Alumni Dissertations and Theses

 
 

Alumni Dissertations and Theses

Filter Dissertations and Theses By:

 
 
  • Translator, traitor: A critical ethnography of a U.S. terrorism trial

    Author:
    Maya Hess
    Year of Dissertation:
    2014
    Program:
    Criminal Justice
    Advisor:
    Diana Gordon
    Abstract:

    Abstract TRANSLATOR, TRAITOR: A CRITICAL ETHNOGRAPHY OF A U.S. TERRORISM TRIAL by Maya Hess Adviser: Professor Diana Gordon Historically, the role of translators and interpreters has suffered from multiple misconceptions. In theaters of war, these linguists are often viewed as traitors and kidnapped, tortured, or killed; if they work in the terrorism arena, they may be prosecuted and convicted as terrorist agents. In United States v. Ahmed Abdel Sattar, a/k/a "Abu Omar," a/k/a "Dr. Ahmed," Lynne Stewart, and Mohammed [sic] Yousry, 02 Cr. 395 (JGK) (S.D.N.Y. 2003), Yousry, an Arabic linguist and scholar of Middle Eastern history, was labeled such an agent, his work as translator/interpreter construed as material support to terrorism, and his expertise recast as dangerous knowledge. Drawing on moral panic theory and applying ethnographic content analysis in combination with the extended case method, this critical ethnography of a terrorism trial (a) analyzes trial records and courtroom observations; (b) relates the findings to secondary data, such as Islamophobia polls and translator-traitor incidents and litigation; and (c) locates the synthesized product within an extralegal and historical context. The verdict's effects on the translation and interpreting community, academia, the terrorism knowledge base, and, by extension, national security are discussed and policy solutions offered.

  • Judicial instructions and the juror's ability to disregard inadmissible evidence: Can varying the timing and content of judicial instructions influence juror decision-making?

    Author:
    Courtney Hougham
    Year of Dissertation:
    2011
    Program:
    Criminal Justice
    Advisor:
    Maureen O'Connor
    Abstract:

    During the course of a trial, a judge will instruct the jury on how they are to act and reach decisions. The purpose of this study was to examine the effect of different judicial instructions on a juror's ability to evaluate testimony. The research looked at how instructions can interact with a juror's ability to disregard a piece of evidence ruled inadmissible for different reasons. The design was a 3x5 complete factorial design. The stimulus material was a murder trial summary with weak evidence against the defendant, with the key piece of testimony being a hair found on the victim that matches the defendant. This evidence was objected to and admitted or not admitted into evidence depending on the condition. The hypotheses test how a juror's decision-making process is influenced by a combination of judicial instructions, including one designed to raise suspicion, the ruling - admitted or not admitted, and the reason behind the ruling.

  • A Study Of The Molecular Chemistry Of Glasses By Infrared Microspectroscopy And Its Use In Forensic Glass Discrimination And Classification

    Author:
    Brooke Kammrath
    Year of Dissertation:
    2012
    Program:
    Criminal Justice
    Advisor:
    Thomas Kubic
    Abstract:

    The analysis of the molecular structure of glass is a novel forensic method that provides knowledge about glass chemistry that is not currently employed by forensic scientists as well as improves the discriminatory power within this class of transfer evidence. Infrared (IR) spectra contain extensive information about the molecular structure of the complex silicates in commercial glasses. This research is based on measuring the attenuated total reflection (ATR) mid-IR spectra of soda-lime silicate glasses to detect variations of the molecular structure to assist in the comparison of glass evidence. The use of ATR mid-IR spectra for the discrimination and classification of glasses was investigated. Discrimination error rates of approximately 5% and classification by end-product (window or container) error rates on the order of 2% were achieved with principal component analysis-canonical variate analysis (PCA-CVA) hold-one-out cross validation (HOO-CV) on the first and second derivative spectra of 153 soda-lime silicate glasses. Silicate glass is a common and valuable class of evidence encountered in criminal and civil litigation. Glass is an excellent source of physical evidence because it is brittle and fractures into fragments that are easily transferred from the source to victim and/or perpetrator. The challenge to the forensic scientist is to reliably associate these fragments with their source with high probative value. The mid-IR microprobe analysis of glass requires only minimal additional sample preparation to that which is already done for refractive index (RI) analysis and uses IR investigated samples that are the same size and smaller than the ablated hole made in laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS). Thus mid-IR microprobe analysis can be used to provide additional discrimination for glass samples that are indistinguishable by RI and too small for elemental analysis. ATR mid-IR spectral analysis provides information about the molecular structure of soda-lime silicate glass to support other traditional analysis and strengthen the association of this evidence.

