Immigration Policy, Border Security and Migrant Deaths - Research Paper Example

Research critique Supervisor: April 21, Research critique Introduction Research defines a systematic approach to knowledge development that applies empirical methods and is characterized by distinct elements and procedures. I, in this paper, evaluate three research articles to demonstrate my developed knowledge of components of research. Each article critique reviews the applied research design, type of applied data, sampling procedure, dependent variable, key findings, and limitations. Review of the first article: Immigration policy, Border security, and migrant deaths: An impact evaluation of life saving efforts under border safety initiatives Research design The study applies a time series design in researches collect and explores longitudinal data on a subject, over a period. In the case, the study considered a two-decade period from the year 1984 but data on control measures are collected from the year 1999. The design is suitable in establishing trends for forecasts into policy developments (Guerette, 2007). Type of data used to conduct analysis The study uses numeric type of data for the variables. Both dependent and independents variables are measured in numeric form as they relate to numbers and rates (Guerette, 2007). Sampling procedure The study is explorative of the entire population and therefore lacks an established sampling procedure. Its reliance on secondary data also eliminates the need for sampling and sampling procedure because it relies on an entire available data (Guerette, 2007). Dependent variable The rate at which immigrants die in their attempt to cross the borders into the United States is the study’s dependent variable. The study retrieved data on the variable from triangulated sources that include registration points at national government level and from state governments. Data on the dependent variable was further obtained from an independent agency that was tasked to investigate the immigrants’ deaths as they cross the nation’s border (Guerette, 2007). Key findings The study’s findings that consider data before a policy into border security and after the policy do not identify any significant difference in the rate at which immigrants’ deaths are reported. The lack of significance indicates that the developed policy has not achieved its intended purpose of improving safety across the border and therefore reducing the reported death cases. While the border security policy does not reports positive results, presumably because of its broad scope, more target specific initiatives such as “BOSTAR search and rescue” and the “Lateral Repatriation Program” reported significant effects in reducing death rates of immigrants across the border (Guerette, 2007, p. 202, 208). Limitations of the study Threats of internal validity are one of the limitations that the study faces. This means that the collected data and its subsequent results and analysis may not be consistent with reality. The involved data in the study is for example not comprehensive and only represent a section of the period in the time series study. Threats of reliability are another set of challenge that the study faces. Reliability defines the degree of dependability that can be placed on a research’s data, results and conclusions. The study’s scope that retrieves data from two different and independent sources identifies this threat, especially with fears of inconsistency in the two data sets. Historical confound is another limitation to the study because of the longitudinal scope. Immigrants, in their attempts to cross the borders without being detected after the security improvement means that the reported number of immigrants could be lower than the actual one and the rate could therefore be compromised (Guerette, 2007). Review of the second article: Striking out as crime reduction policy: The impact of ‘Three Strikes’ laws on crime rates in the United States Cities Research design The researchers applied a multiple time series design, an approach in which several data sets, paired over a period, are used. The design involves corresponding data sets that are collected longitudinally, on same participants for each corresponding data or different participants and analysis done over the involved period. The design is an example of quasi-experimental designs and is applicable to the scope of the research because of its advantages that include assumptions of a natural experiment and ability to clean data for consistency. Other advantages are applicability of large sample sizes for external reliability, ability to incorporate pseudo experiments and potentials of establishing relationships between the variables (Kovandzic, Sloan and Vieraitis, 2004). Type of data used to conduct analysis The research used two types of data for analysis, numeric data, and categorical data. Numeric data defines types of data that assigns quantitative values to variables such as the crime rates and percentage of the population by ethnicity as used in the research while categorical data is descriptive and does not offer quantitative properties. Examples categorical variables in the research include dummy variables (Kovandzic, Sloan and Vieraitis, 2004). Sampling procedure Even though the researchers do not state, explicitly, the sampling procedure that they used for the research, the scope of the research that identifies sampled cities subject to a population size criteria identifies simple random sampling approach of cities that met the inclusion criteria. This implies a simplistic procedure in which cities were sampled at random as long as they had a minimum