Master's thesis

Symbolbild: Texte aus einem Wörterbuch ausschnittsweise auch auf eine CD gedruckt

Image: Jan-Peter Kasper (University of Jena)

This page has been machine translated.

Master's thesis topics

The following Master's thesis topics are currently offered by the Professorship of Psychological Methodology (further or own topics on request):

Topic	Topic description	Are you interested? Please contact:	Further supervisors
Global fit indices in MetaSEM	MetaSEM is a way to estimate meta-analytic structural equation models. This is done in two stages: Stage 1 tests the homogeneity of the data and in stage 2 the assumed models are fitted. However, it is still completely unknown how large the heterogeneity may be in Stage 1, without compromising the quality of the subsequent stage 1 modelling. There are several fit statistics that are handed out in Stage 1, but there are no concrete threshold values backed up by empirical research. As a master's dissertation, a simulation study would therefore be conducted to answer the research question: "How high can RMSEA and SRMR be in Stage 1 without degrading the quality of Stage 2 modelling?"	Lukas Knitter de	Christian Bloszies de
Simulation study on measurement invariance in confirmatory factor analyses with frequentist vs Bayesian model tests	The aim of this BA / MA is to investigate in a simulation study whether (and if so, why) there are different results when testing measurement invariance in structural equation models using (1) frequentist test procedures such as likelihood ratio tests and (2) Bayesian test procedures such as leave-one-out cross-validation or Bayesian factors. The analysis effort can be adapted depending on the BA / MA, but good knowledge of R is required.	Fabian Münch de
Data analysis on a self-selected topic with data from the family panel (pairfam) or socio-economic panel (SOEP)	In this BA / MA, various questions on data from two large German panel studies can be analysed. Many personality traits and other psychological constructs are recorded in one or both studies (e.g. Big 5, loneliness, attachment styles, ...). The question is to be investigated using a suitable statistical model (HLM, SEM, ...), with the focus on modelling.	Fabian Münch de
Model fit in ML-SEM	Multilevel structural equation modelling (ML-SEM) is widely used in Psychology. One problem in the application of ML-SEM lies in the assessment of model quality based on fit statistics (e.g. chi-square model fit test). In previous studies it could shown that the chi-square model fit test for ML-SEM with one-dimensional latent factors at level-2 is biased if the statistical programme the statistical programme Mplus was used. Within the framework of a simulation study whether these biases are also present in models with indicator-specific factors and/or in other and/or in other statistical programmes. The aim is to derive specific recommendations for action based on the simulation results.	Tobias Koch de	N.N.

The Professorship of Psychological Methodology is not (yet) working on any Master's thesis topics. Own topics can be processed on request.