Alumni Dissertations

 

Alumni Dissertations

Filter Dissertations By:

 
 
  • Case Residuals in Structural Equation Modeling

    Author:
    John Cardinale
    Year of Dissertation:
    2011
    Program:
    Educational Psychology
    Advisor:
    John Cardinale
    Abstract:

    From the beginning, lead methodologists in psychometrics and quantitative psychology have been well aware of the problems of fitting structural and confirmatory factor models. The question we approach in our research is how to best detect this misfit and how to identify specific sources of misfit by scrutinizing the data at the case level. Since Anscombe's seminal 1973 paper, detecting problems at the case level in ordinary least-squares regression has become the norm in statistical modeling. In contrast, the usual practice in fitting structural and confirmatory factor models has been to only examine misfit at the variable and sufficient statistic level. This practice ignores a small body of literature that has arisen since the early 1990s about diagnostics of case level and case by variable level misfit. An important paper by Bollen and Arminger (1991) and a follow-up paper by Raykov and Penev (1999), have developed theory behind Individual Case Residuals (ICRs). These papers help lay the ground work for more detailed case and case by variable level diagnostics, without discarding traditional variable oriented procedures. Our goal is to demonstrate uses of multivariate techniques, such as robust Mahalanobis distances, biplots and cluster analysis to analyze the multivariate dataset of ICRs and thereby detect sources of data problems with respect to a target model. We hope to encourage researchers to make better use of case level diagnostics among the various classes of latent variable models, especially with the advent of multivariate tools in packages such as R and SAS.

  • Does Temperament Relate to Sensory Processing Styles in 3 -to 5-year Old Preschoolers with Disabilities

    Author:
    Jeanne Cavanaugh-Todd
    Year of Dissertation:
    2010
    Program:
    Educational Psychology
    Advisor:
    Georgiana Tryon
    Abstract:

    The purpose of the present study was to examine the relationship between temperament, as assessed on the TABC-R Teacher Form (Martin and Bridger, 1999), and sensory processing, as assessed on the Sensory Profile Caregiver Questionnaire (Dunn,1999). In examining the literature, these constructs appear complementary in theory and purpose. While temperament and sensory processing both explain behavior, the former defines how a child reacts and the latter explains why. It has been suggested that similarities exist between the theories of temperament and sensory processing, but few studies have examined this relationship. There is some evidence that a difficult temperament relates to increased sensory processing dysfunction, however, temperament is more complex than a continuum of easy to difficult. Increased knowledge of this relationship would benefit school psychologists when addressing challenging behaviors and creating individualized interventions. The study included 57 children between the ages of 3 and 5 currently receiving preschool special education services. Relationships amongst sensory processing styles and temperament types were investigated. With a few exceptions, correlations were not significant at the customary p < .05 level. Consequently, my hypotheses were not supported by the data. Post hoc analyses, however, revealed a few significant results. These results suggest that a child's adaptive behavior is correlated with sensory processing dysfunction and temperament, and that a low threshold to sensory stimuli is related to temperaments with a stronger inhibited trait. Correlation analyses revealed that delayed development of adaptive skills significantly related to the low threshold sensory processing patterns of Sensitivity and Avoiding. It approached significance with the high threshold patterns of Registration and Seeking. A significant relationship between a low adaptive behavior composite score and the Typical temperament was discovered. The relationship between the adaptive score and the Inhibited and Reticent temperaments also approached significance. After controlling for the Adaptive score, due to its effect on the variables of interest, partial correlations revealed a significant relationship between Sensory Sensitivity and the Inhibited temperament. In addition, Sensory Sensitivity and Sensory Avoiding correlated with the Inhibition dimension. Implications and limitations of results are discussed.

