

231,534,000

5,603.7

571.1

5,032.5

9.1

34.0

238.9

289.1

1,488.8

3,084.9

458.9

1983

233,981,000

5,175.0

537.7

4,637.3

8.3

33.7

216.5

279.2

1,337.7

2,869.0

430.8

1984

236,158,000

5,031.3

539.2

4,492.1

7.9

35.7

205.4

290.2

1,263.7

2,791.3

437.1

1985

238,740,000

5,207.1

556.6

4,650.5

8.0

37.1

208.5

302.9

1,287.3

2,901.2

462.0

1986

240,132,887

5,480.4

620.1

4,881.8

8.6

38.1

226.0

347.4

1,349.8

3,022.1

509.8

1987

242,288,918

5,550.0

609.7

4,940.3

8.3

37.4

212.7

351.3

1,329.60

3,081.3

529.5

1988

245,807,000

5,664.2

637.2

5,027.1

8.4

37.6

220.9

370.2

1,309.2

3,134.9

582.9

1989

248,239,000

5,741.0

663.1

5,077.9

8.7

38.1

233.0

383.4

1,276.3

3,171.3

630.4

1990

248,709,873

5,820.3

731.8

5,088.5

9.4

41.2

257.0

424.1

1,235.9

3,194.8

657.8

Multiple Objective Analytics

(Continued)

TABLE 1.1 (CONTINUED)

Crime Rates (Number of Crimes per 100,000 Population in the United States) from 1960 to 2018

Total

Crime

Violent Crime

Property Crime

Forcible-

Aggravated

Larceny-

Vehicle-

Year

Population

Rate

Rate

Rate

Murder

Rape

Robbery

Assault

Burglary

Theft

Theft

1991

252,177,000

5,897.8

758.1

5,139.7

9.8

42.3

272.7

433.3

1,252.0

3,228.8

658.9

1992

255,082,000

5,660.2

757.5

4,902.7

9.3

42.8

263.6

441.8

1,168.2

3,103.0

631.5

1993

257,908,000

5,484.4

746.8

4,737.7

9.5

41.1

255.9

440.3

1,099.2

3,032.4

606.1

1994

260,341,000

5,373.5

713.6

4,660.0

9.0

39.3

237.7

427.6

1,042.0

3,026.7

591.3

1995

262,755,000

5,274.9

684.5

4,591.3

8.2

37.1

220.9

418.3

987.1

3,043.8

560.4

1996

265,284,000

5,087.6

636.6

4,451.0

7.4

36.3

201.9

390.9

945.0

2,980.3

525.7

1997

267,637,000

4,927.3

611.0

4,316.3

6.8

35.9

186.1

382.1

919.6

2,891.8

505.7

1998

270,296,000

4,615.5

566.4

4,049.1

6.3

34.4

165.2

360.5

862.0

2,728.1

459.0

1999

272,690,813

4,266.5

523

3,743.6

5.7

32.8

150.1

334.3

770.4

2,550.7

422.5

2000

281,421,906

4,124.80

506.5

3,618.3

5.5

32

145

324.0

728.8

2,477.30

412.2

2001

285,317,559

4,162.60

504.5

3,658.1

5.6

31.8

148.5

318.6

741.8

2,485.7

430.5

2002

287,973,924

4,125.00

494.4

3,630.6

5.6

33.1

146.1

309.5

747.0

2,450.7

432.9

2003

290,690,788

4,067.00

475.8

3,591.20

5.7

32.3

142.5

295.4

741.0

2,416.50

433.7

2004

293,656,842

3,977.30

463.2

3,514.1

5.5

32.4

136.7

288.6

730.3

2,362.3

421.5

2005

296,507,061

3,900.50

469.0

3,431.5

5.6

31.8

140.8

290.8

726.9

2,287.8

416.8

2006

299,398,484

3,808.10

473.6

3,334.5

5.7

30.9

149.4

287.5

729.4

2,206.8

398.4

Criminal Justice Systems in the United

(Continued)

TABLE 1.1 (CONTINUED)

Crime Rates (Number of Crimes per 100,000 Population in the United States) from 1960 to 2018

