This article reports findings from a study of school- and college-going adolescent girls and boys in Patna that explored their awareness of sexual matters and safe sex, and their experience of sexual relations. The article correlates this awareness and experience at the individual, family and peer levels. Maternal education and parental supervision of adolescent activities played an important role in promoting awareness among girls, suggesting that parenting issues are critical. Further, programmes that apprise parents of the role they play are needed.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly48Exploring Safe Sex Awareness and Sexual Experiences of Adolescents in PatnaMedha Shekhar, Saswata Ghosh, Pradeep PandaThis article reports findings from a study of school- and college-going adolescent girls and boys in Patna that explored their awareness of sexual matters and safe sex, and their experience of sexual relations. The article correlates this awareness and experience at the individual, family and peer levels. Maternal education and parental supervision of adolescent activities played an important role in promoting awareness among girls, suggesting that parenting issues are critical. Further, programmes that apprise parents of the role they play are needed. There is an increasing recognition in India of the need to address the sexual and reproductive health of adolescents. A small but growing body of literature has begun to exam-ine the reproductive and sexual health situation of this group. Evidence generated thus far from small studies conducted in different parts of the country suggests that the youth remain poorly informed about sexual and reproductive matters, and that significant minorities of youth have engaged in premarital sex, largely unprotected [see for example, Jejeebhoy and Sebastian 2004]. While these studies have provided considerable insight into young people’s lives in certain geographic settings of the country, other regions, however, have been somewhat neglected. The state of Bihar is one such setting in which insight into the sexual and reproductive behaviour of unmarried youth remains sparse. While it is believed that strict social norms that forbid opposite-sex social mixing among young people and close supervision of children tend to prevent pre-marital sexual relations, there is little evidence that supports or reverses such assumptions. In a setting in which about one quarter of the population is adolescent (10-19 years), in which proportion of adolosecent pregnancy is high and in which human immunodeficiency virus (HIV) and sexually transmitted infections (STIs) are increasingly observed, lack of evidence has clearly impeded the development and imple-mentation of programmes that inform adolescents about sexual matters in general and safe sex in particular.The objective of this paper is to fill this gap in evidence. Speci-fically, the paper reports findings from a study of school- and college-going adolescent girls and boys in the city of Patna that explored their awareness of sexual matters and safe sex, experi-ence of sexual relations; and correlates of this awareness and experienceat the individual, family and peer level. BackgroundAvailable studies conducted in India are generally small, using variable methodologies and focusing on young people of differ-ing ages [Abraham and Kumar 1999; Alexander et al 2006; Awasthi, Nichter and Pandey 2000; Jejeebhoy 2000; Mehra, Savithri and Coutinho 2002; Pachauri and Santhya 2003; Pandey et al 2006; Patel and Andrew 2001]. Findings cannot therefore be generalised. Even so, there is a remarkable consistency in findings: they suggest that between 15 per cent and 30 per cent of young men and fewer than 10 per cent of young women report the experience of penetrative sex, and that substantially more report a romantic relationship or other forms of physical intimacy, Medha Shekhar(shekharmedha@yahoo.co.in) is managing trustee, PURWA, Patna; Saswata Ghosh (ghosh.saswata@gmail.com)is lecturer, Institute of Development Studies, Kolkata and Pradeep Panda (pradeep4164@yahoo.com) is director, rural research, Institute of Rural Research and Development, Gurgaon.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200749including handholding, kissing or touching private parts [Alexander et al 2006; Jejeebhoy and Sebastian 2004]. Most of these studies have been descriptive and have rarely focused on the correlates of sexual behaviour. There are, however, exceptions. A study of sexual experience among 966 low-income college-going students in Mumbai [Abraham and Kumar 1999] noted that the sexually experienced were significantly more likely than others to differ on individual characteristics: for example, they were more likely to have access to resources, hold positive attitudes towards premarital sex, report more social interaction with peers and be exposed to pornographic materials. Factors such as parental restriction, family religiosity and family income played no role in influencing whether or not these students had experienced sex. In another study among adolescents in a slum setting in Pune city and the surrounding rural areas, level of self-confidence and peer interaction were significantly as-sociated with reports of the experience of romantic and sexual relations; while parental restriction was not associated with romantic or sexual relations, the extent to which youth reported close interaction with parents was inversely associatedwith these experiences [Alexander et al 2006]. Finally, a study of adolescents in a Delhi slum setting, using en-tirely qualitative methodologies, reports, similarly, that parental restrictiveness does not deter young people from forming roman-tic partnerships or engaging in sexual relations [Mehra, Savithri and Coutinho 2002].This study builds on Abraham and Kumar’s (1999) study of low-income college students in Mumbai. The study instrument and design have borrowed extensively from the Mumbai study, but the focus is on a younger group of adolescents attending school/college.Patna, the site of this study, is different from the settings in which the studies described above were conducted in many res-pects. For one, the state of Bihar in which Patna, the capital city, is located, is relatively poorly developed with below all-India average scores on socio-economic indicators. For example, Bihar has the lowest literacy rate in the country: in 2001, less than half the population aged seven and above were literate (60 per cent of males and 34 per cent of females) against the all-India figure of 75.9 and 54.3 for males and females respectively [Registrar General, India(RGI) 2001a]. School enrolment also lags behind national figures: in 1999-2000, the percentage of adolescents (10-19 years) attending edu-cational institutions was 48 per cent in Bihar, compared to 60 per cent in India as a whole.1 Within Bihar, wide gender disparities in school enrolment are also apparent: in 1999-2000, 56 and 38 per cent of males and females were enrolled in school, compared to 66 and 53 per cent, respectively in India as a whole. School enrol-ment rates are much higher in urban than rural Bihar: 72 and 61 per cent of males and females aged 10-19 years are in school in urban areas, compared to 54 and 35 per cent, respectively, in rural areas. The reproductive health situation is also poorer in Bihar than in India on average. For example, in 2005-06, the total fertility rate in Bihar was considerably higher than all-India figure (4.00 vs 2.68) [International Institute for Population Sciences 2007a; 2007b]. Similarly, during the same period, sub-stantially smaller percentages of women received three antenatal care visits for their last birth (17 per cent vs 51 per cent), reported institutional delivery (22 per cent vs 41 per cent), or fully immu-nised their children (33 per cent vs 44 per cent). The public health infrastructure is grossly inadequate: Only 23 out of 38 districts in the state have a district hospital [Government of Bihar 2006]. In-depth awareness of HIV-related issues among adolescents (15-19 years) in Bihar is poorer than in India more generally: for example, in Bihar 37.5 per cent of rural males and 15.9 per cent of rural females were aware of important methods of HIV prevention (i e, consistent condom use and sex with a faithful uninfected partner) compared to 54.1 per cent and 44.5 per cent of adolescents nationally. Similarly, adolescents in Bihar are much less likely to be correctly aware of some common misconceptions regarding transmission of HIV/acquired immunodeficiency syndrome (AIDS), i e, that mosquito bites and sharing a meal with an infected person do not transmit HIV, and that a healthy looking person can transmit HIV: 8.3 per cent and 7.9 per cent of rural adolescent men and women respectively in Bihar compared to 21.8 per cent and 20.2 per cent nationally [NACO andUNICEF 2002].Among the 38 districts of Bihar, Patna has the largest popula-tion (4.72 million) [RGI 2001b], the largest urban centre (42 per cent urban) and the highest literacy rate (63 per cent overall; 73 per cent for males and 51 per cent for females) (calculated from data inRGI 2001b). As the state capital, Patna is the leading ad-ministrative and educational centre. Adolescents in Patna are as-sumed to have greater access to education, employment opportu-nities, modern lifestyles and new ideas as compared to other districts of the state.Study Design and MethodologyThis study was conducted in Patna city in 22 schools and colleges, government and private, both single-sex and coeducational. We must acknowledge that the selection of educational institutions was somewhat opportunistic. Researchers approached 41 schools and colleges; many institutions (N=15) refused permission to undertake the study or dropped out after giving permission (N=4) and hence the study has been conducted in self-selected schools and colleges that granted researchers access to students.The study comprised two phases. In an initial qualitative phase, eight focus group discussions were held with male and female students, and eight key informant interviews were held with selected students, parents and teachers. The objective of this phase was to inform the design, content and wording of the survey instrument, and specifically to better understand the context in which young students grow up and the terminologies used to describe sexual matters. The second phase comprised a survey conducted in 2005 among students in Classes IX (aged around 14 years) and XI (aged around 16 years). Questionnaires were self-administered.While the selection of educational institutions was, by necessity, opportunistic, the sample of students who responded to the survey was largely representative of students drawn from the two classes within these schools and colleges. The survey was conducted in one randomly selected section of Class IX and Class XI from each
