Purpose. This paper discusses the development, piloting and subsequent use of the Child and Youth Resilience Measure (CYRM) with a diverse sample of 1503 youth in 14 sites globally as part of the International Resilience Project (IRP). This presentation will highlight the challenges associated with the design and development of the primary quantitative measure of child resilience used in the study. Findings summarizing the psychometric properties of the CYRM will be presented along with data regarding the stability of the CYRM's factor structure across culturally diverse settings. Finally, based on the lessons learned from this project, suggestions will be offered for subsequent multi-site international studies desiring to develop cross-cultural quantitative measures.

Method. Over a three-year period, the IRP developed and piloted the CYRM, a 58-item self-report measure. The instrument also includes space for the addition of 15 site-specific questions. A number of team meetings held to develop the instrument resulted in 32 domains being named as potential factors influencing resilience across cultures (e.g. peer relationships, sense of cultural rootedness, self-esteem). A nested ecological model was used to sort items into four groupings, individual, relational, community and culture. Questions were then solicited from each research site in order to complete the design work. The CYRM was administered to youth ranging in age from 12-19 over a one year period from February 2004-February 2005. Each site contributed between 60 and 224 respondents to the study. Data was analysed in Halifax, Tampa and Medellin by research staff and findings disseminated back to each community that participated and the advisory committees that helped to design the study.

Results. Findings from exploratory factor analyses performed to assess the validity of the CYRM demonstrate that 45 of the initial 58 questions load onto 4 components, supporting the 4 rational facets of resilience (i.e. individual, relationships, community and culture) hypothesized by members of the IRP during the CYRM's development. Examination of the internal consistency of these components are acceptable with Alpha reliability values of .85, .76, .76 and .74 respectively. A site-by-site examination of the stability of the factor structure and reliability of the components, however, reveals great cultural variation within the data with differing factor structures evident across minority and majority world contexts and genders.

Implications. Results are as were expected, with resilience being shown to be a culturally embedded construct. While all 32 domains were relevant to the youth internationally, patterns in their responses show that protective factors and mechanisms vary by site. We argue, based on these findings, that a priori assumptions about either cultural homogeneity or cultural heterogeneity are difficult to prove when sensitivity to diversity is shown in both the design and measurement phases of quantitative research on resilience.