Education role in early marriage prevention: evidence from Indonesia’s rural areas

Abstract

Background

Adolescent girls might suffer physical and psychological harm from early marriage. Meanwhile, a good education can make women more independent in making decisions for their good. The study analyzes the role of education level in early marriage among adolescents in Indonesia’s rural areas.

Methods

This cross-sectional study analyzed 4,360 respondents. It used early marriage as an outcome variable, education level as an exposure variable, and six control variables (age, employment, socioeconomic status, age of the first sexual intercourse, access to family planning (FP) information from print media, and access to FP from electronic media). The study used a binary logistics regression in the final test.

Results

The study shows that the average early marriage among adolescent girls in Indonesia’s rural areas was 11.9%. Teenage girls in elementary school were 448.926 times more likely than college girls to experience early marriage (AOR 448.926; 95% CI 36.384-5539.050). Meanwhile, adolescent girls in junior high school were 146.171 times more likely to experience early marriage than college girls (AOR 146.171; 95% CI 13.140-1625.997). Moreover, adolescent girls in senior high school were 21.355 times more likely than those in college to perform an early marriage (AOR 21.355; 95% CI 2.017-226.067).

Conclusions

The study concluded that education level had a role in early marriage in Indonesia’s rural areas. The higher the education, the lower the probability of experiencing early marriage.

Introduction

Marriage generally occurs in the age range of 20 to 30 years or entering early adulthood [1, 2]. However, the reality is that many marriages are carried out at a young age (< 18 years). Early marriage is formal or informal, with one or both spouses under 18 [3, 4]. Marriage in Indonesia used to be regulated in Law Number I of 1974, which stated that marriage could be carried out if the man was at least 19 years old and the woman was at least 16 years old. The government changed Law Number 1 of 1974 with Law Number 16 of 2019 [5], which states that the age of individuals who may marry is at least 19 years for women and men. This change is due to protecting children’s rights, gender discrimination, and the mental maturity of married couples to reduce the risk of maternal and child mortality, produce healthy and quality offspring, and reduce the divorce rate [6].

Globally, more than 12 million girls marry at a young age (under 18 years) every year [7]. About 21% of young women marry before the age of 18. About 37% of these child marriages occur in sub-Saharan Africa and 30% in South Asia [7, 8]. Three countries in South Asia are in the high category for early marriage rates, namely India, Bangladesh, and Nepal [9]. Likewise, in several Southeast Asian countries, early marriage is still widely practiced. Estimates from a recent national household survey show that the proportion of women aged 20–24 who were married or in a union before the age of 18 ranged from 11% in Vietnam to 33% in the Lao People’s Democratic Republic and the Philippines. Southeast Asia’s adolescent birth rate remains high, especially compared to other Asia-Pacific subregions, with an average of 38.2 births per 1,000 women aged 15–19 [10]. In Indonesia, the incidence of early marriage has also increased. According to the Indonesian Statistics report, 1.7 million marriages were found throughout 2022. The early marriage rate among adolescents in Indonesia from 2015 to 2020 has increased by 70% [11]. Indonesia is in second place after Cambodia, with a child marriage rate of 27.6% or around 23 million children who married in Indonesia in 2018; 59% of them lived in Java, or around 668,900 women [12].

Several theories can be used to explain how education affects marriage decisions and patterns. First is human capital theory, which views education as an investment. The higher a person’s education, the greater their potential for getting a better job and higher income. With these increasing economic opportunities, people often delay marriage to pursue a career or further education [13]. Second, the modernization theory explains that modernization brings changes in social structures and societal values, one of which is the importance of education and delaying marriage [14]. Third, the theory of preferences and time allocation explains that education increases an individual’s ability to allocate time and priorities effectively. Education provides an understanding of the importance of life planning, which includes considering postponing marriage until financial or personal stability is achieved [15]. Fourth, rational choice theory emphasizes that individuals make decisions based on cost-benefit calculations. Education influences individuals to consider marriage more rationally, considering the long-term impact on their careers, finances, and quality of life [16]. Finally, gender and emancipation theory explains how education enables women’s emancipation by providing more access to knowledge and skills [17]. The phenomena of each theory can apply alone, alternately, or together in each event.

