Search

This study tests the argument that human resource management in hotels enhances service-recovery performance and job satisfaction through empowering front-line employees to respond to service failures. After an initial phase of qualitative interviewing, dyadic data were gathered through a large-scale survey in thirty hotels in Sri Lanka. The results of structural equation modelling show that the HR practices and management styles adopted in this context help to develop job competence, which is then related to service-recovery performance and job satisfaction. However, they show that service recovery is carefully stage-managed and ‘staircased’ in this hotel context with empowerment strongly related to hierarchical level. Empowerment to address service failures is important in these hotels but it is deliberately graduated according to rank. While employee training shows benefits for both parties, greater job autonomy would enhance the well-being of these service workers.

Researchers who use survey-type instruments to perform cross-cultural studies are frequently confronted with factorial non-invariance. This situation arises when the same set of items is used to measure a construct in different cultures, but the relationships between the items and the construct differ significantly across cultures. These (non-invariant) items are usually eliminated or ignored when calculating an index value representing the level of the construct. This paper proposes that under certain conditions non-invariant items ought to be retained, since they may constitute the only sources of information concerning cross-cultural differences. A new procedure is described that decomposes a cross-cultural difference into a component due to item responses, and a second component due to unequal item-construct relationships. A numerical example based upon a large published data set is presented, and the results of the new procedure are compared with those obtained using simple summated scores and regression-weighted factor scores.