BEYOND THE
EMBARRASSMENT
BARRIERS
Egalitarian values contribution to sexual health
Table of content
Abstract ... 2
1.0 Introduction ... 2
2.0 Literature review ... 3
2.1 Availability, routines and education ... 3
2.2 The impact of traditional values on individual behaviours... 4
3.0 Aim of study... 5
4.0 Method ... 6
4.1 Data... 6
4.2 Variables... 7
Individual level variables ... 7
Macro-level independent variables ... 8
4.3 Strategy of analysis ... 9
4.4 Ethics ... 9
4.5 Descriptive statistics ... 9
5.0 Result ... 10
5.1 Analysing research questions ... 10
5.2 Sensitivity test ... 11 6.0 Discussion ... 12 6.1 Conclusion ... 13 Referential ... 15 Appendix ... 18 Praktikrelevans ... 18 Etiska överväganden ... 18
Validitet och reliabilitet ... 19
Abstract
The aim of this study is to investigate whether egalitarian values or structural gender equality influ-ence adolescents’ sexual activity and condom use. Current research comparing contextual differ-ences effect on sexual behaviours is lacking, and new institutional theory is used to illustrate and discuss the need for further research on contextual factors affecting sexual behaviours. Factors, most prevalent in current research, affecting individual sexual behaviours are availability, routine, education and the sexual double standard. Furthermore, egalitarian gender attitudes seem to be asso-ciated with both positive sexual ideas and experiences. These individual level factors are, in this study, argued to be factors sensitive to the institutional context. Data for the present article was gathered from multiple sources, the comparative individual level data were collected from The Global School-based Student Health Survey and World Values Survey, resulting in a final sample of 19 countries contexts analysed. The data was analysed utilizing multilevel logistic regression analysis, controlling for composition differences between countries. The results showed that egali-tarian values were significantly associated with both sexual activity and condom use. The conclu-sion was that egalitarian values seem to increase both adolescents’ sexual activity and condom use and thus contribute to sexual wellbeing. Future research should investigate how egalitarian values could be implemented in societies and contribute to sexual health beyond, what could be referred to as, the embarrassment barriers.
1.0 Introduction
Sexuality is omnipresent in societies. Sexuality is a part of being human and sexual health is generally regarded as desirable (Casique, 2019; Harden, 2014; Hull, T. et al, 2002). Sexual well-being has been shown to enhance other aspects of health, such as having a more positive body image among male college students and reducing both genders’ psychological distress (Boislard et al., 2016). At the same time, boys are generally allowed more sexual freedom, whereas norms seem to be stricter for females (Boislard et al., 2016; Marston & King, 2006). Unequal sexual norms seem to be factors affecting sexual behaviours, for instance inconsistent contraceptive use (Leung & MacDonald, 2018). Egalitarian values seem to act as a counterweight, affecting consistent condom use, increased sense of responsibility, positive sexual ideas and experiences thereof (Casique, 2019; Curtin et al., 2011; Leung & MacDonald, 2018). Egalitarian values are based on one belief: the belief that all humans are equal and should have equal rights and opportunities (Cambridge international dictionary of English, 1995). The impact of egalitarian values on behaviours has previously been studied at the individual level. Individual level values, however, exist within contextual pre-conditions, conditions which importance for sexual behaviours need to be further understood.
In order to compare nations one important element is to take both within- group and between- group differences and similarities into consideration (Boislard et al., 2016).Since studies exploring the between- group differences in sexual behaviours are lacking, it is important to analyse groups using a comparative approach. The aim of this study is to investigate the relationship between egalitarian values, gender equality, sexual activity and condom use using data from four global surveys in a cross-national sample. The sample is mainly gathered outside the western world. This study
contributes to currents research since different aspects of egalitarian values are tested to increase the understanding of equality in those contexts increasing the overall knowledge about equality.
