Informática 32 (2008) 111-122 111 Urban-Rural Life Setting as the Explanatory Factor of Differences in Fertility Behaviour in Slovenia Majda Cernic Istenic Sociomedical Institute at Scientific Research Centre of Slovenian Academy of Sciences and Arts, Novi trg 2, 1000 Ljubljana, Slovenia University of Ljubljana, Biotechnical Faculty, Agronomy Department, Jamnikarjeva 101, 1000 Ljubljana, Slovenia, E-mail: majdaci@zrc-sazu.si Andrej Kveder Population Activities Unit, UNECE, Palais des Nations, CH-1211 Geneva 10, E-mail: andrej.kveder@unece.org Keywords: survey data, statistical data, fertility behaviour, typology of urban-rural settings, Slovenia Received: March 31, 2008 Contemporary research of fertility behaviour that considers simultaneous longitudinal and prospective inclusion of individual and contextual levels of observation increases its explanatory power of explanation. The only research data of the kind in Slovenian is Fertility behaviour of Slovenians which was a part of international project the Family and Fertility Survey (FFS), carried out in 1995. On the basis of this survey data combined with statistical census data in the form of socio-economic typology of Slovenian countryside the article aims to explain the tendency and reasons of still persistent difference in urban-rural fertility. The study of relationship between fertility decisions of different generations and developmental characteristics of rural and urban areas in Slovenia reveals that this relationship is a complex and dynamic one. Obtained research results call for diversified actions of population policies in Slovenia. In to its field of actions not merely family and social policy should be integrated, but also space and regional measures that will consider different every day's life conditions and needs of people, living in particular space setting, and in this way assist them to fulfil their desired number of children. Povzetek: Narejena je analiza slovenske rodnosti na podeželju in v mestih. 1 Introduction According to Notestein's theory of demographic transition (1953) a high-level fertility regime of premodern societies was replaced with a low-level one owing to the process of modernisation and its accompanying processes: industrialisation and urbanisation. As supposed, the growth of big city agglomerations and mobile urban population disrupted the strength of traditional norms and commitments to a traditional way of life and stimulated individualism and affirmation of personal aspirations. However, as later research revealed (e.g. "The European Fertility Project" (Coale, Watkins, 1986)), the casual relationship between urbanisation and the beginning of demographic transition was not always entirely direct and unproblematic. For instance, the decline in fertility in France "began in the late eighteenth century, long before the appearance of the modern city, while the decline in fertility in England only got underway decades after cities like Birmingham and Manchester become grimy industrial centres" (Sharlin, 1986). In spite of relatively great variety in the beginning of demographic transition and the level of urbanisation causal relations between these two phenomena certainly existed. Research on social-group forerunners of fertility control in Europe (Livi-Bacci, 1986) proved that urban-rural difference in fertility existed even prior to the onset of general decline in fertility. Some groups within the city population (mainly elites: aristocracies and bourgeoisie) were practising effective family limitation which influenced the overall urban fertility level. Other reasons for variation in levels of fertility between urban and rural populations before the demographic transition are also low levels of nuptiality in the cities due to high concentration of servants (it was not easy for them to get married) (Sharlin, 1986). Urban-rural difference in natural fertility was also affected with factors such as infant mortality (higher in the cities) and breast-feeding (more widespread in the country) (van de Walle, 1986). After the completion of the transition from high fertility 112 Informatica 32 (2008) 111-122 M. Černič Istenič et al. to the low one (enforcement of modern demographic behaviour), urban marital fertility remains lower than rural marital fertility. The differences declined in size, but nonetheless continued to exist in most of the countries that experienced demographic transition (Andorka, 1978). As some recent research (Cernic Istenic, Kveder, 2000, Cernic Istenic, Kveder, Obersnel Kveder, 2000) indicates, this 'state of affairs' also holds true for Slovenia: today all social strata that live in the cities still have fewer children than those who live in the countryside despite the fact that the difference in the level of fertility among various social strata is indeed diminishing (Javornik, 1999) due to the predominance of two children per family norm (Malacic, 1990, Cernic Istenic 1994, Obersnel Kveder et al., 2001). To explicate this issue more precisely in the present article we explore the relationship between fertility decisions of different generations and developmental characteristics of rural and urban areas in Slovenia. On the basis of individual survey data combined with statistical census data in the form of socio-economic typology of Slovenian countryside we intend to discover the tendency of this still persistent difference in urban-rural fertility and its reasons. In this vein, the main traits of Slovenian urbanisation and deagrarisation process over the last 150 years are firstly briefly sketched. Secondly, the main analytical frame of our analysis is outlined. It follows the pertinent theoretical observations which indicate that due to urbanization social, economic and political differentiation firstly increased elsewhere, but later on, a kind of homogenisation in social behaviour of the population took place due to further impacts of urban life patterns on the countryside. In the third part the main explanations of the selected data and applied methods are presented. In the forth section, the results are presented according to the outlined hypothesis and in the final part they are discussed and conclusions are outlined. 1.1 Urbanisation and deagrarisation of Slovenia The beginning of urbanisation in Slovenia was relatively late. A growth of cities started only after 1848. In that time Ljubljana, the capital and the biggest city of today's Slovenia, had just 17000 residents. Urban-rural difference in fertility level existed before, during and after the demographic transition in Slovenia. The earliest available data for the level of fertility in Slovenia that make the comparison between urban and rural fertility possible pertain to generations born during 1873-1877 and during 1898-1902. This comparison show that urban women had from 34 to 47 percentage points lower fertility than countryside women did (Sircelj, 1991). The beginning of demographic transition in Slovenia started only in the first quarter of the twenty-century. However, fertility in the cities already began to decline towards the end of the nineteenth-century. Firstly it started to decline in larger cities and then spread onto the smaller ones. At the break of the twentieth century fertility in cities declined for about 40 percentage points and for about 20 percentage points in the countryside (Šircelj, 1991). Slovenian society was still weakly structured in the middle of the nineteenth century. The main occupation was farming. Specifically, in 1868 the share of farm population was 81,4 percent. A the break of the 20th century it decreased to 73,2 percent (Klemenčič 2002). The difference between areas with the highest and the lowest share of farm population was 13 percentage points. In 1931 this difference reached 50 percentage points. The share of farm population decreased especially in the neighbourhoods of Ljubljana. At that time no association between the share of farm population and the level of fertility was observed. It appeared only after the Second World War. "Up to that time the level of fertility was much more influenced by the type of settlement where a woman lived than by the share of farm population" (Šircelj, 1991: 244). Since the Second World War farm women have the highest fertility and today they are the only socio-professional group who assures itself biological reproduction (Šircelj, Ilič, Kuhar, Zupančič, 1990, Šircelj 2007). However, fertility of farm women has actually no effect on the national fertility level due to its very low share compared to the whole population1. In the period 1931-2002 the share of farm population declined from 60 to 6,5 percent, the most rapidly in the period 1948-1981; from 48,9 to 9,4 percent (Klemenčič 2002, Statistical Yearbook 2003). At the same time, the age structure of agrarian countryside deteriorated significantly. Young generation was/is immigrating into the cities whereas older generations remain in the countryside. In some parts of Slovenia (hills, highlands, the Karst and particularly borderlands) the age structure of rural areas is so unfavourable that it causes demographic extinction and stagnation. Namely, only 2 percent of the population lives on the above mentioned areas which represent 20 percent of the surface (Pečan, Ravbar, 1999). On the other hand, countryside has also experienced social strata transformation. Due to abandonment of farming and forestry, moving of provincials into other occupations and owing to permanent or accessional immigration of a part of the urban population into the countryside, its social structure becomes more and more heterogeneous (Barbič, 1991; Kovačič, 1995). In relation to this process the way of life of the countryside population is changing. Achievements of urban society are advancing rapidly into the rural areas; activities and services once typical for urban areas become more and more widespread in rural areas. Thus, rural population is getting similar to the urban one (Pečan, Ravbar, 1999). It can be expected that all these processes are influencing social behaviour of the countryside people and consequently their fertility behaviour as well. 1 However, the correlation between decrease in total fertility rate in Slovenia and the decline in the share of farm population was high after the Second World War period. URBAN-RURAL LIFE SETTING AS. Informatica 32 (2008) 111-122 113 1.2 Theoretical background In now already classical textbook "Determinants of fertility in advanced societies" by Andorka (1978) a relationship between fertility behaviour and place of residence was characterised as a direct linkage. There is a fairly consistent correlation between urban or rural trait of the place of residence and fertility. The place of residence has a property of natural or man-made environment. In this sense highly populated, densely build-up areas heavily loaded with traffic are defined as urban areas. Correspondingly, there is little space left for parks, private gardens and other places where children can safely spend time outside their homes. On the other hand, the trait of rural area is determined by living predominantly in a one-family house with garden or in a relatively small apartment house. Life there is quieter and safer and children have plentiful space for playing outside their homes. According to Andorka, this ecological characteristic of urban-rural differential is also connected with different monetary costs and efforts necessary for raising and educating children that are much greater in urban areas than in the rural ones. Besides, he presupposed different preferences and different consumption alternatives in these two types of residence. To understand urban-rural differences in fertility behaviour of advanced societies it is also necessary to take into consideration the social characteristics of these types of environment. Mackensen (1982) believes that one general theory of fertility that could adequately explain fertility behaviour in all societies and at all periods of time is neither possible nor justifiable. He is convinced that for this reason, every explanation, observation and research of fertility behaviour like any other social behaviour should proceed from the concept of specific structural and cultural characteristics of each society, which is the product of certain historical processes. Hoffnan-Nowotny (1987) who asserts that fertility behaviour of an individual is connected with structural and cultural characteristics of his/her micro, mezzo and macro social environment shares similar view. Importance of geographical variations in place or context in understanding fertility decision-making is further stressed by Boyle (2003). Discussion concerning typical characteristics of urban and rural places is already a long present one. Classics of sociological thought like K. Marx, F. Tonnies, G. Simmel, (European representatives) and L. Wirth, R. Redfield (attached or influenced by the Chicago School) argued that strong distinctions exist between urban and rural societies. E.g. accordingly to Marx's theorization, an individual living in urban place has universal chances to develop all his/her abilities while in the countryside he/she is bounded by constant reiteration of firmly established patterns of thinking and acting derived from direct dependence of men to nature. Quite on the contrary, Tonnies was convinced that life in the countryside (in Gemeinshaft) bounded together by a unity of wills and solidarity with its tradition and social order, gave an individual a supreme opportunity to harmoniously live with others, while life in urban areas (in Gesellschaft), constituted by commodity exchange and rested on "union of rationale wills", led to undermining genuine attachment between people and community (see Bonner, 1999 for more extensive overview of the urban-rural debate). Extensive empirical research of the rural society evolved since the 1950s, showed that there is no clear rural-urban distinction. Concepts like community and locality did not wholly prove their justification. Fieldwork investigations (Williams, 1963, Bell and Newby, 1971, Pahl, 1965, 1966, Newby, 1978, 1980) revealed that there are no close, isolated, harmonious, functioning in traditional manners and closely kinship-bounded communities. They also showed that low population density or certain fixed settlement patterns were insufficient basis to distinguish between urban and rural places. Marsden (1998, 1999), focusing on restructuring of agriculture, drew attention to the varying degrees of influence and interaction of agricultural, residential and commercial interests in shaping differentiating rural areas that could be analyzed by modelling of typologies (see Mahon, 2001 for more extensive review of recent comprehension of rurality). According to Bourdieu (2003) each classification of social world should take into consideration the principle of differentiation in order to theoretically construct empirical reality. Basically, this principle pertains to the distribution of the forms of power or the kinds of capital that vary according to the specific place and moment. This means that the set of agents or institutions which possess sufficient quantum of a specific capital (especially economic and