1,721,014 research outputs found
Nonlinear Models with Moderated Parameters: New Methods and Software for Social Science Applications
No full textIntrinsically nonlinear models are used rarely in psychology and social science research, despite being well suited for informing theory, directly testing key hypotheses, and identifying intraindividual variation in important psychological processes. Additionally, assessing the presence of moderation is often of key importance for psychological theory testing. However, methods for testing, plotting, and probing moderation were developed for linear models, and currently, their use remains largely restricted to linear models. Psychology researchers seeking to implement nonlinear models with moderated parameters currently face a variety of conceptual and logistical challenges. Therefore, the overarching aims of this dissertation are to establish the utility of nonlinear models for social science, and to derive novel methodological and software tools that facilitate the creation and visualization of nonlinear models with moderated parameters. First, a review of the methodological and applied literature demonstrates the enhanced theoretical and substantive contributions that may be made through use of moderated nonlinear models. Second, guidelines for moderated nonlinear model selection, parameterization, and specification are introduced. Third, conceptual and mathematical extensions of the Johnson-Neyman technique – the state-of-the-art method for probing and visualizing moderation – are derived such that this technique can now be applied to moderated nonlinear models. Finally, a new Shiny app is introduced, which enables researchers to fit, evaluate, and visualize nonlinear models with moderated parameters in a code-free environment. Ideally, the methods and software presented in this dissertation will reduce many of the key conceptual and logistical barriers that social scientists currently face when implementing moderated nonlinear models, thereby increasing the use of such models across psychology and social science
Nonlinear Models with Moderated Parameters: New Methods and Software for Social Science Applications
No full textIntrinsically nonlinear models are used rarely in psychology and social science research, despite being well suited for informing theory, directly testing key hypotheses, and identifying intraindividual variation in important psychological processes. Additionally, assessing the presence of moderation is often of key importance for psychological theory testing. However, methods for testing, plotting, and probing moderation were developed for linear models, and currently, their use remains largely restricted to linear models. Psychology researchers seeking to implement nonlinear models with moderated parameters currently face a variety of conceptual and logistical challenges. Therefore, the overarching aims of this dissertation are to establish the utility of nonlinear models for social science, and to derive novel methodological and software tools that facilitate the creation and visualization of nonlinear models with moderated parameters. First, a review of the methodological and applied literature demonstrates the enhanced theoretical and substantive contributions that may be made through use of moderated nonlinear models. Second, guidelines for moderated nonlinear model selection, parameterization, and specification are introduced. Third, conceptual and mathematical extensions of the Johnson-Neyman technique – the state-of-the-art method for probing and visualizing moderation – are derived such that this technique can now be applied to moderated nonlinear models. Finally, a new Shiny app is introduced, which enables researchers to fit, evaluate, and visualize nonlinear models with moderated parameters in a code-free environment. Ideally, the methods and software presented in this dissertation will reduce many of the key conceptual and logistical barriers that social scientists currently face when implementing moderated nonlinear models, thereby increasing the use of such models across psychology and social science
Applications of Exploratory Q-Matrix Discovery Procedures in Diagnostic Classification Models
Get PDFDiagnostic Classification Models (DCM) use a Q-matrix to determine which skills are required to correctly answer items on large-scale assessments. DCMs are fit under the assumption that the Q-matrix is correctly specified. Misspecification of the Q-matrix is problematic for several reasons; problems with model convergence, poor model fit, and inflated model parameters. The current study examines the use of probabilistic estimation of the Q-matrix for cognitive diagnosis modeling in order to allow for uncertainty to help shape the construction of the Q-matrix. Two DCMs, the DINA and the DINO, were estimated for common reading comprehension tests using an EM algorithm and the goodness of fit was checked. Models using a probabilistic Q-matrix showed better fit and lower slip and guess parameters, suggesting that the probabilistic model provided more accurate Q-matrix specification and more accurate prediction of examinee skills
Fit Index Sensitivity in Multilevel Structural Equation Modeling
Get PDFMultilevel Structural Equation Modeling (MSEM) is used to estimate latent variable models in the presence of multilevel data. A key feature of MSEM is its ability to quantify the extent to which a hypothesized model fits the observed data. Several test statistics and so-called fit indices can be calculated in MSEM as is done in single-level structural equation modeling. Accordingly, problems associated with these measures in the single-level case may apply to the multilevel case and new complications may arise. Few studies, however, have examined the performance of fit indices in MSEM. Furthermore, recent findings suggest that evaluating fit at each level separately is advantageous to evaluating fit for the overall model. Therefore, the purpose of the present study was to evaluate the sensitivity of several fit indices to misspecification in the cluster-level model under varying multilevel data conditions including the intraclass correlation coefficient, sample size configuration, and severity of model misspecification. Furthermore, three methods of level-specific fit evaluation were compared. Results from a Monte Carlo simulation study suggest that fit indices are affected by the ICC of model indicators and sample size configurations in MSEM. With the exception of the SRMR, all fit indices were less sensitive to cluster-level model misspecification at low indicator ICCs, large overall sample sizes, and smaller numbers of clusters. Discrepancies in fit information between two methods of level-specific fit were observed at low ICC values. Finally, two fit indices rarely used in SEM applications revealed desirable properties in certain simulation conditions. Implications of the simulation results are discussed and a program for implementing level-specific fit evaluation in the R statistical language is provided
