Development of a Short-Form of the RCOPE for Use with Bereaved College Students.

Abstract

Bereavement is being increasingly recognized as a key issue in college populations (Balk in Mortality 2:207-220, 1997; Balk et al. in Death Stud 34:459-468, 2010). However, there is currently a dearth of research on the impact that the loss of a loved one has on college students and the ways that college students cope during the grieving process. This lack of research, particularly among younger groups, is problematic as researchers have shown that emerging adults experience a surprisingly high number of losses and are an at-risk group for poor post-loss outcomes (Servaty-Seib and Taub 2010; Servaty-Seib and Hamilton 2006; Arnett in Am Psychol 55(5):469, 2000). Religion is a common way that individuals cope with bereavement (Frantz et al. in Pastor Psychol 44(3):151-163, 1996) and may also be commonly used by college students to manage everyday stress (Merrill et al. 2009). The RCOPE (Pargament et al. in J Clin Psychol 56(4):519-543, 2000; J Health Psychol 9:713-730, 2004) is a frequently used measure of religious coping that has recently been evaluated for use with a bereaved undergraduate population. Lord and Gramling (2014) examined the factor structure of the RCOPE and concluded that overlap between the positive and negative religious coping subscales when used with a bereaved undergraduate sample detracted from the predictive utility of the instrument. The researchers provided evidence for the use of a new 2-factor, 39-item version of the RCOPE with the bereaved college student population. The current study replicated Lord and Gramling (2014) with a large follow-up sample of bereaved undergraduates. Participants (N = 677) consisted of individuals who had lost a loved one within the past 2 years, had a mean age of 19.1, and were predominantly female (62%) underclassmen (84% freshman or sophomore status). A majority of participants (68%) were identified as Christian. Exploratory factor analyses closely mimicked the results of the previous study, and the number of items was further reduced in order to provide a briefer version of the scale for use in future research. Hierarchical regression models demonstrated discriminate, convergent, and predictive validity of the instrument.