Evaluation of the Gang Resistance Education and Training (G.R.E.A.T.) Program

 Beginning in 1991, the Phoenix Police Department developed the “Gang Resistance Education and Training (G.R.E.A.T.) program, a school-based, police officer-instructed, prevention program targeting middle-school students” (Esbensen, Osgood, Taylor, Peterson, & Freng, 2001, p. 87). With funding from Congress and the Bureau of Alcohol, Tobacco and Firearms (ATF), this pilot program became the driving force as an intervention for youth long before they become delinquents and joined gangs. Implemented with a strategy of a cognitive approach, it focused on two standards: attitudinal and behavioral characteristics. Modeled after the DARE (Drug Abuse Resistance Education) program, “the dual goals of the G.R.E.A.T. program (as stated in the G.R.E.A.T. Officer Instructor’s Manual) are ‘to reduce gang activity, and to educate a population of young people as to the consequences of gang involvement’” (Esbensen et al., 2001, p. 109). The program encompasses eight lessons taught over a nine-week lesson plan. In 1992, the G.R.E.A.T. program became available nationwide, and in 2009, it expanded to Central America.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

 From 1995 to 1999, Dr. Finn-Aage Esbensen, from the University of Missouri-St. Louis, and his colleagues from the University of Nebraska at Omaha, the University of Wyoming, and Pennsylvania State University, conducted a National Evaluation on the G.R.E.A.T. program. Using student questionnaires, Esbensen et al. developed a longitudinal quasi-experimental research design based on “measures of perceptions about the appropriateness of certain behaviors and measures of peer group conduct” (Esbensen et al., p. 91), and conceptualized on the program’s objectives: delinquency and gang membership. Although not defined in the research, it is important to cover these concepts. Delinquency, as defined by Merriam-Webster, involves crimes or other morally wrong acts, or illegal or immoral behavior, especially by young people. “The (youth) gang is a self-formed association of peers having the following characteristics: a gang name and recognizable symbols, identifiable leadership, a geographic territory, a regular meeting pattern, and collective actions to carry out illegal activities” (Howell, 1997, p. 1).

 The research applied the evaluation to both treatment and control groups for analysis, and surveys consisting of a pre-test, a post-test, and four consecutive annual surveys were completed by both groups. However, the research developed for the operationalization of gang membership was not fully developed—there were only four sections (followed by the number of variables in parentheses) that focused on youth gangs: gang membership (1), positive gang (6) and negative gang outlooks (7), as well as, attitudes about gangs (1). As one of the main concepts, gang membership, lacked content and face validity. Delinquency was much more operationalized, provided an earnest assessment of self-control (8), commitment to negative peers (3), and negative peer behavior (16), as well as, self-reported delinquency (17) which also had content validity covering minor, person, property, status, and total distinctions, as well as, peer delinquency. In other areas, parental monitoring, maternal and paternal attachments, guilt, and self-concept—were not a factor in the G.R.E.A.T. program curriculum, the evaluation consisted of these fundamental elements.

 Although, not the focus of the G.R.E.A.T. program, the evaluation was aimed at the theoretical perspectives of social learning and self-control theories and viewed them as the key points covering the attitudinal section of the study. The researchers evaluated the program by aligning those theoretical concepts with the curriculum of eight lessons. Social learning theory in criminology was applied by Ronald L. Akers (1973) for which he defined how human behavior changed or was altered by an individual’s interactions through cognitive, behavioral, and environmental influences. Applying social learning theory led to the understanding of aggression and psychological disorders. The G.R.E.A.T. program established this concept in Lessons 1, 3, and 4, as defined by “laws, values, norms, and rules supportive of law-abiding behavior” (Esbensen et al., p. 90).

 Self-control theory, credited to Travis Hirschi and Michael Gottfredson (1990), was a condition characterized by the disadvantages of children (ages 7 to 8) who develop low self-control. This characteristic remains within an individual throughout the life cycle. Criminal behavior is believed to be a response to this theory. Lesson 4 (conflict resolution) illustrates anger and coping strategies. Lesson 5 (meeting basic needs) ties in a risk-taking element. Lessons 6, 7, and 8 address “impulsive behavior by teaching responsibility, goal setting, and personal and career goals” (Esbensen et al., p. 90). Other lessons involved crime and victimization, cultural sensitivity and prejudice, drugs and neighborhoods, and their responsibilities for their school and neighborhoods.

