Revisiting the Factor Structure of the Centrality of Event Scale

Abstract

Traumatic experiences that become self-defining impact posttraumatic outcomes, yet exact mechanisms are currently being studied. Recent research has used the Centrality of Event Scale (CES). However, the factor structure of the CES has been in question. We analyzed archival data ( N = 318) split into homogenous groups of participants to examine whether the factor structure of the CES differed based on event type (bereavement vs. sexual assault) or by levels of PTSD (meeting clinical cut off score vs. group of lowest scorers). Exploratory factor analyses with subsequent confirmatory analyses revealed a single factor model in the bereavement group, sexual assault group, and the low PTSD group. A three-factor model emerged in the high PTSD group, with the factors’ themes matching previous findings. Event centrality appears to be a universal theme when people endure and process a diversity of adverse events. These distinct factors may illuminate pathways in the clinical syndrome.

A multilevel factor analysis of the short form of the Centrality of Event Scale

“…The first question touches upon the measurement of differences between persons in the sense of, for example, examining whether persons with post-traumatic stress disorder symptoms judge the centrality of a stressful event higher than persons with no post-traumatic stress disorder symptomatology. This type of investigation, which can be described as examining between-person or interindividual differences in event centrality, is the predominant way the CES has been used in previous studies (e.g., Ionio et al, 2018;Bruce and Handal, 2023).…”

“…Moreover, the three-factor solutions, with a grain of salt, map to the theoretical structure suggested by Berntsen and Rubin (2006). At the same time, factors are typically strongly correlated, which is why some authors opted for a onefactor solution (e.g., Wamser-Nanney, 2019;Bruce and Handal, 2023). Note, however, that the strong correlations among factors (which imply strong inter-item correlations) may result from the fact that mostly traumatic and most negative events were evaluated by participants-one would expect relatively strong endorsement of all CES items in this case.…”

“…In a recent article, Bruce and Handal (2023) examined the CES factor structure in a sample of MTurk participants recruited online for a survey-based study on self-reported experiences posttrauma and a sample of students exposed to trauma. For the data analysis, from both studies those participants who described their trauma as either bereavement (N = 221) or sexual assault (N = 97) were selected, resulting in a sample size of 318 persons.…”