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly50selected institution; all students who were present in the class at the time of the survey and who consented to respond to the sur-vey were eligible for inclusion. A total of 32 sessions were held in the 22 schools and colleges selected for study.Consent procedures were followed at two levels. All students in the selected classes were provided a parental consent form and asked to obtain permission from a parent to participate in the study; only those whose parents con-sented were considered for the study. At the time of administering the ques-tionnaires, consent was sought from students; only consenting students were included in the study. Refusals were few: only eight parents refused to grant consent; not a single student who was in class on that day opted not to participate.The survey was conducted in the presence of a five-member survey team (two male and two female in-vestigators, and the principal investi-gator) who explained the objective of the survey to the class, obtained individual consent from all students and ensured that students had privacy while filling out the question-naire and fully understood the con-tents. Respondents were instructed not to identify themselves in the questionnaire and were assured that only the research team would have access to the data. Teach-ers and school or college authorities were not involved in admin-istering the questionnaire.Survey InstrumentThe survey instrument, largely adapting that used by Abraham and colleagues, was in Hindi and pre-tested. It comprised a total of 100 questions. Following the completion of the survey in each class, the research team invited students who had questions on issues addressed in the questionnaire to approach the research team; many students did take advantage of this offer or subsequently contacted the research team for further information. In case the survey was to be conducted in Classes IX and XI of the same school/college, the session was conducted simultaneously in both classes so as to avoid sharing of infor-mation. The total sample size planned for the survey was 1,000;in total, 1,048 students (461 boys and 587 girls) completed the interview. Some 10 boys and 13 girls were married, and their responses have been excluded from the analysis. In total then, our analysis is restricted to 1,025 unmarried students. Of these, 44 per cent are male (451) and 56 per cent female (574). More-over, 53 per cent were from Class IX and 47 per cent from Class XI. Findings are presented for 255 and 196 boys and 293 and 281 girls in Classes IX and XI, respectively. In order to test the signi-ficance of gender differences, we present z-tests and t-tests com-paring reports of girls and boys in Classes IX and XI, as well as all girls and boys.Our analysis examines awareness of sexual and reproduc-tive health matters and experiences of physical contact (hugging, kissing, touching private parts) and sexual relations with an opposite-sex partner. Awareness of sexual and reproductive health matters is meas-ured by a total of eight questions, five covering aspects of men-struation, pregnancy and contraception, and three covering HIV transmission and prevention. Res-ponses were set to equal 1 if correct and 0 if not. A summary index of awareness was then created that sums responses on all of these eight indicators, and ranges from 0 to 8. Physical contact and sexual experi-ence were measured by asking about the following experiences with an opposite-sex partner: (a) hugging, (b) kissing, (c) touching private parts; and (d) sexual intercourse. Those who experienced sexual inter-course were also asked about condom use. Four indicators were defined. The first is a summary measure ref-lecting any physical or sexual experi-ence, that is the respondent was as-signed a value of 1 if s/he reported the experience of hugging, kissing, touch or sexual intercourse and 0 if not. The second indicator omits hugging and measures the expe-rience of kissing, touching or sex, while the third specifically measures the experience of sexual intercourse with an opposite-sex partner among males as few girls reported the experience of sexual intercourse. The fourth indicator measures unprotected sex (i e, non-use or sporadic use of condoms) among males reporting sexual experience.In line with the literature, our analytic framework posits that awareness and sexual experience may be associated with a number of factors at the individual, family and household, and peer levels. A description of the variables used in this analysis is provided below.Individual level variables that we considered likely to influence awareness and sexual experience included age, class attended at the time of interview (Class IX or XI), type of school or college attended at the time of interview (government/private), wage earning status (yes/no) and exposure to pornographic films, ei-ther onDVD or at theatres or on REN TV (a pornographic televi-sion channel from Russia with reach in Patna). It also included attending a family life education or sex education programme in school or college.Family and household level variables include parents’ educa-tional attainment levels, parents’ occupation, residential status (co-residing with at least one parent or not) and perceived strict-ness of parental supervision: respondents were asked the extent to which, compared to their same-sex peers, they perceived that their parents supervised them very strictly, somewhat strictly or not at all strictly.Table 1: Percentage Distribution of Adolescent Students by Sex, Education Level and Individual Level CharacteristicsCharacteristics BoysGirls PresentEducation Total PresentEducation Total Level Level Class IX Class XI Class IX Class XIN 225 196 451 293 281 574Age 13-14 58.8 -33.3 57.0 -29.1 15-17 37.6 69.9 51.7 41.6 84.0 62.4 18-20 3.5 30.1 15.1 1.4 16.0 8.5Mean age 14.5 16.7 15.5 14.3 16.7 15.5Type of school/college Government 53.7 26.5 41.9 58.4 29.5 44.3 Private 46.3 73.5 58.1 41.6 70.5 55.7Wagework Working 6.7 11.7 8.9 3.4 3.9 3.7Religion Hindu 92.5 89.8 91.4 95.9 83.6 89.9 Muslim 6.7 9.2 7.8 3.1 13.5 8.2 Other 0.8 1.0 0.9 1.0 2.8 1.9Ever attended family life or sex education in school/college Yes 20.0 13.3 17.1 7.5 15.0 11.2Exposure to blue films Yes 14.5 21.4 17.6 6.8 2.8 4.9* Percentages do not add up to 100 because of missing responses.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200751Peer level factors are measured by two indicators reflecting theextent of social interaction with peers. Respondents were asked if they spent time with their friends in a number of social activities (going to the cinema, restaurants, sports clubs; going out on picnics) and the frequency of this interaction. Responses were coded to equal 0 if the respondent never engaged in the ac-tivity, 1 if s/he sometimes participated in the activity, and 2 if frequently. The social interaction index then ranges from 0 to 8. Respondents were also asked whether they socialised with their friends in more “adult” settings like bars, late-night parties or dis-cotheques; a dichotomous indicator was created that was set to equal 1 if the respondent reported that s/he attended any of these events sometimes or often, and 0 if not.Background FactorsIn this section, we explore the individual, family and household, and peer level factors that are assumed to influence sexual aware-ness and experiences of young people. Table 1 (p 50) presents individual level factors. On an average, respondents were aged 15.5 years and over 90 per cent were Hindu. While about 60 per cent of those in Class IX came from government institutions, this percentage was reversed and almost three quarters of respond-ents from Class XI were drawn from private institutions. A small proportion (fewer than 10 per cent) combined schooling with economic activity. A few were exposed to family life or sex education in school or college: 17 per cent of boys and 11 percent of girls. Finally, significant minorities – 18 per cent of boys and 5 per cent of girls – reported that they had seen blue films on DVD, in theatres or on REN TV.Family and household level factors are described in Table 2. Gender and to a lesser extent age differences are stark. It is clear that compared to boys, girls are drawn from households in which parents are well educated and fathers are largely in managerial, professional or business occupations. This selectivity reflects the fact that in a setting in which large numbers of girls are with-drawn from school at puberty, girls who remain in school by Class IX come from families that are better educated and better-off economically, who may place a greater value on the education of their daughters than does the population more generally. Also the finding that parents of those in Class XI tend to be far better educated than the parents of those in Class IX can be explained by the fact that many youth discontinue schooling after Class X (school completion), and it is likely that better educated and better-off parents are more likely than others to educate their children beyond this level. Co-residence with one or both parents also reflects gender and age differences: Just over one-fifth of boys and about one-tenth of girls live separately from a parent; this proportion is considerably higher among Class XI students than Class IX, presumably reflecting the facts that parents are less likely to allow girls to move away than boys, and that many youth from outside Patna who wish to pursue a higher education may not have the opportunity and thus require to relocate to Patna and reside in hostels or with relatives while studying in Class XI.Table 2 also presents adolescent students’ perceptions of the extent of supervision by parents. Some one third of boys and one-quarter of girls reported that they are very strictly super-vised by their parents; about half and three-fifths, respectively, reported they are somewhat strictly supervised