The respondents in this study were 180 adolescents consisting of 60 high school students, 60 vocational school students, and 60 madrasah aliyah students, with a ratio of male to female respondents of 50:50. The average age of respondents was 16.23 years with an age range of 13-18 years. The average age of respondents' fathers (43.33 years) was higher than the average age of mothers (42.97 years). The length of education of father and mother is 10.82 years and 10.35 years. The average family income is IDR. 4,836,388, which means that the average respondent's family is above the poverty line (Tangerang Regency's poverty line is IDR 545,065), with the lowest income being IDR. 200,000 and the highest income IDR 60,000,000 show in the Table 1.

The results of the descriptive test showed an average religiosity index of 86.97 (categorized as high) with an average per dimension, namely for the Islamic belief dimension of 93.75 (categorized as high) and the dimension of obligatory practices in Islam of 82.04 (categorized as high) (see Table 2). The average index is obtained from the total index of all respondents divided by the number of respondents.

Based on the research results on the Table 2, the average index of the dimensions of Islamic belief and the dimensions of obligatory practices in Islam for female adolescents is higher than for male adolescents. In the dimension of Islamic belief, almost all male and female adolescents are categorized as high. Meanwhile, in the dimension of obligatory practices in Islam, which covers religious practices, only 60% of male adolescents and 70% of female adolescents are categorized as high. This indicates that not all adolescents know Islamic beliefs and practice them in their daily lives.

The distribution of the average items that shows on the Figure 2 the highest average is in items 17, 19 and 20. Meanwhile, the lowest average is in items 4 and 8. These results indicate that adolescents close intimate parts (item 17), stay away from adultery (item 19) and consume halal food (item 20). However, adolescents still don't understand well that the Koran is the solution to life's problems (item 4) and paying zakat is an obligation (item 8).

The average sexuality education index carried out by fathers was 56.29 (categorized as low). Meanwhile, sexuality education carried out by mothers was 64.65 (categorized as moderate). The results of the analysis show that the average reproductive health education index carried out by fathers and mothers is 59.40 (categorized as low) and 71.50 (categorized as moderate), respectively. The average reproductive health education index carried out by fathers and mothers is 53.98 (categorized as low) and 57.91 (categorized as low), respectively (see Table 2). These results indicate that mothers are more active in educating their adolescents regarding sexuality than fathers. Parents often experience challenges with their adolescents related to their role as sexuality educators (Byers et al., 2013) (Chibuye & Kumatongo, 2022). Parents as educators experience challenges in conveying sexuality education and cultural influences related to sexual knowledge (for example, biological differences between men and women, sexual behavior, and the consequences of sexual behavior), the age at which to start sexuality education, and providing sources of information about sexuality education to their adolescent (Fevriasanty et al., 2021).

The specificity of the topic of sexuality, which is often considered taboo, makes parents' attitude in providing sex education to adolescents non-committal and carried out in a non-systemic manner. According to (Mahajan & Sharma, 2005), parents avoid mention of sex in their daily correlations with their children. Another reason is that parents themselves lack scientific knowledge about this matter (Zaleski & Schiaffino, 2000). Three out of four parents also lack understanding of sexual behavior and consequences, the biological differences between men and women and more than half (52.5%) of parents also do not consult with their partners regarding sexuality education issues for their adolescent (Lavanya et al., 2011).

The distribution of the average index on the Figure 3 shows that the average index of sexuality education carried out by mothers, both for male and female adolescents, in the dimensions of reproductive health education and risk education is higher than the sexuality education carried out by fathers. These results indicate that mothers discuss and educate their adolescents more often than fathers.

These findings are in line with (Raffaelli & Green, 2003), which found that mothers were more likely to communicate with their children about sexual matters than fathers and that daughters reported higher levels of communication than sons. The average index of father's and mother's sexuality education for female adolescents, in each dimension, shows a higher figure than for male adolescents, except for the dimension of reproductive health carried out by fathers. This indicates that parents are more attentive and tend to think that female adolescents are more at risk than male adolescents, so sexuality education is needed explain in Figure.