In previous studies, early marriage has been associated with social and health impacts, confirmed in various studies [18, 19]. Mainly related to reproductive health impacts on female adolescents, such as experiencing pregnancy more quickly due to high fertility rates, risk of complications in pregnancy and childbirth [12, 19], increased risk of cervical cancer due to frequent sexual activity [20], not getting information about the use of contraceptive methods, and lack of access to general health services, particularly maternal healthcare [19, 21]. Adolescent girls are often in relationships characterized by significant age differences with their partners, which can consequently reduce bargaining power and decision-making [22]. Teenage girls are at higher risk of experiencing intimate partner violence than their counterparts who marry later [18, 19, 23].

Several factors influenced early marriage, including the lack of economic factors, education level [16–18], and religious fanaticism [24, 25]. In addition to financial and educational conditions, the mass media and the number of family members also contribute to the incidence of early marriage [26]. A previous study found that there are parents who felt early marriage occurred because it was considered a solution to solving problems such as escaping poverty [27]. Qualitative studies found factors that cause early marriage, including economic factors and sociocultural factors, namely social habits, cultural beliefs, community encouragement, social learning, and social support; next are individual factors, such as physical characteristics, personality traits, lack of awareness of the impact of early marriage, and fear of an uncertain future, family factors such as weak parental awareness, dysfunctional families, and prevention of harm, and structural factors namely high demand, limited access to educational services, lack of social and legal structures that support [28].

Rural is an area prone to early marriage, and one of the factors is the low opportunity to work [29]. Rural communities are still thick with local culture; some researchers reveal that early marriage is associated with cultural heritage, such as girls’ weakness and inability to refuse [27, 28]. A study in Gambian villages found that culture and fear of premarital sex among girls were essential factors in early marriage [29]. The multifactorial causes of early marriage have been widely studied, but research focusing on education levels and rural locations is still tiny. Based on the background, the study analyzes the role of education level in early marriage among adolescents in Indonesia’s rural areas.

Methods

Data source and study design

This cross-sectional study used secondary data from the 2019 Government Performance and Accountability Survey (GPAS). The 2019 GPAS was conducted in July–September 2019 by a partnership between the National Population and Family Planning Board (BKKBN) and the Indonesian Statistics (BPS), with representative sample coverage at provincial and national levels. The target population for this study included homes, women aged 15 to 49 who could carry children, families, and unmarried adolescent girls aged 10 to 24. The study used a clustering method as an enumeration region. According to the survey, a cluster was defined as a group of census blocks (one or more) that were close to one another, situated along a single stretch, and contained about 200 homes. The sample design for this survey was stratified multistage random sampling.

Sample selection

The objective of the 2019 GPAS was to produce parameters at the provincial and federal levels. The 2019 GPAS sample was intended to include 67,725 households spread throughout 34 provinces, 514 districts/cities, 82,030 villages, and 1,935 clusters. In previous long-term national plans and GPAS surveys, respondents included households, families, and women aged 15 to 49 who could have children; in 2019, the youth module (unmarried) had a younger age range of 10 to 24.

This study used a staged sampling process. Using probability proportionate to size (PPS) sampling, a selection of villages was made in the first step based on the number of households in each village’s list (or in the sample frame of all villages). A random sample of villages from urban and rural areas was picked within a district or city. The second stage involved utilizing PPS sampling to select one cluster from each desired village depending on the number of households in the cluster. Based on the results of the household listings completed by enumerators in the targeted cluster using smartphone applications, 35 households were chosen in the third stage using systematic random selection.