and about 17 per cent of both boys and girls reported that they are not strictly supervised. It is indeed surprising that a larger proportion of boys than girls perceivethemselves to be strictly supervised. We sug-gest that in a setting in which restrictive norms for girls but not boys are widespread, “strict supervision” may be differently interpreted or perceived by boys and girls.Students’ interaction with peers is described in Table 3 (p 52). As mentioned earlier, we have clubbed interaction indicators into those that are more “typical” of adolescents (picnics, films, clubs, eating places) and those that are more adult and may provide opportunities for greater physical contact (for example, bars, discotheques and late night parties). It is clear that interaction is limited. Indeed, many students – about one-third of all boys and two-fifths of all girls – have never interacted with their peers in even such relatively common settings as picnics and films. And frequent interaction is relatively uncommon. Gender disparities are evident: boys are far more likely than girls to interact with their peers in three of the four locations (picnics, films and sports clubs). Scores on the social interaction index suggest that while girls in Class IX are only slightly less likely than boys to interact with their peers (mean scores of 2.2 and 2.5 respectively), those in Class XI are considerably less likely to do so (mean scores of 2.1 and 3.3 respectively).We also probed the extent to which adolescents visited more adult sites with their peers, for example, bars, discotheques and late night parties: Findings suggest that some 29 per cent of boys and 15 per cent of girls have attended such events sometimes or frequently. Table 4 (p 52) presents students’ awareness of sexual and re-productive health matters as described earlier. Findings suggest Table 2: Distribution of Students by Sex, Education and Family and Household CharacteristicsCharacteristics BoysGirls PresentEducation Total PresentEducationTotal Level Level Class IX Class XI Class IX Class XIN 255 196 451 293 281 574Father’s education* Up to middle school 16.9 12.2 14.9 12.6 3.2 8.0 Highersecondary 35.3 26.5 31.5 35.5 18.1 27.0 College/technical 47.1 60.2 52.8 51.5 78.3 64.6Father’s occupation* Low/semiskilled/others 25.1 30.6 27.5 25.3 7.1 16.4 Clerical 31.8 21.4 27.3 21.8 29.5 25.6 Managerial/professional/business 38.8 39.8 39.2 51.2 58.0 54.5Mother’s education* Up to middle school 51.8 37.8 45.7 41.0 17.8 29.6 Highersecondary 32.9 31.1 32.2 32.1 33.1 32.6 College/technical 11.0 27.6 18.2 26.3 44.8 35.4Mother’s occupation* Wageemployment 7.8 11.2 9.3 9.2 10.7 9.9 Home-basedwork 23.9 17.3 21.1 26.3 13.9 20.2 Housewife 63.9 66.8 65.2 63.5 71.5 67.4Co-residence pattern Co-resides with at least one parent 89.8 63.8 78.5 96.9 85.1 91.1Parental supervision* Very strictly supervised 38.4 26.5 33.3 32.4 18.5 25.6 Somewhat strictly supervised 45.5 53.1 48.8 47.4 66.5 56.8 Not strictly supervised 15.7 19.9 17.5 19.1 14.2 16.7* Percentages do not add up to 100 because of missing responses.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly52that awareness is far from universal but that considerably larger proportions of students are aware of HIV/AIDS and its transmis-sion and prevention than of the physiology of menstruation, tim-ing of the cycle when pregnancy is most likely to occur, role of contraception in preventing pregnancy or even heard of the con-dom. Indeed, fewer than 10 per cent of students could correctly indicate the safe period; only two-fifths had heard of the condom and just under half were aware of the role of contraception in preventing pregnancy. In the area of HIV, in contrast, about three-fifths or more were aware of transmission and prevention of HIV. Of the total of eight issues raised, the average student was cor-rectly aware of fewer than four, suggesting poor overall aware-ness of sexual and reproductive health matters.Gender and age differences are wide on individual questions. As expected, students in Class XI were better informed than those in Class IX. As far as gender differences are concerned, boys were more likely to be aware of all HIV-related matters than girls; in matters related to pregnancy and contraception, gender differ-ences were less consistent. Compared to girls, boys were more likely to be aware of contraception and the condom, less likely to be aware of menstruation and about equally as poorly informed about the timing of the cycle during which chances of pregnancy are highest. The overall awareness index, summing the total of the eight matters about which awareness was explored, shows considerable age and gender differences. Despite the fact that two of the eight questions refer to menstruation, boys in Class IX were likely to be aware of significantly more matters than were girls; among those in Class XI, girls and boys were aware of an equal number of matters. Data shows that young girls are parti-cularly poorly informed.In a conservative setting like that prevailing in Bihar, in which social norms prohibit hugging and kissing among the unmarried, those reporting any physical intimacy have clearly engaged in socially prohibited behaviour. Hence in our study, we inquired not only about sexual experience but also about other types of physical contact. As discussed earlier, we asked respondents to report whether they had ever hugged an opposite-sex partner, kissed an opposite-sex partner on the lips, touched or been touched on the private parts by an opposite-sex partner and en-gaged in sexual relations with an opposite-sex partner. Findings and summary measures described earlier are presented in Table 5(p53).As evident from Table 5, some 28 per cent of boys and 12 per cent of girls report the experience of any physical intimacy, and some 23 per cent and 9 per cent respectively report the experience of kissing, touch or sex. There is, moreover, a systematic decline in percentages reporting increasingly intimate behaviours: among boys for example, 24 per cent report hugging, 20 per cent report kissing, 14 per cent report sexual touching and 10 per cent report sexual experience. Among girls, similar percentages are much lower: 10 per cent, 6 per cent, 3 per cent and 1 per cent, res-pectively. As expected, those in Class XI were more likely to report each of these behaviours than those in Class IX, but age specific differences are much narrower than the gender differences des-cribed above. What is clear from this table is that opposite-sex relations are not uncommon among school- and college-going adolescents in the ages 13-19, and sexual relations are experienced by significant minorities of boys, particularly those in Class XI.Almost all sexual experiences are risky. In total, among boys, while 10 per cent report having engaged in sexual relations, 8 per cent report non-use or sporadic condom use. Among girls, almost all (five of six) sexually experienced respondents had not used condoms regularly. We now explore the extent to which selected individual, family and household, and peer factors discussed above are associated Table 3: Awareness of Sexual and Reproductive Health Matters among Adolescent Students by Sex and Level of Education Characteristics Boys Girls Present Education Level Total Present Education Level Total Class IX Class XI Class IX Class XIN 255 196 451 293 281 574Pregnancy/contraception Per cent aware of the physiology of menstruation 11.4 25.5**** 17.5**** 17.1 49.8 33.1 Per cent aware that menstruation normally begins around 12-13 years 14.5**** 22.4**** 18.0**** 45.1 60.5 52.6 Per cent aware of safe period 7.5** 6.1** 6.9 2.0 14.6 8.2 Per cent who have heard of the condom 50.2**** 74.5**** 60.8**** 10.9 37.0 23.7 Per cent aware that pregnancy can be prevented by contraception 39.2* 60.2 48.3* 29.0 55.5 42.0Mean score (0-5) 1.23* 1.89* 1.51 1.04 2.17 1.60HIV/AIDS Per cent aware of the causes of HIV/AIDS 58.4**** 84.2* 69.6*** 40.6 79.4 59.6 Per cent aware of the modes of transmission 68.2**** 86.2 76.1*** 49.1 84.0 66.2 Per cent aware of the means of prevention 73.3**** 81.6*** 76.9**** 44.0 64.8 54.2Mean score (0-3) 2.00 2.52 2.23 1.34 2.28 1.80Overall total score (0-8) 3.23**** 4.41 3.74** 2.38 4.46 3.40* p≤0.10, **p≤0.05, ***p≤0.01, ****p≤0.001, t test applied.Table 4: Sexual Relations Reported by Students, by Sex and Education% Reporting Boys Girlsthe Following Present Education Total PresentEducationTotal LevelLevel Class IX Class XI Class IX Class XIN 255 196 451 293 281 574Any physical or sexual experience 26.7**** 28.6** 27.5**** 6.5 18.1 12.2Kissing, touching or sex 21.6**** 24.5** 22.8**** 5.1 13.2 9.1Sexual intercourse 7.5**** 13.8**** 10.2**** 0.4 2.2 1.2Sexual intercourse and unprotected sex 5.9*** 9.7*** 7.6**** 0.4 1.8 1.1Any physical experience Hugging 24.3****23.5** 23.9**** 4.1 15.3 9.6 Kissing 18.4****23.0*** 20.4**** 3.1 10.0 6.4Touching/being touched on private parts 11.7****17.3****14.2****1.73.62.61 * p≤0.10, ** p≤0.05, *** p≤0.01, **** p≤0.001.2 Data are missing for1-2 cases among boys and 2-9 cases among girls for various types of physical and sexual relations reported in the table. 3 t test applied.