Effects of sadness and hostility on depressive attentional allocation processes
Get PDFA large proportion of individuals with unipolar depression experience predominantly hostile as opposed to sad mood. The consideration of hostile mood states has been virtually ignored, however, in depression research. To gain a better understanding of these processes, the current study explored the impact of hostility and sadness on attentional allocation patterns in dysphoria. Individuals exhibiting dysphoria completed questionnaires as well as a computer task during which they rated themselves and others with respect to trait adjectives. Attention was measured using pupillary response, reaction time, and recall of adjectives. Structural equation modeling was used for analyses, and results indicated that hostility and sadness seem to differentially impact attentional processes in dysphoria
Testing complex correlational hypotheses using structural equation modeling
No full textIt is often of interest to estimate partial or semipartial correlation coefficients as indexes of the linear association between 2 variables after partialing one or both for the influence of covariates. Squaring these coefficients expresses the proportion of variance in 1 variable explained by the other variable after controlling for covariates. Methods exist for testing hypotheses about the equality of these coefficients across 2 or more groups, but they are difficult to conduct by hand, prone to error, and limited to simple cases.Aunified framework is provided for estimating bivariate, partial, and semipartial correlation coefficients using structural equation modeling (SEM).
Within the SEM framework, it is straightforward to test hypotheses of the equality of various correlation coefficients with any number of covariates across multiple
groups. LISREL syntax is provided, along with 4 examples.This work was funded in part by National Institute on Drug Abuse Grant DA16883
EATING DISORDERS AND THE REGULATION OF EMOTION: FUNCTIONAL MODELS FOR ANOREXIA AND BULIMIA NERVOSA
Get PDFDifferent types of eating disorders may be better described and understood in terms of their specific behaviors and the emotion regulatory function these behaviors serve. Individuals may influence their affective states by upregulating or downregulating different emotions. Evidence characterizing eating disordered behavior according to this theory is discussed based on personality research, comorbidity, affect intensity, and neurobiology. An original emotion regulation theory of eating disorders is proposed. This theory centers on individuals' affect intensity and their emotion regulation strategies. Eating disorders are conceptualized by their behavioral components, not by their diagnostic category. Individuals with anorexia nervosa-restricting type (restrictors) were compared to individuals with bulimia nervosa and anorexia nervosa-binge/purge type (binge-purgers). Restrictors were posited to be low in affect intensity, or emotionally constricted. In contrast, binge-purgers were posited to be high in affect intensity, or emotionally labile. Food restriction in restrictors was hypothesized to be a method for increasing positive affect and decreasing negative affect. In binge-purgers, binging was seen as a method for reducing negative affect, and purging was seen as a means to increase positive affect and reduce negative affect. Participants were 63 inpatient females with a clinical diagnosis of an eating disorder. Participants were given an assessment battery measuring various indices of eating behavior and emotionality. Overarchingly, it was hypothesized that women classified as restrictors versus binge-purgers would show different patterns of emotional processing. Results of the present study support the theory that affective differences exist between individuals who solely restrict dietary intake and those who also engage in binge-purge behaviors. It appears that affect intensity may be one of the most important differences. Binge-purgers had marginally higher levels of affect intensity than did restrictors. However, affect intensity moderated the emotional outcomes of disordered eating behaviors in both groups. These preliminary analyses support the emotion regulation theory of eating disorders and warrant further investigation
PHYSICAL ACTIVITY AS A MEDIATOR OF THE RELATIONSHIP BETWEEN SELF-EFFICACY AND BODY MASS INDEX IN A NON-CLINICAL SAMPLE OF CHILDREN
Get PDFThe present study examined the associations among key pediatric overweight prevention and intervention variables: body mass index (BMI), physical activity self-efficacy (PASE), physical activity, and sedentary behavior. The first tested hypothesis purported an association between PASE and BMI that is mediated by physical activity. The second hypothesis stated that this mediation is moderated by sedentary behavior. A community sample of 382 fifth and sixth grade students were measured for height and weight and completed questionnaires. Findings suggest that, for girls only, PASE is negatively associated with BMI via physical activity for children who take part in longer amounts of sedentary behavior. These findings highlight the role of PASE in maintaining beneficial physical activity levels among children who are more sedentary, and the importance of accounting for sex and sedentary behavior within the physical activity literature. These findings are then discussed within the context of informing intervention efforts targeting pediatric overweight
- …
