Psychology - Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

The comparative impact of different forms of violence exposure in youth on long-term adult outcomes

Author: 
Date created: 
2020-02-27
Abstract: 

Violence exposure during childhood and adolescence is associated with a wide range of negative emotional and behavioural outcomes. Despite an extensive body of research, there are numerous problems with respect to how violence exposure has been operationalized and measured; design and methodology (i.e., cross-sectional or short-term longitudinal studies); limited outcome measures; and overall conflicting findings. Further, there is a paucity of research examining the effects of violence exposure during youth on long-term adult outcomes. Given the considerable individual variability that exists with respect to the effects of violence exposure, longitudinal research is needed to clarify the comparative impact of different types of violence exposure across locations. Using a large and racially diverse community sample (n = 753; male = 58%; Black = 46%), the current longitudinal study aimed to elucidate the comparative and cumulative effect of different types of violence exposure (witnessing versus victimization) across different locations (home, school, neighbourhood) occurring during youth (lifetime through grade 8) on long-term adult (age 25) outcomes of internalizing, externalizing, and attention problems; substance use; and intimate partner violence perpetration. Results indicated that victimization, but not witnessing violence, predicted all five adult outcomes. More specifically, being victimized in the home setting was associated with the widest range of negative outcomes (internalizing, externalizing, and attention problems), while school victimization was specifically associated with substance use problems in adulthood. The nature and severity of direct victimization may put youth at greater risk for developing emotional and behavioural dysregulation, and the home and school settings appear to be important contexts for adolescent development. Additionally, when youth experienced multiple types of violence across multiple locations (cumulative violence exposure), they experienced a broader and more diverse range of negative outcomes in adulthood. This study extended existing research on the effects of violence exposure during childhood and adolescence. Taking a life-course perspective, these findings demonstrate that violence exposure has long-term negative effects evident well into adulthood, with victimization at home and school as more robust predictors of negative adult outcomes than exposure to neighbourhood violence. Based on these findings, preventing and effectively addressing youth victimization, especially at home and school, must be a top research, practice, and policy priority.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Robert McMahon
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) Ph.D.

Risk domains and factors of the Multi-Level Guidelines: An updated examination of their support in the literature

Author: 
Date created: 
2019-12-03
Abstract: 

While the field of violence risk assessment has seen tremendous growth in the area of individually oriented violence, risk assessment for group-based violence (GBV) is still in its infancy. One tool developed specifically for assessing this form of violence is the Multi-Level Guidelines (MLG; Cook, Hart, & Kropp, 2015), which is comprised of 16 risk factors nested into four domains: Individual, Individual-Group, Group, and Group-Societal. Given the rapid pace at which the research on terrorism risk factors, in particular, has developed (Gill, 2015b), this review sought to update the systematic review from which the MLG was developed and examine the extent to which its domains and factors are supported by recent empirical research on GBV. A total of 151 studies were reviewed, spanning several forms of GBV. Overall, the content and structure of the MLG appear to be broadly supported by recent empirical research, although the evidence base for the Individual-Group and Group domains is still relatively small, and the strength of much of this research is limited by methodological constraints. The MLG and the HCR-20 V3 (Douglas, Hart, Webster, & Belfrage, 2013) were then applied to a brief series of case studies of GBV in order to demonstrate the applicability of the MLG to this form of violence, and to compare it to the HCR-20 V3 in this respect. While both tools were found to be broadly applicable, it was clear that the MLG captured specific risk factors that are unique to GBV.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Stephen D. Hart
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) M.A.

The effects of cross-examination on children’s reports

Author: 
Date created: 
2020-04-23
Abstract: 