The study’s sample was all females between 15 and 49 who lived in the 35 chosen households in each selected cluster. With a 99.7% response rate, this survey gathered 4,360 adolescent girls (aged 15 to 19) in Indonesia’s rural areas of 59,825 females (aged 15 to 49).

Outcome variables

Early marriage served as an outcome variable in the study. In this study, early marriage is defined as a union that takes place before the age of 19. This age restriction follows Law 16 of 2019, amending Law 1 of 1974 governing marriage [30]. Early marriage comprises two categories: No and Yes.

Exposure variable

We examined education level as an exposure variable. The study defined education level as the last diploma the respondent received. We analyzed three kinds of education: Elementary and under education, junior high school, senior high school, and college.

Control variables

The study included seven parameters as control variables, including the type of residence, age, work status, socioeconomic position, age of the first sexual intercourse, access to family planning (FP) material from print media, and access to FP from electronic media. There are two kinds of residence: urban and rural. Based on the most recent birthday, the study calculated age (in years).

The employment status comprises unemployed and employed. The 2019 GPAS used the wealth index formula to define the socioeconomic level. The wealth index formulated a weighted average of a household’s overall spending. The study produced the wealth index using primary data on household expenditures on health insurance, food, lodging, and other commodities. Furthermore, the survey split socioeconomic status into poor, middle, and rich.

Moreover, we divided the age of the first sexual intercourse into No sexual intercourse, 10–14 years, and 15–19 years. On the other hand, access to family planning (FP) information from print media refers to FP information obtained by accessing print media, various types such as flipcharts, posters, banners, newspapers, magazines, pamphlets/leaflets/brochures, and billboards. We split variables into No (never getting access) and Yes (getting access). Furthermore, access to FP electronic media refers to FP information obtained through various electronic media such as radio, television, and the Internet. We categorized it into No (never getting access) and Yes (getting access).

Data analysis

In the initial stage of their bivariate analysis, the authors performed the Chi-Square test. In the second stage of the investigation, the collinearity test was employed to ensure no significant relationship between the independent variables. We used a binary logistic regression test (enter procedure) in the last scene. Adjusted odds ratios (AOR) with 95% confidence intervals (CI) were presented in the study. The authors used the SPSS 26 version of the program to conduct the statistical analysis. In contrast, the study employed a scatter plot to compare education level (elementary school and college) and early marriage by province in Indonesia.

Results

The results show that the average early marriage among adolescent girls in Indonesia’s rural areas is 11.9%, and the mean age of first marriage in urban areas is 16.94. Moreover, Table 1 is a statistical description of early marriage. The result displays that the higher the education level, the lower the proportion of early marriage. Regarding age group, the results indicate that the older, the higher the early marriage proportion. Moreover, according to employment status, employed girls dominate in the early marriage category.

Table 1.

Descriptive statistics of early marriage among adolescent girls in Indonesia’s rural areas (n = 4,360)

Variables	p-value	Early Marriage
		No (n = 3,801)	Yes (n = 559)
Education Level	***<0.001
Elementary school and Under		54.8%	45.2%
Junior high school		76.8%	23.2%
Senior high school		95.3%	4.7%
College		99.4%	0.6%
Age	***<0.001
15		99.3%	0.7
16		94.8%	5.2
17		93.9%	6.1
18		82.1%	17.9
19		67.8%	32.2%
Employment status	**0.001
Unemployed		88.6%	11.4%
Employed		82.1%	17.9%
Socioeconomic status	***<0.001
Poor		82.2%	17.8%
Middle		90.6%	9.4%
Rich		95.5%	4.5%
Age at first sexual intercourse (years)	***<0.001
No intercourse		99.5%	0.5%
10–14		13.7%	86.3%
15–19		17.7%	82.3%
Access to FP information from print media	0.657
No		88.0%	12.0%
Yes		88.5%	11.5%
Access to FP information from electronic media	0.489
No		88.4%	11.6%
Yes		87.7%	12.3%

Note ***p-value < 0.001; **p-value < 0.01; *p-value < 0.05

Table 1 shows that, based on socioeconomic status, the wealthier, the lower the proportion of early marriage. Meanwhile, regarding age at the first sexual intercourse, 10–14 has the highest ratio of early marriage. On the other hand, the study indicates that they did not find access to FP information from print media, and access to FP information from electronic media was significantly related to early marriage.