Total

Crime

Violent Crime

Property Crime

Forcible-

Aggravated

Larceny-

Vehicle-

Year

Population

Rate

Rate

Rate

Murder

Rape

Robbery

Assault

Burglary

Theft

Theft

2007

301,621,157

3,730.40

466.9

3,263.5

5.6

30

147.6

283.8

722.5

2,177.8

363.3

2008

304,374.846

3,669.00

457.5

3,211.5

5.4

29.7

145.7

276.7

732.1

2,167.0

314.7

2009

307,006,550

3,465.50

431.9

3,036.1

5.0

29.1

133.1

264.7

717.7

2,064.5

259.2

2010

309,330,219

3,350.40

404.5

2,945.90

4.8

27.7

119.3

252.8

701

2,005.80

239.1

2011

311,587,816

3,292.50

387.1

2,905.40

4.7

27

113.9

241.5

701.3

1,974.10

230

2012

313,873,685

3,255.80

387.8

2,868.00

4.7

27.1

113.1

242.8

672.2

1,965.40

230.4

2013

316,497,531

3,112.40

379.1

2,733.30

4.5

25.9

109

229.6

610.4

1,901.60

221.3

2014

318,907,401

2,946.10

372

2,574.10

4.4

26.6

101.3

229.2

537.2

1,821.50

215.4

2015

320,896,618

2,885.10

384.6

2,500.50

4.9

28.4

102.2

238.1

494.7

1,783.60

222.2

2016

323,405,935

2,849.10

397.5

2,451.60

5.4

40.9

102.9

248.3

468.9

1,745.40

237.3

2017

325,147,121

2,757.80

394.9

2,362.90

5.3

41.7

98.6

249.2

429.7

1,695.50

237.7

2018

327,167,434

2,580.10

380.6

2,199.50

5

42.6

86.2

246.8

376

1,594.60

228.9

Multiple Objective Analytics

Murder: $7.3 million. Rape: $185,558, Robbery: $57,300, Aggravated Assault: $74,301, Burglary: $11,154, Larceny-Theft: $1,821, and Motor Vehicle Theft: $7,732.

Admittedly, the actual cost for these crimes would vary greatly from one case to another. However, in terms of obtaining a more accurate view of crime over time, it would seem logical to use these cost values as multipliers for the rates shown in Table 1.1. For example, if one used these cost values in a computation for the violent crime rates, property crime rates, and overall crime rates for the years 1996 through 2001, their values would be (given in units of cost in thousands of dollars per 100,000 people):

Violent crime rates (1996 through 2001): $101369, $95355, $88624, $81136. $78469, $78962.

Property crime rates (1996 through 2001): $20032, $19433, $18131, $16504. $15827, $16129.

Total crime rates (1996 through 2001): $121401, $114788, $106756, $97640. $94297, $95091. "

Note that, as would be expected with this approach, the overall crime rates are more heavily weighted by the violent crimes. In addition, the decrease in the total crime rate from 1996 to 2001 is computed as 21.7%, as opposed to the value of 18.1% from Table 1.1 in which figures are not weighted by the cost of crime.

Data on the Solving of Crimes

An FBI report (FBI: UCR. 2017 Crime in the United States, n.d.) provided data on the percentages of various types of crimes that were solved by the police in 2017:

Murder and Nonnegligent Manslaughter:

61.6%,

Rape:

34.5%,

Robbery:

29.7%,

Aggravated Assault:

53.3%,

Burglary:

13.5%,

Larceny-theft:

19.2%,

Motor Vehicle Theft:

13.7%.

In general, one could say that these figures do not indicate particularly good results; however, at least the higher percentages are associated with what are considered the most important crimes (i.e., the violent crimes as indicated in Table 1.1: murder and nonnegligent manslaughter, rape, robbery, and aggravated assault). Lopez (2018) notes that there is much variation in the solved percentage for murder and nonnegligent homicide according to locality and race. He notes that black victims accounted for the majority of homicides in 52 of the largest cities in the United States over the previous decade, and that they were the least likely of any racial group to have their murders result in an arrest. In particular, 63% of white victims had their homicide cases solved, as compared to only 47% of black victims in the 52-city study.