 Having previously been selected, six cities (Philadelphia, Portland, Phoenix, Omaha, Lincoln, and Las Cruces in New Mexico) were enlisted to recommend schools that would have the classrooms and police officers available for the program. Esbensen et al used a research design based on a nonprobability sampling method. The evaluation applied availability sampling from fifteen schools, and a more purposive sampling from seven schools, and in each, attempts were made to draw a random sampling. The goal was to “minimize the potential differences between the sets of treatment and control classes” (Esbensen et al., p. 93), covering “22 schools, 153 classrooms, and more than 3,500 students” (p. 92). For the probability of attrition, “each modified process sought to ensure the integrity of (the) research objective” (p. 93). During a cross-sectional In Philadelphia, “the pretest and attrition differences between the treatment and control groups were the most pronounced; so, the risk of invalid results is greatest there. Therefore, our third approach to addressing the potential difference between groups was to repeat the analyses using only the other five sites” (Esbensen et al., p. 107). As later discussed in this paper, modeling software also supported the results in sample attrition.

 The University of Nebraska’s Institution Review Board approved passive parental consent (absence of refusal implies consent), and subsequent annual surveys warranted active parental consent (absence of permission implies refusal). No mention of the status of the students falling under restrictions as “Special Populations” in research was found. How the university’s IRB was able to evade that requirement remains a question. Although, there were no less than two mailings with cover letters and consent forms, as well as, teachers and school personnel plying their students with incentives to return consent forms, and at the initial sample, there was an overall retention of 57%.

 With four levels of analysis, the researchers utilized a multilevel modeling software, known as MLwiN, developed at the University of Bristol, United Kingdom, with complex hierarchical (in this instance, state, city, school, classroom, and student) structures supported multivariate regression and analysis of the variance for either discrete or continuous responses. To capture the effects of attrition, the variables can be evaluated in linear terms or as a quadratic function. However, the relationship between these variables was not linear, therefore a quadratic function was needed to estimate the level of sample attrition. It was also necessary to include a variance component for each level, i.e., with-in individual, classroom, and school. “Applying logistic regression to estimate the probability (or propensity) for attrition…we then compared the distributions of the treatment and control groups on these propensity scores, and probabilistically eliminated cases to produce matching distributions for the two groups” (p. 107). Frankly, students who endured the evaluation into the last four annual surveys, most likely, would have sustained little if any change in their responses. Under these conditions, the statistical outlook would be constant.

 In a published cross-sectional evaluation, Esbensen and Osgood found that students who completed the G.R.E.A.T. program, “reported committing fewer delinquent acts and expressed more prosocial attitudes, including, among others, more favorable attitudes toward the police, higher levels of attachment to parents and self-esteem, and greater commitment to school” (1997 1999, p. 89). “Using a multisite, pretest/post-test research design, Dennis J. Palumbo and Jennifer L. Ferguson found the students had a ‘slightly increased ability’ to resist the pressures to join gangs,” however “the lack of a control group prevent(ed) assessments of the internal validity” (1995, p. 600). Whereas, the longitudinal results did not reveal positive outcomes until the third and fourth years.

 The causal mechanism in this program was the student surveys given before, between, and after the G.R.E.A.T. program and its outcomes: elimination of delinquency and prevention of gang membership. In the survey, the researchers applied two filter questions, and gang membership was determined by the student’s responses. For example, “have you ever been a gang member?” and “are you now in a gang?” Any student answering yes to either question was requested to give “information about the gang structure, gang activity, and attitudes about the gang. (Esbensen et al., p. 91). Prior to the “gang section,” the student was asked for a response about his or her perceptions of gangs, to include, perspectives of good and bad aspects of gang membership and satisfaction of gang membership.

 The self-reports in the surveys contributed to outcomes of illegal activity, drug use, and victimization. These measures were adopted for the National Youth Survey (Elliott, Huizinga, & Ageton, 1989) or the Denver Youth Survey (Survey, Huizinga, Esbensen, & Weiher, 1991). Activities related to gangs—school environment, “there are gang fights at my school;” and commitment to negative peers, “if your friends were getting you in trouble at home, how likely is it that you would still hang out with them?” Do you view gang membership as having positive reinforcements, i.e., protection, money, and excitement? Activities related to delinquency—self-control, “sometimes I will take a risk just for the fun of it,” commitment to positive peers, “if your friends told you not to do something because it was against the law, how likely is it that you would listen to them?” and guilt, how guilty would he or she feel if he or she did such things as “hit someone with the idea of hurting them” or “used alcohol.”