Meanwhile, Fig. 1 displays a scatter plot of the relationship between early marriage and elementary school education levels by province among adolescent girls in Indonesia’s rural areas. Figure 1 indicates there is a positive relationship between the two. The higher the proportion of teenage girls with elementary school education in a province, the higher the proportion of early marriages. Furthermore, Fig. 2 shows a scatter plot of the relationship between early marriage and college education levels by province among adolescent girls in Indonesia’s rural areas. Figure 1 indicates there is a negative relationship between the two. The higher the proportion of teenage girls with a college education in a province, the lower the proportion of early marriages.

Fig. 1.

Scatter plot of the relationship between early marriage and elementary school education level by province among adolescent girls in Indonesia’s rural areas in 2021. Source: Visualisation by the author based on the 2019 GPAS

Fig. 2.

Scatter plot of the relationship between early marriage and college education level by province among adolescent girls in Indonesia’s rural areas in 2021. Source: Visualisation by the author based on the 2019 GPAS

The next was a collinearity test. The results show no evidence of a relationship between the independent variables. Additionally, each variable’s variance inflation factor (VIF) value is less than 10.00, and the tolerance value for each variable is more significant than 0.10. The investigation identified no evidence of multicollinearity in the regression model, indicating the test’s foundation for making decisions.

Table 2 displays the binary logistic regression results of early marriage among adolescent girls in Indonesia’s rural areas. The results show that the lower the education, the greater the likelihood of experiencing early marriage. Teenage girls in elementary school are 448.926 times more likely than college girls to experience early marriage (AOR 448.926; 95% CI 36.384-5539.050). Meanwhile, adolescent girls in junior high school are 146.171 times more likely to experience early marriage than college girls (AOR 146.171; 95% CI 13.140-1625.997). Moreover, adolescent girls in senior high school are 21.355 times more likely than those in college to perform an early marriage (AOR 21.355; 95% CI 2.017-226.067).

Table 2.

Results of binary logistic regression (n = 4,360)

Predictors	p-value	Early Marriage
		AOR	95% CI
			Lower Bound	Upper Bound
Education: Elementary school and Under	***<0.001	448.926	36.384	5539.050
Education: Junior high school	***<0.001	146.171	13.140	1625.997
Education: Senior high school	*0.011	21.355	2.017	226.067
Education: College (ref.)	-	-	-	-
Age: 15 (ref.)	-	-	-	-
Age: 16	***<0.001	15.205	3.803	60.786
Age: 17	***<0.001	35.349	8.470	147.522
Age: 18	***<0.001	114.899	28.056	470.552
Age: 19	***<0.001	331.653	79.369	1385.856
Employment: Unemployed	0.883	1.058	0.496	2.258
Employment: Employed (ref.)	-	-	-	-
Socioeconomic: Poor	0.551	1.423	0.446	4.536
Socioeconomic: Middle	0.650	1.308	0.410	4.175
Socioeconomic: Rich (ref.)	-	-	-	-
Age at first sexual intercourse: No intercourse (ref.)	-	-	-	-
Age at first sexual intercourse: 10–14	*<0.001	3735.393	825.774	16897.065
Age at first sexual intercourse: 15–19	*<0.001	914.580	452.181	1849.826

CI: confidence interval of 95%; ***p-value < 0.001; **p-value < 0.01; *p-value < 0.05

Regarding age group, Table 2 displays that all groups are likelier than the youngest (15 y.o.) to experience early marriage. Moreover, the results indicate that the older, the higher the probability of performing an earlier marriage. According to the age at the first sexual intercourse, all types are likelier than no intercourse to achieve an early marriage. Furthermore, the highest possibility of experiencing an early marriage is 10–14. On the other hand, the study did not find employment status and socioeconomic related to early marriage among adolescent girls in Indonesia.