One possible reason for poor performance in solving cases is the way police are utilized—that is with more of an emphasis in preventing crimes rather than solving crimes (Lopez, 2018). This issue of how police are utilized is discussed in more detail in Chapter 4, in the section on the Violent Crime Control and Law Enforcement Act of 1994.

The National Incidence-Based Reporting System, the Law Enforcement Officers Killed and Assaulted Program, and the Hate Crime Statistics Program

As noted in 1.5.2, the Summary Reporting System only gives counts of the number of crimes of each type by jurisdiction. Hence, in recent years more detailed reporting systems have been developed, providing details about the crimes committed. These detailed reporting systems, produced on a semiannual basis, are of three types, namely: (1) reports from the National Incident-Based Reporting System (NIBRS), (2) reports from the Law Enforcement Officers Killed and Assaulted (LEOKA) Program, and (3) reports from the Hate Crime Statistics Program.

The NIBRS evolved from the UCR Program by allowing for greater detail in the data collected for each incident of crime. By allowing for this greater detail, the NIBRS provides for the association of the attributes of the crime, victim information, and offender information with a single offense. The NIBRS groups offenses into two categories: Group A (containing 23 different crimes covering 52 offenses) and Group B (covering 10 offenses). An example of a Group A crime is fraud which would cover the offenses of ATM fraud, impersonation,

wire fraud, identity theft, and hacking/computer invasion (Page 41, Renison and Dodge, 2018). In 2018, only 44% of the agencies that took part in the UCR Program participated in the NIBRS Program; however, thousands of additional agencies have made commitments to participate by 2021.

The expressed purpose of the LEOKA Program is to collect data and then thoroughly analyze this data to provide training to law enforcement to keep the relevant personnel safe. The focus is on why an incident (in which a law enforcement official was injured or killed) occurred so that an appropriate analysis is accomplished. This analysis is supposed to result in training for law enforcement to mitigate the problem (Law Enforcement Officers Killed and Assaulted, 2019).

In 1990, the US Congress passed the Hate Crime Statistics Act, which resulted in the Hate Crime Statistics Program. This program, under the UCR umbrella, provided for the collection of hate crime data under several different bias motivation categories, including race/ethnicity/ancestry bias, religious bias, sexual orientation bias, gender identity bias, disability bias, and gender bias. Data collected in 2018 indicated that approximately 60% of the hate crimes committed in 2018 fell into the category of race/ethnicity/ancestry bias (Oudekerk, 2019).

Another category of UCR Program reporting is the Supplementary Homicides Reports (SHR). As the name indicates, these reports provide specific data on homicides committed in the United States. In addition to the “numbers data” provided by the UCR reports, the SHR system provides data on specific characteristics of both the offender and the victim, such as age, sex, race, ethnicity along with the relationship between the two; the type of weapon used; and the circumstances surrounding the incident (The Nation’s Two Measures of Homicide, 2014).

The above programs and reports have as their source the law enforcement agencies of the United States. The National Crime Victimization Survey (NCVS) on the other hand surveys 49,GOO-77,400 individual households twice per year. The survey is administered by the Bureau of Justice Statistics and is conducted by the US Census Bureau. The survey involves the collection of information about nonfatal personal crimes and household property crimes. In addition to information about the characteristics of the crime, the survey collects data on the victim, the offender, and whether the crime was reported to the police (Data Collection: National Crime Victimization Survey (NCVS), 2018).

In addition to publishing data and reports on crime, the Bureau of Justice Statistics publishes data/reports on court systems. The web page listed under courts enumerates the following reports/programs/ surveys/censuses: Census of Problem-Solving Courts, Census of Public Defender Offices, Census of State Court Organization, Civil Justice Survey of State Courts, Court Statistics Project, Juveniles in Criminal Court, National Judicial Reporting Program, National Survey of Indigent Defense Systems, National Survey of Prosecutors, and State Court Processing Statistics (Bureau of Justice Statistics: Courts, n.d.).