2.0 Literature review
The World Health Organization (WHO) argues that sexual health includes mental, emotional, physical and social well-being, not only the absence of illness or sexual problems (Boislard et al., 2016; Hull, T. et al, 2002). Thus, sexual being is a central aspect of sexual health. Sexual well-being has 4 focal aspects: sexual self-esteem, sexual self-efficacy, well-being able to express sensations concerning sexual feelings or experiences and to experience sexuality in a healthy way, free of anxiety and/or pain (Casique, 2019; Harden, 2014). This includes the right to say “no”, using contraceptives and being able to control how one’s sexuality is explored. Few studies found in preparation for this article compare the impact of structures on macro-level in relation to sexual behaviours. Behaviours are generally measured and compared in the same population within different countries, this includes studies on egalitarian values and sexual behaviours. There is research indicating that people in more egalitarian societies have sex to a larger extent, and research indicating that people in egalitarian societies have sex to the same extent (Baumeister & Mendoza, 2011; Carlson & Soller, 2019). Research investigating other aspects of egalitarian values effects on sexual behaviours such as contraceptive use, safe sexual experiences and control were lacking. Even though there is a need for further research on the role of macrolevel factors in relation to sexual behaviours, comparative research traditions indicate that this is an important level for individual attitudinal structures and behavioural constraints. New institutional theory emphasises the role of social norms, rituals and roles for individual outcomes. In new institutional theory a social institution is a system of patterns describing accepted behaviour and actions an individual can use to satisfy a societal need (Vestheim, G. & Kangas, 2010; Eriksson-Zetterquist, 2012). This system of patterns describing accepted behaviour and actions need to be recurring or permanent over time to be described as a social institution, but the amount of time needed is unclear. This framework could be used to understand sexual behaviours. The macrolevel structures could
potentially indicate accepted behaviours and actions an individual is affected by which in turn could affect the sexual behaviours. In contrast to the lack of research on macro-level factors, differences and similarities between different nations and sexual behaviours, the current body of research concerning individual behaviours is rather large. Condom use among individuals are one such example and the main factors in currents research seems to be availability, routines and education.
2.1 Availability, routines and education
Earlier findings indicate that issues related to availability, routines and education affect sexual be-haviours and consistent condom use. Consistent condom use have also been associated with posi-tive sexual behaviours such as responsibility, taking care of oneself and greater self-esteem
(Casique, 2019). Overall consistent condom use is understood to have a direct impact on the sexual well-being as a positive behaviour (Casique, 2019). Studies show that positive sexual ideas and ex-periences can be measured by more consistent condom use (Casique, 2019; Curtin et al., 2011; Leung & MacDonald, 2018). Unfortunately, the condom as a sign of mistrust and promiscuity has been observed by several studies (Boislard et al., 2016; Casique, 2019; Marston & King, 2006; Van der Geugten et al., 2017) and lower rates of condom use are reported by females in comparison to males (Leung & MacDonald, 2018).
level. The availability of condoms for young people could also be of importance to establish habits. If a condom is used during first intercourse the likelihood of consistent condom use increased for both female and males (Casique, 2019; Havaei et al., 2019).
Another important factor for young peoples’ condom use found in previous research is sexual health education SHE (Bedree et al., 2020; Foshay & O’Sullivan, 2020; Sun et al., 2018). Some research also argues that the actual knowledge is not all that matters, the perceived knowledge also affect whether contraceptives is used (Ryan et al., 2007). If young people, especially but not exclusively males, believe they have high knowledge about contraception the chance of using contraceptives increases (Ryan et al., 2007). Maybe the difference in perceived knowledge can be partially ex-plained by the social importance for men to achieve penetrative sex and thus appear as experienced (Marston & King, 2006). The availability and content of SHE has something in common with con-dom availability. SHE is a factor that could be argued to be outside of the young individuals’ direct control and largely at the macro level. If SHE is taught, how and where it is taught as well as the content of SHE will be related to factors such as public health policies.