Discussion

The finding indicates that girls with lower levels of education were more likely to experience early marriage. The condition is consistent with previous studies conducted in Indonesia, Bangladesh, Zambia, and Nepal, showing that girls’ education is the most decisive protective factor against early marriage, especially in developing countries [24, 31–33]. In these regions, girls with secondary and higher education are significantly less likely to have an early marriage than those with elementary education or no formal education [31]. In particular, a study conducted in Indonesia using a multivariate probit regression analysis with data from the 2012 Indonesian Demographic and Health Survey (IDHS) found that women with secondary education or higher were 9 to 33% points less likely to marry early [24].

The situation could be because education gives girls autonomy and the ability to control their decision-making in deciding the timing of marriage, as they have a better knowledge of the negative consequences of early marriage [32]. A study conducted in Mali also mentioned that girls with higher education generally better understand the recommended age for unions to avoid reproductive health-related problems [34]. Girls’ educational background is a significant factor in changing attitudes and behavior against societal norms, culture, and traditions, as well as misconceptions related to religious beliefs that engage them in early marriage [31, 34]. In addition, prolonged schooling usually increases girls’ desire to participate in occupational career development and find a suitable job rather than marrying early [35].

The findings of our study suggest that the government should provide better educational opportunities and participation for girls. Therefore, the government should consider prolonged schooling and completion of higher education for girls for at least secondary levels of attainment as a long-term strategy to prevent early marriage. Additionally, as a short-term strategy, the government can utilize civil society organizations involving religious leaders and affiliations to correct misconceptions related to early marriage, as early marriage in Indonesia is significantly influenced by culture and religion [24, 36]. Lastly, public campaigns should also continue to improve knowledge and attitudes on children’s rights and the harmful consequences of early marriage through mass media networks. In particular, this practice should also target other family members as parents significantly determine and place their children at a high risk of early marriage [24, 37].

All age groups were likelier than the youngest (15 years) to experience early marriage. This study’s findings on age groups and early marriage were in line with studies conducted in Zambia that have found a high prevalence of early marriages before the age of 18 [32]. However, the present study showed that children aged 16–19 years were more likely to have early marriage than those children aged 15 years. As the age of children increased, the probability of entering into early marriage also increased. In contrast, a study that analyzed socioeconomic factors related to early marriage in Bangladesh, Ghana, and Iraq using data from the Multiple Indicator Cluster Survey (MICS) found that early marriage decreases with age [26]. The disparity may be related to sociocultural differences between the countries studied.

According to the age at the first sexual intercourse, all types are likelier than no intercourse to achieve an early marriage. The situation implies that individuals who engage in premarital sexual activity are more prone to get into marriage at an earlier stage compared to those who abstain from such behavior. This study aligns with a previous study conducted in Bangladesh, which demonstrated that women who were subjected to child marriage exhibited significantly greater odds (AOR: 2.68; 95% CI: 2.20–3.26) of self-reporting engagement in premarital sexual activity compared to women who entered into marriage at more mature ages [38]. Another study has demonstrated that the sexual conduct of women has a significant impact on their marital behavior in various contexts. In the context of Congo, it was observed that women who initiated sexual activity before the age of 16 faced a significantly higher likelihood of entering into early marriages, approximately three and a half times more, compared to those who commenced sexual intercourse after reaching the age of 17 [39]. Early marriage primarily revolves around exerting control and authority, as established by the patriarchal structure and masculine culture, over a particularly vulnerable segment of society. The condition suggests a heightened focus on women’s premarital virginity and purity. These aspects are intricately linked to not just the family’s reputation but also the honor of the tribe [40, 41].

