Berk, Richard A.

Collective behavior by Richard A Berk( Book )

5 editions published in 1974 in English and held by 299 WorldCat member libraries worldwide

A national sample survey : public opinion on sentencing federal crimes by Peter H Rossi( Book )

1 edition published in 1995 in English and held by 274 WorldCat member libraries worldwide

Water shortage : lessons in conservation from the great California drought, 1976-1977( Book )

4 editions published in 1981 in English and held by 254 WorldCat member libraries worldwide

Statistical learning from a regression perspective by Richard A Berk( Book )

32 editions published between 2008 and 2016 in English and held by 214 WorldCat member libraries worldwide

"Statistical Learning from a Regression Perspective considers statistical learning applications when interest centers on the conditional distribution of the response variable, given a set of predictors, and when it is important to characterize how the predictors are related to the response. As a first approximation, this is can be seen as an extension of nonparametric regression. Among the statistical learning procedures examined are bagging, random forests, boosting, and support vector machines. Response variables may be quantitative or categorical." "Real applications are emphasized, especially those with practical implications. One important theme is the need to explicitly take into account asymmetric costs in the fitting process. For example, in some situations false positives may be far less costly than false negatives. Another important theme is to not automatically cede modeling decisions to a fitting algorithm. In many settings, subject-matter knowledge should trump formal fitting criteria. Yet another important theme is to appreciate the limitation of one's data and not apply statistical learning procedures that require more than the data can provide." "The material is written for graduate students in the social and life sciences and for researchers who want to apply statistical learning procedures to scientific and policy problems."--Jacket

Dick Sudhalter and his friends "With Pleasure" by Richard M Sudhalter( )

1 edition published in 1994 in English and held by 79 WorldCat member libraries worldwide

Criminal justice forecasts of risk : a machine learning approach by Richard A Berk( Book )

9 editions published in 2012 in English and held by 31 WorldCat member libraries worldwide

"Machine learning and nonparametric function estimation procedures can be effectively used in forecasting. One important and current application is used to make forecasts of future dangerousness" to inform criminal justice decision. Examples include the decision to release an individual on parole, determination of the parole conditions, bail recommendations, and sentencing. Since the 1920s, "risk assessments" of various kinds have been used in parole hearings, but the current availability of large administrative data bases, inexpensive computing power, and developments in statistics and computer science have increased their accuracy and applicability. In this book, these developments are considered with particular emphasis on the statistical and computer science tools, under the rubric of supervised learning, that can dramatically improve these kinds of forecasts in criminal justice settings. The intended audience is researchers in the social sciences and data analysts in criminal justice agencies."--Publisher's website

Evaluating Alternative Police Responses to Spouse Assault in Colorado Springs : an Enhanced Replication of the Minneapolis Experiment, 1987-1989( )

1 edition published in 1994 in English and held by 25 WorldCat member libraries worldwide

The purpose of this study was to replicate an experiment in Minneapolis (MINNEAPOLIS INTERVENTION PROJECT, 1986-1987 [ICPSR 9808]) testing alternative police response to cases of spouse assault, using a larger number of subjects and a more complex research design. The study focused on how police response affected subsequent incidents of spouse assault. Police responses studied included arrest, issuing emergency protection orders, referring the suspect to counseling, separating the suspect and the victim, and restoring order only (no specific action). Data were obtained through initial incident reports, counseling information, and personal interviews. Follow-up interviews were conducted at three- and six-month periods, and recidivists were identified through police and court record checks. Variables from initial incident reports include number of charges, date, location, and disposition of charges, weapon(s) used, victim injuries, medical attention received, behavior towards police, victim and suspect comments, and demographic information such as race, sex, relationship to victim/offender, age, and past victim/offender history. Data collected from counseling forms provide information on demographic characteristics of the suspect, type of counseling, topics covered in counseling, suspect's level of participation, and therapist comments. Court records investigate victim and suspect criminal histories, including descriptions of charges and their disposition, conditions of pretrial release, and the victim's contact with pretrial services. Other variables included in follow-up checks focus on criminal and offense history of the suspect. The data collection includes separate fichiers for the original, second, and final versions of some of the forms that were used ... Cf.: http://webapp.icpsr.umich.edu/cocoon/ICPSR-STUDY/09982.xml