  • Victimization and Involvement in Social Control: Effects of Neighborhood Conditions

    Author:
    Ji Hyon Kang
    Year of Dissertation:
    2009
    Program:
    Criminal Justice
    Advisor:
    James Lynch
    Abstract:

    This study disentangles the interrelationship between victimization and involvement in social control via participation in voluntary associations for crime prevention. There is a great deal of research on the effects of community organization on crime and relatively little on the effects of crime on community organization, despite the acknowledgement of the impact of crime on social capital in communities. The current study addresses this issue. In particular, this research sets out to contribute to the emerging literature in contextual analyses of victimization effects, social control, and community organization, first by examining the impact of crime on individuals' decisions on their involvement in neighborhood crime prevention organizations (NWGs); second by revealing different change models for individuals who join, leave, and stay in these organizations; and finally by comparing how crime impacts individuals' household-protective behaviors and community-protective behaviors. Specifically, I consider different types of crime at both individual and neighborhood levels and individuals' perception of neighborhood safety. Multiple sources of data, including a telephone survey of over 5,000 individuals from a 1990 Seattle study conducted by Miethe (1991), 1989-1991 crime statistics from the Seattle Police Department, and the 1990 Census, provide the best available information to examine the proposed research questions. The data are unique in that they include a longitudinal survey of individuals with retrospective questions, offer extensive information on individuals' decisions about community protection and household protection, and make it possible to construct the changes of each individual's status with neighborhood crime prevention associations. To develop various measures of neighborhood conditions including crime problems, data are extracted from the 1990 census and the Seattle Police Department. Building from the nature of the research questions and the availability of data, multilevel modeling techniques are used. The findings highlight the importance of crime on individuals' involvement in NWGs. In particular, the results show that models of involvement are different from models of change in involvement status. Crime differentiates individuals' decisions to change their membership with NWGs and those who maintain their membership. Negative perceptions of neighborhood safety and a higher residential burglary rate in communities motivate individuals to join NWGs, while individuals' actual property crime victimization makes them leave NWGs. Crime also positively affects individuals' household-protective behaviors. The impact of individuals' NWG involvement on their household-protective behaviors, however, is only significant for the joiners. Joiners in crime prevention associations are more likely to engage in household-protective behaviors as well.

  • Committed to the Cause? Violent and Financial Criminal Behaviors of Domestic Far-Rightists

    Author:
    Ashmini Kerodal
    Year of Dissertation:
    2014
    Program:
    Criminal Justice
    Advisor:
    Joshua Freilich
    Abstract:

    This study used factor analysis, logistic and multinomial logistic regression analysis to evaluate the effects of an individual's level of commitment to far-right extremism on his / her criminal offending behavior. Agnew's General Strain Theory (2001, 2005), Cloward and Ohlin's Differential Opportunity Theory (1960) and Simi and Futrell's (2010) concept of free / movement spaces were used to address the three research questions: (1) What effect did individual level stressors, significant others, and negative interactions with government officials have on membership in a far-right group, (2) What effect did individual level stressors, significant others, membership in an extremist group, and negative interactions with government officials have on an individual's commitment to rightwing extremism, (3) What effect did an individual's commitment to far-right extremism, and membership in extremist groups have on his / her criminal behavior? This study investigated whether strain factors alone influenced radicalization, or if there was a combination of strain factors - including negative interactions with law enforcement - and interactions with other extremists that influenced levels of commitment to rightwing extremism. This study defined radicalization as "the process by which individuals become violent extremists...[that is] individuals who support or commit ideologically motivated violence to further political, social, or religious goals" (NIJ 2012 Research on Domestic Radicalization Solicitation, p. 4). Commitment to rightwing extremism was conceptualized as commitment to far-rightist norms, similar to Cloward and Ohlin's (1960) definition of commitment to delinquent norms or the extent of indoctrination into a deviant subculture. This variable drew on themes found in previous research on extremism (Aho, 1990; Blazak, 2001; Blee, 2002; Ezekiel, 1995; Hamm, 2004, 1993; McCauley & Moskalenko, 2011, 2008). A factor analysis was used to check the validity of the commitment to far-right extremism scale. Another unique characteristic of this study was that its dependent variable of criminal behavior included both violent (i.e., fatal) incidents and financial schemes. Data were obtained from the US Extremist Crime Database (ECDB), a Department of Homeland Security/START-funded project led by Dr. Joshua D. Freilich and Dr. Steven Chermak. Illegal violent incidents and financial schemes committed by domestic extremist that resulted in criminal charged were included in the ECDB. Violent incidents were defined as homicides, and financial schemes were defined as "illicit financial operation[s] involving a set of activities [i.e. techniques] carried out by one or more perpetrators to obtain unlawful gain or other economic advantage through the use of deliberate deception" (Belli, 2011, p. 64). The study found that GST did not predict membership in extremist groups, but was associated with a higher risk of committing a homicide. Group membership was predicted by access to extremist groups and a possible predisposition or sympathy towards extremist beliefs. However, none of the theories explained levels of commitment to extremism. Instead, differences were found between two types of DFRs: Conspiracy Theorists and Proud Supremacists. Conspiracy Theorists were more likely to have been non-white and employed, while Proud Supremacists were more likely to have been white males who experienced strain and had extremist referent others. Finally, the presence of strain and a prior prison record were associated with violent criminal behavior of DFRs. High levels of commitment to extremism, female gender, and the absence of strain (i.e., held a good job and did not have prior negative interactions with government officials) were associated with an increased risk of financial offending behavior.

