Preventing Crime

The study by Sherman et al. (1998) “Preventing Crime: What Works, What Doesn’t, What’s Promising” presents evidence that much of the research done on crime and prevention remained untapped by their research. They refer to this as “fugitive” research done by such organizations as government bodies and that remained largely unpublished. This poses a threat to the validity of the research, as a large inaccessible body of information leaves major gaps in their research that might have altered the results of the study.

It also hurts the possibility of randomization. The authors report that a bias exists against publishing reports that contain “statistically insignificant differences. ” Though this might not pose a problem for the validity of the result for which programs do work, the inaccessibility of these research reports affects the validity of the results for the section entitled “what does not work” (1998), as many of the unsuccessful programs remain unpublished.

No control variables were used in this study, because it mainly reviewed the literature published in previous studies. However, it strongly considered the control variables of the studies it reviewed in order to determine the validity of including their material in the research. The major threat to the study “Long-term Effects of an Early Childhood Intervention on Educational Achievement and Juvenile Arrest: A 15-Year Follow-up of Low-Income Children in Public Schools” done by Reynolds et al.

(2001) is its reported lack of randomization. The community chosen for intervention was selected because of its reported high drop-out level. Therefore, although the sample size was very large (n = 989), the subjects were not selected randomly, and this usually poses a validity threat. However, a control group was established for this study, which was also very large, numbering 550 participants. Randomization took place for the controls in that their schools (not the participants themselves) were selected randomly.

In order to gauge the actual effects of the intervention (which occurred in the form of the Chicago Child-Parent Program for kindergartens), the control group was chosen as group of pre-school and kindergarten students involved in the regular all-day school programs of their district. The five schools for the control group were randomly selected from the Chicago area. Also, because the CPC intervention provides pre-school care, 176 of the 550 control group participants were chosen because of their lack of pre-school intervention. The design of the Sheley et al.

(1991) study, “Firearms, violence, and youth in California, Illinois, Louisiana, and New Jersey, 1991” contained favorable and unfavorable aspects for the purposes of validity. It contained a very large sample size, while its breadth spanned several states and included both pre-criminal activity and post-criminal activity youths. However, the nature of the study dictated that the subjects were chosen from a pre-selected group, which did not favor randomization. In order to circumvent the effects of this, the authors included as many of the youths in the study as possible.

However, other possible threats to validity exist. The questionnaires given to the youths were self-administered (unless the youths’ inability to read necessitated that interviews be given). This lack of direct control over the response procedure allows for errors and/or misinterpretation. The fact that anonymity was granted to each respondent increases the chances of truthful responses. However, it cannot be thought to have eliminated the possibility—and incorrect (or deliberately untruthful) responses would harm validity.

Furthermore, the fact that the respondents were given a five-dollar inducement to participate in the study may have some implications with regard to validity. No control group was used in this study, as it did not include an intervention but was merely a survey. However, comparison groups existed in the form of the inmate youths versus the school youths of similar communities from which most of the inmates originated. Another possible threat to validity that stems from this is the fact that although the questions administered to the inmates and school youths were the same, they were given in reverse order to the two groups.

The questionnaires given to the incarcerated youths began with the questions about their access to guns and drugs, while the ones given to the students began with questions concerning their possible abuse (eg. physical and emotional abuse or neglect). The two groups might then be considered to have been treated differently in the “intervention,” and this difference may have had an impact on their responses. This has the possibility to harm the validity of the study.

The importance of defining terms can be seen to arise in the study “Increasing rates of school completion: Moving from policy and research to practice. ” This study, conducted by Lehr et al. (2004) sought the reasons for and possible solutions to the high drop-out rates in the United States. However, the validity of the study might have been harmed by the discrepancies found in the definition of “drop-out” given by the many studies reviewed in the research. Many variables were involved and hypothesized to have an impact on the drop-out rates of students.

The several definitions hinge on the inclusion or exclusion of some of these variables, and the number and proportion of “drop-outs” changes with each definition. So does the population that is considered as having dropped out of high school. Therefore the reasons and possible intervention solutions change along with the population. The validity of the study’s findings is therefore threatened by the lack of uniformity in the definition of “drop-out” across organizations that conducted the research and published the literature reviewed for the Lehr et al. (2004) study.

Outline of Introduction I. Several reasons exist why students decide to drop out of high school. a. Lack of adequate early-childhood preparation b. Lack of appropriate financial support c. Lack of appropriate emotional and family support d. Lack of intellectual aptitude or the existence of learning disabilities II. Many who do drop out are disadvantaged compared to their counterparts who possess diplomas. a. They are more likely to be unemployed b. They are more likely to be underemployed c. They are more likely to be on welfare d. They are more likely to become incarcerated

III. Programs exist that center on the rehabilitation of drop outs because. a. They are considered more likely to be desperate. b. They are considered more likely to be persons with disabilities. c. They are considered more likely to have emotional/behavioral problems. d. They are considered a drain on the government due to i. Lost revenue from taxes ii. Cost of welfare programs iii. Cost of prison programs Hypothesis: Lack of education as demonstrated by dropping out of high-school leads to an increased likelihood of criminal arrests in young people.


Lehr, C. A. , Johnson, D. R. , Bremer, C. D. , Cosio, A. , & Thompson, M. (2004). “Increasing rates of school completion: Moving from policy and research to practice. ” Essential tools. Minneapolis, MN: University of Minnesota, Institute on Community Integration, National Center on Secondary Education and Transition. Reynolds, A. , Temple, J, Robertson, D. , Mann, E. (2001). “Long-term Effects of an Early Childhood Intervention on Educational Achievement and Juvenile Arrest. ” [Electronic Version] JAMA. 285(18).

2339-2346 Retrieved January 15, 2007. Sheley, J. F. , J. D. Wright, and M. D. Smith. (1991). “Firearms, Violence, and Youth in California, Illinois, Louisiana and New Jersey. ” [Electronic Version] National Archive of Criminal Justice Data. Retrieved on January 31, 2007 from http://www. icpsr. umich. edu/NACJD/ Sherman, L. , Gottfredson, D. , MacKenzie, D. , Eck, J. , & Reuter, P. , Bushway, S. (1998) [Electronic Version] U. S. Department of Justice Office of Justice Programs National Institute of Justice Research in Brief National Institute of Justice