Matching Items (16)
Filtering by
- All Subjects: psychology
- Genre: Academic theses
Description
This project studied a four-variable single mediator model, a single mediator model: X (independent variable) to M (mediator) to Y (dependent variable), and a confounder (U) that influences M and Y. Confounding represents a threat to the causal interpretation in mediation analysis. For instance, if X represents random assignment to control and treatment conditions, the effect of X on M and the effect of X on Y have a causal interpretation under certain reasonable assumptions. However, the randomization of X does not allow for a causal interpretation of the M to Y effect unless certain confounding assumptions are satisfied. The aim of this project was to develop a significance test and an effect size comparison for two sensitivity to confounding analyses methods: Left Out Variables Error (L.O.V.E.) and the correlated residuals method. Further, the project assessed the accuracy of the methods for identifying confounding bias by simulating data with and without confounding bias.
ContributorsAlvarez Bartolo, Diana (Author) / Mackinnon, David P. (Thesis advisor) / Grimm, Kevin J. (Committee member) / McNeish, Daniel (Committee member) / Arizona State University (Publisher)
Created2022
Description
A record number of Latino students are enrolling in higher education in the U.S., but as a group Latinos are the least likely to complete a bachelor’s degree. Cultural factors theoretically contribute to Latino students’ success, including orientation toward ethnic heritage and mainstream cultures (i.e., dual cultural adaptation), feeling comfortable navigating two cultural contexts (i.e., biculturalism), and the degree of fit between students’ cultural backgrounds and the cultural landscapes of educational institutions (i.e., cultural congruity). In a two-part study, these cultural factors were examined in relation to the hypothalamic-pituitary-adrenal (HPA) axis stress response (indexed by salivary cortisol), a physiological mechanism that may underlie how psychosocial stress influences academic achievement and health. First, Latino students’ cortisol responses to stress were estimated in their daily lives prior to college using ecological momentary assessment (N = 206; 64.6% female; Mage = 18.10). Results from three-level growth models indicated that cortisol levels were lower following greater perceived stress than usual for students endorsing greater Latino cultural values (e.g., familism), compared to students endorsing average or below-average levels of these values. Second, cortisol and subjective responses to a standard public speaking stress task were examined in a subsample of these same students in their first semester of college (N = 84; 63.1% female). In an experimental design, viewing a brief video prior to the stress task conveying the university’s commitment to cultural diversity and inclusion (compared to a generic campus tour) reduced cortisol reactivity and negative affect for students with greater Latino cultural values, and also reduced post-task cortisol levels for students with greater mainstream U.S. cultural values (e.g., competition). These findings join the growing science of culture and biology interplay, while also informing initiatives to support first-year Latino students and the universities that serve them.
ContributorsSladek, Michael R. (Author) / Doane, Leah D (Thesis advisor) / Gonzales, Nancy A. (Committee member) / Grimm, Kevin J. (Committee member) / Luecken, Linda J. (Committee member) / Arizona State University (Publisher)
Created2018
Description
There is a growing trend among community samples of young, adult women to initiate drug use for weight loss (Boys, Marsden, & Strang, 2001; Mendieta-Tan, Hulbert-Williams, & Nicholls, 2013). Research has suggested that consequential weight loss may maintain drug use (Cohen, et al., 2010; Ersche, Stochl, Woodward, & Fletcher, 2013; Sirles, 2002), which is compounded by women's perception that drugs are convenient and guarantee weight loss (Mendieta-Tan, et al., 2013). Stimulants, including cocaine, amphetamine, methamphetamine, and ecstasy, are notable drugs of use among college students (Johnston, et al., 2014; Teter, McCabe, LaGrange, Cranford, & Boyd, 2006). With known appetitive and metabolic effects, stimulants may be particularly attractive to college women, who are at elevated risk for increased body dissatisfaction and experimenting with extreme weight loss techniques (Grunewald, 1985; National Eating Disorder Association, 2013). A preliminary epidemiological study of 130 college women between 16- and 24-years old (Mage = 18.76, SDage = 1.09) was conducted to begin to investigate this phenomenon. Results showed women who reported use for weight control (n = 19, 14.6 %) predominantly used stimulants (68.4%), and this subgroup was severely elevated on global and subscales of eating pathology compared with college norms. Moreover, the odds of stimulant use were doubled when women engaged in a compensatory behavior, such as excessive exercise, self-induced vomiting, and laxative use. Although preliminary, these results suggest that a desire for weight control may be associated with stimulant use among college women. Women engaging in more extreme weight loss behaviors are at high risk for initiating and maintaining illicit stimulant use for weight-related reasons.
