Matching Items (15)
Description
In investigating mediating processes, researchers usually use randomized experiments and linear regression or structural equation modeling to determine if the treatment affects the hypothesized mediator and if the mediator affects the targeted outcome. However, randomizing the treatment will not yield accurate causal path estimates unless certain assumptions are satisfied. Since randomization of the mediator may not be plausible for most studies (i.e., the mediator status is not randomly assigned, but self-selected by participants), both the direct and indirect effects may be biased by confounding variables. The purpose of this dissertation is (1) to investigate the extent to which traditional mediation methods are affected by confounding variables and (2) to assess the statistical performance of several modern methods to address confounding variable effects in mediation analysis. This dissertation first reviewed the theoretical foundations of causal inference in statistical mediation analysis, modern statistical analysis for causal inference, and then described different methods to estimate causal direct and indirect effects in the presence of two post-treatment confounders. A large simulation study was designed to evaluate the extent to which ordinary regression and modern causal inference methods are able to obtain correct estimates of the direct and indirect effects when confounding variables that are present in the population are not included in the analysis. Five methods were compared in terms of bias, relative bias, mean square error, statistical power, Type I error rates, and confidence interval coverage to test how robust the methods are to the violation of the no unmeasured confounders assumption and confounder effect sizes. The methods explored were linear regression with adjustment, inverse propensity weighting, inverse propensity weighting with truncated weights, sequential g-estimation, and a doubly robust sequential g-estimation. Results showed that in estimating the direct and indirect effects, in general, sequential g-estimation performed the best in terms of bias, Type I error rates, power, and coverage across different confounder effect, direct effect, and sample sizes when all confounders were included in the estimation. When one of the two confounders were omitted from the estimation process, in general, none of the methods had acceptable relative bias in the simulation study. Omitting one of the confounders from estimation corresponded to the common case in mediation studies where no measure of a confounder is available but a confounder may affect the analysis. Failing to measure potential post-treatment confounder variables in a mediation model leads to biased estimates regardless of the analysis method used and emphasizes the importance of sensitivity analysis for causal mediation analysis.
ContributorsKisbu Sakarya, Yasemin (Author) / Mackinnon, David Peter (Thesis advisor) / Aiken, Leona (Committee member) / West, Stephen (Committee member) / Millsap, Roger (Committee member) / Arizona State University (Publisher)
Created2013
Description
Item response theory (IRT) and related latent variable models represent modern psychometric theory, the successor to classical test theory in psychological assessment. While IRT has become prevalent in the assessment of ability and achievement, it has not been widely embraced by clinical psychologists. This appears due, in part, to psychometrists' use of unidimensional models despite evidence that psychiatric disorders are inherently multidimensional. The construct validity of unidimensional and multidimensional latent variable models was compared to evaluate the utility of modern psychometric theory in clinical assessment. Archival data consisting of 688 outpatients' presenting concerns, psychiatric diagnoses, and item level responses to the Brief Symptom Inventory (BSI) were extracted from files at a university mental health clinic. Confirmatory factor analyses revealed that models with oblique factors and/or item cross-loadings better represented the internal structure of the BSI in comparison to a strictly unidimensional model. The models were generally equivalent in their ability to account for variance in criterion-related validity variables; however, bifactor models demonstrated superior validity in differentiating between mood and anxiety disorder diagnoses. Multidimensional IRT analyses showed that the orthogonal bifactor model partitioned distinct, clinically relevant sources of item variance. Similar results were also achieved through multivariate prediction with an oblique simple structure model. Receiver operating characteristic curves confirmed improved sensitivity and specificity through multidimensional models of psychopathology. Clinical researchers are encouraged to consider these and other comprehensive models of psychological distress.
