Differentials in Female Age at Marriage in Pakistan: Have they Changed or Not?

Age at marriage is an important factor to study the marriage patterns of a country because of strong relationship between age at first marriage and childbearing. In this study, effects of one major socio-cultural factor female education and place of residence have been illustrated. The present study attempted to find differentials in female age at marriage in Pakistan since Pakistan has begun to face early demographic transition since 90’s. The investigation of differentials in female age at marriage in this study is generally based on data analysis of Pakistan Demographic and Health Survey (PDHS) 2006-07. While comparative analysis also discussed in a separate section with findings of PDHS, 2012-13. The age at first marriage of women lived in major urban areas seemed to carry with them the influence of the urban lifestyle and are also influenced by the educational enrollment opportunities.


INTRODUCTION
Female age at first marriage which is an important dimension of population dynamics in demographic perspectives is found to be significantly associated with potential period of childbearing in a society. In the traditional Pakistani society, marriage at an early age and early childbearing are strongly encouraged. It is found in several studies that female age at marriage is affected by different socio-economic and cultural factors like; education, urbanization and work patterns [Afzal, et al. (1976) and Hashmi and Zafar (2000)]. Hence, there is a need to find out the influence of education and urban residence on female age at marriage and delayed marriage decisions. The present study thus focuses on the question of how urbanization and education influences the female age at marriage in Pakistani society.
In the cultural context of Pakistani society, a marriage generally means a relationship between families rather than only between the couple concerned [Fazalbhoy (2006)]. Further, marriages in Pakistani society require the Nikah (marriage contract), which is binding on both parties, and the Mehr (An agreed amount paid by groom to bride as per Islamic law) which is a gift given by the groom to the bride on the occasion of marriage, for a marriage to be considered legally and socially acceptable.
It is preferred that marriages should be arranged by parents rather than by the couple concerned [Shaw (2001) and Hussain (1999)]. Apart from sectarian differences, South Asian Muslims are divided into communities, called Biraderi (Group of community members belonging to same caste), Jamaat (group), and Khandan (Family) to name a few. These communities or groups in South Asian region have different traditions which arise out of the caste to which they had belonged prior to their migrational history or the region to which they belong may practice exogamy. Endogamy is preferred by the previously discussed communities, the stipulation being more strongly enforced for women, so as to preserve community identity [Hussain (1999)].
In general, marriage is universal for girls, and early marriage is preferred [Fazalbhoy (2006)]. Marriage is seen as the best security for a girl. Chastity is highly valued for girls and is a reason for preferring early marriage. Marriages are preferably arranged by parents, and such marriages have a higher status in the opinion than love marriages, where the couples have exercised their own choice [Fazalbhoy (2006)].
While the legal age at marriage is 18 in Pakistan the average age for first marriages among women is 18.8 [ PDHS (1990)]. It is estimated that less than 50 percent of women in Pakistan marry before the age of 15 years as mentioned in PDHS 1990-91. Age at marriage is most important to study as; Pakistani society takes marriage as a natural and expected part of being a grown-up, and they continue to get married for family formation and childbearing in early ages [Afzal, et al. (1994) and Sathar and Kiani (1986)].
Age at marriage is an important factor to study the marriage patterns of a country because of strong relationship between age at first marriage and childbearing. In most of the Muslim societies, childbearing starts after marriage. For instance, in North Sudan pre-marital sex is forbidden by Islamic culture and marriage is necessary for family formation and so childbearing [WFS Report (1975)]. Various studies in marriage have shown that marriage patterns are affected by a number of socio-economic factors, cultural factors, religious and ethnicity factors. For instance, in the Republic of Korea several socio-economic, cultural and ethnic factors have considerably affected the individuals age at marriage [Kim and Stinner (1980)]. Similar positive relationship have been found for these determinants of age at marriage in different traditional societies like Bangladesh, India and Tunisia as well as in modernized societies like Japan, Malaysia and Thailand [WFS (1975), Aryee (1985), Goyal (1988) and Badrud Duza and Baldwin (1976)]. Some other major factors which can affect female age at marriage includes urbanization as, in many countries participated in the World Fertility Survey [WFS (1985)] found that significant differentials exists in female age at marriage by urban and rural areas. The above discussion shows that urbanization and female education, considerably effects female age at first marriage in a society. In the present study these factors are considered to explain the effects on female age at first marriage in Pakistan using the Pakistan demographic and Health Survey 2006-07 while in comparison with findings from PDHS, 2012-13.
In Pakistani society marriage is a universal phenomenon and consanguineous marriage is a major feature of the society [Afzal, et al. (1994), Shaw (2001) and Bittles (1994)]. Mostly marriages are practiced as a religious obligation for family formation and child bearing because most of the childbearing occurs within marriages in Pakistan [Sathar and Kiani (1986)]. Moreover, marriages are particularly arranged by parents or guardians of a family in Pakistan [Shaw (2001)]. It is reported in a number of demographic studies that in Pakistan, age at marriage is particularly lower in consanguineous marriages which are contracted between close kins as compared to non-consanguineous marriages [Afzal, et al. (1994), Shaw (2001), Bittles (1994) and Hussain (1999)]. Age at marriage is affected by several socio-economic variables and leads to marital postponement [Duza and Baldwin (1976)]. It is documented in many marriage studies that other modernity factors like the socio-economic factors, female education, women's participation in labour market and increasing urbanization are the leading factors to increase the female age at marriage [Goyal (1988) and Duza and Baldwin (1976)].