cultural), that enables them to possess a dominant social position, conserve or transform the "exchange rate" between different kinds of capital through more or less administrative measures. Of course, this holds true for relationships between city and country as well. Characteristics of today's urban and rural societies are undoubtedly strongly related to their specific position and mutual relationship in the last century and a half, which has often taken the framework of city domination over the countryside (Hays 1993). An urbanized society evolved out of the former rural society by exploiting the material and human resources of the countryside intensively, which led to considerable economic, social and political inequalities between these two settings. Along with this, cities changed or gave up thoroughly the culture and the way of life, which long prevailed in the countryside. However in this process all parts of countryside were not equally affected. Hypothesis 1: on the basis of the above statements, we expected in our analysis that fertility behaviour of individuals is closely linked with economic and social characteristics of their life settings observed through the selected typology of rural areas. Over the last decades, the social structure and culture of rural areas in Europe and other industrially developed countries has changed significantly. Due to massive abandonment of agriculture by a great part of rural 114 Informatica 32 (2008) 111-122 M. Černič Istenič et al. population and their engagement in other occupations, rural areas became multifunctional and multistructural (Djurfeldt, 1999). Migrations from urban to rural areas contributed to this heterogeneity to some extent as well. With rapid development of new technologies (information and computer sciences) and improved traffic connections among urban and rural areas, the entire societies became increasingly urbanised, "infected" with urban values and the urban way of life. Owing to these changes, it could be supposed that urban-rural difference in fertility behaviour is diminishing or even vanishing. However, changes in social behaviour do not occur quickly, or at the run of one single generation. According to Inglehart's (1989) theory of value changes they occur with the exchange of generations. This theory is based on two hypotheses: 1. deficiency effect: An individual most strongly appreciates the things that are relatively rare in his/her socio-economic setting. 2. socialisation effect: Value priorities of an individual are not direct reflections of his/hers present socioeconomic setting but are the reflections of conditions in which he/she grew up. Therefore it could be supposed that due to increased impact of globalisation (preferring urban life style) every younger rural generation will be more similar in its social and consequently fertility behaviour to urban population than previous rural generations were. Hypothesis 2: in this respect we expected in our analysis smaller differences in fertility behaviour in younger generation of the respondents from different space settings than in the older ones. Cohorts who voluntarily limited marital fertility, enforced demographic transition, which took place during the second part of the 19th and the first part of the 20th century in industrialised world, from high to low fertility. Their belief in marriage and family was still very strong, although they practised contraception from preventive reasons according to Van de Kaa (1999). They wanted to give their children a good start in life and stemmed towards limiting the number of children to be correspondent with that goal. They disciplined themselves to remain married even when love was lost. However, these features of "modern demographic behaviour" did not last very long. Since 1965, new demographic changes have been observed in many European countries: decrease of marriages, increase of cohabitations, postponement or abandonment of parenthood, increase in divorces and single parent households. These changes also denoted as the second demographic transition (Van de Kaa, 1987) presumably occur due to a shift of value orientation from a modern trend to post-modern that signifies further enforcement of the individual's free choice principle, which was introduced during the time of Renaissance and Enlightenment centuries ago. The motto of this trend is that individuals can and should make their own choices. "Post-modern demographic behaviour" of contemporary reproductive cohorts presumably corresponds to the individualistic "lifestyle, where it is understood that sex and marriage/union are no longer closely related, and that contraception is only interrupted to have a self-fulfilling conception" (Van de Kaa, 1999:31). This new pattern of behaviour is seemingly reflected in the changes of the life course of young generations - earlier entry into first sexual intercourse but later achievement of economic and housing autonomy and formation of own families (Iedema et al., 1997; Cordon, 1997; Nave-Herz, 1997). Hypothesis 3: in this vein in our analysis we supposed considerable differences in the life course, especially during the transition into adulthood between younger and older generations of the respondents from different space settings. 2 Methods 2.1 Data The analysis was based on the Family and Fertility Survey data collected in Slovenia between December 1994 and December 1995. A representative area sample of the inhabitants of the Republic of Slovenia in their reproductive age period (i.e. aged 15 to 45) of both genders was drawn. Face to face interviewing was used as the data collection method. Realised sample consisted of n=4.559 respondents. The data were weighted according to survey design and population adjustment based on a specified set of socio-demographic variables: gender, age and size of the settlement (see Obersnel Kveder et al., 2001 for more detailed description of the data set). Considering our objective to explore differences between generations, two birth cohorts were emphasized in the present analysis: • respondents aged 20-24 years of both genders -the "younger" generation at the entrance to their reproductive period and also representing contemporary fertility patterns, • respondents aged 40-45 years of both genders -the "older" generation at the end of their reproductive period and also presenting the immediate past fertility patterns. The main interest of the present article was in the variables concerned with the timing of childbearing as well as other events relating to the entry of an individual into adulthood such as sexual debut, partnership history, leaving parental home, education and employment spells. All these events were measured as retrospective event histories. For all strictly reproductive events (i.e. childbearing, use of contraception), the age at the first sexual intercourse was taken as the threshold of becoming at risk, while for all others (i.e. partnership, education, employment, and leaving parental home) the threshold was the respondents' birth. Right censoring was determined by the date of the interview. Besides, the variables describing respondent's preferences, values, attitudes and status characteristics were considered. Some of them: household structure, attitudes towards URBAN-RURAL LIFE SETTING AS. Informatica 32 (2008) 111-122 115 abortion, gender roles and marriage were constructed from the set of other variables from the survey data set. The variable 'Household structure' was derived from the household grid. It has 13 distinct attributive values depicting various household types in which the respondents can live: 