ContributorsBruening, Amanda B (Author) / Perez, Marisol (Thesis advisor) / Grimm, Kevin (Committee member) / Chassin, Laurie (Committee member) / Arizona State University (Publisher)
Created2016
Description
Statistical mediation analysis allows researchers to identify the most important the mediating constructs in the causal process studied. Information about the mediating processes can be used to make interventions more powerful by enhancing successful program components and by not implementing components that did not significantly change the outcome. Identifying mediators is especially relevant when the hypothesized mediating construct consists of multiple related facets. The general definition of the construct and its facets might relate differently to external criteria. However, current methods do not allow researchers to study the relationships between general and specific aspects of a construct to an external criterion simultaneously. This study proposes a bifactor measurement model for the mediating construct as a way to represent the general aspect and specific facets of a construct simultaneously. Monte Carlo simulation results are presented to help to determine under what conditions researchers can detect the mediated effect when one of the facets of the mediating construct is the true mediator, but the mediator is treated as unidimensional. Results indicate that parameter bias and detection of the mediated effect depends on the facet variance represented in the mediation model. This study contributes to the largely unexplored area of measurement issues in statistical mediation analysis.
ContributorsGonzález, Oscar (Author) / Mackinnon, David P (Thesis advisor) / Grimm, Kevin J. (Committee member) / Zheng, Yi (Committee member) / Arizona State University (Publisher)
Created2016
Description
Researchers who conduct longitudinal studies are inherently interested in studying individual and population changes over time (e.g., mathematics achievement, subjective well-being). To answer such research questions, models of change (e.g., growth models) make the assumption of longitudinal measurement invariance. In many applied situations, key constructs are measured by a collection of ordered-categorical indicators (e.g., Likert scale items). To evaluate longitudinal measurement invariance with ordered-categorical indicators, a set of hierarchical models can be sequentially tested and compared. If the statistical tests of measurement invariance fail to be supported for one of the models, it is useful to have a method with which to gauge the practical significance of the differences in measurement model parameters over time. Drawing on studies of latent growth models and second-order latent growth models with continuous indicators (e.g., Kim & Willson, 2014a; 2014b; Leite, 2007; Wirth, 2008), this study examined the performance of a potential sensitivity analysis to gauge the practical significance of violations of longitudinal measurement invariance for ordered-categorical indicators using second-order latent growth models. The change in the estimate of the second-order growth parameters following the addition of an incorrect level of measurement invariance constraints at the first-order level was used as an effect size for measurement non-invariance. This study investigated how sensitive the proposed sensitivity analysis was to different locations of non-invariance (i.e., non-invariance in the factor loadings, the thresholds, and the unique factor variances) given a sufficient sample size. This study also examined whether the sensitivity of the proposed sensitivity analysis depended on a number of other factors including the magnitude of non-invariance, the number of non-invariant indicators, the number of non-invariant occasions, and the number of response categories in the indicators.
ContributorsLiu, Yu, Ph.D (Author) / West, Stephen G. (Thesis advisor) / Tein, Jenn-Yun (Thesis advisor) / Green, Samuel (Committee member) / Grimm, Kevin J. (Committee member) / Arizona State University (Publisher)
Created2016
Description
Currently, there is a clear gap in the missing data literature for three-level models.
To date, the literature has only focused on the theoretical and algorithmic work
required to implement three-level imputation using the joint model (JM) method of
imputation, leaving relatively no work done on fully conditional specication (FCS)
method. Moreover, the literature lacks any methodological evaluation of three-level
imputation. Thus, this thesis serves two purposes: (1) to develop an algorithm in
order to implement FCS in the context of a three-level model and (2) to evaluate
both imputation methods. The simulation investigated a random intercept model
under both 20% and 40% missing data rates. The ndings of this thesis suggest
that the estimates for both JM and FCS were largely unbiased, gave good coverage,
and produced similar results. The sole exception for both methods was the slope for
the level-3 variable, which was modestly biased. The bias exhibited by the methods
could be due to the small number of clusters used. This nding suggests that future
research ought to investigate and establish clear recommendations for the number of
clusters required by these imputation methods. To conclude, this thesis serves as a
preliminary start in tackling a much larger issue and gap in the current missing data
literature.