LITERATURE REVIEW
The present paper studies age at first marriage because of its strong relationship with population dynamics like fertility and also its prevalence in demographic literature. Women's years of schooling was chosen because there is a natural correlation between age at marriage and education level (i.e., higher the education, higher will be the age at first marriage). The second variable place of residence has been described as the other best predictor of marital postponement. The question analyzed in this study whether differences in female age at marriage in Pakistan can be predicted by female education level and by type of place of residence. Differentials in female age of first marriage for educated females from urban and rural background are also analyzed to find out any significant changes in female age at first marriage in Pakistan. Therefore, the focus of this study is whether place of residence as well as differences in education level are correlated with female age of first marriage and if not so then what can be other associated factors. Karim (1984) observed variations in age at marriage of evermarried women from Pakistan Fertility Survey (PFS) and concluded in his study that female age at marriage has increased in Pakistan during the past decades especially during the 1960's. During the decade of 1960's Pakistan also experienced an expansion of rural economy through the growth of non-agricultural sector due to pace of increasing urbanization. The author identified a number of factors that were associated and differentials in age at first marriage of females such as; region of residence, and childhood place of residence, provincial and ethnic affiliations, women's educational attainment and the levels of female labour force participation. Moreover, the author argued that a variety of social changes lead to variations in marriage pattern so females. Another study of age at marriage examined these social changes in the Pakistani society using the term social change [Afzal, et al. (1976)]. Sathar and Kiani (1986) examined the data from the migration module of the Population, Labour Force and Migration (PLM) Survey of 1979 to explore the impact of modernity factors like education, modern occupations, urbanization and migration on age at marriage by current age groups in Pakistan. The authors concluded that the study showed a substantial increase in proportion never married during the 1960s and subsequently until 1981. The results of the study showed that proportion never married for females in the age 15-29 age groups has increased and this increase was more pronounced for females than males in the same age-group. Moreover, delayed marriages occurred to a greater extent among young educated females living in urban areas than among the females of rural background.
Changes in marriage patterns have been at the core of demographic transitions in Europe and also in the Asian countries [Caldwell (1993)]. It is based on the assumption that most of the childbearing occurs within marriage in Pakistan and delayed marriages of females has direct impact of delaying the age of sexual initiation and the age at first birth [Sathar and Kiani (1998)]. The important findings of the study were that rising female age at marriage is the main factor for decreasing fertility levels especially a decline in marital fertility. However, an emerging trend of marriage squeeze is also introduced in the study that has its own implications for the process of population growth rate in Pakistan. The authors also showed that marriages are delaying for females in Pakistan and conversely there is a rise in nevermarried proportions in the age-groups of 15-29.
Female age at marriage is strongly associated with fertility and therefore levels of fertility in a society. The focus of this study is on determining factors which affect directly female age at first marriage in Pakistan. As women make up half the country's population so it is important to examine whether their delayed age at marriage affects levels of number of children and fertility in different spheres of society undermine the county's potential for reduction in fertility. Specifically, this study analyzed effect of female education and place of residence on women's age at marriage to measure factors of delay in age of first marriage for women in Pakistan.
On the other hand Sathar and Kiani (1998), in a study it has also been observed that the delay in marriages does not show any differentials by education as it is hard to conclude using cross-sectional data that rise in age at marriage is caused by pursuit of educational opportunities particularly because it is occurring even among uneducated people. While in recent times education and particularly female education is rising. In the context of this research study the differentiating factors of interest were assumed to be influencing female age of first marriage in the country and therefore considered to analyze using Pakistan Demographic and Health survey [PDHS (2006-07)].
Women's higher education and participation in decision-making in Pakistani society seems to be an indicator of progress and linked to delay in their age of first marriage and number of children. This happened in Pakistan whenever the evolving indicators of gender equality paved the way of Pakistani society for any transformation. Similarly, delay in female age at first marriage is associated with socioeconomic indicators particularly women empowerment and that marriage is usually associated with the formation of a separate household from the family of origin and the creation of a new family in our society.
In demographic terms comparison with developed nations Maubrigades (2015), reported that the changing pattern of marriage in European societies has been enormously important as a regulatory mechanism in population growth. In this context, during the demographic transition the European marriage pattern contributed to a decline in fertility because people married later and many people did not marry at all. According to an important current of thought in historical demography, the age at which people first marry at least in Western Europe is an important determinant of population growth. While in the context of Pakistan, focusing on demographic theory which has one main feature of delay in women's age at first marriage this study has studied effects of some modern factors like female education and urban place of residence. Further, all independent variables are discussed briefly below: • Urbanization Urbanization is positively associated with age at a marriage of individuals as different studies on marriage patterns reported higher age at marriage for urban females than rural women [Smith (1980) and Goyal (1976)]. However, there are also major differentials in female age at marriage by type of place of residence, i.e., Major urban, other urban and rural. Goyal (1976) reviewed various studies of Indian age at marriage patterns and found that women living in major urban areas generally have higher age at first marriage than the females living in the other urban areas or town. The author also documented state variations in female age at marriage as females from metropolitan areas like Mumbai mostly have higher age at first marriage than the females from other urban and rural areas. Thus, it is argued that women living in metropolitan areas and big cities are more likely to delay marriages than the females from towns and rural areas [Mahmood and Ali (1984)]. The final factor of focus in this study is place of residence prior to marriage. This is further broken down into major urban, other urban and rural regions, as this method has not been used in previous research. Place of residence has been studied infrequently in relation to marital postponement, which are also less prevalent than the second variable of education level. •

Role of Education
Education is the major factor associated with female age at marriage in Pakistan. A number of studies [Karim (1984), Afzal, et al. (1985), Aryee (1985), Gangadharan and Maitra (2000) and Psacharopoulos and Woodhall (1985)] have shown that women's education levels are closely associated with age at first marriage of females. It is also argued that a rise in women's education level leads to increase in the age at marriage and at the age at which women have their first birth are key features of demographic transition in any country which further leads to nuptiality transition [Gangadharan and Maitra (2000) and Duza and Baldwin (1976)]. These studies have shown that education considerably affects the marriage timing and leads to higher age at marriage for educated females. Given that education level generally increases as age increases and that age is positively correlated with marital postponement, it seems reasonable to suggest that education level would be positively correlated with later age at marriage. While studies have explored whether higher education for women could be predictive of marital postponement has been explored, no studies have been devoted specifically to whether any factors other than education level are related to delayed marriages. • Type of Residence Nuptiality transition in various developing countries shows that there are significant urban-rural differentials in the timing of first marriage for females [Smith and Karim (1978) Karim (1994), Goyal (1976), Mahmood and Ali (1984) and UN (1981)]. It can be argued that urbanization through the expansion of rural economy brings a social change in a society that can influence marriage timing of individuals in a country [Afzal, et al. (1976)]. Urban-rural differentials in female age at marriage in Pakistan have increased due to the pace of urbanization in Pakistan over the past decades [Abbasi (1987) and Arif and Ibrahim (1998)]. Aryee (1985) in his study argued that an urban location is much different from a town or rural counterpart because its population tends to be more heterogeneous in its characteristics such as ethnic variations, education levels, labour force opportunities, etc. This heterogeneity also affects the attitudes, beliefs and behaviour patterns of a population towards marriage decisions than that of rural households. Afzal, et al. (1976) in a study of marriage patterns in Lahore suburbs has shown that women living in urban areas marry later than women in rural areas. Therefore, urban women particularly show higher age at marriage as compared to females living in rural areas.