 In the self-report measures of delinquency, the researchers asked for the number of times in the past twelve months (pretest), six months (post-test), and six months (annual surveys) that the student had committed an activity involved in theft, damaged or destroyed property, carried a hidden weapon, stolen or attempted to steal an item worth less than $50 and an item worth more than $50, gone into or attempted to enter a building to steal, stole or attempted to steal a vehicle, hit someone to hurt them, attacked someone with a weapon, used a weapon or force people to give you their money and personal items, been involved in gang fights, shot at someone because you were asked to do so, sold cannabis, or sold other illegal drugs. One self-report measured the use of alcohol, cannabis, paint, glue or other substance to inhale to get high, or other illegal drugs; the last, victimization: did someone attempt to hurt you, did someone use a weapon or force to steal from you, did someone attack with a weapon or did someone attempt to seriously hurt or kill you, have some of your belongings been stolen.

 In this qualitative evaluation, interval validity was influenced by varied matters, some that are indicative of a longitudinal study, such as, endogenous change, subjects (students) mature, on the flip side, they regress—fall victim to their own desires and become afflicted by psychological disorders in early adulthood. They experience interactions with society’s threats: familial disruption, intimate relationships falter, and early adult friendships do not develop into trusting life-long connections. Perhaps, the G.R.E.A.T. program lessons involving responsibility and goal setting could be extended to two-week sessions for each topic. It is extremely important that youth between the elementary school and high school levels begin thinking about their lives beyond a traditional education. Will they continue their education and plan for college? Or a trades school? What other options do they consider in early adulthood?

 Having completed mail questionnaires, parents, school personnel, and G.R.E.A.T. officers were asked to comment on their level of satisfaction and witnessed the effectiveness of the program (Esbensen et al., p. 90). Overall, there was a high level of satisfaction with the program by the three factions (p. 90). If the parents believed that the program made a difference in their children’s lives, could there have been an unseen correlation between the parent and child? If so, would the evaluation’s modeling program reveal a link to causality?

 With an advanced structured plan and greater prior funding established, the theoretical perspectives utilized in the evaluation and the goals established in the lessons could be synthesized. The operationalization could be based on rational choice theory and a well-defined conceptualization, to preclude any misunderstanding, would be discussed prior to any testing. Using the construct of a sampling frame of pre-determined states and/or schools with randomized samples to achieve the probability sampling method and eliminate systematic bias, specifically at the student-level, who have the perquisite parental consent, would necessitate an extensive number of school personnel and police officers. The number of researchers at the inception would be higher than the current program, however, this number would decrease after the post-test. If the schools and/or the researchers could maintain strong bonds with the students and the parents after the post-test, the levels of attrition and endogenous changes could be lower.

 Using the current G.R.E.A.T. program guidelines, generalizability would not be a sound endeavor. However, the previously mentioned alterations may be the answer to the program’s long-term viability and its long-lasting effectiveness.

References

Elliott, D. S., Huizinga, D., & Ageton, S. S. (1989). Explaining delinquency and drug use. Newbury Park, CA: Sage Publications.

Esbensen, F., Osgood, D., Taylor, T., Peterson, D., & Freng, A. (2001). How great is G.R.E.A.T.? Results from a longitudinal quasi-experimental design. Criminology & Public Policy, 1(1), 91.

Howell, J. C. (1997). OJJDP fact sheet: Youth gangs (72). Retrieved from The Office of Juvenile Justice and Delinquency Prevention website: https://www.ncjrs.gov/pdffiles/fs-9772.pdf.

Palumbo, D. J., & Ferguson, J. L. (1995). Evaluating Gang Resistance Education and Training (Great). Evaluation Review, 19(6), 600. doi:10.1177/0193841×9501900601.

Survey, D. Y., Huizinga, D., Esbensen, F., & Weiher, A. W. (1991). Are there multiple paths to delinquency? The Journal of Criminal Law and Criminology, 82(1), 83. DOI:10.2307/1143790