  • Working Alliance with Adolescents who Receive Mandated School-based Counseling Services

    Author:
    Amanda Cenerelli
    Year of Dissertation:
    2013
    Program:
    Educational Psychology
    Advisor:
    Georgiana Shick Tryon
    Abstract:

    Working alliance is widely recognized as one of the most important factors in understanding therapeutic outcomes. However, there is a lack of published research on the relationship between alliance and treatment outcome in the school setting. In particular, there is no empirical evidence of this relationship with mandated counseling in the schools. The purpose of this study was to investigate the relationship between working alliance and treatment outcome in mandated school-based counseling with adolescents. It also examined the relationship between adolescent development and working alliance, as well as other process variables that predict a strong working alliance in this type of counseling environment. This information is intended to provide a first step toward more effective, empirically-based counseling in the schools. Participants were recruited from both public non-public schools in suburban and urban regions of New York. Student participants ranged in age from 12 - 20 and represented grades 6 through 12; approximately 62% of the student participants came from public school settings, while 38% came from non-public settings. The majority of the participating counselors (73%) identified as cognitive-behavioral in theoretical orientation; and the majority were female (82%). Adolescent participants were asked to complete several questionnaires regarding their demographic features (Demographic Questionnaire), level of autonomy (Adolescent Autonomy Questionnaire), therapeutic alliance with their counselor (Working Alliance Inventory - Short Form), and a Current Versus Ideal Counseling Questionnaire. Counselors were asked to complete a demographic questionnaire, as well as an outcome measure (Counseling Outcome Measure) for each student. Results indicate that working alliance is significantly related to counseling outcome/progress in the mandated school-based setting. Regression analyses suggest that client-rated working alliance can be predicted by examining adolescents' level of cognitive autonomy, the consistency between their current and ideal counseling scenario (i.e., expectations of counseling), their number of years in school counseling, and the counselor's level of experience. Additionally, student's number of sessions with the current counselor and the student's ratings of working alliance can be used together to predict counseling progress or outcome, as rated by the counselor.

  • Parental Involvement of Chronically Ill Mothers and Its Impact on the Child's Education

    Author:
    Yung-Chi Chen
    Year of Dissertation:
    2009
    Program:
    Educational Psychology
    Advisor:
    Marian Fish
    Abstract:

    This study examined how maternal chronic illnesses may affect children's academic functioning through parental involvement. Levels of maternal demands of illness were measured in order to see if they affect the levels of parental involvement and children's grades. Four research questions are addressed in this study. Do the maternal demands of illness affect children's educational achievement? Do the maternal demands of illness impact the extent of parental involvement? Does parental involvement of mothers with chronic illness influence their children's academic achievement? Does positive parental involvement mediate or moderate the impact of maternal chronic illness on children's educational performance? One hundred fifty mothers diagnosed with Multiple Sclerosis (MS), diabetes, cancer, HIV/AIDS, Myelodysplasic Syndrome, and Fibromyalgia and with a child in middle school or high school (aged 10-18) participating in this study were recruited from national organizations, clinics, and social support groups serving patients with chronic illnesses. Participants completed a 184-item questionnaire that was composed of measures of 1) parent demographic information, 2) parent medical information, 3) child demographic information, 4) Demands of Illness Inventory (DOII), 5) parental self-efficacy, 6) parental educational aspirations, 7) grade expectations, 8) school contact and participation, 9) Parent Involvement in School Interview, 10) home supervision, and 11) children's educational outcomes. Each participant was compensated ten dollars for completing the questionnaire. Overall, the results suggest that the majority of students of mothers with chronic illness were able to function adequately in terms of academic achievement. However, children's academic functioning may be at risk when their mothers experienced high levels of illness demands as a result of their chronic illness. Children's grades were found negatively related to levels of demands of illness their mothers experienced. This study also revealed that levels of demands of illness imposed on the mothers with chronic illness and disruption in normal family functioning were negatively related to parental self-efficacy in helping their children succeed in education. Moreover, this study found that parental self-efficacy mediated the effects of maternal demands of illness on children's academic achievement. Children of chronically ill mothers with higher academic efficacy tended to do better academically than those of mothers with lower levels of efficacy. Finally, among different forms of parental involvement, parental educational aspirations and grade expectations were positively related to children's educational performance in terms of grades.