  • Cyber-surveillance: A Case Study in Policy & Development

    Author:
    Richard Kim
    Year of Dissertation:
    2010
    Program:
    Criminal Justice
    Advisor:
    F Benton
    Abstract:

    The dissertation examines the historical developement of surveillance, electronic surveillance, and cyber-surveillance from the Colonial times to the present. It presents the surveillance laws, technologies, policies as a balance between national security and privacy. To examine more recent developments, the dissertation includes case-studies of three cyber-surveillance tools: Carnivore, Magic Lantern, annd NARUS; describing the operational functions, logistic search functions, minimizaation capacities of these tools. The closing chapters assess the dynamic balance between the achievement of national security and public order and the need to preserve rights and the expectatation of privacy.

  • IS BURGLARY A VIOLENT CRIME? AN EMPIRICAL INVESTIGATION OF CLASSIFYING BURGLARY AS A VIOLENT FELONY AND ITS STATUTORY IMPLICATIONS

    Author:
    Phillip Kopp
    Year of Dissertation:
    2014
    Program:
    Criminal Justice
    Advisor:
    Jon Shane
    Abstract:

    Under the common law, burglary is defined as a crime committed against the property of another, and is listed as a property offense for purposes of statistical description by the Uniform Crime Reports (UCR) and the National Crime Victimization Survey (NCVS). However, burglary is prosecuted and sentenced as a violent crime under habitual offender laws at the federal level, and can be regarded as violent in state law, depending on varied circumstances. Using a mixed methods approach, the current study compared state and federal burglary and habitual offender statutes to an empirical description of the offense. First, a comprehensive content analysis of the provisions of state burglary and habitual offender statutes showed that burglary is often treated as a violent crime, instead of prosecuting and punishing it as a property crime and then separately charging and punishing any violent acts that occasionally co-occur with it. Second, using data from the period1998-2007 from the NCVS and the National Incident Based Reporting System (NIBRS), results showed that in contrast to its statutory classification, burglary is overwhelmingly a non-violent offense. The reported incidence of actual violence or threats of violence during a burglary ranged from a low of 0.9% in rural and suburban areas, to a high of 7.6% in highly urban areas. Additionally, a victim was present during only 26% of all burglaries. These findings led the present study to recommend reform for state and federal burglary and habitual offender statutes to comport with the empirical description of the burglary characteristics provided. Furthermore, it is suggested that federal law should be amended to remove non-violent burglaries as a violent felony under habitual offender statutes, and instead, that burglary should be prosecuted and punished at a level equal with other non-violent property crimes, unless actual violence occurred during the offense.

