The relative effects of a scandal on member engagement in rites of integration and rites of passage: evidence from a child abuse scandal in the Catholic Archdiocese of Philadelphia

Stroube, B and Zavyalova, A (2024) The relative effects of a scandal on member engagement in rites of integration and rites of passage: evidence from a child abuse scandal in the Catholic Archdiocese of Philadelphia. Organization Science. ISSN 1047-7039 (In Press)

Abstract

Organizational research has documented that scandals lead to negative aggregate stakeholder reactions. There is little reason to believe, however, that the effects of a scandal are homogenous across different types of engagement. We therefore compare the effects of a scandal on member engagement in two types of rites at normative organizations: rites of integration and rites of passage. Rites of integration focus on the community, celebrate organizational values, and help strengthen organizational identification; they are thus enacted more by core members. Rites of passage focus on the individual, celebrate transition between social roles, and require only occasional engagement; they are thus enacted by core and peripheral members. Because of these differences, we hypothesize that a normative organization’s implication in a scandal affects rites of passage more negatively than rites of integration, but that this effect depends on scandal prevalence among neighboring organizations, organizational age, and organizational size. We test our hypotheses in the context of a child abuse scandal in the Catholic Archdiocese of Philadelphia. Using yearly parish-level data from 1990 to 2010, we find that a parish’s implication in the scandal was associated with a larger decline in rites of passage (marriages, baptisms, and funerals) than in rites of integration (mass attendance). This difference was reversed with the increase in scandal prevalence. Furthermore, rites of integration were more resilient than rites of passage at older and larger parishes. To help rule in the plausibility of our organization-level theory, we present a simulation grounded in individual-level polling data from the context.