To date, the literature has only focused on the theoretical and algorithmic work
required to implement three-level imputation using the joint model (JM) method of
imputation, leaving relatively no work done on fully conditional specication (FCS)
method. Moreover, the literature lacks any methodological evaluation of three-level
imputation. Thus, this thesis serves two purposes: (1) to develop an algorithm in
order to implement FCS in the context of a three-level model and (2) to evaluate
both imputation methods. The simulation investigated a random intercept model
under both 20% and 40% missing data rates. The ndings of this thesis suggest
that the estimates for both JM and FCS were largely unbiased, gave good coverage,
and produced similar results. The sole exception for both methods was the slope for
the level-3 variable, which was modestly biased. The bias exhibited by the methods
could be due to the small number of clusters used. This nding suggests that future
research ought to investigate and establish clear recommendations for the number of
clusters required by these imputation methods. To conclude, this thesis serves as a
preliminary start in tackling a much larger issue and gap in the current missing data
literature.
ContributorsKeller, Brian Tinnell (Author) / Enders, Craig K. (Thesis advisor) / Grimm, Kevin J. (Committee member) / Levy, Roy (Committee member) / Arizona State University (Publisher)
Created2015
Description
Pediatric obesity is a public health concern due to its elevated prevalence rates and its relation to concurrent and long-term physical and psychosocial consequences. Pediatric obesity has been found to be associated with problem behaviors, albeit with inconsistent findings. The mechanism of this relation is unclear. It is possible that they have a shared etiology. Self-regulation and parenting practices are two factors that have been implicated in the development of problem behaviors and are garnering evidence for their relation with pediatric obesity. The goal of the present study was to examine whether self-regulation (SREC), positive behavior support (PBSEC), and coercive limit-setting (CLSEC) in early childhood are shared etiological factors of pediatric obesity and problem behaviors. Using multinomial logistic regression the likelihood of belonging to four outcome groups (Comorbid, Problem behavior only, Overweight only, and Typically developing) at age 10 based on these factors was assessed. Analyses controlled for intervention group assignment, child gender, child African-American or Bi-racial, child Hispanic, cumulative risk, child body size impression at age 2, and parent body size impression at baseline. In the models examining SREC alone, for every 1 standard deviation increase in SREC, there was a reliable reduction in the odds of the child belonging to the comorbid and problem behavior only groups at age 10, compared to the typically developing group (OR = 0.386, 95% CI [0.237, 0.628], OR = 0.281, 95% CI [0.157, 0.503], respectively). This relation was maintained when SREC was in the same model as PBSEC and CLSEC. PBSEC and CLSEC alone did not impact the likelihood of belonging to any of the outcome groups. A significant interaction was found between SREC and CLSEC, such that at high levels of both SREC and CLSEC the odds of a child belonging to the overweight only group at age 10 increased, compared to the typically developing group. Results highlight CLSEC as a parenting practice that may place a highly regulated child at risk for becoming overweight. Overall, the findings suggest that problem behaviors and pediatric obesity do not have a shared etiology.
ContributorsMontano, Zorash (Author) / Dishion, Thomas J (Thesis advisor) / Gonzales, Nancy (Committee member) / Perez, Marisol (Committee member) / Enders, Craig (Committee member) / Arizona State University (Publisher)
Created2016
Description
The comparison of between- versus within-person relations addresses a central issue in psychological research regarding whether group-level relations among variables generalize to individual group members. Between- and within-person effects may differ in magnitude as well as direction, and contextual multilevel models can accommodate this difference. Contextual multilevel models have been explicated mostly for cross-sectional data, but they can also be applied to longitudinal data where level-1 effects represent within-person relations and level-2 effects represent between-person relations. With longitudinal data, estimating the contextual effect allows direct evaluation of whether between-person and within-person effects differ. Furthermore, these models, unlike single-level models, permit individual differences by allowing within-person slopes to vary across individuals. This study examined the statistical performance of the contextual model with a random slope for longitudinal within-person fluctuation data.
