Theses and Dissertations
This collection includes both ASU Theses and Dissertations, submitted by graduate students, and the Barrett, Honors College theses submitted by undergraduate students.
Displaying 1 - 10 of 534
Filtering by
- All Subjects: Machine Learning
- All Subjects: Developmental Psychology
Description
Communication skills within dating contexts are developed during the adolescent years, and are associated with a lifelong ability to have satisfying, enduring, and non-violent partnerships. As such, they are currently and increasingly implemented into both more general forms of healthy relationship education, as well as that targeting the prevention of teen dating violence specifically. Reaching Mexican American youth with culturally and developmentally appropriate relationship education, including communication skills, may be particularly important given their earlier transitions to marital and parenting relationships, acculturative stressors that present them with unique coupling challenges, and their higher rates of teen dating violence as compared to European American youth. We know very little about how Mexican American dating couples communicate about areas of conflict. This dissertation research utilizes Bell and Naugle's (2008) framework of interpersonal violence to explore how cultural and developmental considerations may be integrated in order to better understand how communication behaviors contribute to Mexican American middle adolescents' experiences with dating conflict. I use an observational study design in order to 1.) Qualitatively explore the communication strategies used by a sample of committed couples, including integration of culturally- and developmentally-relevant contexts, 2.) Quantitatively examine whether couple-level discrepancies in acculturation are associated with observed negativity, including whether this relationship may be mediated by dissimilar gender-related beliefs, and to 3.) Review empirical findings pertaining to the communication behaviors of Mexican American adolescents and to integrate ecodevelopmental theory in said framework as informed by Papers 1, 2, and literature specific to this topic area. The ultimate aim of this dissertation research is to generate findings that may improve the dating health of Mexican American adolescents living in the United States.
ContributorsAdams, Heidi (Author) / Rankin Williams, Lela (Thesis advisor) / Marsiglia, Flavio (Committee member) / Anthony, Elizabeth (Committee member) / Arizona State University (Publisher)
Created2013
Description
Sparsity has become an important modeling tool in areas such as genetics, signal and audio processing, medical image processing, etc. Via the penalization of l-1 norm based regularization, the structured sparse learning algorithms can produce highly accurate models while imposing various predefined structures on the data, such as feature groups or graphs. In this thesis, I first propose to solve a sparse learning model with a general group structure, where the predefined groups may overlap with each other. Then, I present three real world applications which can benefit from the group structured sparse learning technique. In the first application, I study the Alzheimer's Disease diagnosis problem using multi-modality neuroimaging data. In this dataset, not every subject has all data sources available, exhibiting an unique and challenging block-wise missing pattern. In the second application, I study the automatic annotation and retrieval of fruit-fly gene expression pattern images. Combined with the spatial information, sparse learning techniques can be used to construct effective representation of the expression images. In the third application, I present a new computational approach to annotate developmental stage for Drosophila embryos in the gene expression images. In addition, it provides a stage score that enables one to more finely annotate each embryo so that they are divided into early and late periods of development within standard stage demarcations. Stage scores help us to illuminate global gene activities and changes much better, and more refined stage annotations improve our ability to better interpret results when expression pattern matches are discovered between genes.
ContributorsYuan, Lei (Author) / Ye, Jieping (Thesis advisor) / Wang, Yalin (Committee member) / Xue, Guoliang (Committee member) / Kumar, Sudhir (Committee member) / Arizona State University (Publisher)
Created2013
Description
This study examined the relations between cognitive ability, socioemotional competency (SEC), and achievement in gifted children. Data were collected on children between the ages of 8 and 15 years (n = 124). Children were assessed via teacher reports of SEC, standardized cognitive assessment, and standardized achievement assessment. Composite achievement significantly correlated with all areas of SEC on the Devereux Student Strengths Assessment (DESSA). Cognitive ability significantly correlated with all areas of SEC as well. Composite cognitive ability significantly correlated with all composite achievement, as well as with achievement in all subject areas assessed. Achievement scores tended to be higher in older age groups in comparison to younger age groups. When gender differences were found (in some areas of SEC and in language achievement), they tended to be higher in females. Gender moderated the relation between SEC and composite achievement. The areas of SEC that best predicted achievement, over-and-above other SEC scales, were Optimistic Thinking, Self-Awareness, and Relationship Skills. While cognitive scores did not significantly predict achievement when controlling for SEC, SEC did significantly predict achievement over-and-above cognitive ability scores. Overall findings suggest that SEC may be important in children's school achievement; thus it is important for schools and families to promote the development of SEC in gifted children, especially in the areas of optimism and self-awareness.