1 alone 2 with a partner 3 with a partner and others (e.g. his/hers parents and siblings, partner's parents and siblings, other relatives)* 4 with a partner and children 5 with a partner, children and others* 6 with children 7 with children and others* 8 with parents and siblings 9 with parents, siblings and other relatives 10 with one parent 11 with non-relatives 12 with other relatives 13 with other relatives and non-relatives The variable 'Abortion' was derived by taking into account "approval" answers from the set of the following dichotomous statements: • An abortion is approved/not approved when mother's life is in danger due to pregnancy. • An abortion is approved/not approved when the risk of the birth of an abnormal child is great. • An abortion is approved/not approved when a woman is not married. • An abortion is approved/not approved when a married couple does not want to have another child. • An abortion is approved/not approved when a woman does not want to have a child at the time being. The variable 'Gender roles' was constructed by taking into account answers "strongly agree" and "agree" from the Likert 5- item scale of the following attitudes: • An employed mother can create as warm and safe relationship with her children as a mother who is not employed. • Employment is the best way for a woman to achieve independence. • Being a housewife is as fulfilling for a woman as being employed. • Both man and woman should equally contribute to their household's budget. • Preschool child would most probably suffer if his mother was employed. • It is acceptable for a woman to be employed, but what most of women really want is home and children. The variable 'Marriage vis-a-vis Cohabitation' was constructed by taking into account responses "very favourable" and "favourable" from the Likert 5- item scale regarding advantages of cohabitation over marriage to achieve the following aims: • general happiness, • economic security, • friendly relationship with others, • personal freedom, • stable relationship, • having children, • social acceptance. Following the aim of this article, fertility behaviour should be put in the perspective of the contextual micro, mezzo and macro level variables. Microenvironment is defined as the immediate living surrounding of the individual varying from their family, household to the neighbourhood. Mezzo environment is by definition broader than the micro and thus can encapsulate a variety of geographical units from the settlement to the municipality and region. Macro socio-economic context is usually associated with the national level indicators. This article focuses on the use of one exemplary indicator measured at mezzo level; the respondent's place of residence determined by the basis of Slovenian Census in 1991. In its essence, the indicator reflects the urban-rural dichotomy and defines 4 possible types of living surroundings: urban, suburban, typical rural and rural depopulation areas (Kovacic et al. 2000, Kovacic, et al. 2002): • Those places that have an urban management character according to the space planning documents have been classified as urban environments. Additionally, the function of centrality of each geographical unit was considered as criteria of demarcation between urban and rural areas. Slovenian geographers classify settlements into seven groups according to their centrality (Vriser, 1998). Those settlements with the centrality index between 7 and 3 were also defined as urban environments. The settlements with the lowest centrality index 3 were additionally bounded by the minimal size of 3.500 inhabitants. The share of population living in this type of settlements is estimated at 39,20 percent. • All local communities (slo. krajevne skupnosti) with the density of the population greater than 200/km2 were considered as suburban. In addition, the areas with the density lower than 100/mk2, with the index of population growth in the period 1981/91 greater then 110, were also included in this type of settlements. The share of population living in this type of settlements is estimated at 14,80 percent. • All local communities with the long-term (1961/919 and short-term (1981/91) population growth index below 97,5 were considered as depopulation areas. In addition, local communities with the non-negative short-term population growth and with the index of population ageing above the absolute demographic threshold (i.e. 72) were also considered as depopulating. The share of population living in this type of settlements is estimated at 14,87 percent. 116 Informatica 32 (2008) 111-122 M. Černič Istenič et al. • All rural areas between suburban and depopulation were considered as typically rural. The share of population living in this type of settlements is estimated at 31,22 percent. Additional analysis which applied the above typology (Perpar, Kovacic, 2002) revealed that these types of rural settlements significantly differ among each other in terms of socio-economic development, infrastructure and natural resources. Taking into account several indicators pertaining to statistical data from 1991, suburban areas are in the most favourable position and depopulation areas are in the least favourable one. Considering the employment structure of population, the highest share of population in suburban and typically rural areas works in secondary sector, whereas in depopulation areas population mostly works in the 2 primary sector . Daily migrations additionally indicate the level of engagement in gainful employment. Its share is again the highest in suburban areas (770 per 1000 inhabitants) and the lowest in depopulation areas (540 per 1000 inhabitants). Further, significant differences among the areas are also observed concerning the education. The highest level of education is reached in suburban areas (43 percent of inhabitants finished at least high school), whereas in typically rural and especially in depopulation areas this share is considerably smaller. The same picture is indicated by the proportion of students per 1000 inhabitants; it is again the most favourable in suburban areas (25 students per 1000 inhabitants), less favourable in typical rural areas (21 to 25 students/1000 inhabitants) and the least favourable in depopulation areas (20 students/1000 inhabitants). Furthermore, indicators pertaining to economic situation confirm already indicated differences. E.g. density of business entities that indicates economic development of the area is the most favourable again in suburban settlements with more than 14 business entities per 1000 inhabitants, less favourable in typical rural areas (12 to 14 business entities per 1000 inhabitants) and the least favourable in depopulation areas (less then 12 business entities per 1000 inhabitants). Considering the infrastructure, the analysis (Perpar, Kovacic, 2002) demonstrated that all areas are relatively well equipped with basic infrastructure, but the best equipped are the suburban ones. In typical and depopulation areas the population is still frequently faced with the problems of drinking water supply, unsettled canalisation and purifying plants and maintenance of local roads. An additional problem of the depopulation areas is abandonment of farming and consequently the forest over-growing. 