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200753with sexual and reproductive health awareness and physical and sexual experiences respectively. Dependent variables include (a)number of sexual and reproductive health matters about which the respondent is correctly aware; and (b) whether or not the respondent has experienced (i) any physical or sexual relationship, including hugging, kissing, touching or sex; (ii) kissing, touching or sex; and (iii) sex, and unprotected sex in the case of boys. Awareness of sexual and reproductive health matters is associated with individual, family and household, and peer level factors but the strength of these factors differs between boys and girls (Table 6). Among boys, awareness is positively influenced by age, exposure to both family life or sex education and blue films, and peer interaction; of note is that boys reporting peer interaction in relatively adult locations are less likely than others to be aware of sexual and reproductive health matters. Family and household level factors are unrelated to awareness among boys. Among girls in contrast, all three sets of factors appear to influence awareness: age, attendance at private school/college (perhaps reflecting the quality of education received) and exposure to family life or sex education clearly increase girls’ awareness of sexual and reproductive health matters; so too does peer interaction of any sort. Similarly, girls who reported that they were somewhat strictly supervised appear to report better awareness than either those who report being very strictly supervised or those reporting not being strictly supervised at all.The analysis of correlates of physical and sexual relations in-cludes by and large the same set of independent factors. It is clear that this set of variables better explains sexual and physical experiences among boys than among girls (Table 7, p 54).As far as any physical and sexual relations are concerned, gender differences are evident. Among girls, it is individual and family and household level variables that are associated with physical or sexual relations: those more likely to report physical or sexual relations include older girls (Class XI), those in private schools/colleges and those with less educated mothers. Neither the frequency of peer contact nor the extent of parental supervision is associated with either of the two dependent variables. Among boys, in contrast, all three sets of indicators are signi-ficantly associated with one or more of the four outcome indi-cators. A look at the factors associated with the two indicators also available for girls (any physical or sexual experience; and kissing, touching or sex) suggests that as in the case of girls, maternal edu-cation is significantly associated with both outcomes. Other asso-ciations differ considerably: unlike girls, for boys, among individu-al level factors, exposure to blue films and regular interaction with peers is significantly and positively associated. Of the two outcome indicators reflecting sexual experience (ever had sex; and unpro-tected sex), a larger set of factors is significant for boys. These in-clude exposure to blue films and extent of peer interaction; at the family level, again we find that maternal education is significantly and inversely associated with boys’ sexual experience. What is no-table is the extent to which parental supervision influences whether or not boys have engaged in sexual activity: findings clearly suggest that those who report somewhat strict supervision, and particularly those who report very strict supervision, are sig-nificantly less likely than those reporting that they were not at all strictly supervised to have engaged in sexual relations. What the Findings SuggestFindings suggest that students in Patna lack complete awareness about the kinds of sexual and reproductive matters that are likely to protect them from unwanted pregnancy and infection. At the Table 5: Frequency and Location of Interaction with PeersCharacteristics BoysGirls PresentEducationTotal PresentEducation Total Level Level Class IX Class XI Class IX Class XIN 255 196 451 293 281 574A Social interaction with peers% been to the cinema* Never 40.424.533.545.450.547.9 Sometimes 42.7 41.8 42.4 43.7 40.2 42.0 Often/very often 16.1 32.7 23.3 8.9 6.4 7.7% eating out at restaurants* Never 51.4 52.6 51.9 44.0 34.9 39.5 Sometimes 34.5 33.2 33.9 35.5 45.2 40.2 Often/veryoften 12.9 12.8 12.9 19.1 16.4 17.8% attending sports clubs/ health clubs* Never 75.3 63.8 70.3 90.1 89.3 89.7 Sometimes 12.9 14.8 13.7 5.8 3.9 4.9Often/very often 9.8 18.9 13.7 1.7 2.1 1.9% going out on picnics* Never 31.0 28.6 29.9 35.5 38.4 36.9 Sometimes 45.9 31.1 39.5 50.9 46.3 48.6 Often/very often 20.4 37.2 27.7 11.9 10.3 11.1Social interaction index (0-8) 2.5 3.3 2.9 2.2 2.1 2.2B Interaction with peers in more “adult” settings % who go sometimes/often to a bar, discotheque or late night parties 25.5 32.7 28.6 17.1 12.8 15.0* Percentages do not add up to 100 because of missing responses.Table 6: Correlates of Reproductive Health Knowledge Boys Girls Beta BetaPresent education level Class IX (rc) ClassXI 1.00**** 1.61****Type of school/college Government (rc) Private 0.07 1.06****Exposure to blue films: ever watched 0.78*** 0.43Family life or sex education classes: ever attended 0.77*** 0.66**Mother’s education Up to middle school (rc) Highersecondary 0.28 0.23 College/technical 0.18 0.29Father’s occupation Manual/semiskilled/others(rc) Clerical 0.11 -0.37 Managerial/professional/business 0.20 -0.21Parental supervision Very strictly supervised (rc) Somewhat strictly supervised 0.12 0.53*** Not strictly supervised 0.12 0.05Co-resides with at least one parent -0.20 -0.70**Peer interaction Social interaction index (0-8) 0.15*** 0.11** Sometimes/often visit adult settings (bars, discotheques or late night parties) -0.36 0.46**Adjusted R2 0.18 0.36N# 394 5191 * p≤0.10, ** p≤0.05, *** p≤0.01, **** p≤0.001. 2 rc: reference category.3 # Data are missing for 57 boys and 55 girls for some explanatory variables.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly54same time, findings suggest that even in this relatively traditional setting, school- and college-going students do find opportunities to engage in physical and even sexual relationships: a substantial proportion of adolescent girls and boys have experienced some form of physical intimacy, and a significant minority of young boys have experienced sexual relations, usually unprotected. Despite the fact that questionnaires were anonymous and self-administered, the familiar gender gaps in reports of physical and sexual experience are evident. We acknowledge that, as other studies have observed [Jejeebhoy and Sebastian 2004], this could reflect possible under-reporting among females and/or over-reporting among males. Our findings suggest, moreover, that in this sample of students, selected individual, peer and family level factors are significantly associated with awareness of sexual and reproduc-tive health, and physical and sexual experience. Gender differ-ences in the correlates of physical and sexual behaviour are evident, although the directions of effects are generally similar. Among individual level correlates, as expected, the older the students are or the longer they are exposed to education, more the opportunities they have to form relationships and engage in both physical and sexual relations; however, this is significant only among girls, for whom Class XI may involve more freedom of movement and unsupervised time than Class IX, compared to boys who may have freedom from an earlier age. As others have noted, moreover, exposure to other risky influences like pornographic films, is significantly associated with physical or sexual experience among boys. Peer contact likewise is power-fully and positively associated with physical or sexual experi-ence; in contrast, among girls, who display considerably less peer contact than boys, the closeness of peer contacts appears only weakly associated. At the family level, maternal education appears to inversely influence the experience of physical intimacy and sexual relations among adolescents. What is notable is the extent to which parental supervision influences young people’s behaviour. While among girls, the extent of supervision is only weakly associated with physical intimacy indicators, among boys there is a strong and powerful relation-ship; findings highlight that those reporting somewhat strict and very strict parental supervision are indeed significantly less likely to report sexual relations.While many of the same variables were associated with awareness of sexual and reproductive health matters, some clear differences are noted. Adolescents who have undergone family life or sex education and have frequent peer contact are more likely to be aware of these issues than others. Among boys, exposure to blue films appears positively associated with awareness of sexual matters, raising questions about whether or not the information provided in these is always inaccurate. Family and household level factors appear, by and large, to be unrelated to awareness about sexual matters – indeed, those living away from their parents appear to be better informed than those who co-reside with their parents.Table 7: Correlates of Physical or Sexual Intimacy, among Adolescent Students: Odds Ratios from Logistic Regression Analysis Any Physical or Sexual Experience@ Kissing, Touching or Sex Ever Had Sex Ever Had Unprotected Sex Boys Girls Boys Girls Boys BoysPresent education level Class IX (rc) Class XI 1.12 CI (0.65 – 1.96) 3.01***CI (1.53 – 5.91) 1.18CI (0.65 – 2.13) 2.79**CI (1.29 – 6.07) 1.61CI (0.70 – 3.70) 1.49 CI (0.60 – 3.68)Type of school/college Government (rc) Private 1.09 CI (0.62 – 1.91) 2.09**CI (1.01 – 4.33) 0.89CI (0.49 – 1.63) 1.82CI (0.81 – 4.10) 2.34CI (0.89 – 6.16) 2.14CI (0.76 – 6.03)Exposure to blue films No (rc) Yes 6.59****CI (3.60 – 12.06) 1.79CI (0.54 – 5.95) 8.23****CI (4.48 – 15.12) 1.95CI (0.51 – 7.39) 5.37****CI (2.43 – 11.86) 3.90***CI (1.62 – 9.43)Mother’s education Up to middle school (rc) Higher secondary 0.55*CI (0.30 – 1.00) 0.62CI (0.28 – 1.38) 0.69CI (0.37 – 1.30) 0.71CI (0.29 – 1.72) 0.34**CI (0.13 – 0.89) .36**CI (0.13 – 1.02) College/technical 0.68 CI (0.33 – 1.38) 0.47*CI (0.20 – 1.08) 0.90CI (0.43 – 1.89) 0.45CI (0.17 – 1.17) 0.51CI (0.18 – 1.44) .55CI (0.18 – 1.70)Father’s occupation Low/semiskilled/others (rc) Clerical 1.25CI (0.61 – 2.57) 0.53CI (0.20 – 1.41) 1.01CI (0.47 – 2.20) 0.49CI (0.17 – 1.39) 0.95CI (0.33 – 2.78) 0.95CI (0.32 – 2.86) Managerial/professional/business 1.75*CI (0.90 – 3.38) 0.50CI (0.21 – 1.23) 1.56CI (0.78 – 3.14) 0.41*CI (0.15 – 1.08) 0.98CI(0.37 – 2.56) 0.60CI (0.21 – 1.70)Parental supervision Very strictly supervised (rc) Somewhat strictly supervised 1.72*CI (0.92 – 3.22) 1.66CI (0.72 – 3.80) 1.30CI (0.67 – 2.53) 1.35CI (0.55 – 3.32) 2.85**CI (0.97 – 8.36) 2.95*CI (0.90 – 9.67) Not strictly supervised 1.21CI (0.54 – 2.68) 1.42CI (0.51 – 3.94) 0.96CI (0.41 – 2.25) 1.13CI (0.36 – 3.52) 4.80**CI (1.46 – 15.77) 5.91***CI (1.66 – 20.97)Residence pattern Co-resides with at least one parent (rc) Co-resides with neither parent 1.41 CI (0.69 – 2.87) 0.69CI (0.28 – 1.68) 1.17CI (0.55 – 2.49) 0.63CI (0.23 – 1.71) 1.35CI (0.51 – 3.56) 2.00CI (0.68 – 5.92)Interaction with peers 1.20**CI (1.04 – 1.40) 1.11CI (0.93 – 1.33) 1.19**CI (1.02 – 1.39) 1.12CI (0.91 – 1.37) 1.30**CI (1.05 – 1.62) 1.26**CI (1.0 – 1.59)Interacts sometimes or often with peers in adult locations No (rc) Yes 1.66CI (0.91 – 3.05) 1.30CI (0.59 – 2.87) 1.37CI (0.71 – 2.63) 1.22CI (0.49 – 3.03) 2.04CI (0.83 – 5.03) 1.54 CI (0.57 – 4.19) R2 0.26 0.110.270.090.310.22N# 394 510 394 510 394 3941 rc: reference category. 