  • The Effect of Story Contexts on Complex Verb Learning in Third Grade Students

    Author:
    Molly Chilton
    Year of Dissertation:
    2013
    Program:
    Educational Psychology
    Advisor:
    Linnea Ehri
    Abstract:

    The purpose of this study was to compare the impact of more and less connected semantic contexts on third graders' learning of complex verb meanings. An experimental design was used. Middle class SES third grade students (N=40) were assigned to one of two conditions to learn complex verbs. Students were matched based on word reading ability (WRMT) and members of pairs were randomly assigned to one of the two conditions. The experimental group (cohesive story context) was presented with four sets of sentences. Each set consisted of a brief synopsis of a story followed by six connected sentences containing target verbs telling the story. The comparison group (unconnected context) was exposed to the same sentences without the synopsis and with different agents presented in a different order to minimize connections between the sentences. Based on connectionist theories, it was hypothesized that students presented with target verbs in a cohesive story context would learn more verb meanings than those assigned to the unconnected context. Students completed word reading and receptive and expressive vocabulary pretests prior to being assigned to a learning condition. Performance during learning as well as on five posttests one day after training were examined in order to assess students' abilities to spell, define, and use target verbs in sentences. The contribution of students' existing vocabulary knowledge and word reading skill to their verb learning was also examined. Students in the cohesive context condition outperformed students in the unconnected condition on most learning and posttest measures. Significant differences between the groups were detected on the more demanding posttest measures (definition production and sentence generation), with students in the connected context condition outperforming those in the unconnected condition. Word reading skills but not vocabulary explained significant unique variance on several measures. Results are discussed in regard to various learning theories, and recommendations are made for vocabulary and general reading instruction.

  • Improving the Acquisition and Retention of Science Material by Fifth Grade Students Through the Use of Imagery Interventions

    Author:
    Marisa Cohen
    Year of Dissertation:
    2011
    Program:
    Educational Psychology
    Advisor:
    Helen Johnson
    Abstract:

    A strong base of knowledge in vocabulary is imperative for all students as they are exposed to a great deal of novel words throughout their academic careers, especially in content areas such as science. By devising effective interventions to teach science vocabulary, literacy and science can be integrated and students' mastery of novel words will improve. This study examined the effect of imagery interventions for the presentation of novel science vocabulary to fifth grade learners. Eighty-nine students from two schools in Long Island participated in this study and were randomly assigned to four different instructional interventions: a Picture Presentation method, in which a word was paired with a picture; an Image Creation- No Picture method, in which the participants were told to create an image of the word and draw it on paper; an Image Creation- Picture method, in which the students were presented with the picture and then told to draw it; and a Word Only method, which involved the simple verbal presentation of the word. These interventions were developed taking into account the ability of images to facilitate vocabulary learning, the theory of dual coding, and depth of processing. Participants' acquisition of the words was measured one day after instruction and retention was examined two weeks later. The students were given word fill-in and definition word match tasks at both time points. Results demonstrated that students in the imagery intervention groups (Picture Presentation, Image Creation- No Picture, and Image Creation- Picture) scored higher on the outcome measures at both immediate and delayed recall. It was also shown that the deeper the students processed the "to be learned" vocabulary words, the higher they scored on the outcome measures. Based on the mean outcome measure scores at both time points, students in the Image Creation- Picture intervention scored the highest, followed by the students in the Image Creation- No Picture intervention, those in the Picture Presentation intervention, and finally the Word Only intervention students. Such a study has implications as to the most effective way to integrate science and literacy and successfully present novel concepts in the classroom.