The distribution of the average items shows that the highest average education received by the father is in items 1, 2, and 4 explain in Figure 4. Meanwhile, the lowest average is in items 10 and 17. These results indicate that the adolescent has received a good education from his father regarding the concept of mahram and non-mahram (item 1), the concept of hadats (item 2), and the boundaries of women's and men's private parts (item 4). The topics with a high average seem closely related to religion, but other issues related to health still need to be maximally addressed. Adolescents also do not receive sufficient education from fathers regarding the process of pregnancy (item 10) and the topic of abortion (item 17). This could be because the topic of pregnancy is better discussed with the mother.

Almost the same as the education carried out by the father, the distribution of the average items shows that the highest average education received by the mother is in items 2, 4, and 8. Meanwhile, the lowest average is in items 10, 11 and 17 (see Figure 3). These results are informative that adolescents have received a good education from their mothers regarding the concept of hadats (item 2), the boundaries of women's and men's private parts (item 4), and parts that members of the opposite sex should not touch (item 8). Adolescents also do not receive sufficient education from mothers regarding the process of pregnancy (item 10), the prohibition and sin of masturbation (item 11), and the topic of abortion (item 17). The topic of pregnancy is close to mothers' experiences, but it is not taught well enough to adolescents.

The average referencereference of the social media index is 74.33 (categorized as moderate). The average index of positive references on social media in was obtained at 68.67 (categorized as moderate). Meanwhile, negative references on social media average index is 79.04 (categorized as moderate) (see Table 2).

The references from social media are also a source of information that adolescents widely use to get information about sexuality. Today's phenomenon illustrates that adolescents' self-control over content or information about sexuality on social media needs more attention. Data from the (Child Protection, 2016) revealed that of 4,500 adolescents in 12 cities in Indonesia, 97 percent had seen pornography, including students. Of the 2,818 students, 60 percent had seen this indecent show. Apart from poor self-control and filtering power among adolescents, external factors such as poor quality sexual content and being allowed to do so by the government also play a role. Kominfo, on its website, says that of the 28-30 million pornographic websites, blocking has only reached 700 thousand sites.

The research by (Pratiwi et al., 2023) found that 42% of educators agreed that children were exposed to information about sexuality from sources that were not credible. The results of this study show that the average index of control over negative references on social media for adolescent girls is higher (81.63) than for male adolescents (76.44). This indicates that female adolescents control the social media content that contains pornography, sexual harassment, content regarding excessive dating activities, adultery, and sexual deviations better than male adolescents.

The distribution of item averages shows that the highest average is for items 14, 16, and 20. Meanwhile, the lowest averages are for items 1, 5, and 17 explain in Figure 1. These results indicate that adolescents have a good sense because they reject harassment on social media (item 14); adolescents can also control themselves so they are not influenced by lousy content on social media (items 16 and 20). Unfortunately, adolescents also do not get enough education from social media regarding information about the anatomy of the female and male body (item 1) and the characteristics of pregnancy (item 5). Social media provides information that adolescents are not ready to accept, for example, information about tools to prevent pregnancy (item 17).

The descriptive test results show an average peer attachment index of 71.86 (categorized as moderate). The average index of trust dimension obtained was 78.06 (categorized as moderate). The average closeness index was found to be 71.91 (categorized as moderate). The average involvement index is 69.00 (categorized as moderate), and the average guidance index was found to be 70.51 (categorized as moderate) (see Table 2).

This research showed that the average index for all dimensions of male adolescents on the peer attachment variable was higher than that for female adolescents except for the guidance variable. Female adolescents (70.93) had a higher average guidance index than male adolescents (70.09) (categorized as moderate) These results show on Figure 5 that female adolescents are more easily influenced to follow the behaviour of their peers than male adolescents.

The result on the Figure 5 shows that male adolescents have a higher average score than female adolescents in all dimensions except the guidance dimension. (Sijtsma et al., 2023) found that male adolescents show higher levels of initial trusting behavior and more muscular adaptation of trusting behavior during untrustworthy interactions than female adolescents. In the guidance dimension, female adolescents have a higher average score than male adolescents. These results indicate that female adolescents are more likely to follow the behavior of their peers’ behavior than male adolescents.

The score distribution for item averages shows that the highest average is for items 6, 13, and 14 (see Figure 6). Meanwhile, the lowest averages are for items 5, 7, 10 and 11. These results indicate that adolescents invite each other to valuable activities (item 6) and follow each other's peers away from pornographic content (item 13) and promiscuity (item 14). However, on the Figure 6 shows that adolescents do not yet discuss unwanted pregnancies (item 5), reproductive health (item 7), and puberty (item 11). Adolescents also have not yet advised their friends who are dating excessively (item 10).

The average sexuality index was found to be 87.89 (categorized as high). Meanwhile, the average index for gender identity is 92.92 (categorized as high). This means that both male and female adolescents believe and affirm their bodies and gender very well.

The average index for gender expression is 83.33 (categorized as high), this means that almost all adolescents believe and affirm their appearance and behavior according to their gender and believe that dominant masculine or feminine traits are in line with their gender very well. However, 6.7% of male and 14.4% of female adolescents are categorized as low. This percentage is contributed by one of the statements with the lowest average (item 3) explain in Figure 6, which indicates that there are still adolescents who think that boys do not have to have dominant masculine characteristics and girls do not have dominant feminine characteristics.

The average index for sexual orientation is 74.07 (categorized as moderate). The result found that 2.78% of adolescents answered "very inappropriate" and "not appropriate" to the statement "I am attracted to the opposite sex." As many as 12.2% of adolescents answered "very inappropriate" and "not appropriate" to the statement "liking the same sex is a wrong tendency." As many as 16.11% of adolescents also answered "very unsuitable" and "not suitable" to the statement that "same-sex lovers can be cured or returned to their natural state." This statement also has the lowest average (item number 6 and 7) explain in Figure 6. Nevertheless, almost half of male (60%) and female (54.4%) adolescents are categorized as high in this dimension. These results mean that male adolescents who understand that liking the same sex is a wrong tendency, same-sex attraction, can be healed back to their natural nature—well. The average indices of the two are 77.22 for males and 70.9 for females.

In the sexual ideology dimension, the number of female adolescents (95.6%) categorized as high was greater than that of male adolescents (84.4%). This means that female adolescents understand very well the prohibition of sex outside marriage, sexual harassment, and the issue of abortion more than male adolescents. The average indices of the two are 90.37 and 95.96, respectively. Nevertheless, as many as 3.89% of adolescents answered "very inappropriate" and "not appropriate" to the statement "unmarried adolescents are not allowed to have sexual relations even though both parties mutually agree." Data from the 2017 Indonesian Demographic and Health Survey reported that 8% of men and 2% of women had sexual relations. Among women and men who had premarital sexual relations, 59% of women and 74% of men reported starting sexual relations for the first time at the age of 15-19 years. The highest percentage occurred at 17 years, as much as 19%. Among adolescents who have sexual relations, it is reported that 12% of women experience unwanted pregnancies.

The results show that the dimensions of Islamic belief are positively related to adolescent sexuality (r= 0.128) (See Table 3). This means that the higher an adolescent's belief in Islam, the higher their understanding of good and correct sexuality. The results of this correlation test are in line with previous studies; for example, research by (Zaleski & Schiaffino, 2000) found that religiosity is a strong predictor of sexuality in the family environment. (Lefkowitz et al., 2004) examined the correlation between religiosity and sexuality in Eastern United States students. Lefkowitz found that respondents whose religion was more prominent in their daily lives had more conservative attitudes and behavior regarding sex. (Odimegwu, 2005) also found that there was a strong relationship between religiosity and adolescent sexual attitudes and behaviour. In the research conducted, a religious commitment was more important than a religious affiliation in influencing adolescents' sexual attitudes and behaviour.

The variables of sexuality education carried out by fathers and mothers, references to social media and peer attachment, and their respective dimensions also have a significant positive correlation with adolescent sexuality. This means that the higher the sex education carried out by fathers and mothers, both reproductive health education and education about risks, the higher adolescents' understanding of good and correct sexuality. The more adolescents get positive references on social media and the higher adolescents control negative references on social media, the higher they understand good and correct sexuality. Likewise, the higher the attachment of peers, both in the dimensions of trust, closeness, involvement, and guidance, the higher the adolescent's understanding of good and correct sexuality.

The validity and reliability analysis results showed that the AVE value was ≥ 0.50, the composite reliability value was ≥ 0.70, and the SRMR value was ≥ 0.10. These results mean that the measurement model meets the good fit criteria and has good validity and reliability. The research model hypothesis test uses a significance of 0.05 (95% confidence level), indicating that religiosity has a significant positive effect on references of social media, sexuality education variable carried out by fathers and mothers has a significant positive effect on peer attachment, references of social media and peer attachment have a significant positive effect on adolescent sexuality. This means that the higher the adolescent’s religiosity, the higher positive references on social media they get, and the higher the control of negative references on social media they have. The more sex education carried out by fathers and mothers, the more the attachment to peers. The higher peer attachment and references from social media, the more directly the adolescents' understanding of good and correct sexuality will increase.

The results of this study found that, through references from social media, religiosity indirectly effect adolescent sexuality. This shows that the better the understanding and practice of Islam by adolescents, which is driven by adolescents' self-control regarding good social media references, the better the adolescents' understanding of good and correct sexuality will be. The results of this study also are in line with (Sekar Arumi & Faradilla, 2023) research that contributes to premarital sexual research by explaining that religiosity does not reduce premarital sexual behavior if the person is permissive or has a supportive attitude towards it. In this study, the attitude can be influenced by references of social media.

Sexuality education carried out by fathers and mothers significantly positively affects peer attachment. These results show that the better the sex education carried out by fathers and mothers, the better the peer attachment between adolescents will be in relation to a good and correct understanding of sexuality. (Lansford, 2021-06-15) discusses parental education's influence on peer attachment, who says that parents can influence adolescents' peer relationships indirectly through parent-adolescent relationships and directly through mediation, education, and providing advice to adolescents regarding their same age relationships. Adolescents learn a lot from their parents about many things, for example, regulating emotions, solving problems and conflicts, or sharing the values they get from their parents with their peers. Included in the context of this research is information about good and correct sexuality to their peers. (Miller et al., 1998), which found that, for both male and female adolescents, the quality of parent-adolescent communication regarding sexuality had no direct effect on adolescent sexual behavior. However, there was a significant indirect effect through the sexual value that adolescents believe.

Peer attachment is also a mediator of sex education carried out by fathers and mothers so both have a significant indirect effect on adolescent sexuality. This means that better sexuality education carried out by fathers and mothers will have a significant influence on adolescents' understanding of good and correct sexuality if it is encouraged by the better peer attachment that occurs among adolescents.

The results of the influence test show that the references from social media have a significant positive effect on adolescent sexuality. This means that the better control adolescents have in using social media, the better their sexuality will be. This research is in line with research by (Shashi Kumar et al., 2013) on 586 Indian male and female adolescents in grades 9 to 12, which found a strong positive correlation between sexual interactions and unsupervised internet use among male adolescents. Research by (Ouytsel et al., 2014) also found that adolescents' sexting habits were closely correlated with pornography consumption, age, gender, school path, and internet use. (Laporte et al., 2020) also said that various types of sexual content on social media influence a person's sexual behavior in different ways.

It was also found that peer attachment had a significant positive effect on adolescent sexuality. This means that the better the quality of peer attachment, the better the adolescent's sexuality will be. The more adolescents open up with their peers about complaints, share stories about people they like, closely discuss the dangers of adultery and unwanted pregnancies, get involved in inviting positive activities and helping peers understand reproductive health, and limiting correlations with the opposite sex, and adolescents follow the positive behavior of their peers regarding limits on correlations with the opposite sex, staying away from pornographic content, this will make adolescents have better sexuality. This research is in line with (Peci, 2017), who found that adolescents are more likely to have sex if their friends and peers are older, use alcohol or drugs, or engage in other harmful behaviors. Likewise, they were more likely to have sex if they believed their friends had more positive attitudes toward childbearing, holding permissive values about sex. (Bingenheimer et al., 2015) also found that affiliation with antisocial peers and perceiving peer norms as supportive of sex directly increased the likelihood that adolescents engaged in sexual activity for the first time.

Based on the results of the analysis of the influence model of religiosity, parent-adolescent sexuality education, references from social media, and peer attachment on adolescent sexuality, an R-Square value of 0.457 was obtained explain in Figure 7. It can be interpreted that the independent latent variable in this study influences adolescent sexuality by 45.7%. was obtained in Figure 7. It can be interpreted that the independent latent variable in this study influences adolescent sexuality by 45.7%. Meanwhile, the remaining 54.3% was influenced by other variables outside this research.