The Census of Problem-Solving Courts addresses the operation of the so-called specialty courts. These courts are defined as those which “use therapeutic justice to reduce recidivism” (Data Collection: Census of Problem-Solving Courts, 2012). An example of such a court would be a drug court. The Census of Problem-Solving Courts provides data on the number of participants, services provided, and benefits associated with the court.

The Court Statistics Project provided annual data from 1975 to 2007 on state courts, such as structure jurisdiction, caseload volume, and trends. One of the advantages of this project is that, since the data is translated into a common framework, it allows for the comparison of the various states with respect to the operation of their state courts.

Reports published by the Bureau of Justice Statistics related to corrections include Jail Inmates in 2018 (Zeng, 2018), and the 2018 Update on Prisoner Recidivism: A 9-Year Follow-Up Period (2005-2014) (Alper, Durose. and Markman, 2018).

Some interesting facts from the report, Jail Inmates in 2018, include the following: the number of jail inmates at midyear 2018 (738,400 inmates) declined by 12% from 2008 (785,500 inmates) and the jail population in 2018 was 50% white, 33% black, and 15% Hispanic. While the percentage of the jail population that was black in 2018 was still disproportionately high compared to the percentage in the entire population (at 12.3%), this still represents a significant reduction from 45.8% of the jail population from 2008 (a 28% reduction from 45.8% to 33%).

In addition to the large number of people in jail or prison, there is typically even a larger number on probation or parole in the United States. For example, in 2014, there were 3,864,100 people on probation and 856,900 parolees (Kaeble et al, 2015).

The statistics regarding recidivism are complex in nature because of the different ways that recidivism can be defined, among other things. For example, one can address recidivism among those released from prison/jail, or among those on probation. One can also classify recidivists as (1) those who are rearrested for a technical violation, (2) those who are rearrested for a new crime, or (3) those who are rearrested and returned to prison.

Examples of technical violations for those on probation include failure to report, moving without permission, and failing a blood/ urine test. Something like failing to report is sometimes just a problem with transportation. Many reformists have suggested some relaxation of the restrictions resulting from these technical violations.

Some of the notable information obtained from the Update on Prisoner Recidivism: A 9-Year Follow-Up Period includes the fact that 44% of the released prisoners from the study were re-arrested during the first year following release. This fact provides a clear indication of the need for greatly enhanced rehabilitation programs for prisoners, both from the standpoint of benefitting the convicted and the public.

Sipes (2018) provides an excellent summary of many of the reports on recidivism, Examples of some of the data from his summary report include the following: 68% of released prisoners were re-arrested for a new crime within three years of release from prison, and 77% were re-arrested within five years. Of course, not everyone who is re-arrested is convicted, but 49.7% of inmates released were returned to prison within three years because of either a technical violation or conviction of a new crime, while 55.1% were returned to prison within five years. These figures point out the importance of rehabilitation.

While there is much useful information on the wide variety of reports that are available from the Bureau of Justice Statistics, there are obvious difficulties with some of the reports. For example, much of the data obtained to provide some of the reports (e.g., the NIBRS reports) was self-reported, and obtained on a voluntary basis; of course, given the large number of agencies (with already heavy workloads) from which data was obtained, such an approach was almost a necessity. One would expect however that agencies with good performances would more likely provide the relevant data.

Second, for many reports, the numbers provided would only be a small part of the overall story. For example, follow-ups on the prisoner recidivism reports would involve extensive questioning of both recidivists and non-recidivists to determine why recidivism occurs in certain cases, and why not in other cases. Results from such a process would be useful in the design of rehabilitation programs. While there certainly are reports/papers in existence which involve the questioning of former prisoners in this regard, it appears that more effort along these lines would be extremely useful.

Finally, it appears that the generation of reports which involved the use of longitudinal data would be extremely useful. Such an approach was suggested by Blumstein and Cohen (1987) a few decades ago in terms of predicting criminal behavior. Most of the reports discussed above address one area of the criminal justice system (crime and arrests, courts, or corrections). It appears that data collected from individuals as they proceed through the entire system would be useful in establishing appropriate correlations.

Of course, there are trade-offs involved in addressing the three difficulties mentioned above—useful reports generated from addressing these issues would require much additional effort and increased staffing levels.

 
Source
< Prev   CONTENTS   Source   Next >