2 The share of farm population in depopulation areas presents 200 per 1000 inhabitants, in typical rural areas this proportion counts 120 per 1000 inhabitants and in suburban areas only 40 per 1000 inhabitants (Perpar, Kovacic, 2002). Each respondent from the survey was ascribed a settlement type according to his/her residence at the time of the interview. The key for information matching was the local community that could be matched to the survey data as well as the settlement typology specification. The analysed sub-samples were as follows: Table 1: The size of sub-samples (N). Settlements type 20-24 40-45 Urban 192 302 Suburban 109 182 Typical rural 204 295 Depopulation rural 75 131 2.2 Analysis Bivariate associations were analyzed using either the analysis of contingency tables either using comparisons among means. Timings of observed events were analyzed through event history models (Allison, 1995), which enable the estimation of the differences in the individuals' life course. Since only the distribution of the probability of time (T) was taken into account, it was described through cumulative distribution function perspective (survivor function): S(t) = Pr{r > t}= 1 -F(t) (1) The result of the survivor function is a probability that an individual "survives" in the process beyond time t. It is defined on the interval from 0 to 1. Life-table method was used for the estimation of the survivor function. The estimate is calculated using the conditional probability of failure (i.e. the probability for an event within a certain interval, given that an individual made it to the start of interval): S (t )=n(1 - q) (2) j=1 where qj is the conditional estimation of failure and is calculated as the ratio of number failed over the effective sample size. Major events in a life history were depicted by the means of calculating the quartiles of the survivor function. 2.3 Results In order to evaluate the importance of contextual information in explaining fertility and family behaviour, the major moments of the reproductive life period, structural and cultural characteristics of individuals were compared against the regional typology. At first, the processes of entry into the first, second and third parenthood were compared among the regions, taking into account both generations together (Figure 1, 2 and 3). The major differences that can be observed at the first birth are related to the differentiation of the urban areas from the rest. As survivor functions show, inhabitants of the urban areas tend to delay the birth of the first child more than the inhabitants of the other regions. There are only 15 percent of women from suburban and rural areas that are still childless 10 years URBAN-RURAL LIFE SETTING AS. Informatica 32 (2008) 111-122 117 after their sexual debut, while the proportion of nullipara women in the urban areas is above 30 percent. At the second birth, the differentiation of the urban areas from the rest still prevails; with more than 30 percent of inhabitants that did not experience the birth of the second child 10 years after the entry into the first parenthood. However, in this case the suburban areas also differ significantly from typical rural and depopulation rural areas; approximately 24 percent of its inhabitants did not experience the birth of the second child 10 years after the first birth, whereas in typical rural and depopulation rural areas only 18 and 14 percent of inhabitants did respectively. In the case of the third birth, the difference between areas got the character of polarity. The share of inhabitants that experienced the third birth decreased significantly in all regions in comparison with the share of inhabitants that experienced the second birth, but particularly in urban and suburban ones, where more than 80 percent of dwellers did not experience that event at all. In typical rural and depopulation rural areas the comparable shares are 69 and 66 percent respectively. 1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 duration (inyears) "Durban suburban typical rural depopulation rural Figure 1: Time to the first birth. i 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 duration (in years) -D-urban * suburban -*- typical rural -e- depopulation rural Figure 2: Time to the second birth. 1 3 5 7 9 17 19 21 11 13 15 duration (in years) — urban — suburban typical rural -©- depopulation rural Figure 3: Time to the third birth. Table 2 shows times in years from the first sexual intercourse to the first, second and third parenthood for both generations separately. The older generation relates a more homogenous picture then the younger one, where the differences among the rural and urban areas are more transparent. The most evidently, the 20-24 urban generation postpones the parenthood in comparison to the urban older one and their rural coevals as well. Among the younger generation in urban areas, 25 percent enters into first parenthood more then 4 years later (7,1 years in comparison to 2,9 years) then the first quarter of the older generation and approximately 3 years later than their counterparts in typical and depopulation rural areas. Furthermore, the postponement of parenthood among younger urban generation is manifested also in the absence of any childbirth at 50th and 75th percentile and consequently any second or third birth at all. Results pertaining to the younger generation also show that the first births reach their maximum level the fastest in typical rural areas. Table 2: Time to parenthood (in years). First Birth Second birth Third birth 20-24 40-45 20-24 40-45 20-24 40-45 25th percentile Urban 7.1 2.9 - 3.0 - - Suburban 4.3 2.3 3.0 2.3 - - Typical rural 3.3 2.2 2.6 2.3 3.2 6.0 Depopulation 3.9 2.4 3.6 2.0 - 3.8 50th percentile Urban - 7.2 - 5.1 - - Suburban 11.9 4.3 4.3 4.4 - - Typical rural 6.6 6.6 4.2 3.7 - - Depopulation 6.8 4.1 - 3.3 - - 75th percentile Urban - 10.3 - - - - Suburban - 7.7 6.9 9.3 - - Typical rural 8.8 7.0 - 6.0 - - Depopulation 6.5 - 6.4 - - 118 Informatica 32 (2008) 111-122 M. Černič Istenič et al. Although different patterns of entering into parenthood can be observed within generations, substantial differences in the 40-45 generation can be observed in the 75th percentile values. At this level the second births are rare among urban dwellers. There are also scarcely any families with three or more children. The third births are presented only at the level of 25th percentile among families in typical and depopulation rural areas. In general, the intervals between childbirths are considerably shorter in rural then in urban areas, what has an important implication on the overall number of children in each of the regions. 25 urban suburban typical rural depopulation rural "All ^40-45 years n20-24yK«s Figure 4: Actual number of children (means with confidence intervals). As a consequence of these events' and spells' patterns, in general, families in both types of rural areas 3 have statistically significantly (F=27.595*** , Bonferroni post-hoc mean difference tests for Urban compared to other types are -0.216***, -0.308*** and -0.358*** respectively) more children then families in urban areas have (Figure 4). Suburban areas do not differ significantly from urban or the rest of rural areas, however they are more similar to urban than rural areas. In the case of 20-24 generation, the differentiation is significant among urban and typical rural areas (F=8.562***, Bonferroni post-hoc mean difference tests for Urban compared to other types are -0.240**, -0.248*** and -0.209 respectively). It is also observed that in this generation, the number of children slightly decreases in depopulation areas in comparison with typical rural and suburban areas. Considering the perception of the ideal number of children per family, there are no great differences among rural and urban areas (Figure 5). The overall perceived ideal family size is above two children, though it is a bit diverse concerning the region type and age. In particular, the members of the younger generation from urban and suburban areas want more children then their older counterparts do. In this regards they are quite similar to their rural counterparts. Of course, more conclusive observations concerning the match between preferred and actual number of children can only be drawn from the older generation, which more or less already finished its reproductive period. It seems that among all older 3 p < 0.01 - ***, 0.01 < p < 0.05 - **, 0.05 < p < 0.10 - * respondents, those from suburban areas came the most near to their desired number. Among older generation, concerning preferred number of children, urban and suburban areas create one pattern (lower number) and typical and depopulation rural areas create the other one (higher number). This observation is consistent with previous as well as with subsequent findings. 