Research Seminar
To talk or not to talk? An illustration of Lord’s paradoxKaren BransKU Leuven | |
| Abstract: | A frequently observed action that people undertake to regulate their emotions is social sharing. Moreover, a widespread commonsense belief reads that merely talking about emotions contributes to emotional recovery. However, in several previous studies no significant relationship between social sharing and emotional recovery was found. Yet, an important limitation of these studies is the fact that the intensity of the emotional experience was only assessed twice in them; emotional recovery then was defined as the difference in intensity between the two assessments. To address this limitation, we set up a large scale experience sampling study in which emotional intensity and social sharing were assessed repeatedly over time in the course of a number of anger and sadness episodes. In this talk we will focus on the results of some initial analyses of the resulting data to examine the effect of social sharing on emotional intensity. For this purpose, we made use of two different methods: Gain Score Analysis (GSA) and Analysis of Covariance (ANCOVA). Unfortunately, the methods yielded contradictory results. For example, in the case of anger, ANCOVA suggested a negative effect of sharing on emotional intensity whereas GSA suggested a positive one. In literature, the phenomenon in which GSA and ANCOVA yield conflicting results when analyzing change in natural groups is known as Lord’s paradox. In this talk I will explain the paradox in question by means of the collected data, and I will briefly point to the different assumptions underlying the GSA and ANCOVA methods. I hope that by sharing these results and by exchanging ideas, we will be able to find out which technique is best suited to address our research question in this type of studies. |
| Date: | Tue Dec 21, 12:15 pm - 1:15 pm |
| Place: | room 03.60 (Department of Psychology, Tiensestraat 102, 3000 Leuven) |