ABSTRACT
In the present era of quick, single-focus research, it seems appropriate to remind others of the gains achieved by instigating, carrying out and publishing research using complex data from more than one source and multiple predictors from more than one domain when building models for predicting behaviour. We have gone into our methodology in some detail as we feel it is inexorably inter-twined with the substance of a study. We have used seven culturally diverse samples, four sources of information, three adjustment areas, and predictive variables from several content domains. This parallels closely Jessor’s (1993) requirements for a developmental behavioural science based on multi-disciplinary approaches, akin to our domains of antecedents (demographics, family, peers and culture); and data based on internal experiences and overt behaviour, which we called subjective and objective sources of information. We were led to pick adolescents as the focus of study as human maturation requires adaptation to changing roles in multiple areas of life at puberty, irrespective of culture. Of course, other equally stressful events requiring adaptation to more than one role in life could have been selected as the focus of this cross-cultural study, such as birth of the first child, leaving home or migrating.