The study did not find employment status and socioeconomic related to early marriage among adolescent girls in Indonesia. Child marriage is a transgression against an individual’s fundamental right to exercise informed decision-making and choice, undermining their potential opportunities and prospects for the future. Previous studies have shown that employment is linked to early marriage, especially in women from wealthy families, so it can support the needs of women who get married early [42]. However, the results of this study contradict previous studies. It can be assumed that additional variables moderate the influence of individual traits on the choice to enter into early marriage, including women’s ability to secure employment and the responsibilities associated with domestic work. The topic of discussion pertains to the accessibility of maternity leave and childcare services [43]. The significant impact of women’s household duties on their time allocation is crucial concerning their capacity to engage in paid employment [43].

Assessing the socioeconomic consequences of early marriage on women’s later life has inherent challenges due to the many experiences women encounter after their premature marital union. Parents frequently viewed early marriage as a potential remedy for financial difficulties. Therefore, many parents advocate quick marriage arrangements even if the children are still not old enough to get married. Families below the poverty line often choose to arrange early marriages for their children to reduce dependence on parents [44]. Poverty constitutes a significant determinant contributing to the prevalence of early marriages within emerging nations such as Bangladesh. In regions characterized by severe poverty, it is common for a young female child to be perceived as a source of burden. In such instances, parents arrange early marriages for their daughters, as marrying off a daughter reduces the financial burden of providing for one less individual [45, 46].

Nevertheless, the findings of this study indicate a contrasting scenario wherein economic factors do not serve as a justification for the occurrence of early marriage. The presence of diverse tribal, cultural, and religious backgrounds in Indonesia significantly impacts The practice of early marriage is well entrenched within several groups due to sociocultural norms and religious convictions [47]. In many places in Indonesia, the patriarchal system continues to endure, encompassing factors contributing to sexual impropriety and the associated stigma, irrespective of consent. In Indonesia, the prevailing conservative culture and perspective regarding sex and sexuality contribute to the continuation of child marriage as a perceived viable strategy for poverty avoidance. It is evident that while traditional cultural norms are present across several cultures and religions in Indonesia, Islam is frequently seen and advocated as a faith that exhibits limited tolerance. The media continually promotes and highlights premarital sexual activity as the primary transgression in Islam that should be strictly avoided [48].

Strength and limitation

One of the study’s advantages was using big data as analytical material to show generalized work up to the national level. However, because the author used a quantitative research design, we cannot examine the factors contributing to early marriage phenomena identified in earlier studies, such as family, child, and religious values [49–52].

Conclusions

Based on the results, the study concluded that education level had a role in early marriage in Indonesia’s rural areas. The higher the education, the lower the probability of experiencing early marriage. This study implies that the government must target girls from an early age by giving greater attention from the beginning of education.

Given that human behavior constantly evolves, it is essential to consider and incorporate societal perceptions and cultural values ​​to make the paper more impactful. For this reason, we recommend further studies with a qualitative approach to explore the role of education in early marriage and the perception and cultural factors related to this situation.

Author contributions

MF and ADL developed the proposal, analyzed the data, and interpreted the results. IMS and RDW were significant contributors in conducting the study, interpreting the data, and writing the manuscript. YA and RM contributed substantially to the research, analyzing the data and writing the manuscript. All authors read and approved the final manuscript.

Funding

Not applicable.

Data availability

The National Population and Family Planning Board has data that support this study’s conclusions. These data, which the researcher used under license for the current research, are not publicly available due to restrictions on their release. The National Population and Family Planning Board has permitted the authors to make the data available via the website https://e-ppid.bkkbn.go.id/page/prosedur-permohonan-informasi (phone + 62 21 8098018; fax. +62 21 8008554) and upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the national ethics committee from the Family Planning and Reproductive Health (BKKBN) Committee of Research Ethics (No. 454/LB.02 /H4/2019). The survey was carried out following the regulations of the BKKBN’s Head. Written informed consent was used, and respondents or their legal guardian(s) provided written informed permission to account for the voluntary and confidential components of the data-gathering approach.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.