ContributorsKong, Tiffany (Author) / Caterino, Linda (Thesis advisor) / Naglieri, Jack (Committee member) / Brulles, Dina (Committee member) / Arizona State University (Publisher)
Created2013
Description
This work presents two complementary studies that propose heuristic methods to capture characteristics of data using the ensemble learning method of random forest. The first study is motivated by the problem in education of determining teacher effectiveness in student achievement. Value-added models (VAMs), constructed as linear mixed models, use students’ test scores as outcome variables and teachers’ contributions as random effects to ascribe changes in student performance to the teachers who have taught them. The VAMs teacher score is the empirical best linear unbiased predictor (EBLUP). This approach is limited by the adequacy of the assumed model specification with respect to the unknown underlying model. In that regard, this study proposes alternative ways to rank teacher effects that are not dependent on a given model by introducing two variable importance measures (VIMs), the node-proportion and the covariate-proportion. These VIMs are novel because they take into account the final configuration of the terminal nodes in the constitutive trees in a random forest. In a simulation study, under a variety of conditions, true rankings of teacher effects are compared with estimated rankings obtained using three sources: the newly proposed VIMs, existing VIMs, and EBLUPs from the assumed linear model specification. The newly proposed VIMs outperform all others in various scenarios where the model was misspecified. The second study develops two novel interaction measures. These measures could be used within but are not restricted to the VAM framework. The distribution-based measure is constructed to identify interactions in a general setting where a model specification is not assumed in advance. In turn, the mean-based measure is built to estimate interactions when the model specification is assumed to be linear. Both measures are unique in their construction; they take into account not only the outcome values, but also the internal structure of the trees in a random forest. In a separate simulation study, under a variety of conditions, the proposed measures are found to identify and estimate second-order interactions.
ContributorsValdivia, Arturo (Author) / Eubank, Randall (Thesis advisor) / Young, Dennis (Committee member) / Reiser, Mark R. (Committee member) / Kao, Ming-Hung (Committee member) / Broatch, Jennifer (Committee member) / Arizona State University (Publisher)
Created2013
Description
This study examined four research questions investigating relationships among the experience of trauma, identity development, distress, and positive change. There were 908 participants in the study, ranging in age from 18 to 24 which is known as the period of emerging adulthood. Participants completed an online survey regarding their exposure to trauma and reactions to these experiences. The first research question examined the experience of trauma for the sample. The second question examined group differences among the participant's identity status, gender, and posttraumatic stress disorder (PTSD) diagnostic status on the hypothesized variables. In general, comparisons among the four identity status groups found participants who experienced greater identity exploration (diffused and moratorium) experienced more distress, whereas the identity status groups that reported greater identity commitments (foreclosed and achieved) were associated with positive change. Similar findings were found for PTSD diagnostic status indicating more distress and identity exploration for participants with the diagnosis and more positive change and identity commitments for participants without the diagnosis. Female participants were found to experience more PTS symptoms, centrality of the trauma event, and positive growth than males. Examination of the relationships between trauma severity and posttraumatic growth revealed an inverted U-shaped relationship (quadratic) that was a significant improvement from the linear model. An S-shaped relationship (cubic) was found for the relationship between trauma exposure and posttraumatic growth. Regression analyses found the centrality of the trauma event to one's identity predicted identity distress above and beyond the experience of trauma. In addition, identity distress and the centrality of the trauma contributed to the variance for identity exploration, while only identity distress contributed to identity commitments. Finally, identity development significantly predicted positive change above and beyond, identity distress, centrality of the trauma event, and the experience of trauma. Collectively, these results found both distress and growth to be related to the experience of trauma. Distress within one's identity can contribute to difficulties in the psychosocial stage of identity development among emerging adults. However, the resolution of identity exploration towards commitments to goals, roles, and beliefs, can help trauma survivors experience resilience and growth after stressful experiences.