of 100000 people (Kovandzic, Sloan and Vieraitis, 2004). Dependent variable Crime rate is the dependent variable and was measured by different types of crimes such as “homicide, robbery, assault, rape, larceny, and motor vehicle theft” to identify a multiple dependent variable set up (Kovandzic, Sloan and Vieraitis, 2004, p. 219). Key findings One of the major findings is a positive correlation between the three strike laws and reported homicides rates in the states that have implemented the laws. The results meant that existence of the laws leads to increased cases of homicides in the states that apply the laws. Another major finding is lack of significant association between existence of the laws in the state that enact and apply them and the general crime rates (Kovandzic, Sloan and Vieraitis, 2004). Limitations of the study Major limitations of the study, based on its design, are historical confounds, seasonality, and timing of measurements. Historical confounds defines developments, apart from treatment effects, which occur over time to influence results. Developed knowledge over time such as social awareness campaigns against criminal activities could compromise validity of the collected data. Changing trends by seasons such as insecurity trends along variations in the political environment, and poor timing such as collection of control and treatment data at different times identify reliability-based limitations and reduces the confidence in the data and its analysis results (Kovandzic, Sloan and Vieraitis, 2004). Review of the third article: “Eyes on the street”: The impact of Tennessee’s emergency cellular telephone program on alcohol related fatal crashes Research design The researchers applied a multiple times series design in which it collected different data sets over a period. The design considered data sets for both treatment and controlled that were simultaneously collected over a period and the corresponding data sets were considered by months (D’ Alessio, Stolzenberg and Terry, 1999). Type of data used to conduct analysis The analysis used both categorical and numeric data. Categorical type of data, in its descriptive role was is used through dummy variable to describe implementation the Tennessee Star program while numeric type of data is used to measure accident rates (D’ Alessio, Stolzenberg and Terry, 1999). Sampling procedure The sampling procedure is arbitrary as the researchers do not apply any sampling strategy but considers data from the entire population of Tennessee’s reported cases. Absence of a defined sampling approach is also identified from the researchers’ reliance on an existing data base that already established the required data for the study (D’ Alessio, Stolzenberg and Terry, 1999). Dependent variable The research studies multiple dependent variables at two levels, “high way alcohol related fatal crush rate” and “non high way alcohol related fatal crush rate” (D’ Alessio, Stolzenberg and Terry, 1999, p. 257). The crush rates on the high ways are expressed on a monthly basis and as a percentage of deadly accidents that were reported on “interstates, U.S. highways, state highways, and county roads that involves a single vehicle” (D’ Alessio, Stolzenberg and Terry, 1999). Non-highway crush rates are similarly measured on a monthly basis but on municipal roads (D’ Alessio, Stolzenberg and Terry, 1999). Key findings Research results establish a significant relationship between the program’s implementation and the rate of alcohol related accidents in the region. Based on the results, the program can reduce the type of accidents by about 2.5 percent as opposed to the realized rates of alcohol related accidents on roads that the program does not regulate. The results also identify possibility of the program’s application as an economical strategy controlling alcohol related accidents on roads (D’ Alessio, Stolzenberg and Terry, 1999). Limitations of the study Reliability threat is one of the study’s limitations and defines reduced confidence in data and its applicability. The narrow scope of the study that was limited to a single state challenges reliability of the findings that cannot be inferred to the entire nation. The research design, being a time series design that is implemented over a time period also identifies historical confound as a possible limitation to the study’s validity. The threat relates to factors that are independent of a research’s scope but occur within the research period to influence results. The threat limits the study because of possible environmental factors, such as evoked morality among road users, which could influence the results instead of the policy’s effect. Threat of internal validity, loss of credibility, is another limitation emanates from the applied sample size that was considerably small, as compared to the total number of road users is another challenge to the study. This is because only a few road users in the region have been exposed to the policy (D’ Alessio, Stolzenberg and Terry, 1999). References D’Alessio, S; Stolzenberg, L., and Terry, W. Clinton III (1999). “Eyes on the street”: The impact of Tennessee’s Emergency Cellular Telephone Program on alcohol related fatal crashes.” Crime and Delinquency, 45(4): 453- 466. Guerette, Rob T. (2007). “Immigration policy, Border Security and migrant deaths: An impact evaluation of life saving efforts under the border safety initiative.” Criminology & Public Policy, 6(2): 201-222. Kovandzic, T., Sloan, J., and Vieraitis, L. (2004. “’Striking out’ as crime reduction policy: The impact of ‘three strikes’ laws on crime rates in U.S. cities.” Justice Quarterly, 21(2): 207-239. Read More