Methodenzentrum

Forschung

Aktuelles aus unserer Forschung

Ein neues nicht-lineares Framework für latente Variablen

Augustin Kelava und Holger Brandt veröffentlichten im Structural Equation Modeling Journal ein neues nichtlineares Strukturgleichungsmodell (NDLC-SEM) der dynamischen latenten Klasse. Das NDLC-SEM ist in der Lage, intraindividuelle psychologische Prozesse (z.B. Veränderungen in affektiven Zuständen als Trajektorien in der Mathematik) durchzuführen, die beispielsweise einen Abbruch vorhersagen könnten. Diese Prozesse werden in Teile zerlegt, die individuelle spezifische Komponenten (z.B. Schwachstellen, stabile Risikofaktoren wie Persönlichkeitsfaktoren oder kognitive Fähigkeiten) und zeitspezifische Komponenten beinhalten. Das NDLC-SEM fungiert als ein sehr umfassendes Framework, das es ermöglicht, Informationen aus verschiedenen Datenebenen und flexible Beziehungen zwischen Variablen (z.B. spezifische Interaktionen) zu integrieren.

Kelava, A. & Brandt, H. (2019). A nonlinear dynamic latent class structural equation model. Structural Equation Modeling: A Multidisciplinary Journal, 26(4), 509-528. doi: 10.1080/10705511.2018.1555692


Call for Abstracts

Die internationale und interdisziplinäre Konferenz "Gender, Work and Organization" mit dem Titel "Transforming Contexts, Transforming Selves: Gender in New Times" (24.-26. Juni 2020) bittet um Beiträge. Deadline ist der 01.11.2019.

Aus der Ankündigung:

"Our stream invites contributions that aspire to bring together theoretical, methodological and empirical reflections of researching doings and undoings of gender as well as other doings of differences. The following questions may be relevant, but are of course not exhaustive:

How can we conceptualize and empirically investigate different possible ranges of undoing and not doing gender? We suggest that understandings of not doing and undoing gender need to be specified with regard to different levels of analysis, such as materialities, discourses, institutions and interactions. Even if only separable analytically, empirical investigations should differentiate in order to determine and empirically depict possible ranges of undoing and not doing gender.

How can we empirically capture the complexity of differentiation processes (by gender, age, class,race, ability, citizenship, etc.) and conceptualize them? Moreover, how can “(un)doing differences”be conceptualized and taken up empirically?

How should we conceive the relation between theory and methods? As our above considerations show, theory and methods always constitute each other and can best be thought of as theory-methods packages, which „produce“ or generate their very subjects. Which “packages” can tacklethe complexities of persistence and change relevant for understanding recent transformations?

What kind of methods and methodologies have been developed for the analysis of (un)doing? Where can we identify needs for future developments, and how can we address these needs?

How and in what ways are questions of undoing and not doing gender related to questions of structure and agency? What are conceptual expansions that are promising and fruitful?

What are the affective practices and different intensities of (un)doing gender? How useful is the recent turn to affect for studying doing gender and differences?

Abstracts of approximately 500 words (submitted direct to stream leaders, ONE page, WORD NOT PDF, single spaced, excluding any references, no headers, footers or track changes) are invited by Friday 1st November 2019. [...]"

Call for Abstracts