ContributorsWiley, Rachel (Author) / Robinson-Kurpius, Sharon (Thesis advisor) / Stamm, Jill (Committee member) / Kinnier, Richard (Committee member) / Arizona State University (Publisher)
Created2013
Description
Accumulating evidence implicates exposure to adverse childhood experiences in the development of hypocortisolism in the long-term, and researchers are increasingly examining individual-level mechanisms that may underlie, exacerbate or attenuate this relation among at-risk populations. The current study takes a developmentally and theoretically informed approach to examining episodic childhood stressors, inherent and voluntary self-regulation, and physiological reactivity among a longitudinal sample of youth who experienced parental divorce. Participants were drawn from a larger randomized controlled trial of a preventive intervention for children of divorce between the ages of 9 and 12. The current sample included 159 young adults (mean age = 25.5 years; 53% male; 94% Caucasian) who participated in six waves of data collection, including a 15-year follow-up study. Participants reported on exposure to negative life events (four times over a 9-month period) during childhood, and mothers rated child temperament. Six years later, youth reported on the use of active and avoidant coping strategies, and 15 years later, they participated in a standardized psychosocial stress task and provided salivary cortisol samples prior to and following the task. Path analyses within a structural equation framework revealed that a multiple mediation model best fit the data. It was found that children with better mother-rated self-regulation (i.e. low impulsivity, low negative emotionality, and high attentional focus) exhibited lower total cortisol output 15 years later. In addition, greater self-regulation in childhood predicted greater use of active coping in adolescence, whereas a greater number of negative life events predicted increased use of avoidant coping in adolescence. Finally, a greater number of negative events in childhood predicted marginally lower total cortisol output, and higher levels of active coping in adolescence were associated with greater total cortisol output in young adulthood. Findings suggest that children of divorce who exhibit better self-regulation evidence lower cortisol output during a standardized psychosocial stress task relative to those who have higher impulsivity, lower attentional focus, and/or higher negative emotionality. The conceptual significance of the current findings, including the lack of evidence for hypothesized relations, methodological issues that arose, and issues in need of future research are discussed.
ContributorsHagan, Melissa (Author) / Luecken, Linda (Thesis advisor) / MacKinnon, David (Committee member) / Wolchik, Sharlene (Committee member) / Doane, Leah (Committee member) / Arizona State University (Publisher)
Created2013
Description
The purpose of the current study was to use structural equation modeling-based quantitative genetic models to characterize latent genetic and environmental influences on proneness to three discrete negative emotions in middle childhood, according to mother-report, father-report and in-home observation. One primary aim was to test the extent to which covariance among the three emotions could be accounted for by a single, common genetically- and environmentally-influenced negative emotionality factor. A second aim was to examine the extent to which different reporters appeared to be tapping into the same genetically- and environmentally-influenced aspects of each emotion. According to mother- and father-report, moderate to high genetic influences were evident for all emotions, with mother- and father-report of fear and father-report of anger showing the highest heritability. Significant common environmental influences were also found for mother-report of anger and sadness in both univariate and multivariate models. For observed emotion, anger was moderately heritable with no evidence for common environmental variance, but sadness, object fear and social fear all showed modest to moderate common environmental influences and no significant genetic variance. In addition, cholesky decompositions examining genetic and environmental influences across reporter suggested that despite considerable overlap between mother-report and father-report, there was also reporter-specific variance on anger, sadness, and fear. Specifically, there were significant common environmental influences on mother-report of anger- and sadness that were not shared with father-report, and genetic influences on father-report of sadness and fear that were not shared with mother-report. In-home observations were not highly correlated enough with parent-report to support multivariate analysis for any emotion. Finally, according to both mother- and father-report, a single set of genetic and environmental influences was sufficient to account for covariance among all three negative emotions. However, fear was primarily explained by genetic influences not shared with other emotions, and anger also showed considerable emotion-specific genetic variance. In both cases, findings support the value of a more emotion-specific approach to temperament, and highlight the need to consider distinctions as well as commonalities across emotions, reporters and situations.
ContributorsClifford, Sierra (Author) / Lemery, Kathryn (Thesis advisor) / Shiota, Michelle (Committee member) / Eisenberg, Nancy (Committee member) / Arizona State University (Publisher)
Created2013
Description
Although anxiety may be developmentally appropriate, it can become problematic in some youth. From an ecological perspective, social systems, like family and friendships, are theorized to influence developmental trajectories toward (mal)adjustment, but empirical evidence is scant with regard to the relative impact of subjective satisfaction with family and friendship on anxiety problem development. This thesis study used a subsample of approximately 50% Hispanic/Latino clinic-referred youth (n = 71, ages 6-16 years). Overall, results suggest that the effect of friendship satisfaction on anxiety varied as a function of age but not ethnicity, such that there was a significant negative relationship between child-reported friendship satisfaction and anxiety levels for older children (approx. 9 years and older) but not for younger children. The effect of family satisfaction on anxiety also varied as a function of age, such that older children showed a positive relation between child reported family satisfaction and parent reported anxiety. Furthermore, a positive relation between family satisfaction and anxiety was found only for the H/L children. Post hoc analyses regarding cultural underpinnings of this finding and implications for future research are discussed, as are the results regarding differences between parent and child reports of anxiety.
ContributorsHumphrey, Julia (Author) / Pina, Armando A (Thesis advisor) / Doane, Leah (Committee member) / Bradley, Robert (Committee member) / Arizona State University (Publisher)
Created2013
Description
Image understanding has been playing an increasingly crucial role in vision applications. Sparse models form an important component in image understanding, since the statistics of natural images reveal the presence of sparse structure. Sparse methods lead to parsimonious models, in addition to being efficient for large scale learning. In sparse modeling, data is represented as a sparse linear combination of atoms from a "dictionary" matrix. This dissertation focuses on understanding different aspects of sparse learning, thereby enhancing the use of sparse methods by incorporating tools from machine learning. With the growing need to adapt models for large scale data, it is important to design dictionaries that can model the entire data space and not just the samples considered. By exploiting the relation of dictionary learning to 1-D subspace clustering, a multilevel dictionary learning algorithm is developed, and it is shown to outperform conventional sparse models in compressed recovery, and image denoising. Theoretical aspects of learning such as algorithmic stability and generalization are considered, and ensemble learning is incorporated for effective large scale learning. In addition to building strategies for efficiently implementing 1-D subspace clustering, a discriminative clustering approach is designed to estimate the unknown mixing process in blind source separation. By exploiting the non-linear relation between the image descriptors, and allowing the use of multiple features, sparse methods can be made more effective in recognition problems. The idea of multiple kernel sparse representations is developed, and algorithms for learning dictionaries in the feature space are presented. Using object recognition experiments on standard datasets it is shown that the proposed approaches outperform other sparse coding-based recognition frameworks. Furthermore, a segmentation technique based on multiple kernel sparse representations is developed, and successfully applied for automated brain tumor identification. Using sparse codes to define the relation between data samples can lead to a more robust graph embedding for unsupervised clustering. By performing discriminative embedding using sparse coding-based graphs, an algorithm for measuring the glomerular number in kidney MRI images is developed. Finally, approaches to build dictionaries for local sparse coding of image descriptors are presented, and applied to object recognition and image retrieval.
ContributorsJayaraman Thiagarajan, Jayaraman (Author) / Spanias, Andreas (Thesis advisor) / Frakes, David (Committee member) / Tepedelenlioğlu, Cihan (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2013
Description
The current study expands prior work on children's coping with peer victimization by employing person-centered analyses to identify discrete classes of coping behavior, associations with children's maladjustment, and patterns of stability and change over time. Specifically, data were collected at two longitudinal time points from 515 middle school children who reported experiencing at least occasional peer victimization (284 girls, 231 boys; Mage = 8 years, 5 months, SDage = 10.38 months). Three active, behavioral coping strategies were examined: support seeking from teachers, support seeking from friends, and retaliation. A series of cross-sectional latent profile analyses suggested that coping styles may be characterized by 3 distinct classes: (1) support seeking, (2) retaliation, or (3) a combination of these strategies, labeled mixed strategy coping. Peer victimization, depression, and loneliness were included as concurrent covariates of class membership and results indicate that mixed strategy coping may put children at greater social and emotional risk, whereas both support seeking and retaliation may pose potential benefits in the face of victimization. Further, longitudinal latent transition analyses were conducted to examine the stability and change in coping over time, indicating that coping is largely dispositional, though has the potential to change, particularly among children who experience shifts towards greater maladjustment over time. Results emphasize mixed strategy coping - a coping style that is underrepresented in the current research - as both an important factor that may contribute to greater social and emotional difficulties and also as a potential transitioning point during which change in children's coping may be addressed.
ContributorsVisconti, Kari Jeanne (Author) / Kochenderfer-Ladd, Becky (Thesis advisor) / Ladd, Gary W (Thesis advisor) / Valiente, Carlos (Committee member) / Kochel, Karen (Committee member) / Arizona State University (Publisher)
Created2013