2.2 The impact of traditional values on individual behaviours
There is, as argued above, good reason to believe contextual factors such as the availability of con-doms and SHE affect individual sexual behaviours. In line with new institutional theory, these structures form normative institutions through political processes. Such normative institutions could also matter for the individual level sexual behaviour through perceived normative constraints linked to sexual behaviours. One example of perceived normative constraints affecting sexual behaviours could be traditional values. Traditional values stress traditional family values (like the nuclear-fam-ily with two heterosexual parents), significance of religion, parent-child ties and respect for author-ity. Examples of values that are commonly rejected are abortion, suicide, divorce and euthanasia (Inglehart, R., C.et al, 2014). Adherence to traditional gender roles and similar attitudes can become an obstacle to sexual health, for example by diminishing women's sexual autonomy and thereby re-ducing satisfaction, discourage contraceptive use and increasing the risk of sexual violence
(Casique, 2019; Grose et al., 2014; Pearson, 2018; Shannon et al., 2012). Inconsistent condom use and less responsibility concerning preventing and reducing unwanted pregnancy are two additional examples of risky behaviour that seem to be connected to traditional gender roles (Casique, 2019). Furthermore, traditional gender roles can make males’ unpredictable in terms of methods for sexual protection and by having multiple partners since the males sexual desires can be seen as an uncon-trollable force driven by biology (Casique, 2019; Grose et al., 2014; Jovanovic & Williams, 2018; Pearson, 2018). This can be further reinforced by females becoming subordinated to men in order to appear as a decent woman which leads to a restricted sexuality, to be expressed within the borders of a stable and loving relationship (Casique, 2019; Pearson, 2018). Finally, the ability to access and utilize sexual-risk knowledge, advocate their sexuality and condom self-efficacy seems to be
know nothing about contraceptives in order to make sure their reputation does not become dam-aged (Marston & King, 2006). Egalitarian gender attitudes could act as a counterweight; egalitarian gender attitudes seem to be associated with both positive sexual ideas and experiences, as shown by studies conducted in Bolivia, The United States and Ecuador (Casique, 2019). It seems clear that values impact how individuals behave, even though there are mechanisms controlling in what way, yet to be explained. The margin for manoeuvre seems affected by traditional values, leading to dif-ferences in the margin for manoeuvre in sexual situations. Thus, a possible consequence of tradi-tional values is inequality based on a person's gender.
Inequality based on a person's gender as part of a macro-structure could hypothetically make women assume a passive role, men being the condom provider and in charge of making a decision whether a condom should be used (Leung & MacDonald, 2018; van der Geugten et al., 2017; Vasilenko et al., 2015). Hostile attitudes toward women seem to influence men to resist condom use in heterosexual intercourse (Leung & MacDonald, 2018). Several studies propose that there are what could be referred to as “embarrassment barriers” for women, barriers inhibiting caring, access-ing and purchasaccess-ing of condoms (Bell, 2009; Leung & MacDonald, 2018; Marston & Kaccess-ing, 2006). Embarrassment seems to be connected to seeking sex, according to beliefs held by females and males (Leung & MacDonald, 2018). Thus, women have been observed to express fewer intentions to carry condoms (Jellema et al., 2013; Leung & MacDonald, 2018). Supporting the embarrassment barriers theory, some research shows that males’ attitudes are more important than females’ atti-tudes. Positive contraceptive attitudes and increasing desire to use condom among male adolescence seem to affect female adolescents and make them use condom to a larger extent (Vasilenko et al., 2015). A strong indicator for females’ contraceptive and condom use is whether the female believes becoming pregnant would be a disaster at this point in her life. Believing pregnancy to be a disaster is associated with higher likelihood of using contraceptives (Ryan et al., 2007). However, it is im-portant to note that preventing pregnancy, generally, is considered to be females responsibility (Marston & King, 2006)
In sum, the use of contraceptives is influenced by social factors, something that researchers have identified beyond the simplistic reasons such as “ignorance” and “barriers to access contraception” (Marston & King, 2006). The WHO clearly states that it is a sexual right and a part of healthy sexu-ality to be able to experience and express sexusexu-ality - including the right to pleasure. Sexual health also enhances communication, personality and love (Hull, T. et al, 2002). There is also reason to consider the sex-positive framework, thinking of sexuality as something that is natural and part of life, leading to increased health overall (Boislard et al., 2016; Harden, 2014; Hull, T. et al, 2002). Experience and exploring sexuality are something that can enhance the individual’s general health, whereas unwanted pregnancy and disease generally can affect health negatively. Thus, safe inter-course with contraceptives is most likely to generate health
3.0 Aim of study
The aim of this study is to investigate whether egalitarian values influence adolescents’ sexual ac-tivity and condom use, increasing the possibility for sexual health. To fulfil the aim of this study four research questions are asked. Those research questions are:
1. Is sexual activity among adolescents influenced by egalitarian values at the country level? 2. Is the effect of egalitarian values at the country level larger, smaller or the same for young
females’ and males’ sexual activity?
4. Is the effect of egalitarian values at the country level larger, smaller or the same for both genders’ condom use?
The second and the fourth research question are elaborations on the first and the third and focuses on whether the effect of egalitarian values at the country level are the same for both genders’ sexual activity and condom use. There are reasons to believe males and females sexual behaviours differ from each other in many areas such as sexual freedom and considering the effect of the double standard (Boislard et al., 2016; Marston & King, 2006). Furthermore, there are reasons to believe that different factors play an important role for females’ and males’ condom use. The responsibility for preventing pregnancy, the double standard and attitudes regarding contraceptives are a few ex-amples (Boislard et al., 2016; Casique, 2019; Leung & MacDonald, 2018; Manlove et al., 2004; Marston & King, 2006; Ryan et al., 2007). Thus, a lack of egalitarian values seems to increase the differences between genders. The gender most affected by these differences could potentially be af-fected to a larger extent by increased egalitarian values. Investigating genders separately make it possible to evaluate if egalitarian values are a better predictor for females’ or males’ sexual health. Investigating whether egalitarian values enhance activity and condom use is a way to investigate these relationships within the sex-positive framework and acknowledging WHO:s view on sexual health.
4.0 Method
4.1 Data
The aim of this article is to investigate the role of country level factors (egalitarian values) on individual sexual behaviour. This requires access to individual level data containing comparable information on young peoples’ sexual behaviours from multiple country contexts. The Global School-based Student Health Survey (GSHS) offers a unique data source providing a harmonized comparative individual level data set on students’ health behaviours. GSHS was initiated by the WHO and the Centers for Disease Control and Prevention (CDC) in collaboration with UNAIDS, UNESCO, and UNICEF (Centers for Disease Control and Prevention, 2013). GSHSs are conducted in approximately 110 countries, fact sheets and reports available for 103 of them, with the majority being low- or middle-income areas. The GSHS questionnaire is a self-reported survey conducted primarily among students between the ages of 13–17 (some students being a couple years older or younger). The samples are selected nationally, regionally or sometimes both (Assarsson et al., 2018; Centers for Disease Control and Prevention, 2013).
To measure contextual country level egalitarian values the emancipatual values index from World Values Survey (WVS) and the Gender Inequality Index (GII) created by United Nations
would be higher. GII is generally regarded as an accurate measurement of structural differences between genders (Klugman, 2010).
Countries present in both GSHS, GII and WVS were matched by year. In most cases the surveys were not conducted in the same year. Countries were, however, matched as closely as possible. Some countries were excluded since the time difference in data collection between GSHS and WVS were judged to be too big. The final sample only contains countries where the data from GSHS, WVS and GII were collected within a time span of 6 years. During the matching process it was possible to match some countries from GSHS, GII and WVS twice. During the cleaning of the data it could be observed that the indexes, the macro variables and variables measuring sexual
behaviours changed substantially within the same country in between the years separating matching I and II. Since the aim of this study was to investigate the effect of egalitarian values on sexual behaviours matching I and matching II were included to not lose valuable information.
Furthermore, to observe differences between groups, which in this case consists of different countries, it was important to have enough groups in the final sample to make the analyses more reliable. Sensitivity tests were done to evaluate matching I and matching II.
In sum the countries present in the final sample were chosen for several reasons. Firstly, they were present in both GSHS, WVS and GII. Secondly, to ensure the quality of the analyses and have enough groups to be able to observe differences between groups all possible matches were included. Thirdly, to contribute to the understanding of egalitarian values effect on sexual behaviours outside the western world. Finally, alternative surveys investigating sexual behaviours in a comparable way was not found within the timeframe of this study, resulting in no additional countries on top of the matching described above. The final sample includes 15 different countries, with 4 being included twice for a total of 19 country contexts analysed. The total amount of individuals analysed are 238 980. The country being represented by the most individuals is Argentina II containing 56 736 individuals and the country being represented by least individuals is Argentina I containing 3 958 individuals.
4.2 Variables
Individual level variables
All the individual level variables were collected from GSHS. In line with the aim of this article two main dependent variables measuring sexual behaviours were used. To measure sexual activity, the question “have you ever had sexual intercourse?” was selected with the possible answers of yes and no. To measure condom use, the question “The last time you had sexual intercourse, did you or your partner use a condom?” was selected with the possible answers yes, no and no intercourse. Small variations existed since the word condom was translated to multiple languages and surveys utilizing verified synonyms for condom was included. It would have been good if multiple contraceptive practices could have been studied. A few of the survey waves had multiple questions focusing on different contraceptive methods, but the questions were asked in several different ways and deemed to be not comparable. To ensure the quality of the analyses and keep a high consistency these where excluded.