2 *p≤0.10 **p≤0.05 ***p≤0.01 ****p≤0.001. 3 # Data are missing for 57 boys and 64 girls for some explanatory variables.4 @ Includes hugging, kissing, touching or sex with an opposite-sex partner. 5 CI= lower and upper limit of confidence interval.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly48Exploring Safe Sex Awareness and Sexual Experiences of Adolescents in PatnaMedha Shekhar, Saswata Ghosh, Pradeep PandaThis article reports findings from a study of school- and college-going adolescent girls and boys in Patna that explored their awareness of sexual matters and safe sex, and their experience of sexual relations. The article correlates this awareness and experience at the individual, family and peer levels. Maternal education and parental supervision of adolescent activities played an important role in promoting awareness among girls, suggesting that parenting issues are critical. Further, programmes that apprise parents of the role they play are needed. There is an increasing recognition in India of the need to address the sexual and reproductive health of adolescents. A small but growing body of literature has begun to exam-ine the reproductive and sexual health situation of this group. Evidence generated thus far from small studies conducted in different parts of the country suggests that the youth remain poorly informed about sexual and reproductive matters, and that significant minorities of youth have engaged in premarital sex, largely unprotected [see for example, Jejeebhoy and Sebastian 2004]. While these studies have provided considerable insight into young people’s lives in certain geographic settings of the country, other regions, however, have been somewhat neglected. The state of Bihar is one such setting in which insight into the sexual and reproductive behaviour of unmarried youth remains sparse. While it is believed that strict social norms that forbid opposite-sex social mixing among young people and close supervision of children tend to prevent pre-marital sexual relations, there is little evidence that supports or reverses such assumptions. In a setting in which about one quarter of the population is adolescent (10-19 years), in which proportion of adolosecent pregnancy is high and in which human immunodeficiency virus (HIV) and sexually transmitted infections (STIs) are increasingly observed, lack of evidence has clearly impeded the development and imple-mentation of programmes that inform adolescents about sexual matters in general and safe sex in particular.The objective of this paper is to fill this gap in evidence. Speci-fically, the paper reports findings from a study of school- and college-going adolescent girls and boys in the city of Patna that explored their awareness of sexual matters and safe sex, experi-ence of sexual relations; and correlates of this awareness and experienceat the individual, family and peer level. BackgroundAvailable studies conducted in India are generally small, using variable methodologies and focusing on young people of differ-ing ages [Abraham and Kumar 1999; Alexander et al 2006; Awasthi, Nichter and Pandey 2000; Jejeebhoy 2000; Mehra, Savithri and Coutinho 2002; Pachauri and Santhya 2003; Pandey et al 2006; Patel and Andrew 2001]. Findings cannot therefore be generalised. Even so, there is a remarkable consistency in findings: they suggest that between 15 per cent and 30 per cent of young men and fewer than 10 per cent of young women report the experience of penetrative sex, and that substantially more report a romantic relationship or other forms of physical intimacy, Medha Shekhar(shekharmedha@yahoo.co.in) is managing trustee, PURWA, Patna; Saswata Ghosh (ghosh.saswata@gmail.com)is lecturer, Institute of Development Studies, Kolkata and Pradeep Panda (pradeep4164@yahoo.com) is director, rural research, Institute of Rural Research and Development, Gurgaon.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200749including handholding, kissing or touching private parts [Alexander et al 2006; Jejeebhoy and Sebastian 2004]. Most of these studies have been descriptive and have rarely focused on the correlates of sexual behaviour. There are, however, exceptions. A study of sexual experience among 966 low-income college-going students in Mumbai [Abraham and Kumar 1999] noted that the sexually experienced were significantly more likely than others to differ on individual characteristics: for example, they were more likely to have access to resources, hold positive attitudes towards premarital sex, report more social interaction with peers and be exposed to pornographic materials. Factors such as parental restriction, family religiosity and family income played no role in influencing whether or not these students had experienced sex. In another study among adolescents in a slum setting in Pune city and the surrounding rural areas, level of self-confidence and peer interaction were significantly as-sociated with reports of the experience of romantic and sexual relations; while parental restriction was not associated with romantic or sexual relations, the extent to which youth reported close interaction with parents was inversely associatedwith these experiences [Alexander et al 2006]. Finally, a study of adolescents in a Delhi slum setting, using en-tirely qualitative methodologies, reports, similarly, that parental restrictiveness does not deter young people from forming roman-tic partnerships or engaging in sexual relations [Mehra, Savithri and Coutinho 2002].This study builds on Abraham and Kumar’s (1999) study of low-income college students in Mumbai. The study instrument and design have borrowed extensively from the Mumbai study, but the focus is on a younger group of adolescents attending school/college.Patna, the site of this study, is different from the settings in which the studies described above were conducted in many res-pects. For one, the state of Bihar in which Patna, the capital city, is located, is relatively poorly developed with below all-India average scores on socio-economic indicators. For example, Bihar has the lowest literacy rate in the country: in 2001, less than half the population aged seven and above were literate (60 per cent of males and 34 per cent of females) against the all-India figure of 75.9 and 54.3 for males and females respectively [Registrar General, India(RGI) 2001a]. School enrolment also lags behind national figures: in 1999-2000, the percentage of adolescents (10-19 years) attending edu-cational institutions was 48 per cent in Bihar, compared to 60 per cent in India as a whole.1 Within Bihar, wide gender disparities in school enrolment are also apparent: in 1999-2000, 56 and 38 per cent of males and females were enrolled in school, compared to 66 and 53 per cent, respectively in India as a whole. School enrol-ment rates are much higher in urban than rural Bihar: 72 and 61 per cent of males and females aged 10-19 years are in school in urban areas, compared to 54 and 35 per cent, respectively, in rural areas. The reproductive health situation is also poorer in Bihar than in India on average. For example, in 2005-06, the total fertility rate in Bihar was considerably higher than all-India figure (4.00 vs 2.68) [International Institute for Population Sciences 2007a; 2007b]. Similarly, during the same period, sub-stantially smaller percentages of women received three antenatal care visits for their last birth (17 per cent vs 51 per cent), reported institutional delivery (22 per cent vs 41 per cent), or fully immu-nised their children (33 per cent vs 44 per cent). The public health infrastructure is grossly inadequate: Only 23 out of 38 districts in the state have a district hospital [Government of Bihar 2006]. In-depth awareness of HIV-related issues among adolescents (15-19 years) in Bihar is poorer than in India more generally: for example, in Bihar 37.5 per cent of rural males and 15.9 per cent of rural females were aware of important methods of HIV prevention (i e, consistent condom use and sex with a faithful uninfected partner) compared to 54.1 per cent and 44.5 per cent of adolescents nationally. Similarly, adolescents in Bihar are much less likely to be correctly aware of some common misconceptions regarding transmission of HIV/acquired immunodeficiency syndrome (AIDS), i e, that mosquito bites and sharing a meal with an infected person do not transmit HIV, and that a healthy looking person can transmit HIV: 8.3 per cent and 7.9 per cent of rural adolescent men and women respectively in Bihar compared to 21.8 per cent and 20.2 per cent nationally [NACO andUNICEF 2002].Among the 38 districts of Bihar, Patna has the largest popula-tion (4.72 million) [RGI 2001b], the largest urban centre (42 per cent urban) and the highest literacy rate (63 per cent overall; 73 per cent for males and 51 per cent for females) (calculated from data inRGI 2001b). As the state capital, Patna is the leading ad-ministrative and educational centre. Adolescents in Patna are as-sumed to have greater access to education, employment opportu-nities, modern lifestyles and new ideas as compared to other districts of the state.Study Design and MethodologyThis study was conducted in Patna city in 22 schools and colleges, government and private, both single-sex and coeducational. We must acknowledge that the selection of educational institutions was somewhat opportunistic. Researchers approached 41 schools and colleges; many institutions (N=15) refused permission to undertake the study or dropped out after giving permission (N=4) and hence the study has been conducted in self-selected schools and colleges that granted researchers access to students.The study comprised two phases. In an initial qualitative phase, eight focus group discussions were held with male and female students, and eight key informant interviews were held with selected students, parents and teachers. The objective of this phase was to inform the design, content and wording of the survey instrument, and specifically to better understand the context in which young students grow up and the terminologies used to describe sexual matters. The second phase comprised a survey conducted in 2005 among students in Classes IX (aged around 14 years) and XI (aged around 16 years). Questionnaires were self-administered.While the selection of educational institutions was, by necessity, opportunistic, the sample of students who responded to the survey was largely representative of students drawn from the two classes within these schools and colleges. The survey was conducted in one randomly selected section of Class IX and Class XI from each 
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly50selected institution; all students who were present in the class at the time of the survey and who consented to respond to the sur-vey were eligible for inclusion. A total of 32 sessions were held in the 22 schools and colleges selected for study.Consent procedures were followed at two levels. All students in the selected classes were provided a parental consent form and asked to obtain permission from a parent to participate in the study; only those whose parents con-sented were considered for the study. At the time of administering the ques-tionnaires, consent was sought from students; only consenting students were included in the study. Refusals were few: only eight parents refused to grant consent; not a single student who was in class on that day opted not to participate.The survey was conducted in the presence of a five-member survey team (two male and two female in-vestigators, and the principal investi-gator) who explained the objective of the survey to the class, obtained individual consent from all students and ensured that students had privacy while filling out the question-naire and fully understood the con-tents. Respondents were instructed not to identify themselves in the questionnaire and were assured that only the research team would have access to the data. Teach-ers and school or college authorities were not involved in admin-istering the questionnaire.Survey InstrumentThe survey instrument, largely adapting that used by Abraham and colleagues, was in Hindi and pre-tested. It comprised a total of 100 questions. Following the completion of the survey in each class, the research team invited students who had questions on issues addressed in the questionnaire to approach the research team; many students did take advantage of this offer or subsequently contacted the research team for further information. In case the survey was to be conducted in Classes IX and XI of the same school/college, the session was conducted simultaneously in both classes so as to avoid sharing of infor-mation. The total sample size planned for the survey was 1,000;in total, 1,048 students (461 boys and 587 girls) completed the interview. Some 10 boys and 13 girls were married, and their responses have been excluded from the analysis. In total then, our analysis is restricted to 1,025 unmarried students. Of these, 44 per cent are male (451) and 56 per cent female (574). More-over, 53 per cent were from Class IX and 47 per cent from Class XI. Findings are presented for 255 and 196 boys and 293 and 281 girls in Classes IX and XI, respectively. In order to test the signi-ficance of gender differences, we present z-tests and t-tests com-paring reports of girls and boys in Classes IX and XI, as well as all girls and boys.Our analysis examines awareness of sexual and reproduc-tive health matters and experiences of physical contact (hugging, kissing, touching private parts) and sexual relations with an opposite-sex partner. Awareness of sexual and reproductive health matters is meas-ured by a total of eight questions, five covering aspects of men-struation, pregnancy and contraception, and three covering HIV transmission and prevention. Res-ponses were set to equal 1 if correct and 0 if not. A summary index of awareness was then created that sums responses on all of these eight indicators, and ranges from 0 to 8. Physical contact and sexual experi-ence were measured by asking about the following experiences with an opposite-sex partner: (a) hugging, (b) kissing, (c) touching private parts; and (d) sexual intercourse. Those who experienced sexual inter-course were also asked about condom use. Four indicators were defined. The first is a summary measure ref-lecting any physical or sexual experi-ence, that is the respondent was as-signed a value of 1 if s/he reported the experience of hugging, kissing, touch or sexual intercourse and 0 if not. The second indicator omits hugging and measures the expe-rience of kissing, touching or sex, while the third specifically measures the experience of sexual intercourse with an opposite-sex partner among males as few girls reported the experience of sexual intercourse. The fourth indicator measures unprotected sex (i e, non-use or sporadic use of condoms) among males reporting sexual experience.In line with the literature, our analytic framework posits that awareness and sexual experience may be associated with a number of factors at the individual, family and household, and peer levels. A description of the variables used in this analysis is provided below.Individual level variables that we considered likely to influence awareness and sexual experience included age, class attended at the time of interview (Class IX or XI), type of school or college attended at the time of interview (government/private), wage earning status (yes/no) and exposure to pornographic films, ei-ther onDVD or at theatres or on REN TV (a pornographic televi-sion channel from Russia with reach in Patna). It also included attending a family life education or sex education programme in school or college.Family and household level variables include parents’ educa-tional attainment levels, parents’ occupation, residential status (co-residing with at least one parent or not) and perceived strict-ness of parental supervision: respondents were asked the extent to which, compared to their same-sex peers, they perceived that their parents supervised them very strictly, somewhat strictly or not at all strictly.Table 1: Percentage Distribution of Adolescent Students by Sex, Education Level and Individual Level CharacteristicsCharacteristics BoysGirls PresentEducation Total PresentEducation Total Level Level Class IX Class XI Class IX Class XIN 225 196 451 293 281 574Age 13-14 58.8 -33.3 57.0 -29.1 15-17 37.6 69.9 51.7 41.6 84.0 62.4 18-20 3.5 30.1 15.1 1.4 16.0 8.5Mean age 14.5 16.7 15.5 14.3 16.7 15.5Type of school/college Government 53.7 26.5 41.9 58.4 29.5 44.3 Private 46.3 73.5 58.1 41.6 70.5 55.7Wagework Working 6.7 11.7 8.9 3.4 3.9 3.7Religion Hindu 92.5 89.8 91.4 95.9 83.6 89.9 Muslim 6.7 9.2 7.8 3.1 13.5 8.2 Other 0.8 1.0 0.9 1.0 2.8 1.9Ever attended family life or sex education in school/college Yes 20.0 13.3 17.1 7.5 15.0 11.2Exposure to blue films Yes 14.5 21.4 17.6 6.8 2.8 4.9* Percentages do not add up to 100 because of missing responses.
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200751Peer level factors are measured by two indicators reflecting theextent of social interaction with peers. Respondents were asked if they spent time with their friends in a number of social activities (going to the cinema, restaurants, sports clubs; going out on picnics) and the frequency of this interaction. Responses were coded to equal 0 if the respondent never engaged in the ac-tivity, 1 if s/he sometimes participated in the activity, and 2 if frequently. The social interaction index then ranges from 0 to 8. Respondents were also asked whether they socialised with their friends in more “adult” settings like bars, late-night parties or dis-cotheques; a dichotomous indicator was created that was set to equal 1 if the respondent reported that s/he attended any of these events sometimes or often, and 0 if not.Background FactorsIn this section, we explore the individual, family and household, and peer level factors that are assumed to influence sexual aware-ness and experiences of young people. Table 1 (p 50) presents individual level factors. On an average, respondents were aged 15.5 years and over 90 per cent were Hindu. While about 60 per cent of those in Class IX came from government institutions, this percentage was reversed and almost three quarters of respond-ents from Class XI were drawn from private institutions. A small proportion (fewer than 10 per cent) combined schooling with economic activity. A few were exposed to family life or sex education in school or college: 17 per cent of boys and 11 percent of girls. Finally, significant minorities – 18 per cent of boys and 5 per cent of girls – reported that they had seen blue films on DVD, in theatres or on REN TV.Family and household level factors are described in Table 2. Gender and to a lesser extent age differences are stark. It is clear that compared to boys, girls are drawn from households in which parents are well educated and fathers are largely in managerial, professional or business occupations. This selectivity reflects the fact that in a setting in which large numbers of girls are with-drawn from school at puberty, girls who remain in school by Class IX come from families that are better educated and better-off economically, who may place a greater value on the education of their daughters than does the population more generally. Also the finding that parents of those in Class XI tend to be far better educated than the parents of those in Class IX can be explained by the fact that many youth discontinue schooling after Class X (school completion), and it is likely that better educated and better-off parents are more likely than others to educate their children beyond this level. Co-residence with one or both parents also reflects gender and age differences: Just over one-fifth of boys and about one-tenth of girls live separately from a parent; this proportion is considerably higher among Class XI students than Class IX, presumably reflecting the facts that parents are less likely to allow girls to move away than boys, and that many youth from outside Patna who wish to pursue a higher education may not have the opportunity and thus require to relocate to Patna and reside in hostels or with relatives while studying in Class XI.Table 2 also presents adolescent students’ perceptions of the extent of supervision by parents. Some one third of boys and one-quarter of girls reported that they are very strictly super-vised by their parents; about half and three-fifths, respectively, reported they are somewhat strictly supervised and about 17 per cent of both boys and girls reported that they are not strictly supervised. It is indeed surprising that a larger proportion of boys than girls perceivethemselves to be strictly supervised. We sug-gest that in a setting in which restrictive norms for girls but not boys are widespread, “strict supervision” may be differently interpreted or perceived by boys and girls.Students’ interaction with peers is described in Table 3 (p 52). As mentioned earlier, we have clubbed interaction indicators into those that are more “typical” of adolescents (picnics, films, clubs, eating places) and those that are more adult and may provide opportunities for greater physical contact (for example, bars, discotheques and late night parties). It is clear that interaction is limited. Indeed, many students – about one-third of all boys and two-fifths of all girls – have never interacted with their peers in even such relatively common settings as picnics and films. And frequent interaction is relatively uncommon. Gender disparities are evident: boys are far more likely than girls to interact with their peers in three of the four locations (picnics, films and sports clubs). Scores on the social interaction index suggest that while girls in Class IX are only slightly less likely than boys to interact with their peers (mean scores of 2.2 and 2.5 respectively), those in Class XI are considerably less likely to do so (mean scores of 2.1 and 3.3 respectively).We also probed the extent to which adolescents visited more adult sites with their peers, for example, bars, discotheques and late night parties: Findings suggest that some 29 per cent of boys and 15 per cent of girls have attended such events sometimes or frequently. Table 4 (p 52) presents students’ awareness of sexual and re-productive health matters as described earlier. Findings suggest Table 2: Distribution of Students by Sex, Education and Family and Household CharacteristicsCharacteristics BoysGirls PresentEducation Total PresentEducationTotal Level Level Class IX Class XI Class IX Class XIN 255 196 451 293 281 574Father’s education* Up to middle school 16.9 12.2 14.9 12.6 3.2 8.0 Highersecondary 35.3 26.5 31.5 35.5 18.1 27.0 College/technical 47.1 60.2 52.8 51.5 78.3 64.6Father’s occupation* Low/semiskilled/others 25.1 30.6 27.5 25.3 7.1 16.4 Clerical 31.8 21.4 27.3 21.8 29.5 25.6 Managerial/professional/business 38.8 39.8 39.2 51.2 58.0 54.5Mother’s education* Up to middle school 51.8 37.8 45.7 41.0 17.8 29.6 Highersecondary 32.9 31.1 32.2 32.1 33.1 32.6 College/technical 11.0 27.6 18.2 26.3 44.8 35.4Mother’s occupation* Wageemployment 7.8 11.2 9.3 9.2 10.7 9.9 Home-basedwork 23.9 17.3 21.1 26.3 13.9 20.2 Housewife 63.9 66.8 65.2 63.5 71.5 67.4Co-residence pattern Co-resides with at least one parent 89.8 63.8 78.5 96.9 85.1 91.1Parental supervision* Very strictly supervised 38.4 26.5 33.3 32.4 18.5 25.6 Somewhat strictly supervised 45.5 53.1 48.8 47.4 66.5 56.8 Not strictly supervised 15.7 19.9 17.5 19.1 14.2 16.7* Percentages do not add up to 100 because of missing responses.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly52that awareness is far from universal but that considerably larger proportions of students are aware of HIV/AIDS and its transmis-sion and prevention than of the physiology of menstruation, tim-ing of the cycle when pregnancy is most likely to occur, role of contraception in preventing pregnancy or even heard of the con-dom. Indeed, fewer than 10 per cent of students could correctly indicate the safe period; only two-fifths had heard of the condom and just under half were aware of the role of contraception in preventing pregnancy. In the area of HIV, in contrast, about three-fifths or more were aware of transmission and prevention of HIV. Of the total of eight issues raised, the average student was cor-rectly aware of fewer than four, suggesting poor overall aware-ness of sexual and reproductive health matters.Gender and age differences are wide on individual questions. As expected, students in Class XI were better informed than those in Class IX. As far as gender differences are concerned, boys were more likely to be aware of all HIV-related matters than girls; in matters related to pregnancy and contraception, gender differ-ences were less consistent. Compared to girls, boys were more likely to be aware of contraception and the condom, less likely to be aware of menstruation and about equally as poorly informed about the timing of the cycle during which chances of pregnancy are highest. The overall awareness index, summing the total of the eight matters about which awareness was explored, shows considerable age and gender differences. Despite the fact that two of the eight questions refer to menstruation, boys in Class IX were likely to be aware of significantly more matters than were girls; among those in Class XI, girls and boys were aware of an equal number of matters. Data shows that young girls are parti-cularly poorly informed.In a conservative setting like that prevailing in Bihar, in which social norms prohibit hugging and kissing among the unmarried, those reporting any physical intimacy have clearly engaged in socially prohibited behaviour. Hence in our study, we inquired not only about sexual experience but also about other types of physical contact. As discussed earlier, we asked respondents to report whether they had ever hugged an opposite-sex partner, kissed an opposite-sex partner on the lips, touched or been touched on the private parts by an opposite-sex partner and en-gaged in sexual relations with an opposite-sex partner. Findings and summary measures described earlier are presented in Table 5(p53).As evident from Table 5, some 28 per cent of boys and 12 per cent of girls report the experience of any physical intimacy, and some 23 per cent and 9 per cent respectively report the experience of kissing, touch or sex. There is, moreover, a systematic decline in percentages reporting increasingly intimate behaviours: among boys for example, 24 per cent report hugging, 20 per cent report kissing, 14 per cent report sexual touching and 10 per cent report sexual experience. Among girls, similar percentages are much lower: 10 per cent, 6 per cent, 3 per cent and 1 per cent, res-pectively. As expected, those in Class XI were more likely to report each of these behaviours than those in Class IX, but age specific differences are much narrower than the gender differences des-cribed above. What is clear from this table is that opposite-sex relations are not uncommon among school- and college-going adolescents in the ages 13-19, and sexual relations are experienced by significant minorities of boys, particularly those in Class XI.Almost all sexual experiences are risky. In total, among boys, while 10 per cent report having engaged in sexual relations, 8 per cent report non-use or sporadic condom use. Among girls, almost all (five of six) sexually experienced respondents had not used condoms regularly. We now explore the extent to which selected individual, family and household, and peer factors discussed above are associated Table 3: Awareness of Sexual and Reproductive Health Matters among Adolescent Students by Sex and Level of Education Characteristics Boys Girls Present Education Level Total Present Education Level Total Class IX Class XI Class IX Class XIN 255 196 451 293 281 574Pregnancy/contraception Per cent aware of the physiology of menstruation 11.4 25.5**** 17.5**** 17.1 49.8 33.1 Per cent aware that menstruation normally begins around 12-13 years 14.5**** 22.4**** 18.0**** 45.1 60.5 52.6 Per cent aware of safe period 7.5** 6.1** 6.9 2.0 14.6 8.2 Per cent who have heard of the condom 50.2**** 74.5**** 60.8**** 10.9 37.0 23.7 Per cent aware that pregnancy can be prevented by contraception 39.2* 60.2 48.3* 29.0 55.5 42.0Mean score (0-5) 1.23* 1.89* 1.51 1.04 2.17 1.60HIV/AIDS Per cent aware of the causes of HIV/AIDS 58.4**** 84.2* 69.6*** 40.6 79.4 59.6 Per cent aware of the modes of transmission 68.2**** 86.2 76.1*** 49.1 84.0 66.2 Per cent aware of the means of prevention 73.3**** 81.6*** 76.9**** 44.0 64.8 54.2Mean score (0-3) 2.00 2.52 2.23 1.34 2.28 1.80Overall total score (0-8) 3.23**** 4.41 3.74** 2.38 4.46 3.40* p≤0.10, **p≤0.05, ***p≤0.01, ****p≤0.001, t test applied.Table 4: Sexual Relations Reported by Students, by Sex and Education% Reporting Boys Girlsthe Following Present Education Total PresentEducationTotal LevelLevel Class IX Class XI Class IX Class XIN 255 196 451 293 281 574Any physical or sexual experience 26.7**** 28.6** 27.5**** 6.5 18.1 12.2Kissing, touching or sex 21.6**** 24.5** 22.8**** 5.1 13.2 9.1Sexual intercourse 7.5**** 13.8**** 10.2**** 0.4 2.2 1.2Sexual intercourse and unprotected sex 5.9*** 9.7*** 7.6**** 0.4 1.8 1.1Any physical experience Hugging 24.3****23.5** 23.9**** 4.1 15.3 9.6 Kissing 18.4****23.0*** 20.4**** 3.1 10.0 6.4Touching/being touched on private parts 11.7****17.3****14.2****1.73.62.61 * p≤0.10, ** p≤0.05, *** p≤0.01, **** p≤0.001.2 Data are missing for1-2 cases among boys and 2-9 cases among girls for various types of physical and sexual relations reported in the table. 3 t test applied. 
REPRODUCTIVE HEALTHEconomic & Political Weekly december 1, 200753with sexual and reproductive health awareness and physical and sexual experiences respectively. Dependent variables include (a)number of sexual and reproductive health matters about which the respondent is correctly aware; and (b) whether or not the respondent has experienced (i) any physical or sexual relationship, including hugging, kissing, touching or sex; (ii) kissing, touching or sex; and (iii) sex, and unprotected sex in the case of boys. Awareness of sexual and reproductive health matters is associated with individual, family and household, and peer level factors but the strength of these factors differs between boys and girls (Table 6). Among boys, awareness is positively influenced by age, exposure to both family life or sex education and blue films, and peer interaction; of note is that boys reporting peer interaction in relatively adult locations are less likely than others to be aware of sexual and reproductive health matters. Family and household level factors are unrelated to awareness among boys. Among girls in contrast, all three sets of factors appear to influence awareness: age, attendance at private school/college (perhaps reflecting the quality of education received) and exposure to family life or sex education clearly increase girls’ awareness of sexual and reproductive health matters; so too does peer interaction of any sort. Similarly, girls who reported that they were somewhat strictly supervised appear to report better awareness than either those who report being very