Cross-examination is a fundamental aspect of the adversarial legal system; it is meant to test the reliability of the evidence. To date, all research has examined the effect of cross-examination with children who have experienced a unique event. However, many children who testify have experienced multiple similar instances of abuse. In two experiments I investigated the effects of cross-examination on the consistency, accuracy, and perceived credibility of children’s reports. In Experiment 1(N = 222), younger children (kindergarten or grade 1) and older children (grade 3, grade 4, or grade 5) participated in either one (single-event; SE) or five similar (repeated-event; RE) magic shows. One-week later, children received a baseline interview which used best practice techniques. Next, half the children were cross-examined (cross) and the other half answered all cued recall questions again (direct-direct; DD). Finally, all children received a re-direct interview (Interview 3) and were encouraged to answer the questions as they remembered the details occurring during the show. In most analyses children in the cross condition were less accurate during Interview 2 than children in the DD condition. The re-direct interview helped to rehabilitate accuracy for variable details (vary in predictable ways); however, the negative effects of cross endured for accuracy of deviation details (vary in unpredictable ways) and when accuracy was defined broadly. In Experiment 2, recorded interviews of the children were randomly sampled and matched on accuracy. Undergraduate participants (N = 532) rated the children’s perceived honesty, accuracy, credibility, and susceptibility to suggestive questions. Younger children and RE children were perceived as less honest, accurate, and credible than older children and SE children. Children who were cross-examined were rated as less susceptible to suggestive questions than children in the DD condition. These results show that cross-examination has detrimental effects on children’s reports and that RE children may be disadvantaged when testifying in court.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Deborah Connolly
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) Ph.D.

Is depression a risk factor for adolescent offending? A meta-analytic review

Author: 
Date created: 
2020-04-17
Abstract: 

The literature is mixed about whether depression is a risk or protective factor for violence and general offending in adolescents. A meta-analytic review was conducted on 29 studies reporting on 27 unique prospective samples, with a total of 97,316 participants. The majority of samples were community (non-offender) or population samples (77.8%, k = 21), with a smaller proportion being justice-involved (e.g., incarcerated, probation, or history of arrest) samples (22.2%, k = 6). Overall, depression was associated with an increased risk for general offending (OR = 1.58, p < .001), and violent offending (OR = 1.45, p < .001). For community adolescents, depression was a significant risk factor for general offending; however, in justice-involved youth depression was not a significant risk factor. Gender, study quality, publication year, and country of publication did not moderate any of the results.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Jodi Viljoen
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) M.A.

Singlehood and social identity: Does identifying with other singles mitigate the negative effects of perceived discrimination on wellbeing?

Author: 
Date created: 
2019-11-22
Abstract: 

Single people tend to experience lower wellbeing than coupled people. One explanation for this phenomenon is that single people experience discrimination, which can negatively impact psychological wellbeing. Using data across two studies composed of single university students (N = 128) and single people from the community (N = 67), I examined whether identifying strongly with other singles might buffer the adverse effects of discrimination on wellbeing. The results demonstrated some inconsistent effects: greater singlehood commonality mitigated the negative impact of perceived discrimination on wellbeing in a sample of young singletons (Study 1), but exacerbated the negative impact of perceived discrimination on wellbeing in a community sample of singletons (Study 2). These findings hold implications for understanding the potential costs versus benefits of single people’s group identification.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Yuthika Girme
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) M.A.

Adverse childhood experiences in Indigenous and Caucasian female and male youth on probation: Rates and interventions

Author: 
Date created: 
2020-01-17
Abstract: 

Even though Indigenous peoples are overrepresented in the Canadian justice system, little research has studied potential mechanisms for this overrepresentation. To address this gap, the current dissertation examined the association between adverse childhood experiences (ACEs) and recidivism and investigated whether youth probation officers (YPOs) considered trauma in their case formulations. The sample comprised 187 justice-involved Indigenous and Caucasian female and male youth. ACEs, reoffense records, risk statements and trauma focused interventions were coded from justice files with a follow-up period of 1.51 years. Results showed that compared to Caucasian youth, Indigenous youth had significantly higher ACE scores. Indigenous males violently recidivated significantly more than Caucasian males and Indigenous females had significantly more any recidivism than Caucasian females. ACEs predicted and shortened time to any recidivism, added incremental validity above the Structured Assessment of Violence Risk in Youth (SAVRY; Borum, Bartel & Forth, 2006) for any recidivism and mediated the relationship between Indigenous ethnicity and any recidivism. YPOs screened for ACEs but infrequently linked ACEs to recidivism. Finally, trauma focused interventions like therapy referrals were rare for youth with high ACEs. Overall, the current study found that ACEs were especially important for Indigenous youth. As such, screening for ACEs alongside the SAVRY and including Indigenous specific trauma interventions (e.g., Honoring Children interventions) may reduce recidivism for Indigenous youth. Future research should incorporate participatory action approaches and focus on how colonialism plays into justice overrepresentation for Indigenous youth.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Jodi Viljoen
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) Ph.D.