RATIONALE
In the context of rise in women's education levels, an increase in the age at marriage and an increase in the age at which they have their first child are key features of demographic transition in any country. Education is considered to be an essential component in this process because increases in educational attainment are likely to significantly affect both age at marriage and the duration between marriage and first birth -in particular increasing both the age at marriage and the time to first child. Similarly, place of residence and life style significantly effects timing of first marriage. As in urban areas women's age at first marriage is generally higher than those living in rural areas. While both of these discussed factors have strong influence on female's age of first marriage in a community.
The present study attempted to find differentials in female age at marriage in Pakistan since Pakistan has begun to face early demographic transition since 1990s which generally starts from delayed marriages through rising female age at marriage. The female age at first marriage is strongly associated with reproductive life span or years of child bearing. In Pakistan, where maternal death toll is still on 76 deaths per 100000 live births and infant mortality is still high to about 78 deaths per 1000 live births. Female age at marriage effects years of child bearing, age at first pregnancy, mother and child health and also average number of children. There is a need to provide improved counseling to married women who married at younger age about child bearing and health care. This study will measure factors of interest which can contribute to rise age at marriage in the culture of Pakistan. Particularly in this study effect of education and place of residence (urban/rural) has been measured.
In Pakistan where a sufficient percentage of women married to age group 15-24, maternal and child health (MNCH), child bearing, number of children, demand and practice of family planning methods and unmet need for contraception are critically associated components which could be studied in separate context.

Objectives of Study
This study only focused on differentials in female age at first marriage in Pakistan and estimated influencing of female education which is a socio-cultural factor. Therefore, the major objectives of the present study are to: • Study the patterns and levels of female age at first marriage in Pakistan; • Analyze the differentials in female age of marriage by ruralurban place of residence in Pakistan; and • Examine the differentials in female age at marriage by level of education.

METHODOLOGY 4.1 Data Source
The investigation of differentials in female age at marriage in this study is generally based on primary and to some extent secondary data analysis which particularly obtained from Pakistan Demographic and Health Survey (PDHS) 2006-07. PDHS is a national level survey of evermarried women of reproductive aged 15-49 years and used as primary source of data in this study. It is the second survey of its type which was conducted in Pakistan since its unification in 1990-91 [NIPS (1990-91) and (2006-07)]. This survey collected information on child health, marriage, fertility, family planning, fertility preferences, prenatal care and maternal mortality. 10,601 ever-married married were identified as eligible respondents in the survey while 10,023 successful interviews were conducted from eligible women in the households and the response rate was 94.5 percent ( Figure.1). While comparative analysis also discussed in a separate section with findings of PDHS, 2012-13. Geographical representation of respondents in PDHS 2006-07 shows that Punjab and Sindh have highest percentage, i.e., 57.9 percent and 24 percent respectively while NWFP and Balochistan have lower percentage of ever-married women that were interviewed in PDHS 2006-07.

