Category: Research Methods and Statistics
Boer, D., Hanke, K., & He, J. (2018). On detecting systematic measurement error in cross-cultural research: A review and critical reflection on equivalence and invariance tests. Journal of Cross-Cultural Psychology, 49(5), 713-734.
,Olaru, G., Schroeders, U., Hartung, J., & Wilhelm, O. (2019). Ant colony optimization and local weighted structural equation modeling: A tutorial on novel item and person sampling procedures for personality research. European Journal of Personality, 33(3), 400–419. https://doi.org/10.1002/per.2195
,Costello, T. H., & Patrick, C. J. (2023). Development and initial validation of two brief measures of left-wing authoritarianism: A machine learning approach. Journal of Personality Assessment, 105(2), 187–202. https://doi.org/10.1080/00223891.2022.2081809
,Kelsey Lowman, B.A.
Florida State University
Tallahassee, Florida
Aidan Wright, Ph.D.
University of Michigan
Pittsburgh, Pennsylvania
Min Jeon, M.S. (she/her/hers)
Graduate Student
Florida State University
Tallahassee, Florida
Frederick Schubert, B.S. (he/him/his)
Clinical Psychology Graduate Student
Florida State University
Tallahassee, Florida
Rochelle Stewart, M.S. (she/her/hers)
Graduate Student
Florida State University
Tallahassee, Florida
Kelsey Lowman, B.A.
Florida State University
Tallahassee, Florida
Quantitative, dimensional models of psychopathology (e.g., HiTOP; Kotov et al., 2017) have increasingly gained favor over categorical models given their capacity to resolve issues such as diagnostic comorbidity, instability, and heterogeneity. Support for these models suggests that psychopathology is structured hierarchically, with broad dimensions of internalizing and externalizing cascading down toward more specific phenotypic expressions (e.g. symptoms). Applications of these models require replication efforts across research settings and diverse identity groups, which may benefit from cutting-edge analytic and psychometric approaches. These modern approaches may complement traditional methods by improving modeling, assessment, and inclusivity in research and clinical practice, thus bolstering construct validity as a whole. We will review some examples of modern techniques that represent new and exciting ways of advancing theory and measurement which will allow for more comprehensive case conceptualization and more effective and inclusive treatments.
Min Eun Jeon will present on the application of dimensional approaches in incorporating the role of minority stress in relation to externalizing and internalizing disorders, as well as suicidal ideation. Implications that will be discussed include structural invariance of internalizing, externalizing, minority stress, and suicidal ideation across race, ethnicity, and sexual orientation, as well as externalizing and internalizing serially accounting for the correlation between minority stress and suicidal ideation.
Ted Schubert will discuss results and implications from an analysis of measurement invariance in the Anxiety Sensitivity Index-3 (ASI-3) across age using traditional invariance methods complemented by locally weighted structural equation modeling (LSEM). Implications for use of the ASI-3 across age and LSEM as a technique for evaluating model parameters for measures of dimensional constructs across continuous moderators will be discussed.
Rochelle Stewart will present evidence from structural equation models which indicate that three widely used questionnaires developed to assess physical appearance concerns in relation to eating disorders, social anxiety disorder, and body dysmorphic disorder are indicators of a common latent construct. Findings from model-based multidimensional item response theory methods, which were used to derive a brief scale measure with optimal measurement properties for indexing variations in the general appearance concerns trait continuum, will be presented. The current findings and broader utility of the quantitative methods will be discussed in relation to dimensional approaches to psychopathology.
Kelsey Lowman will introduce an innovative machine learning approach to scale construction - a modified genetic algorithm - that allows for flexible integration of theory into an otherwise data-driven scale development process. An application to the triarchic model of psychopathy will be reviewed and future directions will be discussed, including the potential for this technique to integrate multimodal data in the operationalization of dimensional constructs.Learning Objectives: