In spite of specific programmes in many European countries designed to reduce inequality and social exclusion (Mackenbach and Bakker, 2002, Marmot, 2003 and Woodward and Kawachi, 2000), studies of regional differences within certain countries have shown that inequalities in health and social well-being still remain. Social inequality in some countries has even increased during the past 10 years in spite of policies aimed at reducing it (Shaw, Davey Smith, & Dorling, 2005). Leyland (2004), for instance, indicated that while disease-specific mortality rates in Great Britain decreased between 1979 and 1998, differences in premature mortality persisted between regions and even increased between districts within regions. Also in Finland, in the northern part of the country, life expectancy for both men and women has remained about one year lower than in other parts of the country (Näyhä and Hassi, 1999 and Näyhä and Järvelin, 1998).
There is an extensive literature on the influence of socioeconomic and psychosocial factors on health and social well-being. Despite strenuous efforts, the reasons for the inverse gradient between socioeconomic status and overall mortality and morbidity remain incompletely explained (Delzell, 1996 and Marmot, 2003). However, socioeconomic indicators such as education, employment, occupational social class and income have been used to explain social and health inequalities, and also differences in health behaviour (Blank and Diderichsen, 1996, Droomers and Westert, 2004, Ecob and Davey Smith, 1999, Lahelma et al., 2004, van Lenthe et al., 2004, Mackenbach et al., 1997, Mackenbach et al., 2003, Marmot, 2003, Pekkanen et al., 1995, Pikhart et al., 2003, Power and Manor, 1992, Rahkonen et al., 2000, Ross and Wu, 1995, Smith et al., 1997 and Wadsworth et al., 1999). Social support is also known to serve as a protective factor for various health problems (Kawachi et al., 1996). The amount of actual and perceived social support is very much influenced by the behaviour and social skills of the individual (Sarason & Sarason, 1990).
Psychosocial mechanisms such as coping strategies, attitude towards the future and social support also probably have an influence on regional differences in health and social well-being. Most researchers have observed three basic dimensions in ways of coping: problem-focused, evaluative and passive, or emotion-focused coping. Problem-focused coping behaviour attempts to change the situation (e.g. “I made a plan of action and followed it”). Passive coping, i.e. emotion-focused coping means not confronting the problem but trying to confront the feelings associated with it. Active coping can be hypothesised to greatly affect a decision to migrate. Optimistic attitude towards the future is regarded as generalised expectation of positive outcomes in the future (Scheier & Carver, 1985), and as such, it also can be assumed to influence major personal decisions such as one's concerning migration. Optimism starts to develop from infancy on and is greatly influenced, for example, by being born wanted, childhood socioeconomic status in the family and school achievement (Ek, Remes, & Sovio, 2004). Although positive experiences later on such as occupational achievement further improve optimism, optimistic attitude towards future seems to have developed to a great extent already before adulthood. There is a substantial amount of research on children's coping strategies (for a review see Compas, Worsham, & Ey, 1992) which indicate that the ways of coping develop to a great extent in childhood family and peer contexts and are also affected by environmental factors (e.g. Tolan, Guerra, & Montaini-Klovdahl, 1997).
In the two northernmost provinces of Finland, the setting for this study, agriculture has been, and still is locally, an important sector of the economy. The total area of the provinces is very large, with a sparsely scattered population (161 000 km2 with a population density in 2005 of 4.0 inhabitants per km2) and migration to expanding urban centres is lowering the population density in rural areas. The population is widely spread in small village communities, and distances to service providers are long, the means of public transportation are restricted and auxiliary or supplementary services are available only to a limited extent. The population tends to be sicker in the northern parts of the country than in the southern parts, and premature deaths and early retirement seem to be concentrated there (Lahelma, 1991, Näyhä and Järvelin, 1998 and Ohinmaa et al., 1996). In addition to remote place of residence, low social class and poor education seem to be cumulative risk factors for health in the north (Lahelma, 1991).
Since the decision to migrate results from discrepancies between the possibilities offered by the current place of residence and the individual's own aspirations, the regional differences that have been found in health and well-being may be at least partly due to the fact that in western societies those who have more personal resources tend to migrate (e.g. Ritsilä & Ovaskainen, 2001). Consequently, in addition to education and occupational status, we set out to study other personal resources such as coping strategies, social support and attitude towards the future as possible mediating factors.
We aimed to study here how migration and current place of residence were independently associated with subjective well-being (self-reported health and life satisfaction) while taking individual resources such as education, occupation, work history and psychosocial resources into account. Since it can be assumed that the reasons for migration and its effects are gender-specific, we analysed men and women separately.