A Monte Carlo simulation was used to generate data based on the contextual multilevel model, where sample size, effect size, and intraclass correlation (ICC) of the predictor variable were varied. The effects of simulation factors on parameter bias, parameter variability, and standard error accuracy were assessed. Parameter estimates were in general unbiased. Power to detect the slope variance and contextual effect was over 80% for most conditions, except some of the smaller sample size conditions. Type I error rates for the contextual effect were also high for some of the smaller sample size conditions. Conclusions and future directions are discussed.
A Monte Carlo simulation was used to generate data based on the contextual multilevel model, where sample size, effect size, and intraclass correlation (ICC) of the predictor variable were varied. The effects of simulation factors on parameter bias, parameter variability, and standard error accuracy were assessed. Parameter estimates were in general unbiased. Power to detect the slope variance and contextual effect was over 80% for most conditions, except some of the smaller sample size conditions. Type I error rates for the contextual effect were also high for some of the smaller sample size conditions. Conclusions and future directions are discussed.
ContributorsWurpts, Ingrid Carlson (Author) / Mackinnon, David P (Thesis advisor) / West, Stephen G. (Committee member) / Grimm, Kevin J. (Committee member) / Suk, Hye Won (Committee member) / Arizona State University (Publisher)
Created2016
Description
Low-income Mexican American women face significant risk for poor health during the postpartum period. Chronic stressors are theorized to negatively impact mental and physical health outcomes. However, physiological factors associated with increased self-regulatory capacity, such as resting heart rate variability, may buffer the impact of stress. In a sample of 322 low-income Mexican American women (mother age 18-42; 84% Spanish-speaking; modal family income $10,000-$15,000), the interactive influence of resting heart rate variability and three chronic prenatal stressors (daily hassles, negative life events, economic stress) on maternal cortisol output, depressive symptoms, and self-rated health at 12 weeks postpartum was assessed. The hypothesized interactive effects between resting heart rate variability and the chronic prenatal stressors on the health outcomes were not supported by the data. However, results showed that a higher number of prenatal daily hassles was associated with increased postpartum depressive symptoms, and a higher number of prenatal negative life events was associated with lower postpartum cortisol output. These results suggest that elevated chronic stress during the prenatal period may increase risk for poor health during the postpartum period.
ContributorsJewell, Shannon Linda (Author) / Luecken, Linda J. (Thesis advisor) / Lemery-Chalfant, Kathryn (Committee member) / Perez, Marisol (Committee member) / Arizona State University (Publisher)
Created2015
Description
Research has consistently shown that gay/lesbian/bisexual (GLB) or sexual minority youth are at an increased risk for adverse outcomes resulting from the stress caused by continual exposure to negative events (e.g., victimization, discrimination). The present study used a nationally representative sample of adolescents to test mechanisms that may be responsible for the differences in offending behaviors among sexual minority and heterosexual adolescents. Specifically, this study tested whether bisexual adolescents received less maternal support than did heterosexual adolescents because of their sexual orientation, thus increasing the likelihood that they run away from home. This study then examined whether the greater likelihood that bisexual adolescents running away would lead to them committing a significantly higher variety of income-based offenses, but not a significantly higher variety of aggression-based offenses. This study tested the hypothesized mediation model using two separate indicators of sexual orientation measured at two different time points, modeled outcomes in two ways, as well as estimated the models separately for boys and girls. Structural equation modeling was used to test the hypothesized direct and indirect relations. Results showed support for maternal support and running away mediating the relations between sexual orientation and offending behaviors for the model predicting the likelihood of committing either an aggressive or an income offense, but only for girls who identified as bisexual in early adulthood. Results did not support these relations for the other models, suggesting that bisexual females have unique needs when it comes to prevention and intervention. Results also highlight the need for a greater understanding of sexual orientation measurement methodology.
ContributorsMansion, Andre (Author) / Chassin, Laurie (Thesis advisor) / Barrera, Manuel (Committee member) / Grimm, Kevin J. (Committee member) / Toomey, Russell B (Committee member) / Arizona State University (Publisher)
Created2018