Disputation: Jonas Moss
Doctoral candidate Jonas Moss at the Department of Mathematic, Faculty of Mathematics and Natural Sciences, is defending the thesis Psychometrics and the modelling of publication bias for the degree of Philosophiae Doctor.
Doctoral candidate Jonas Moss
The University of Oslo is closed. The PhD defence and trial lecture will therefore be digital and streamed directly using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.
Ex auditorio questions: the chair of the defence will invite the audience to ask questions ex auditorio at the end of the defence. If you would like to ask a question, click 'Raise hand' and wait to be unmuted.
- The webinar opens for participation just before the disputation starts, participants who join early will be put in a waiting room.
3rd of December, 10:15, Zoom
"Assessment of model fit for confirmatory factor analysis with ordinal data"
Join the trial lecture
The webinar opens for participation just before the trial lecture starts, participants who join early will be put in a waiting room.
Main research findings
Most research findings should be taken with a grain of salt. Researchers routinely play up their results, sometimes to the degree that we cannot trust them.
p-values are often in the focus when overplaying results, where the goal of the researcher is to find p-values smaller than 0.05. It’s often feasible to force a p-value below this barrier. The researcher might experiment with different statistical methods, remove some study participants, or add information about the participants. In my dissertation, we propose a method to tell us what the research findings would have been if they were not played up. Our method yields reasonable results when applied to fields that we know are severely overplayed.
Another topic of my dissertation is psychological testing. Psychological testing does not work without making assumptions about unobserved quantities, such as intelligence. Several popular models of psychological testing assume these quantities are multivariate normal. We investigate how much we can say about the unobserved quantities without making this assumption. As it turns out, we can barely say anything at all.