strictly supervised or those reporting not being strictly supervised at all.The analysis of correlates of physical and sexual relations in-cludes by and large the same set of independent factors. It is clear that this set of variables better explains sexual and physical experiences among boys than among girls (Table 7, p 54).As far as any physical and sexual relations are concerned, gender differences are evident. Among girls, it is individual and family and household level variables that are associated with physical or sexual relations: those more likely to report physical or sexual relations include older girls (Class XI), those in private schools/colleges and those with less educated mothers. Neither the frequency of peer contact nor the extent of parental supervision is associated with either of the two dependent variables. Among boys, in contrast, all three sets of indicators are signi-ficantly associated with one or more of the four outcome indi-cators. A look at the factors associated with the two indicators also available for girls (any physical or sexual experience; and kissing, touching or sex) suggests that as in the case of girls, maternal edu-cation is significantly associated with both outcomes. Other asso-ciations differ considerably: unlike girls, for boys, among individu-al level factors, exposure to blue films and regular interaction with peers is significantly and positively associated. Of the two outcome indicators reflecting sexual experience (ever had sex; and unpro-tected sex), a larger set of factors is significant for boys. These in-clude exposure to blue films and extent of peer interaction; at the family level, again we find that maternal education is significantly and inversely associated with boys’ sexual experience. What is no-table is the extent to which parental supervision influences whether or not boys have engaged in sexual activity: findings clearly suggest that those who report somewhat strict supervision, and particularly those who report very strict supervision, are sig-nificantly less likely than those reporting that they were not at all strictly supervised to have engaged in sexual relations. What the Findings SuggestFindings suggest that students in Patna lack complete awareness about the kinds of sexual and reproductive matters that are likely to protect them from unwanted pregnancy and infection. At the Table 5: Frequency and Location of Interaction with PeersCharacteristics BoysGirls PresentEducationTotal PresentEducation Total Level Level Class IX Class XI Class IX Class XIN 255 196 451 293 281 574A Social interaction with peers% been to the cinema* Never 40.424.533.545.450.547.9 Sometimes 42.7 41.8 42.4 43.7 40.2 42.0 Often/very often 16.1 32.7 23.3 8.9 6.4 7.7% eating out at restaurants* Never 51.4 52.6 51.9 44.0 34.9 39.5 Sometimes 34.5 33.2 33.9 35.5 45.2 40.2 Often/veryoften 12.9 12.8 12.9 19.1 16.4 17.8% attending sports clubs/ health clubs* Never 75.3 63.8 70.3 90.1 89.3 89.7 Sometimes 12.9 14.8 13.7 5.8 3.9 4.9Often/very often 9.8 18.9 13.7 1.7 2.1 1.9% going out on picnics* Never 31.0 28.6 29.9 35.5 38.4 36.9 Sometimes 45.9 31.1 39.5 50.9 46.3 48.6 Often/very often 20.4 37.2 27.7 11.9 10.3 11.1Social interaction index (0-8) 2.5 3.3 2.9 2.2 2.1 2.2B Interaction with peers in more “adult” settings % who go sometimes/often to a bar, discotheque or late night parties 25.5 32.7 28.6 17.1 12.8 15.0* Percentages do not add up to 100 because of missing responses.Table 6: Correlates of Reproductive Health Knowledge Boys Girls Beta BetaPresent education level Class IX (rc) ClassXI 1.00**** 1.61****Type of school/college Government (rc) Private 0.07 1.06****Exposure to blue films: ever watched 0.78*** 0.43Family life or sex education classes: ever attended 0.77*** 0.66**Mother’s education Up to middle school (rc) Highersecondary 0.28 0.23 College/technical 0.18 0.29Father’s occupation Manual/semiskilled/others(rc) Clerical 0.11 -0.37 Managerial/professional/business 0.20 -0.21Parental supervision Very strictly supervised (rc) Somewhat strictly supervised 0.12 0.53*** Not strictly supervised 0.12 0.05Co-resides with at least one parent -0.20 -0.70**Peer interaction Social interaction index (0-8) 0.15*** 0.11** Sometimes/often visit adult settings (bars, discotheques or late night parties) -0.36 0.46**Adjusted R2 0.18 0.36N# 394 5191 * p≤0.10, ** p≤0.05, *** p≤0.01, **** p≤0.001. 2 rc: reference category.3 # Data are missing for 57 boys and 55 girls for some explanatory variables.
REPRODUCTIVE HEALTHdecember 1, 2007 Economic & Political Weekly54same time, findings suggest that even in this relatively traditional setting, school- and college-going students do find opportunities to engage in physical and even sexual relationships: a substantial proportion of adolescent girls and boys have experienced some form of physical intimacy, and a significant minority of young boys have experienced sexual relations, usually unprotected. Despite the fact that questionnaires were anonymous and self-administered, the familiar gender gaps in reports of physical and sexual experience are evident. We acknowledge that, as other studies have observed [Jejeebhoy and Sebastian 2004], this could reflect possible under-reporting among females and/or over-reporting among males. Our findings suggest, moreover, that in this sample of students, selected individual, peer and family level factors are significantly associated with awareness of sexual and reproduc-tive health, and physical and sexual experience. Gender differ-ences in the correlates of physical and sexual behaviour are evident, although the directions of effects are generally similar. Among individual level correlates, as expected, the older the students are or the longer they are exposed to education, more the opportunities they have to form relationships and engage in both physical and sexual relations; however, this is significant only among girls, for whom Class XI may involve more freedom of movement and unsupervised time than Class IX, compared to boys who may have freedom from an earlier age. As others have noted, moreover, exposure to other risky influences like pornographic films, is significantly associated with physical or sexual experience among boys. Peer contact likewise is power-fully and positively associated with physical or sexual experi-ence; in contrast, among girls, who display considerably less peer contact than boys, the closeness of peer contacts appears only weakly associated. At the family level, maternal education appears to inversely influence the experience of physical intimacy and sexual relations among adolescents. What is notable is the extent to which parental supervision influences young people’s behaviour. While among girls, the extent of supervision is only weakly associated with physical intimacy indicators, among boys there is a strong and powerful relation-ship; findings highlight that those reporting somewhat strict and very strict parental supervision are indeed significantly less likely to report sexual relations.While many of the same variables were associated with awareness of sexual and reproductive health matters, some clear differences are noted. Adolescents who have undergone family life or sex education and have frequent peer contact are more likely to be aware of these issues than others. Among boys, exposure to blue films appears positively associated with awareness of sexual matters, raising questions about whether or not the information provided in these is always inaccurate. Family and household level factors appear, by and large, to be unrelated to awareness about sexual matters – indeed, those living away from their parents appear to be better informed than those who co-reside with their parents.Table 7: Correlates of Physical or Sexual Intimacy, among Adolescent Students: Odds Ratios from Logistic Regression Analysis Any Physical or Sexual Experience@ Kissing, Touching or Sex Ever Had Sex Ever Had Unprotected Sex Boys Girls Boys Girls Boys BoysPresent education level Class IX (rc) Class XI 1.12 CI (0.65 – 1.96) 3.01***CI (1.53 – 5.91) 1.18CI (0.65 – 2.13) 2.79**CI (1.29 – 6.07) 1.61CI (0.70 – 3.70) 1.49 CI (0.60 – 3.68)Type of school/college Government (rc) Private 1.09 CI (0.62 – 1.91) 2.09**CI (1.01 – 4.33) 0.89CI (0.49 – 1.63) 1.82CI (0.81 – 4.10) 2.34CI (0.89 – 6.16) 2.14CI (0.76 – 6.03)Exposure to blue films No (rc) Yes 6.59****CI (3.60 – 12.06) 1.79CI (0.54 – 5.95) 8.23****CI (4.48 – 15.12) 1.95CI (0.51 – 7.39) 5.37****CI (2.43 – 11.86) 3.90***CI (1.62 – 9.43)Mother’s education Up to middle school (rc) Higher secondary 0.55*CI (0.30 – 1.00) 0.62CI (0.28 – 1.38) 0.69CI (0.37 – 1.30) 0.71CI (0.29 – 1.72) 0.34**CI (0.13 – 0.89) .36**CI (0.13 – 1.02) College/technical 0.68 CI (0.33 – 1.38) 0.47*CI (0.20 – 1.08) 0.90CI (0.43 – 1.89) 0.45CI (0.17 – 1.17) 0.51CI (0.18 – 1.44) .55CI (0.18 – 1.70)Father’s occupation Low/semiskilled/others (rc) Clerical 1.25CI (0.61 – 2.57) 0.53CI (0.20 – 1.41) 1.01CI (0.47 – 2.20) 0.49CI (0.17 – 1.39) 0.95CI (0.33 – 2.78) 0.95CI (0.32 – 2.86) Managerial/professional/business 1.75*CI (0.90 – 3.38) 0.50CI (0.21 – 1.23) 1.56CI (0.78 – 3.14) 0.41*CI (0.15 – 1.08) 0.98CI(0.37 – 2.56) 0.60CI (0.21 – 1.70)Parental supervision Very strictly supervised (rc) Somewhat strictly supervised 1.72*CI (0.92 – 3.22) 1.66CI (0.72 – 3.80) 1.30CI (0.67 – 2.53) 1.35CI (0.55 – 3.32) 2.85**CI (0.97 – 8.36) 2.95*CI (0.90 – 9.67) Not strictly supervised 1.21CI (0.54 – 2.68) 1.42CI (0.51 – 3.94) 0.96CI (0.41 – 2.25) 1.13CI (0.36 – 3.52) 4.80**CI (1.46 – 15.77) 5.91***CI (1.66 – 20.97)Residence pattern Co-resides with at least one parent (rc) Co-resides with neither parent 1.41 CI (0.69 – 2.87) 0.69CI (0.28 – 1.68) 1.17CI (0.55 – 2.49) 0.63CI (0.23 – 1.71) 1.35CI (0.51 – 3.56) 2.00CI (0.68 – 5.92)Interaction with peers 1.20**CI (1.04 – 1.40) 1.11CI (0.93 – 1.33) 1.19**CI (1.02 – 1.39) 1.12CI (0.91 – 1.37) 1.30**CI (1.05 – 1.62) 1.26**CI (1.0 – 1.59)Interacts sometimes or often with peers in adult locations No (rc) Yes 1.66CI (0.91 – 3.05) 1.30CI (0.59 – 2.87) 1.37CI (0.71 – 2.63) 1.22CI (0.49 – 3.03) 2.04CI (0.83 – 5.03) 1.54 CI (0.57 – 4.19) R2 0.26 0.110.270.090.310.22N# 394 510 394 510 394 3941 rc: reference category. 2 *p≤0.10 **p≤0.05 ***p≤0.01 ****p≤0.001. 3 # Data are missing for 57 boys and 64 girls for some explanatory variables.4 @ Includes hugging, kissing, touching or sex with an opposite-sex partner. 5 CI= lower and upper limit of confidence interval.