3j0 -| urban suburban typical rural depopulation "All n 40-45years n20-24y«ars Figure 5: Preferences towards the number of children (means with confidence intervals). Table 3 distinctly shows extremely interlaced differences in the ages of entering certain life events among the two observed generations and the four area types. The prevailing pattern of sequence of events can be described as starting with the entry into sexual life, finishing education, getting the first job, leaving the parental home and as the last step, entry into partnership union. The entry into partnership union is consistently the last event to be experienced by any observed group. Some resemblances can be observed between both urban and suburban areas, as well as between typical rural and depopulation areas. In rural areas the conclusion of education process tends to precede the sexual debut. The main difference between the two generations irrespectively of the region is in the earlier sexual debut and the postponing of all other events of the younger population, especially setting up a partnership union and an independent life away form parents, which is most evident in the young urban generation. The end of Table 3: Age at life events (in years). First sexual intercourse First partnership The end of education Leaving parental home First employment 20-24 140-45 20-24 1 40-45 20-24 1 40-45 20-24 | 40-45 20-24 140-45 25th percentile Urban 16.0 17.9 22.3 20.8 17.8 16.9 19.3 18.4 19.0 18.2 Suburban 16.0 17.0 20.5 20.0 17.2 15.8 19.7 18.9 18.5 17.7 Typical rural 17.0 17.0 20.5 20.6 15.9 14.5 19.3 18.3 18.0 17.3 Depopulation 17.0 17.0 20.1 20.3 15.9 14.5 18.9 18.0 17.9 17.2 50th percentile (Median) Urban 17.0 18.9 24.8 23.3 20.6 20.5 22.7 20.3 21.3 19.8 Suburban 18.0 18.0 - 22.1 18.3 18.3 24.4 20.8 19.8 18.8 Typical rural 18.0 18.0 22.9 22.5 17.4 17.5 22.8 20.5 18.8 18.5 Depopulation 18.0 18.0 23.3 22.4 17.1 16.7 20.4 19.3 18.6 18.6 75th percentile Urban 18.0 20.0 - 26.7 - 25.8 - 24.3 - 22.5 Suburban 18.0 20.0 - 25.4 23.7 24.9 - 24.8 23.9 21.5 Typical rural 19.0 20.0 - 25.3 19.1 22.7 - 24.7 20.2 21.2 Depopulation 19.0 20.0 - 24.8 18.8 18.4 - 24.0 20.2 21.6 URBAN-RURAL LIFE SETTING AS. Informatica 32 (2008) 111-122 119 education process tends to differentiate most strongly among generations and regions. The older generation in depopulation areas is the first to finish their education, whereas the young urban generation is the last. As in case of entry into parenthood, the intervals between above listed events which are considerably shorter in rural areas then in urban ones, additionally contribute to a different fertility level in observed areas. Table 3 can be observed in relation to the contraceptive use (Figure 6). Among the younger generation, the most common contraception method used at the time of the interview was hormonal contraception following by condom and withdrawal. The rarest method used was injection. Quite substantial differences can be observed among different regions. Depopulation areas have a very low condom or diaphragm usage (6,7 percent) compared to other regions (14 percent in urban to 23 percent in suburban areas), but on the other hand, they have the highest usage of hormonal contraception (33,3 percent). However, the key finding is higher proportion of non users among typical rural dwellers (40 percent) than in other areas (24 to 33 percent). Thus, the differences among the regions in relation to the entries into certain life events, as well as in the birth of the first child, are reflected in variations of contraceptive use. Share of respondents in % 10,0 15,0 20,0 25,0 30,0 35,0 Pill = Condom Current non-user Ncn-user less than 3 months jq No sex d -g Ever non-user Withdrawal Periodical abstinence UID Pregnancy Diaphragm Injection □ urban □ suburban □ typical rural I Depopulation rural Figure 6: Current usage of contraception (20-25 years). urban suburban typical rural depopulation rural O All CH 40-45 years A20-24 years Figure 7: Attitudes towards abortion (means with confidence intervals). Attitudes towards abortion are in line with the respondents' expressed religiosity. Figure 8 shows quite high and significant differences across observed regions. Only one third of urban dwellers defined themselves as being religious, while more than a half of the inhabitants of the other three regions did. Typically, rural areas have the highest proportion of religious people. Within the younger generation, these differences are even more expressed (x2 sig. p=0.000; Adjusted Standardized Residuals for Urban 20-24 are -5.6 "Religious" and 6.4 "Not Religious"). Do not know Not religious 2025 Nether nor ^ PâlmiAiic Do not know 4045 Not religious years Nether nor 30,0 40,0 Share in % □ urban □ suburban □ typical rural ■depopulation rural 70,0 Figure 8: Religiosity. Explanation for the difference in the number of children can also be indicated in relation with the attitudes towards abortion. The results in Figure 7 show, that abortion is not a very acceptable phenomenon among the respondents. However, there is a small, but a consistent downward trend in acceptability of abortion from urban to depopulation areas. Abortion tends to be more often considered fairly unacceptable in rural areas, whereas in urban areas, the average attitude is more pro-choice oriented. In general, there are no statistically significant differences between the two observed generations, except in typical rural areas where the young show significantly greater (F=10.370***, Bonferroni post-hoc mean difference tests for Typical Rural compared to urban is -0.613***) opposition towards the abortion as the free choice act than the older generation does. The results concerning the attitudes towards marriage and gender roles reveal a rather different picture. As Figure 9 shows, irrespective of the region and the age among the respondents, there are no significant differences in the preference of either cohabitation or institution of marriage. Only the 40-45 generation in urban areas shows a significant tendency (F=8.620***; Bonferroni post-hoc mean difference tests for Urban compared to Suburban and Depopulation Rural urban are -0.2379** and -2562** respectively) towards favouring the cohabitation over marriage in comparison to other regions, whereas younger generation everywhere is more inclined towards cohabitation than marriage. These results are consistent with the share of respondents living in cohabitation unions: in urban areas the share is 15 percent, in suburban 10,8, in typical rural 12,4 and in depopulation rural 13,7 percent. All respondents also share the same tendency in attitudes towards supporting 3,6 3,4 3,2 3 2,8 2,6 2,4 2,2 2 120 Informatica 32 (2008) 111-122 M. Černič Istenič et al. the equality of gender roles (Figure 10). There are no major differences between generations or regions. 