Race and the Decision to Seek the Death Penalty in Federal Cases, 1995-2000 [United States]( )

1 edition published in 2006 in English and held by 24 WorldCat member libraries worldwide

The purpose of this project was to examine possible defendant and victim race effects in capital decisions in the federal system. Per the terms of their grant, the researchers selected cases that were handled under the revised Death Penalty Protocol of 1995 and were processed during Attorney General Janet Reno's term in office. The researchers began the project by examining a sample of Department of Justice Capital Case Unit (CCU) case files. These files contained documents submitted by the United States Attorney's Office (USAO), a copy of the indictment, a copy of the Attorney General's Review Committee on Capital Cases (AGRC's) draft and final memorandum to the Attorney General (AG), and a copy of the AG's decision letter. Next, they created a list of the types of data that would be feasible and desirable to collect and constructed a case abstraction form and coding rules for recording data on victims, defendants, and case characteristics from the CCU's hard-copy case files. The record abstractors did not have access to information about defendant or victim gender or race. Victim and defendant race and gender data were obtained from the CCU's electronic files. Five specially trained coders used the case abstraction forms to record and enter salient information in the CCU hard-copy files into a database. Coders worked on only one case at a time. The resulting database contains 312 cases for which defendant- and victim-race data were available for the 94 federal judicial districts. These cases were received by the CCU between January 1, 1995 and July 31, 2000, and for which the AG at the time had made a decision about whether to seek the death penalty prior to December 31, 2000. The 312 cases includes a total of 652 defendants (see SAMPLING for cases not included). The AG made a seek/not-seek decision for 600 of the defendants, with the difference between the counts s ... Cf.: http://dx.doi.org/10.3886/ICPSR04533

Specific Deterrent Effects of Arrest for Domestic Assault: Minneapolis, 1981-1982 by Richard A Berk( )

4 editions published in 1984 in 3 languages and held by 22 WorldCat member libraries worldwide

This data collection contains information on 330 incidents of domestic violence in Minneapolis. Part 1, Police Data, contains data from the initial police reports filled out after each incident. Parts 2-5 are based on interviews that were conducted with all parties to the domestic assaults. Information for Part 2, Initial Data, was gathered from the victims after the incidents. Part 3, Follow-Up Data, consists of data from follow-up interviews with the victims and with relatives and acquaintances of both victims and suspects. There could be up to 12 contacts per case. Suspect interviews are the source for Part 4, Suspect Data. An experimental section, Part 5, Repeat Data, contains information on repeat incidents of domestic assault from interviews with victims. Parts 2-5 include items such as socioeconomic and demographic data describing the suspect and the victim, relationship (husband, wife, boyfriend, girlfriend, lover, divorced, separated), nature of the argument that spurred the assault, presence or absence of physical violence, and the nature and extent of police contact in the incident. The collection also includes police records, which are the basis for Parts 6-9. These files record the date of the crime, ethnicity of the participants, presence or absence of alcohol or drugs and weapons, and whether a police assault occurred ... Cf.: http://webapp.icpsr.umich.edu/cocoon/ICPSR-STUDY/08250.xml

Race and the decision to seek the death penalty in federal cases by Stephen P Klein( Book )

2 editions published in 2006 in English and held by 10 WorldCat member libraries worldwide

This study examined the relationship between the federal government's decision to seek the death penalty in a case and that case's characteristics, including the defendant's and victim's races. This research began by identifying the types of data that would be appropriate and feasible to gather. Next, case characteristics were abstracted from Department of Justice Capital Case Unit (CCU) files. Defendant- and victim-race data were obtained from electronic files. Finally, three independent teams used these data to investigate whether charging decisions were related to defendant or victim race. The teams also examined whether these decisions were related to case characteristics and geographic area. There are large race effects in the raw data that are of concern. However, all three teams found that controlling for nonracial case characteristics eliminated these effects, and that these characteristics could predict the seek decision with 85 to 90 percent accuracy. These findings support the view that decisions to seek the death penalty were driven by heinousness of crimes rather than by race. Nevertheless, these findings are not definitive because of the difficulties in determining causation from statistical modeling of observational data