  • Evaluation of the Effects of Shift work Assignment: A Survey of Motivation in Police Officers

    Author:
    Rainer Kroll
    Year of Dissertation:
    2010
    Program:
    Criminal Justice
    Advisor:
    Maria Haberfeld
    Abstract:

    The detrimental health effects of a varied shift schedule on personnel were researched extensively. In fact, the culmination of this work was substantive policy changes, especially within the law enforcement field. While these policy changes were sound in principal, the implementation of invariable shift assignments to meet organizational requirements and the subsequent impact on personnel was documented less frequently. There was little research on employee motivation as a consequence of shift assignments. Partly a consequence of the implementation of organizational mandates with little regard for employee welfare, it was this employee/organization nexus that was inherently important to personal and organizational success. This relationship between employee motivation and shift assignments needed to be researched, and was the focus of this proposal. A Motivation Index, comprised of the various survey questions grouped together according to a specific factor, was created to allow analysis of specific effects of shift assignments on the employee's motivation. These factors were based on external-personal and internal-institutional variables. The analysis began with descriptive statistics of the data--mean, median, mode, range, minimum, maximum, standard deviation, bar chart, histogram, and significant correlations were presented. Additional analysis was conducted using principal components analysis, used to help reduce the data to outline patterns of relationships between the survey questions. These results were used to identify which clusters of variables shared variance and were considered most important to the survey respondent. The purpose of the analysis was twofold--to identify which factors were most meaningful to the survey respondent, and to incorporate these results into policy development in organizational behavior, specifically within the law enforcement community.

  • A meta-analysis of the prediction of violence among adults with mental disorders

    Author:
    Hing Po Lam
    Year of Dissertation:
    2014
    Program:
    Criminal Justice
    Advisor:
    Karen Terry
    Abstract:

    The study of the risk for violence among persons with mental disorders has received substantial scientific attention over the past few decades; however, many uncertainties and controversies remain due to the wide disparities in the reported results. Using the state-of-the-art perspective of public health, a meta-analysis was conducted to clarify the ambiguities by synthesizing quantitative findings from 85 research reports (completed between January 1970 and May 2010) on violence risk assessment among mentally disordered adults. Results of this meta-analytic study revealed that the estimates of the prevalence of violence among the psychiatric population varied considerably from 1.1% to 78.4% with a combined mean rate of 19.3% (95% CI = 15.7-23.5%, k = 68, N = 160,206). Additionally, a total of 290 effect sizes were computed for 36 risk factors of interest and their relative strength in relation to violence was compared. Most importantly, this review demonstrated that mentally disordered patients were no more likely than their non-mentally disordered counterparts to commit violent acts. Overall, the findings have significant implications for clinicians, policy makers, researchers, and the general public, including the psychiatric patients. Lastly, a "Global Public Health-Comprehensive Meta-Analysis" (GPH-CMA) approach is proposed as a new direction for risk assessment and management.

  • Offending in Karachi's neighborhoods: An empirical test of the systemic model of social disorganization

    Author:
    Zainab Latif
    Year of Dissertation:
    2012
    Program:
    Criminal Justice
    Advisor:
    Amy Adamczyk
    Abstract:

    The systemic model of social disorganization posits that structural challenges impede the development of neighborhood networks responsible for informal social control, and thereby increase residents' involvement in crime. However recent studies suggest that in severely disadvantaged and isolated communities, even well-functioning networks may be unable to prevent offending, as a result of cultural and political economy factors such as legal cynicism and the legitimization of violence. This research examines the utility of the ecological framework in Karachi, and assesses the impact of key social disorganization indicators - ethnic heterogeneity, residential mobility and concentrated poverty - on offending. It also examines the extent to which these impacts are mediated by variation in state legitimacy across Karachi's neighborhoods. The project draws on three sources of secondary data: census of Pakistan, election data, and jail admission records. The sample consists of 453 of Karachi's census tracts. Analyses include descriptive techniques to assess the utility of the ecological framework, multivariate regressions to examine direct and mediation effects, and structural equation models to test the significance of the entire complex models. Results display considerable variation across neighborhoods in offending, voting behavior and other sociological indicators. Two social disorganization factors - concentrated disadvantage and ethnic heterogeneity - as well as legitimacy, are each seen to have significant, positive, direct effects on total offending, violent offending and robbery. Results further suggest that concentrated disadvantage and mobility contribute to voting behaviors. These results have important implications for the application of ecological theories in general, and social disorganization in particular, in Karachi. The study also points to the importance of macro-social indicators such as voting behavior in explaining offending. Finally, results speak to the need for careful selection of multivariate regression models and the benefits of supplementing these techniques with SEM. However, it is suggested that further investigation into the nature of the micro-social processes underlying social disorganization, and a more rigorous understanding of the determinants of voting behavior and its ability to adequately represent attitudes toward state legitimacy will allow for more accurate application of these findings to the systemic model.