¶ … experimental design feasible? Why or why not? • What suggestions can you make for future studies of the DARE program?

The aims of DARE are long-term in nature, namely to encourage students to not abuse drugs over the course of their lifetimes. The only way to test this aim is to conduct a longitudinal study of a representative body of DARE graduates over at least a twenty-year period, to see if the intervention had a lasting effect upon their drug use habits. The control groups would be a group of students from similar demographics and geographical locations who did not have DARE or any other anti-drug program in their schools and a group of students who experienced an anti-drug education intervention substantially different than DARE. The selection of students would have to be balanced in terms of factors such as race, gender, and neighborhood, given that graduates of DARE programs might be more apt to come from either more or less affluent locations than their non-DARE peers.

This is a feasible experimental design because it takes place in the 'real world,' without the need for artificially-imposed constraints. It also contains a control group, which is vitally necessary to ensure that causality is established, not mere correlation (Maxwell & Babbie 2011). So long as the control and experimental groups are variable and genuinely similar, the presence of absence of a DARE program is the only significant variable that should or should not indicate the development of negative behavioral patterns towards drugs. A previous study of 1,000 twenty-year-olds who went through the DARE program did not yield promising results: "20-year-olds who'd had DARE classes were no less likely to have smoked marijuana or cigarettes, drunk alcohol, used illicit drugs like cocaine or heroin, or caved in to peer pressure than kids who'd never been exposed to DARE" and even had lower self-esteem (frequently associated with drug use)" but a larger sampling with a more diverse control group might possibly yield different results (Reaves 2001). Also, this comparative study did not assess the efficacy of DARE in comparison with other anti-drug campaigns, merely those who did not experience an intervention at all.

Reference

Maxfield, M.G. & Babbie, E.R. (2011). Research methods for criminal justice and criminology. Belmont, CA: Wadsworth, Cengage Learning.

Reaves, J. (2001). Just say No to DARE. Time Magazine. Retrieved:

http://www.time.com/time/nation/article/0,8599,99564,00.html

• What are the advantages and disadvantages of the sampling methods used in these two studies?

• How might you use this data to develop an experimental design?

The study "AIDS-Related Written Court Decisions in Federal and State Courts, 1984-1989: [United States]" assessed all court decisions within this specific period of time involving AIDS, including cases in which the plaintiff or defendant had AIDS. It assessed such factors as size and resources of plaintiff vs. defendant; respective power roles; which decisions made reference to constitutional values, to equal treatment, and so forth. Obviously, not all cases involving AIDS are rights-based cases, but an experimental study could be constructed to determine the extent to which cases involving persons with AIDS (either plaintiffs or defendants) that hinged upon certain conceptual modalities such as constitutional rights were decided in favor or against persons with AIDS. Obviously, many cases in which one or more participant had AIDS might not solely revolve around this issue, so further segmentation of the data would be necessary.

The study "ABC News Poll of Public Opinion on Crime, December 1982" was a random sampling of the entire population of the United States, selected by random digit dialing. It asked the study participants a series of questions about their perceptions of the extent to which crime was occurring in the United States; the severity of different types of crimes; their perceptions of the economy, approval of Ronald Regan's handling of the economy,...

The study design was deliberately broad and random, designed to paint a fairly complete picture of U.S. attitudes. At the time of the study, most U.S. households had a landline phone and little else, although it could be argued that the study was biased in favor of persons who had time and were home long enough to give information to the person doing the survey. The attitudes regarding perceptions of crime in conjunction with other issues could be used to construct an experimental study to determine whether persons who tend to hold liberal vs. conservative views on non-crime related issues are more or less likely to perceive crime as a growing problem in the U.S.
References

ABC News Poll of Public Opinion on Crime, December 1982. NACJD. Retrieved:

https://www.ncjrs.gov/pdffiles1/Digitization/198406NCJRS.pdf

ABC News Poll of Public Opinion on Crime, December 1982. NACJD data sets. Retrieved:

http://www.icpsr.umich.edu/icpsrweb/NACJD/ssvd/studies/08100/variables?q=

AIDS-Related Written Court Decisions in Federal and State Courts, 1984-1989.