3,6 3,4 2,6 2,4 2,2 1 ] 1 ] J 1 1 > suburban typical rural depopulation rural OAll [U40-45 years A20-24 years Figure 9: Marriage vis-a-vis cohabitation (means with confidence intervals). 3,6 3,4 3 2,8 others with other relatives ncn-relatives with one parent parents and others parents and siblings children and others children partner, children and others partner and children urban suburban typical rural depopulation rural OAll □ 40-45 years A20-24 years Figure 10: Gender (means with confidence intervals). Substantive differences among regions (x2 sig. p=0.000) are revealed in the household composition (Figure 11). The largest differences are related to the rate of nuclear and extended families - of orientation and procreation -which represent two predominant household composition types in all regions and both generations. The extended family in depopulation regions of Slovenia represents one third of households, while in the urban areas, this proportion is less then 10 percent. Nuclear family is the most representative composition type where the majority of the respondents live. However, in this regards the difference among the regions is significant (Adjusted Standardized Residuals for Depopulation Rural are -5.4 "Partner and Children" and -2.5 "Parents and siblings"); depopulation areas are just above the 50 percent mark, whereas in other regions the share of this type of households is around 70 percent, being the highest in suburban areas (over 75 percent). Although one-parent family households are relatively rare, they can only be observed in the urban and suburban areas (6,9 - 4,1 percent), while in the rural areas they are more an exception then a rule (1,9 - 0,9 percent). Very similar proportions among regions hold true for those living alone. Thus, the variation in number of children in various areas is reflected in the household composition and consequently in their life style. partner _ alone 10 20 30 40 Share of respondents in % □ urban □ suburban □ typical rural ■ depopulation rural Figure 11: Household composition. 2.4 Conclusions Based on the presented results, the first and the third hypothesis can fully be confirmed. The analysis revealed that fertility behaviour of Slovenian population is significantly related to the socio-economic context of diverse socio-geographical regions; the more developed is one setting in terms of favourable and diverse economic activities, available infrastructural capacities and social services, the lower is fertility of its dwellers. Nevertheless, the application of regional typology also revealed that relations of urban-rural division are not uniform and static. Population of urban areas with the pattern of postponement of first births considerably differs from the other regions, whereas at the second and even more at the third birth, the population of suburban areas is getting gradually more similar to the urban ones. The typical rural areas are the most stable in relatively high fertility pattern, whereas in depopulation areas, this pattern is no longer very firm, most probably due to less favourable developmental conditions. Changing similarities and differences among the regions are particularly related to behavioural patterns of the younger generation. The results revealed that tempo and sequences of events in entering into adulthood changed significantly between generations. In older generation among all regions, the life events, that normatively precede parenthood, followed each other quicker than in the case of the younger generation. In older generation among the regions the variations in transition to adulthood certainly exist. But contrary to our expectations, formulated in the second hypothesis, these variations are even more pronounced among the younger generation, particularly due to distinctive behaviour of urban youth. According to the theory of second demographic transition, this group shows an indicative pattern of post-modern demographic life course both through their behaviour and expressed attitudes, whereas rural young generation, for the most part those living in typical rural areas, mainly follow the pattern of their older counterparts. As our analysis indicates, variations in fertility behaviour among the regions and generations can be more pertinently explained by social structural factors (education, household type) then values and attitudes pertaining to the post-modern life patterns. 3,2 3 2.8 2 3,2 2,6 2,4 2,2 2 URBAN-RURAL LIFE SETTING AS. Informatica 32 (2008) 111-122 121 The obtained results call for greater attention to more contextualised approaches in demographic research. Our analysis demonstrated that rural setting as a social space is not a homogeneous category, but a grouping of various structural and cultural characteristics that might specifically determine fertility behaviour. By neglecting this reality, too much information needed for better understandings of variations in fertility behaviour could be lost unduly. The residues of the past urban-rural differences in fertility behaviour are still present today despite the effects of global economic and social trends that carry structural and ideational changes into the countryside and will, as our analysis indicates, also remain in the near future. From this point of view, it was unreasonable that international project Fertility and Family Surveys in ECE countries under supervision of United Nation gave so little attention to urban-rural dimension. Among 22 European countries included in this project, only five of them (Poland, Estonia, Lithuania, Switzerland and Slovenia) took into consideration the respondents' place of residence. In the new Generations and Gender Programme (GGP), which started in the year 2000, the need for more contextual approach is fully considered; all micro, mezzo and macro levels are taken into account. On that basis pertinent results are expected. Hopefully, Slovenian researchers will also have the opportunity to join these endeavours. The observed differences in fertility behaviour patterns among generations living in various geographical regions with different socio-economic characteristics also call for diversified actions of population policies. In the near past it was anticipated that family and social policy measures would contribute to uninterrupted population reproduction in Slovenia the most successfully. However, as our results indicate, population policy also has to integrate space and regional measures in to its field of actions; i.e. urban and rural developmental programmes that will consider different every day's life conditions and needs of people living in particular space setting. To scientifically support such actions, a new research data, e.g. in the frame of GGP, that will provide the picture of the present generations of reproductive age from various viewing angles, are urgently needed. References [1] Allison, P.D. (1995) Survival Analysis Using The SAS System. A Practical Guide. Cary. SAS Institute Inc. [2] Andorka, F. (1978) Determinants of Fertility in Advanced Countries. Methuen & Co. Ltd. London. [3] Barbič, A. (1991) K "novemu" pojmovanju ruralnosti (Towards a "New" Comprehension of Rurality) In: Barbič, A. (ed.) Prihodnost slovenskega podeželja. Prostor, prebivalci, gospodarske dejavnosti (A future of Slovenian Countryside. Space, Inhabitants and Economic Activities). Dolenjska založba. Novo mesto. [4] Bell, C., Newby, H. (1971) Community Studies. Allen and Unwin Ltd. London. [5] Bonner, K.M. (1999) A Great Place to Raise Kids. Interpretation, Science, and the Urban-Rural Debate. McGill-Queen's University Press, Montreal & Kingstone • London • Ithach. [6] Boyle, P. (2003) Population geography: does geography matter in fertility research? Progress in Human Geography, 27(5): 615-626. [7] Bourdieu, P. (2003) Practical Reason. On the Theory of Action. Polity Press. Cambridge. [8] Coale, A. J., Watkins, S. C.