Validation of the Crime and Violence Scale (CVS) Against the Rasch Measurement Model Including Differences by Gender, Race, and Age

Validation of the Crime and Violence Scale (CVS) Against the Rasch Measurement Model Including Differences by Gender, Race, and Age

Summary

Due to an unavailability of interval measures, researchers used counts of crimes and arrests to assess criminality. Depending on counts of convictions and crimes provides inaccurate estimates because the approach weights all crimes committed equally (Conrad, Riley, Conrad, Chan & Dennis, 2010). The research focused on examining the accuracy of crime and violence scale with regards to psychometric quality by employing the item response theory (Conrad, Riley, Conrad, Chan & Dennis, 2010). Similarly, the authors used invariance of the crime seriousness hierarchy for groups such as age, gender, and racial (Conrad, Riley, Conrad, Chan & Dennis, 2010). The 77 studies performed on individuals beings screened for substance abuse included a sample of 7435 participants. Independent investigators and a variety of institutional settings were used to perform studies. The interviewers conducted the interviews involving 3 to 4 days of training.

Results regarding content validity show that CVS covers crimes of higher seriousness and they include forced sex, arson, and homicide. Conversely, SRDS includes mid-range items. In the SRDS hierarchy, sexual assault was the sixth highest item. With dimensionality, CVS measure provided a substantial variance of 45% (Conrad, Riley, Conrad, Chan & Dennis, 2010). The CVS had two item with problematic fit figures. Discuss calmly settle it had 1.52 infit and 3.01 outfit. Left rather than argue had 1.32 infit and 3.36 outfit (Conrad, Riley, Conrad, Chan & Dennis, 2010). Gender DIF indicates that it was easy for females to endorse forced sex compared to males. In conclusion, although the Crime and Violence Scale pose to be useful in measuring criminality, it may require improvements or elimination of some items. Analysis of differential item functioning showed that the seriousness of crime varies with gender and age.

Evaluation Goals

The program’s goals included the following. First, the program ensured it offered a comprehensive evaluation of the CVS with regards to psychometric quality employing the item response theory. The model allows placement of people on the ruler according to the crime committed (Conrad, Riley, Conrad, Chan & Dennis, 2010). The relationship between the ability of the person and the difficulty of the item is estimated using the probability estimator. Second, the program evaluated the CVS with regards to the invariance of the crime seriousness hierarchy for groups such as gender, age and racial (Conrad, Riley, Conrad, Chan & Dennis, 2010). In a situation when CVS items were not invariant, the evaluator would implement some changes such as dropping items and replacing them with new ones.

The author proposed to evaluate the program in terms of its goals by collecting data from 7,435 participants using information from the 77 research that involved individuals being screened for substance abuse. Independent investigators performed more than two-thirds of these studies. The authors used the item response theory to determine the psychometric quality. With the IRT measurement, the authors were able to assess and adjust for demographic variations in the seriousness of the crime. The authors used differential item functioning to observe people from different demographic groups and measure their score even though they portray the same level of traits.

Theoretical Perspectives

In this context, the program theory of change connects the program to the goals. With the theory of change, the evaluators expect the key elements of the program to produce the outcome. For instance, regarding the program to validate the crime and violence scale, the authors designed the evaluation to support program’s goals and validate that CVS is useful in measuring the construct of crimes and violence.

The rival explanation that was left out was that change happens in stages. Although the goals of many programs aim to change types of behavior, this program focused on validating the crime and violence scale. Therefore, people will start using better measures of crime after changes such as the development of new skills and new information have occurred. The theory guided the authors by helping them to identify the desired goals and conditions that must be in place for the program to achieve the goals. Additionally, the theory offers the outcome framework to help in determining activities.

Research Methods

The authors used qualitative research methods such as interviews and quantitative research methods such as a survey. The evaluation process entailed using data from a sample of 7,435 participants. The research involved 77 studies on individuals being screened for substance abuse (Conrad, Riley, Conrad, Chan & Dennis, 2010). Interviewers conducted interviews with three to four days of training.

The unintended consequence that occurred includes the following. The evaluators found that Crime and Violence Scale was less reliable for adults compared to adolescents. Similarly, the measure is least reliable for female adults recording 0.72 compared to male adolescents at 0.85 (Conrad, Riley, Conrad, Chan & Dennis, 2010). From the results, the authors concluded that measures for female adults had more errors. This is due to limited qualitative cognitive work with female adults.

Conclusion

The authors concluded that although the CVS is convenient at measuring criminality, some areas need improvement. The CVS indicated that the person reliability is stronger for adolescent males with the figure of 0.85 while the adolescent female is 0.83. Moreover, it is weaker for adults males (0.76) and adults females (0.72). The program achieved its intended goal in part. At the end of the evaluation process, the authors found that CVS is convenient in measuring criminality; however, some items need improvements or elimination. According to the authors, they may drop two items from scoring. Precisely, the authors may drop discuss calmly settle it and left rather than argue. The authors will drop the two items because they are not crimes. Additionally, by not being crimes, the two items showed high misfit estimates.

The observation of various hierarchies for age and gender is useful because it helps in understanding that one size cannot be used to calibrate the seriousness of the crime. You cannot calibrate youth, adults, male, female, gender and age. The authors recommended that due to the findings, there is need to develop better items and related measures for female adults.

Evaluation Concepts

First, in an evaluation process, input refers to the materials or resources that an evaluator use in a program to provide activities. The evaluator can use supplies, financial resources, volunteers, staff, and facilities. In the article, the input that the authors used include the services of independent investigators, funds, volunteers, a variety of institutional settings, interviewers and Chestnut’s institutional review board. Second, activities refer to the services provides by the program. Regarding the article, the services provided by the program include measuring criminality.

Third, output entails the amount of activity provided. Besides, outputs help the evaluators to monitor and report the evaluation plan. In the articles, the CVS indicated that the person reliability is stronger for adolescent males with the figure of 0.85 while the adolescent female is 0.83. Lastly, outcome refers to benefits that the participants receive during and after the program. In the article, the outcome of the program is that CVS is convenient in measuring criminality.

 

Reference

Conrad, K. J., Riley, B. B., Conrad, K. M., Chan, Y. F., & Dennis, M. L. (2010). Validation of the Crime and Violence Scale (CVS) against the Rasch measurement model including differences by gender, race, and age. Evaluation review, 34(2), 83-115.

 
Do you need an Original High Quality Academic Custom Essay?