NACJD data sets. Retrieved: http://www.icpsr.umich.edu/icpsrweb/NACJD/ssvd/studies/06502/variables?q=&paging.startRow=1

Article review: Evaluating the BRI program

Introduction

While many programs have been developed with the specific intention of transitioning former criminals into mainstream society, the best way to accomplish this objective remains hotly debated amongst policy-makers and criminologists. The article "Controlling violent offenders released to the community: An evaluation of the Boston Reentry Initiative" specifically studies the Boston Reentry Initiative (BRI) used to transition violent adult offenders into Boston neighborhoods "through mentoring, social service assistance, and vocational development" (Braga, Piehl, & Hureau 2008: 1). The quasi-experimental design compares rates of recidivism with a control group (Braga, Piehl, & Hureau 2008: 1).

Summary

There is a general consensus in support of the need for programs to help inmates make the difficult transition from life 'inside' to life 'outside' jail walls. There has been considerable financial investment devoted such initiatives but it is often extremely difficult to measure program efficacy. Character traits (including drug abuse) as well as the character of the environments into which the inmates are released all put certain individuals at high risk for recidivism (Braga, Piehl, & Hureau 2008: 2). To combat these negative influences that make it very easy to fall back into old ways, "in-prison programs designed to increase educational levels, job skills, and social functioning" are designed to reduce recidivism rates (Braga, Piehl, & Hureau 2008: 2). However, there have been efforts to do more inter-agency collaboration and constructions of community partnerships to ensure that meaningful structures of support can enable prisoners to put these values into action.

The BRI program actively involves participants long before their scheduled release. Offenders must create a "transition accountability plan" (Braga, Piehl, & Hureau 2008: 4). BRI program participants are violent offenders between the ages of 17-34, all labeled as high risk for violent recidivism. Prisoners are assigned mentors from day 1 of their release. Faith-based mentors who have worked with the inmates pre-release continue their support; caseworkers connect workers to meaningful jobs in the community.

To measure effectiveness, "a non-randomized quasi- experimental design was used to compare recidivism patterns among BRI participants to the recidivism patterns of an equivalent control group" using data derived from Massachusetts Criminal Offender Record Information (CORI) (Braga, Piehl, & Hureau 2008: 6). One problem, however, was finding an appropriate control group of violent offenders that did not receive any supportive treatment, since the 'worst' offenders were selected to be in the program. "This selection process makes program evaluation difficult since any contemporaneous control group will not be as high risk to fail" as the experimental group (Braga, Piehl, & Hureau 2008: 6).

Ethically, the worst offenders could not be excluded from the program, given the potential benefits they could accrue. However, this biased the quasi-experimental design against the program, since those excluded were by definition at a lower risk for recidivism. Although through statistical modeling a reasonably similar control group was assembled, "BRI subjects were somewhat more likely to have a current conviction for violent offense, and a current or past arrest for a non-violent gun offense (i.e., an arrest for illegal gun possession)" (Braga, Piehl, & Hureau 2008: 7).

However, despite these potentially troubling variables that could limit the full display of the efficacy of the BRI program, overall the program was found to be successful. "Relative to the comparison group subjects, BRI participants were found to have 30% lower rates of recidivism" (Braga, Piehl, & Hureau 2008: 11). Gang membership was found to be the most significant obstacle for members to overcome, also underlining the need for effective support structures.

Analysis

From a purely qualitative description, the BRI sounds like a valuable and noteworthy program. However, there are certain problematic components to the experimental design. The fact that the control group received no intervention at all may be one reason that recidivism was reduced in the group receiving the intervention. 'Something' may be indeed better than nothing when it comes to interventions, but that does not necessarily mean that the BRI program is the optimal intervention. There may be better programs to deal with the problems of violent offenders' recidivism, but all the quasi-experimental design proved was that the BRI was better than nothing, an unsurprising finding given the high rates of recidivism for violent criminals and the need for some support structures to prevent re-offending.

Critique

Although the experiment design is flawed at best, it is admittedly difficult to compare different programs. A single…