(1986) The Fertility Decline in Europe. Princeton University Press. Princeton. New Jersey. [9] Cordon, J. A. F. (1997) Youth residential independence and autonomy: A comparative study. Journal of Family Issues. 18 (6): 576-607. [10] Černič Istenič, M. (1994) Rodost v Sloveniji. (Fertility in Slovenia) Znanstveno in publicistično središče. Ljubljana. [11] Černič Istenič, M., Obersnel-Kveder, D., Kveder, A. (2000) Differences in contraceptive behaviour of men and women in Slovenia regarding their partnership and parenthood history. Paper presented at the Flagship Conference, Brussels, May 29-31. [12] Černič Istenič, M., Kveder, A. (2000) Urban-rural differences in fertility behaviour in Slovenia. Paper presented at X. World Conference of Rural Sociologists, Rio de Janeiro; July 30 - August 5. [13] Djurfeldt, G. (1999) From CAP to CAR? New Directions in European Rural Sociology. http://www.soc.lu.se/esrs. [14] Hays, S.P. (1993) From the history of the city to the history of the urbanized society. Journal of Urban History. 19(4): 3-23. [15] Hoffman-Novotny, H.-J. (1987) The Future of the Family. European Population Conference. Issues and Prospects. Plenary Séances. Central Statistical Office of Finland. Helsinki. pp. 113-199. [16] Iedema, J., Becker, H. A., Sanders, K. (1997). Transition into independence: A comparison of cohorts born since 1930in The Netherlands. European Sociological Review, 13 ( 2): 117-137. [17] Ingleahrt, R. (1989). Kultureller Umbruch. Campus Verlag. Frankfurt/New York. [18] Javornik, J.(1999). Slovenija v 90-ih. Prebivalstvo (Slovenia in the 1990s). In: Hanžek, M. (ed.) Poročilo o človekovemu razvoju Slovenije 1999 (Human Development Report for Slovenia 1999). Urad R Slovenije za makroekonomske analize in razvoj. Ljubljana. pp. 65-67. [19] Kaa, Van de, D. (1987) Europe's second demographic transition. Population Bulletin, 42(1): 1-47. [20] Kaa, Van de, D. (1999) Europe and its population: the long view. In: Kaa, Van de, D., Leridon, H., Gesano, G., Okolski, M. (eds.) European Populations Unity in Diversity. European Population Studies no.6. Kluwer Academic Publishers. Dordrecht/Boston / London. pp. 1-49. [21] Klemenčič, V. (2002) Procesi deagrarizacije in urbanizacije slovenskega podeželja (Processes of deagrarisation and urbanisation of Slovenian 122 Informatica 32 (2008) 111-122 countryside). Podeželje na prelomu tisočletja (Countryside at the break of the century/ Mednarodna znanstvena geografska konferenca (International scientific geographic conference), Ljubljana, 19.-21. September 2001. Ljubljana. Oddelek za geografijo Filozofske fakultete. Accessible on: (http://www.ff.uni-lj.si /oddelki /geo / publikacije/dela/files/Dela_17/Vladimir_Klemencic .pdf) [22] Kovačič, M, Perpar, A., Gosar, L. (2002) Členitev podeželja v Sloveniji (Differentiation of countryside in Slovenia) Sodobno kmetijstvo. 35(2): 62-66. [23] Kovačič, M., Gosar, L., Fabijan, R., Perpar, A. (2000). Razvojno - tipološka členitev podeželja v Republiki Sloveniji (Developmentally -Typological Differentiation of Rural Areas in the Republic of Slovenia.). Institute of Agricultural Economics, Biotechnical Faculty, University of Ljubljana. [24] Kovačič, M. (1995) Idejne in teoretske osnove za razvoj podeželja (Ideal and theoretical basis for the countryside development) In: Kovačič, M. (ed.) Izhodišča, sestavine in problemi celovitega razvoja podeželja v Sloveniji (Starting-points, components and problems of integral countryside development in Slovenia) Biotechnical Faculty, University of Ljubljana, pp. 3-20 [25] Livi-Bacci, M. (1986) Social-Group Forerunners in Fertility Control in Europe. In: Coale, A. J., Watkins, S. C. (ed.): The Fertility Decline in Europe. Princeton University Press. Princeton. New Jersey, pp.182-200. [26] Mackensen, R. (1982) Social Change and Reproductive Behaviour - on Continuos Transition. In: Hohn, C., Mackensen, R. (eds.) Determinants of Fertility Trends: Theories Re-examined. Ordina ed.. Liege. pp. 251-279. [27] Mahon, M. (2001) Reconceptualising rural change. A literature review. Paper presented at XIX European Congress for Rural Sociology. Dijon, September 3-7. [28] Malačič, J. (1990) Načrtovanje zdravega potomstva - odgovornost posameznika do družbe (The planning of healthy future generations - the individual's responsibility towards society). Zdravstveni vestnik. 59(7-8): 349-354. [29] Marsden, T. (1998) New rural territories: Regulating the differentiated rural space. Journal of Rural Studies. 14(1): 107-117. [30] Marsden, T. (1999) Rural futures: The consumption countryside and its regulation. Sociologia Ruralis. 39(4): 501-520. [31] Nave-Herz, R. (1997) Still in the nest: The family and young adults in Germany. Journal of Family Issues. 18(6): 671-689. [32] Newby, H. (1978) Property, Paternalism and Power: Class and Control in Rural England. Hutchinson. London. M. Černič Istenič et al. [33] Newby, H. (1980) Urbanisation and the Rural Class Structure: Reflections on a Case Study. In The Rural Sociology of Advanced Societies: Critical Perspectives. Croom Helm. London. [34] Notestein, F. (1953) Economic problems of population change. In: Proceedings of the Eight International Conference of Agricultural Economists. London. Oxford University Press. pp. 13-31. [35] Obersnel Kveder, D., Kožuh Novak, M., Černič Istenič, M., Šircelj, V., Vehovar, V., Rojnik, B. (2001) Fertility and Family Surveys in Countries of the ECE Region - Standard Country Report, Slovenia. Economic Studies No. 10r, UNECE, UNFPA, New York/Geneva. [36] Pahl, R.E. (1965) Urbs in Rure: The Metropolitan Fringe in Hertfordshire. Wiedenfeld & Nickelson. London. [37] Pahl R.E. (1966) The rural-urban continuum. Sociologia Ruralis. 6(3): 299-329. [38] Pečan, J., Ravbar, M. (1999) Regionalni razvoj Slovenije (Regional Development of Slovenia). In: Hanžek, M. (ed.) Poročilo o človekovemu razvoju Slovenije 1999 (Human Development Report for Slovenia 1999). Urad R Slovenije za makroekonomske analize in razvoj. Ljubljana. pp. 31-63. [39] Perpar, A., Kovačič, M. (2002) Razvojno stanje, značilnosti in problemi podeželskih območij v Sloveniji (Development situation, characteristics and problems of rural areas in Slovenia). Sodobno kmetijstvo. 35( 2): 52-61. [40] Sharlin, A. (1986) Urban-Rural Difference in Fertility in Europe during the Demographic Transition. In: Coale, A. J., Watkins, S. C. (ed.): The Fertility Decline in Europe. Princeton University Press. Princeton. New Jersey, pp.234257. [41] Statistical Yearbook (2003) Statistical Office of Republic of Slovenia. Ljubljana. [42] Šircelj, V. (2007) Fertility and educational attainment in Slovenia. Anthropological Notebooks. 13(2): 11-34. [43] Šircelj, V., Ilič, M., Kuhar, A., Zupančič, M. (1990) Projekcije prebivalstva R Slovenije 1986-2006 (Population projection of republic of Slovenia 1986-2006). Statistical Office of Republic of Slovenia. Ljubljana. [44] Šircelj, V. (1991) Determinante rodnosti v Sloveniji (Determinants of fertility in Slovenia): Doctoral Dissertation. Faculty of Arts, University of Ljubljana. [45] Vrišer, I. (1998) Središčna (centralna) naselja. (Central (capital) settlements). In: Geografski atlas Slovenije. Ljubljana, DZS. pp. 308-309. [46] Williams, W.M. (1963) A West Country [47] Village: Ashworthy: Family, Kinship, and Land. Routledge Kegan Paul. London.