ContributorsThomas, Michael Lee (Author) / Lanyon, Richard (Thesis advisor) / Barrera, Manuel (Committee member) / Levy, Roy (Committee member) / Millsap, Roger (Committee member) / Arizona State University (Publisher)
Created2011
Children's coping efforts and coping efficacy: effects of parenting, surgency, and effortful control
Description
Stress responses play a central role in the development of psychopathology. Coping efforts, one subset of stress responses, have been shown to influence the relations between stress and adjustment. Although the relations between youths' coping and emotional and behavioral outcomes are well-documented, less is known about the factors that predict youths' coping. Given their importance for adaptation, understanding influences on youths' coping has important implications for developmental theories and preventive interventions. The current study examined the main and interactive effects of positive parenting and youths' temperament on youths' coping efforts and coping efficacy one year later in a sample of 192 youth aged 9-15 years when assessed initially. Data used were from the first and third waves of a four-wave, prospective, longitudinal study of families where one or both parents recently became unemployed. Positive parenting was measured with a combination of mother-report, child-report, and observational measures. Temperament was assessed with mother-report, child-report, and/or teacher-report measures. Children reported on their coping. It was hypothesized that positive parenting, effortful control, and surgency would be positively associated with active coping and coping efficacy, and negatively associated with avoidant coping. Further, it was hypothesized that the relations between positive parenting and youths' coping would be stronger for youths low in effortful control or surgency. Structural equation modeling with latent variables revealed no significant main effects of positive parenting, effortful control, or surgency on youths' coping efforts or coping efficacy. Path analyses revealed no significant positive parenting by temperament interactions in the prediction of youths' coping efforts or coping efficacy. Several significant correlations between measures of positive parenting or surgency and youths' coping emerged. The pattern of correlations provided some support for the hypothesized relations. For example, aspects of positive parenting (e.g., maternal acceptance) and youth surgency were associated with more adaptive coping both concurrently and longitudinally, whereas an aspect of negative parenting (i.e., maternal rejection) was associated with less adaptive coping both concurrently and over time. Potential explanations of the unexpected findings and future directions for understanding the role of parenting and youths' temperament in youths' coping efforts and coping efficacy are discussed.
ContributorsVélez, Clorinda Eileen (Author) / Wolchik, Sharlene (Thesis advisor) / Eisenberg, Nancy (Thesis advisor) / Ayers, Tim (Committee member) / Millsap, Roger (Committee member) / Sandler, Irwin (Committee member) / Arizona State University (Publisher)
Created2010
Description
The purpose of this thesis study was to evaluate the nature of social anxiety in clinic-referred African American children versus their Caucasian counterparts. In particular, social anxiety symptom endorsement along the Social Phobia and Anxiety Inventory Scale for Children (SPAI-C; Beidel, Turner, & Morris, 1995) was examined in a sample of 107 African American and 364 Caucasian children (ages 7- to 17-years old) referred for anxiety. To evaluate symptom endorsement, simple descriptive analyses were conducted whereas measurement invariance tests were examined using confirmatory factor analyses. For the most commonly endorsed items, African American and Caucasian children shared seven of the top 10 most commonly identified social anxiety symptoms. Similar social fears across ethnicity focused on "assertiveness in situations perceived to be difficult" and ""speaking to large groups of peers they do not know." Findings also showed that African American children were more likely to report symptoms of "shaking when in social situations" than Caucasian children, and Caucasian children were more likely to report symptoms of "embarrassment when in front of adults" compared to African American children, but this was also on the basis of two items. When it came to the five factors of the SPAI-C, results showed measurement invariance across African American and Caucasian children. Overall, there were more similarities than differences between African American and Caucasian children in social anxiety symptoms based on the SPAI-C. Findings from this thesis study shed light on how to best accurately identify social anxiety among African American children compared to Caucasians, a contribution that can potentially impact assessment, treatment planning, and program response evaluation.
ContributorsWynne, Henry (Author) / Pina, Armando (Thesis advisor) / Gonzales, Nancy (Committee member) / Millsap, Roger (Committee member) / Arizona State University (Publisher)
Created2013
Description
This study used growth mixture modeling to examine attendance trajectories among 292 Mexican–American primary female caregivers enrolled in a universal preventive intervention and the effects of health beliefs, participation intentions, cultural influences, and intervention group cohesion on trajectory group membership as well as trajectory group differences on a distal outcome, immediate posttest teacher report of child externalizing (T2). Results supported four trajectory groups—early terminators (ET), mid-program terminators (MPT), low-risk persistent attenders (LRPA), and high-risk persistent attenders (HRPA). Compared with LRPAs, caregivers classified as HRPAs had weaker familism values, less parenting efficacy, and higher externalizing children with lower GPAs. Caregivers in the two persistent attender groups reported strong group cohesion and providers rated these caregivers as having strong participation intentions. Children of caregivers in the LRPA group had the lowest T2 child externalizing. Children of caregivers in the MPT group had lower T2 externalizing than did those of the ET group, suggesting partial intervention dosage can benefit families. Despite high levels of attendance, children of caregivers in the HRPA had the highest T2 externalizing, suggesting this high-risk group needed either more intensive services or a longer period for parents to implement program skills to evidence change in child externalizing.
ContributorsMauricio, Anne (Author) / Tein, Jenn-Yun (Author) / Gonzales, Nancy (Author) / Millsap, Roger (Author) / Dumka, Larry (Author) / Berkel, Cady (Author) / College of Liberal Arts and Sciences (Contributor)
Created2014-12-01