Public Support for Rehabilitation in Ohio, 1996 (ICPSR 2543)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Hide

Study Purpose View help for Study Purpose

Because of the important role that public opinion plays regarding decisions about punishment and sentencing policy, an accurate appraisal of the public's views is critical. The most pressing issue concerning public attitudes toward rehabilitation is the lack of specificity in the existing research. Several researchers have made significant advances in identifying the factors that shape treatment attitudes, but the relationships reported have been observed by collating results across numerous studies rather than in a single project. Also, these characteristics often have been examined in isolation, without considering contextual features that likely influence citizens' opinions. Further, only preliminary evidence is available on how the attributes of the criminal, the crime, and the provision of treatment can shape public perceptions. Therefore, the main focus of this research was on identifying the conditions under which the public support for rehabilitation varies. A single, multivariate analysis method was used so that the influence of each respondent, criminal, crime, and treatment characterististic could be determined within the context of all other factors. The research also explored differences between global and specific attitudes toward rehabilitation.

Study Design View help for Study Design

Data for this study were collected through a mail survey of 1,000 Ohio residents. Distribution of the surveys followed Dillman's (1978) total design method. The initial mailing was sent to all 1,000 members of the sample on May 28, 1996, and included a copy of the questionnaire and a business reply envelope. Accompanying these materials was a personalized letter from the project director requesting that the respondent complete and return the survey. The letter emphasized the importance of the study and noted that the project was sponsored by the University of Cincinnati and the United States Department of Justice. A reminder letter was sent to the entire sample one week later. After another two weeks, a replacement survey, accompanied by a reminder letter and return envelope, was mailed to all those who had not yet responded. A final mailing was sent to all nonrespondents seven weeks after the initial mailing and included all of the survey materials. To ensure the confidentiality of the responses, a number was printed on the return envelope. Each number was removed from the mailing list when it was received. Each new letter took a slightly different approach to encourage responses, and the mailings to nonrespondents became slightly more urgent with each follow-up. The first three mailings were sent using first-class postage, but to further emphasize the importance of the final mailing, it was sent by registered mail. Data collection efforts ended on August 26, 1996. Questionnaire items elicited demographic, experiential, and attitudinal information from each respondent. To assess the potential influence of offender, offense, and treatment characteristics on the respondent's support for rehabilitation, several variables were combined to create a factorial vignette. This method allowed the independent effects of each factor on support for rehabilitation to be determined. The respondents were asked to express their agreement or disagreement with five statements following the vignette: (1) general support for rehabilitation, (2) effectiveness of intervention, (3) basing release decisions on progress in rehabilitation programs, (4) individualizing sentences to fit treatment needs, and (5) expanding treatment opportunities for offenders. The computer program, Vig-Write, was used to generate the vignettes and was programmed to screen out illogical combinations and to weight the probability that the computer would select any particular level within each dimension to create a universe of vignettes that more accurately represented reality. Types of offenses included in the vignettes were robbery, burglary, aggravated assault, larceny, motor vehicle theft, fraud, drug sales, and drug use. These offenses were selected since they are well-known to the public, offenders are arrested for these offenses fairly frequently, and the offenses are potentially punishable by a sentence of either prison or probation. Regarding the particular offenses included within each offense category in the vignettes, several attributes within offenses were designed to assess the influence of different levels of harm, either financial or physical. Offender characteristics and offense selection for use in the vignettes were weighted by their frequency of arrests as reported in the Federal Bureau of Investigation's 1995 Uniform Crime Report data. A rating of the seriousness of each offense was assigned using a separate survey of 118 undergraduate university students, and the resulting seriousness score was used in the analysis of the vignettes. For the crime seriousness survey, respondents were asked to rate each offense based on a short event description on a scale of 1 to 9, where 1 equalled "not at all serious" and 9 equalled "extremely serious". Additional items on the mail survey instrument assessed the respondent's global and specific attitudes toward treatment.

Sample View help for Sample

Part 1: Random sampling from the database of Survey Sampling, Incorporated, in Ohio. Part 2: Student volunteers at the University of Central Florida.

Universe View help for Universe

All residents in Ohio.

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Description of Variables View help for Description of Variables

Independent variables from the mail survey include the respondent's age, education, income category, sex, race, political party, rating of political conservativism, personal contact with offenders, religious identity salience, religiosity, attitudes toward biblical literalness and religious forgiveness, fear of crime, and victimization. Variables from the vignettes examined whether support for rehabilitation was influenced by offender age, race, sex, type of offense committed, employment status, substance use, prior record, sentence, and treatment program. Global support for rehabilitation was measured by responses to two questions: what the respondent thought the main emphasis in most prisons was (to punish, to rehabilitate, to protect society), and what the main emphasis should be. Ten items assessed variations in the respondent's attitudes toward rehabilitation by offender's age, sex, and prior record, location of treatment and the type of treatment provided. Variables from the crime seriousness survey recorded the respondent's rating of various crime events including assault and robbery (with or without a weapon, with varying degrees of injury or no injury to the victim), burglary, larceny, and auto theft (with varying values of the property stolen), drug dealing, drug use, and writing bad checks.

Response Rates View help for Response Rates

The attempts to encourage participation resulted in 559 completed or nearly-completed questionnaires being returned. Sixty-seven unanswered questionnaires were returned by the United States Postal Service because the addressee had moved and left no forwarding address or the forwarding order had expired. Thirty-eight surveys were returned unanswered because the intended respondent was deceased or was too ill to complete the questionnaire. The resulting response rate for those members of the sample who received a survey and were capable of completing it was 62.4 percent. Although this response rate is generally considered adequate, some concerns may be raised about the representativeness of the resulting sample. Typical of mailed surveys, this sample overrepresents males, whites, older residents of Ohio, and individuals with higher education and thus higher incomes. A response rate is not applicable for the seriousness survey, since all respondents were volunteers.

Presence of Common Scales View help for Presence of Common Scales

Several Likert-type scales were used.

Hide

Original Release Date View help for Original Release Date

1999-02-17

Version History View help for Version History

Hide