Measurements
The dependent variable in the present study was female age at first marriage based on a question for ever-married women's questionnaire that measured in years that; "how old were you when you first started living with husband?" Month and year of first marriage are also asked from the respondents in the survey.
The independent variables include the social and some other characteristics of ever-married women which are hypothesized to influence female age at marriage in Pakistan. The control group variables include women's education level and work experience before and after marriage, province, region, type of residence and husband's education and occupation. Table 1 presents the measurements of all these variables.
ANOVA technique was used to examine the influence of variables such as; female education and place of residence on female age at marriage in Pakistan. The dependent variable in ANOVA analyses examined the changes and differentials in female age at first marriage by women' years of schooling, and by type of place of residence separately. Secondly, logistic regression was also applied for further analysis in this study, in dichotomous form of dependent variable of female age at first marriage. The purpose of this analysis was to measure the effect of certain independent variables of social factors which affecting female age at marriage in Pakistan including; years of schooling, and type of place of residence.

• Bivariate Analysis
In addition to evaluate the relationships between variables, researchers are often interested in comparing two or more groups of people. As members of the same group changed over time or if there are some differences among the groups on defined criteria. One-way analysis of variance (ANOVA) is mathematically a ratio (the F value), comparing the distribution of values from a variable among three or more groups in terms of the variance within the groups. Similarly, variances are deviations from the means hence we can use ANOVA to compare means. When there are greater differences between the categories than within the categories then we can conclude that differences in the means of a variable among the three or more groups being compared are statistically significant. When the variances between categories are similar to the variances within each category and the ratio is around 1.0 which resulting in a conclusion of no significant differences in means among the groups being compared [Nardi (2006)].
In this section bivariate analysis is conducted to find the effects of key factors associated with the female age at first marriage. All statistical analysis was conducted using SPSS 13.0 and 16.0. In order to find whether a relationship exists between the independent variables of urban-rural place of residence and education level, and the dependent variable of age at first marriage ANOVA technique is used in this study. This method of analysis is appropriate because it illustrates whether there are relationships between variables which is the major approach of this study.
The dependent variable in the analysis was female age at first marriage measured in years. Factors associated with the delayed marriages have been assigned two conceptual categories of whether the ever-married sampled women had attended school which measured in years of schooling, and the factors associated, whether they lived in major urban areas, other urban areas or rural region.
While from a total of 10,023 married women of reproductive age (15-49) years, more than half the sample of women showed no-education prior to marriage, more than 50 percent of women aged between 15 and 49 years started living with their husband i.e. got married with no education, about 20 percent women married with primary education or higher and only 06 percent completed secondary or more education at the time when they were living with their husbands.
The results revealed that with almost secondary or higher years of schooling, age at first marriage, increased for women in Pakistani society. While looking at the relationship between the timing of first marriage and type of place of residence, we find that women who lived in major urban areas prior to marriage have a significantly higher age at first marriage of over one year in comparison to the 66 percent of the sample whose age at first marriage is one year lower than the women living in major urban areas. The next higher age at first marriage found among women who lived in major urban areas and at the same time had higher years of schooling with an average of more than 03 years. It is interesting to note that almost a half of the sample had no education at all and had the lowest age at first marriage among all the education groups (Table 2). Bivariate analysis presented that both the years of schooling and place of residence are significantly associated with female age at first marriage. Women with more years of schooling have age of first marriage about 5 years greater, while women with secondary years of schooling have a mean of about one year lower. On the other hand, women who did not attend school or with limited education have major difference in mean age at first marriage as compared to women with secondary years of schooling or higher, as shown in the following table. Similarly, women from major urban areas have higher age at first marriage within the sample of about (0.517 years) than women who did not live in major urban areas prior to marriage. The results have shown that rural women married about one years (-0.985 years) before than the women living in major urban or other urban areas (0.469 years).
Overall, the above given bivariate analysis presented in Table. 02 show that woman's education in years and place of residence are strongly associated with the length of her age at first marriage. Women who lived in major urban areas have the highest age of first marriage with an average of 21 years, while those who lived in rural areas have the lowest age of marriage. As expected, the age at first marriage of women lived in major urban areas seemed to bring with them the influence of the urban environment and are also influenced by the school enrolling opportunities. Women who spent their life in other urban counterparts have a slightly lower average age of first marriage of 20 years than women who lived in rural areas 18.5 years. Following table indicated that women with no schooling showed greater difference than women who have middle, secondary and higher education. As 64 percent women were married with lower average age at first marriage than women with more years of schooling. Table 3 also presents the percent distributions of the variables used in the analysis as well as the mean difference in age at first marriage for females according to the variables of interest, using ANOVA to test the significance of the differentials in female age at first marriage in Pakistan. The results shown that women's age at first marriage are significantly affected by place of residence as women who have lived in major urban areas have higher age at first marriage, with an average of almost one year, while those women who reported to be lived in rural areas have almost one year lower age at first marriage. As expected, age at first marriage of the women in the other urban category fall in between these two figures. Hence, women seemed to influence by their living environment. Women who lived in urban areas have a slightly higher average age at first marriage (0.5 years) than women who lived in rural areas (-0.985 years). The findings of this study highlighted the significant effects of education and urban/rural place of timing of female marriage in traditional society of Pakistan. Although, data in this study collected from all four provinces of Pakistan, however, variations in the study were not addressed by region rather overall findings have been presented which represented age at first marriage for population groups from diverse backgrounds in the society of Pakistan. Secondly, results focused only on the differentials in female age of first marriage and reported findings are similar with results of other studies of Pakistan and other regions that female age at marriage strongly is strongly influenced by higher education and urban lifestyle. Therefore, the results of this study illustrate the fact that female age at first marriage is generally influenced by different cultural and social factors in community settings as discussed in several previous studies and in the literature.