A comparison of accuracy in old versus new memory reports across interview techniques: Which technique elicits the most accurate reports?

Author: 
Date created: 
2020-03-27
Abstract: 

The present study was designed to investigate reports of old and new memories with eye-closure and mental context reinstatement interview techniques. Seventy-eight undergraduate students were asked to recall a scene from a movie that they had seen two years ago and from a movie they had seen up to one week ago. Memory reports were coded for narrative breadth details, general and specific details, and accuracy. Results indicated that in comparison to new reports, old reports contained proportionately fewer conversation details, fewer object details, and proportionately more setting/circumstance details. New reports contained proportionally more specific details than old reports. Old reports were also shown to be generally accurate overall, however, less accurate in comparison to new reports. Mental context reinstatement was helpful in enhancing accuracy for old memory reports, while no effects were found for eye-closure.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Deborah Connolly
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) M.A.

Illuminating sleep in the dark: The relationship between electric light and sleep on Tanna Island, Vanuatu

Author: 
Date created: 
2019-04-05
Abstract: 

Ideal sleep duration has been a topic of debate for centuries. In industrialized nations, access to electric lighting has had impacts on our daily circadian rhythms and, consequently, our sleep. Because of this, desire to understand human “ancestral” sleep is increasing. It is hypothesized that industrialization, complete with 24h access to electric lighting, delays sleep onset and shortens sleep period. This could suggest that in industrialized societies, people may be getting insufficient sleep, which is important to overall health. Conversely, study of some non-industrialized societies without electricity has shown sleep durations that are shorter, or no different, from those in Westernized societies. To further investigate the direct effect of light exposure on sleep, actigraphy was used to measure these in individuals living traditional subsistence lifestyles, with or without access to electricity, on Tanna Island, Vanuatu. Bedtime, wake time, and rise time were similar between villages, however sleep onset was delayed in electrically lit villages, leading to shorter sleep duration. This effect was strongly influenced by mothers with infants, who were up throughout the night, and therefore exposed to more light at night in villages with electricity. Comparatively, sleep durations measured on Tanna were long relative to those reported in industrialized nations. The results support a hypothesis that exposure to artificial light after sunset can delay sleep onset and reduce sleep duration. Lifestyle differences appear to play a large role in human sleep, and continued investigation of varying levels of industrialization should uncover other industrialization-related impacts on sleep, and subsequently, health.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Ralph Mistlberger
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) Ph.D.

Truthiness and law: Non-probative information biases perceived credibility in forensic contexts

Author: 
Date created: 
2019-10-28
Abstract: 

Non-probative but related photos have been shown to increase the perceived truth value of statements relative to when no photo is presented. In 2 experiments, I tested whether this truth bias generalizes to judgements of credibility in a forensic context. Participants read short vignettes in which a witness viewed an offence. The vignettes were presented with or without a non-probative, but related photo. In both experiments, participants gave higher witness credibility ratings on average in photo-present vignettes compared to photo-absent vignettes. In Experiment 2, some vignettes included additional non-probative information in the form of text. I replicated the effect of photo presence in Experiment 2, but the non-probative text did not have a significant effect on witness credibility. The results suggest that non-probative photos can increase the perceived credibility of witnesses in legal contexts.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Deborah Connolly
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) M.A.

Examining ethnic and cultural differences in the prediction of violence risk among male former offenders

Author: 
Date created: 
2019-10-16
Abstract: 

The use of violence risk assessment instruments to estimate an offender’s likelihood of recidivism has become commonplace. However, questions abound regarding the cross-cultural validity of these tools, and whether ethnic differences can jeopardize their predictive accuracy. Furthermore, many static risk factors included in these tools are highly associated with race and class, potentially overestimating the risk scores of underprivileged minorities. The current study examined ethnic/cultural differences in the predictive validity of 10 commonly used historical risk factors, and whether certain race-correlated risk factors can be considered “proxies” for race. Caucasian, African American, and Hispanic adult ex-offenders (N = 270) completed a series of risk rating scales and reported lifetime engagement in criminal activity. While no risk factors emerged as proxies for race, several risk scores were found to misclassify ethnic minorities as high risk. These findings bear implications for the ethical use of risk assessment with cultural minority groups.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Kevin Douglas
Department: 
Arts & Social Sciences: Department of Psychology
Thesis type: 
(Thesis) M.A.