  • The Relationship of Primary Caregiver Perceptions of Language and Behavioral Levels of Children with Autism to Primary Caregiver Stress and Ratings of Family Climate

    Author:
    Abigail Connolly
    Year of Dissertation:
    2011
    Program:
    Educational Psychology
    Advisor:
    Georgiana Shick Tryon
    Abstract:

    The present study explored language and behavioral levels in a sample of 85 children on the Autism Spectrum between the ages of 6 to 12 years. It also studied the relationship of these levels to the primary caregiver's ratings of parental stress and family environment. Participants were primary caregivers of children diagnosed on the Autism Spectrum. Language levels were measured by primary caregiver ratings of non verbal, pre verbal, phrase speech and verbal as defined by the Autism Diagnostic Observation Schedule (ADOS; Lord, Rutter, DiLavore & Risi, 2002). Primary Caregivers of verbal children also completed the Children's Communication Checklist-2 (CCC-2; Bishop, 2006). Behavioral levels were measured as Externalizing and Internalizing by the Children's Behavior Checklist (CBCL; Achenbach & Rescorla, 2001) completed by the primary caregivers and 71 of the children's teachers. Primary Caregivers also completed the Parenting Stress Index-Short Form (PSI-SF; Abidin, 1995) and the Familly Environment Scale (FES; Moos & Moos, 2002) to measure primary caregiver stress and family environment respectively. Results found a wide range of language functioning. Both primary caregiver and teacher behavior ratings did not reach Borderline Range. Externalizing and Internalizing behaviors were highly correlated among primary caregiver scores, but not for teacher scores. CCC-2 General Communication Composite (GCC) score yielded moderate correlations with both Exteralizing and Internalizing primary caregiver ratings. Teachers' Externalizing scores yielded a mild negative relationshipwith primary caregiver rated language levels and teacher Internalizing scores were negatively assosciated with the CCC-2 Social Interaction Difference Index Score (SIDI). In terms of Primary Caregiver stress, CCC-2 language scores suggested a differential effect of language functioning on Primary Caregiver stress. GCC scores produced moderate negative correlations with 3 of the 4 stress measures. The CCC-2's SIDI score produced a negative moderate correlation with the fourth stress measure: Parent Distress. Primary Caregiver behavior scores yielded moderate correlations across all parental stress measures. Teacher behavior scores paralleled the pattern of the language scores: Externalizing scores correlated significantly with 3 of 4 stress measured while Internalizing scores were associated with Parental Distress. Language functioning did not seem to play a significant part in family climate. However, behavioral ratings suggested there was some link between the children's behavior ratings and family communication among its members. Overall, this study explored the differential effects of varying language and behavior levles on Primary Caregiver stress and ratings of family functioning in a sample of children on the Autism Spectrum

  • Bridging the Gap: Relational Aggression in the College Environment

    Author:
    Irene Delgado
    Year of Dissertation:
    2011
    Program:
    Educational Psychology
    Advisor:
    Helen Johnson
    Abstract:

    The purpose of this study is to expand the limited research on bullying behaviors and their consequences in the college environment; specifically, frequency of exposure, explanation for bullying behaviors, attribution of blame and help seeking responses. The negative trajectory that has been seen throughout elementary school, middle school and high school is expected to continue through into the college environment. This investigation was conducted in two parts. In Study A, participants were 348 college students from an ethnically diverse, suburban private college which consisted of 203 females and 153 males. Students self-selected to participate in filling out a sixty-item survey to gain demographic and basic knowledge about their frequency of exposure to bullying in the college environment. Study A provided key evidence that females and males reported bullying/relational aggression, as a witness, in the college environment. In addition, females did not engage in relational aggressive behaviors in higher frequency than males, and females did not blame the victim more often for relationally aggressive victimization as compared to males. In Study B, participants were thirty-two college students who self-disclosed their interest in participating in in-depth interviews regarding personal accounts of bullying and its effects. Questions were aimed at understanding students' explanations for bullying behaviors, and evaluating their personal experiences with bullying and help seeking as they matured. Each student was individually interviewed by this investigator. The main conclusions drawn from Study B are: 1) females endorsed jealousy as the most relevant reason for bullying as compared to males who endorsed gaining acceptance as the most frequently cited explanation; 2) in elementary school, high school, and college, females reported a higher percentage of help seeking as compared to males; and, 3) college students reported experiencing negative feelings from being exposed to bullying in the college environment, in addition to feeling negatively in the classroom. This study illuminates the continued need to provide bullying intervention programs at the college level, including but not limited to: training professors and staff on how to identify and respond to bullying episodes, provide bullies and victims with individualized support, and, develop and enforce a disciplinary code of conduct.