Multivariate Analysis
Logistic regression is applied for further analysis in this study, in dichotomous form of dependent variable of female age at first marriage. The purpose of this analysis is to measure the effect of certain independent variables which include social factor of female years of schooling, and type of place of residence in Pakistani society. Table 4 presents the OLS estimates from the multivariate analysis of female age at first marriage place of residence and years of schooling. This analysis is based on two categories of age at first marriage, i.e., female age at first marriage <20 years and >20 years. We find whether years of schooling and place of residence affect woman's age of first marriage and how selected variables are significantly associated with the gap.

i. Age at First Marriage and Female Education
Pakistani society is a traditional society with very low literacy rates particularly for females. Research has shown that women are more likely to have delayed age at first marriage with higher education [Gangadharan and Maitra (2000) and Duza and Baldwin (1976)].
The analysis of selected variables in the following showed that both female years of schooling and place of residence are found to be strongly associated with female age at first marriage. Compared to women who have no or limited education with educated females have a significantly longer interval by two years. Women with secondary years of schooling also have significantly later age at first marriage than the reference group.

ii. Female Age at First Marriage and Place of Residence
Similarly, moderate relationship between place of residence and female age of first marriage is explained in several demographic and social studies [Mahmood and Ali (1984), Smith (1980) and Goyal (1976)]. The analysis of selected variable in Table 4 presented the effects of urban-rural residence on timing of female marriage in Pakistan. It shows that place of residence is significantly associated with female age of first marriage in Pakistani society. Compared to women in major urban areas, those who live in rural regions have age at first marriage of about one year lower, while those who live in other urban have shorter difference in female age of first marriage than the reference group. However, the difference in female age at first marriage does vary significantly in some regions. Women in other urban counterparts have greater difference in age at first marriage than those who live in rural areas and shorter difference with the reference category.