  • A microanalytic study of self-regulated learning processes of expert, non-expert, and at-risk science students

    Author:
    Maria DiBenedetto
    Year of Dissertation:
    2009
    Program:
    Educational Psychology
    Advisor:
    Barry Zimmerman
    Abstract:

    The present investigation sought to examine differences in the self-regulated learning processes and beliefs of students who vary in their level of expertise in science and to investigate if there are gender differences. Participants were 51 ethnically diverse 11th grade students from three parochial high schools consisting of 34 females and 17 males. Students were grouped as either expert, non-expert,or at-risk based on the school's classification. Students were provided with a short passage on tornados to read and study. The two achievement measures obtained were the Tornado Knowledge Test : ten short-answer questions and the Conceptual Model Test : a question which required the students to draw and describe the three sequential images of tornado development from the textual description of the three phases. A microanalytic methodology was used which consists of asking a series of questions aimed at assessing students' psychological behaviors, feelings, and thoughts in each of Zimmerman's three phases of self-regulation: forethought, performance, and reflection. These questions were asked of the students while they were engaged in learning. Two additional measures were obtained: the Rating Student Self-Regulated Learning Outcomes: A Teacher Scale (RSSRL) and the Self-Efficacy for Self-Regulated Learning (SELF). Analysis of variance, chi square analysis, and post hoc test results showed significant expertise differences, large effect sizes, and positive linear trends on most measures. Regarding gender, there were significant differences on only two measures. Correlational analyses also revealed significant relations among the self-regulatory subprocesses across the three phases. The microanalytic measures were combined across the three phases and entered into a regression formula to predict the students' scores on the Tornado Knowledge Test. These self-regulatory processes explained 77% of the variance in the Tornado Knowledge Test, which was a significant and substantial effect. Prior to this investigation, there have been no studies which have tested Zimmerman's three phase model on an academic task, such as science, within an expertise framework. Implications from the present study suggest that students varying in expertise level in science achievement also vary in self-regulatory behavior, and that gender is not a significant factor.

  • Using the Internet With A Structured Think-Aloud Methodology to Enhance College Students' Vocabulary

    Author:
    Rachel Ebner
    Year of Dissertation:
    2012
    Program:
    Educational Psychology
    Advisor:
    Linnea Ehri
    Abstract:

    The present study built upon an earlier study by Ebner and Ehri (in press), which examined the Internet's potential as a learning tool for enhancing college students' vocabularies. The current research sought to extend that study by determining how to make online vocabulary learning more effective. An experiment was conducted to investigate a structured think-aloud methodology that encouraged participants' metacognitive focus on an online vocabulary task. Participants were 70 students from a New York City public university. They were randomly assigned to either a treatment condition to learn about particular terms contained in an online text using a structured think-aloud method, or to a control condition using an unstructured think-aloud method. Analyses of variances revealed that structured think-aloud participants demonstrated significantly greater vocabulary gains, both overall and within specific dimensions of word knowledge, compared to the control group. Both quantitative and qualitative analyses revealed that differences between conditions in vocabulary gains were attributable to structured think-aloud participants' greater metacognitive focus on the task (e.g., greater instances of reminding themselves about the online vocabulary goal; planning and evaluating their online actions in relation to achieving the goal). Correlations and regression analyses also showed that participants showing the best performance in the online vocabulary task had more extensive vocabularies going into the activity, had some prior familiarity with the terms, and were assigned to the structured think-aloud condition. Results offer strong support for the structured think-aloud methodology as a scaffold for making online vocabulary learning more effective.