DISCUSSION AND COMPARISON WITH PDHS 2012-13
Findings from PDHS, 2006-07 indicated that women who lived in urban areas have the highest average age of first marriage, i.e., 21 years, while those who lived in rural areas have the lowest age of marriage. As expected, the age at first marriage of women lived in major urban areas seemed to carry with them the influence of the urban lifestyle and are also influenced by the educational enrollment opportunities. Women who spent their life in other urban counterparts have a slightly lower average age of first marriage of 20 years than women who lived in rural areas (18.5 years).
Statistics about female age at first reported in PDHS 2012-13 illustrated similar levels and patterns as women living in urban areas have almost the same age at first marriage and women from rural counterparts married in younger age as data reported from PDHS 2006-07. Only one major difference in age of marriage is for women with secondary education (22.3 years), which seemed higher than in retrospective PDHS 2006-07 survey.
In order to understand the practices of marriage trends among women in Pakistan, there has been a link with trends and indicators in the literature cited above, which considers the women's education levels and area distribution of the population.

i. Age at Marriage and Female Education
Education is the main factor that promotes wide options among women regarding the personal choices because it is related to increased personal autonomy in terms of developing their human capital in the labour market as well to greater scope for independent search in the marriage market. Hence increased formal education tends to make women for delayed marriages and decide not to have children early. In modern societies education is seen as more important than the social status that men and women attain through marital relationship. However, the connection between years of schooling and changes in decision about marriage is clear. In case of Pakistan, the findings of study showed that women's increased years of education are positively correlated with delayed marriage. Even though general trends show a positive relation between the two independent variables there are many differences within effect of these variables of female age of first marriage in Pakistani society.

ii. Age at Marriage and Urbanization
Effect of the place of residence on women's age at first marriage could be combined in diverse types of approach. The urban-rural differentials in marriage patterns can be explained not only by place of residence but also by differences in the structure of the population that live in urban/rural areas. In Pakistan, rural population have a relatively high proportion of people who maintain their cultural and family patterns and social control in their communities is strong to much extent in personal matters such as marriage. These traditional marriage patterns imitate more formal marriage trends and also an early female age at marriage. However, in urban settings social control is less evident for women and they could be more exposed to modern life styles that can lead to delay decision of marriage. However, findings from this study showed that other factors on the other hand may play an important role in explaining the observed differences between rural/urban areas. As discussed above, one is easy access to formal education which led individuals to postpone their entry into formal marriage and second is major urban place of residence.

CONCLUSION
The results of bivariate analysis emphasized the key factors associated with women's age at first marriage in Pakistan. Important variables associated with delayed age at first marriage were whether women achieved higher education and whether they were living in urban areas. We can conclude that rural place of residence affects female age at first marriage as rural women married about one year before than the women living in urban areas.
Conversely, the findings indicate that attending school increases the time period till women stay unmarried. On the other hand, education in itself is a positive outcome as it is the associated delay in marriage such that it allows women greater opportunities for autonomous development in achieving higher education and capacity building. Moreover, delayed age at marriage might mean that women have longer pause between marriage and first child birth.
The associations between factors like place of residence and delayed marriages are strong as the direct correlation between female education and age at first marriage as discussed above. Even though the distribution of women's age at first marriage by their school attending years shows that higher educated women in Pakistan are more likely to have higher age at marriage, and the association is highly significant as it directly affects the other factors of women's individual development. This is true for major urban residence as well.
To end this discussion, more efforts are required to be devoted in understanding the issues involved in women's age at first marriage in Pakistan which beginning with research on female age of first marriage as it is directly associated with child bearing and thus population growth rate of Pakistan . Currently, national level surveys such as the PDHS collect demographic and health information on ever-married women only. However, the data of the survey do not include variables that measure the respondents' role in decision-making process, dowry and about ideal age at marriage in our society. Moreover, information about family's socio-economic characteristics can help to understand the influences of parent as well as women's decisions about age at first marriage in Pakistan.