To control for composition differences between countries three additional individual level
indicate different material assets which have been shown to effect the desire for democracy and freedom of choice (Inglehart, R., C. et al, 2014). Survival tend to be prioritized over democratic rights. Age was measured with a categorical scale: 11 years or younger, 12, 13, 14, 15 and finally 16 years or older. Since the two youngest categories were small in comparison to the other
categories, they were combined into the category 12 years or younger. Gender was measured with the following question “What is your sex?” and the possible answers male or female. Hunger due to lack of food at home were measured with the question “During the past 30 days, how often did you go hungry because there was not enough food in your home?” and the possible answers were sorted as an categorical scale ranging from “never” to “always” with 5 alternatives. During the cleaning of the data this scale were changed from categorical to numeric scale. This was tested during the sensitivity tests and did not change the overall results but made interpretation of the results clearer.
Macro-level independent variables
In order measure structural differences between males and females GII were chosen. GII measures loss in potential human development between males’ and females’ procurement (Gunkelman, 2002). GII covers the following factors: maternal mortality rate, adolescent birth rate, share of
parliamentary seats held by each sex, population with at least some secondary education (SE) and labour force participation rate (Gunkelman, 2002). The index ranges from 0 to 1. If women and men fare equally in those elements the GII score is 1 and thus GII 0 indicates complete inequality, with one gender doing poorly in all measured elements (Gunkelman, 2002). To adjust the scale
considering only parts of the scale being represented in the final sample and make interpretation of the analyses easier GII scores were multiplied with 100. Other indexes were considered such as The Inequality-adjusted Human Development Index (IHDI), the Human Development Index (HDI) and Gender Equality Measurement (GEM). After consideration IHDI and HDI were disqualified since important measurements of the political dimensions were lacking, in combination with having strong similarities with GII (Klugman & Choi, 2011). GEM was replaced by GII in 2010 and GII is regarded as more accurate and correct in relation to some of the drawbacks with GEM (Klugman, 2010).
GII measures structural differences between genders, but egalitarian values could possibly have their own impact within the structural conditions. A high level of equality measured by one measurement is not a guarantee for every part of a society to be equal. Within new institutional theory institutional factors are known to evolve and change. Exactly how change happens is unclear. Thus, social institutions are complex and GII is one way of measuring the structural spectrum of a social institution, giving an over-view or a frame for the context examined. Attitudes, norms and social values could be viewed as another aspect of social institutions, offering a chance to observe movements within the context. To include attitudes and social values that potentially could shape norms emancipatual values index from WVS was included. Furthermore, to advance the
empowerment of women emancipative values have been proven to be the single most important factor (Welzel, 2013). The emancipative values index measure freedom of choice and equal
opportunities such as lifestyle liberties, gender equality and personal autonomy (Inglehart, R., C. et al, 2014). The index range are from 0 to maximum 1.0, low score indicating weaker values, high score indicating stronger values ( Inglehart, R., C. et al, 2014; Welzel, 2013). This index was also multiplied with 100 for the same reasons as GII. During the analyses, the scores for GII and emancipatual values index seemed to have different effects. In addition to this, some countries had low scores on GII and high scores on the emancipatual index. Maybe these scores are signs of movements within the social institution within those countries. In this study GII and emancipatual index are considered to be different and important aspects of gender equality.
GII or emancipatual values index. GDP/ppp measure gross domestic product based on purchasing power parities (World bank, 2019). This measurement is usually used to measure standards of living between countries. To simplify the understanding of GDP/ppp all scores were divided by 1000.
4.3 Strategy of analysis
The data used for this analysis were pooled from many countries and different sources. Due to this it has a nested structure with individuals (level 1) nested in different countries (level 2).
Consequently, each observation or individual could possibly be affected by the nested structure, rendering ordinary least square regression analysis sub-optimal (Högberg et al., 2019). Since the data are nested individual observations are not independent of each other. Multilevel regression analysis allows observations of how level 1 variables and level 2 variables affect the dependent variable. These steps allow for analysis of how clusters are affected possible, since level 1 independent variables and the dependent variable can be conditional on the value of level 2 variables (country in this case).
Within the multilevel framework logistic regression was used as the dependent variable was binary. Several metrics can be used to interpret results from logistic regression models, for this article log odds coefficients (b) were chosen. The predicted log odds should be interpreted as the log of the odds between the observed category and the reference category within the dependent variable in relation to the studied event (individual in x country had sex/used condom) when all other individual predictors are set to zero and the random effect is held consistent at zero (Heck et al., 2012).
Level 1 variables used for the analyses were gender, age and level of hunger. Level 2 variables used for the analyses were GDP/ppp, GII and emancipative values index. To observe the best fit-model and compare the models log likelihood was utilized combined with Akaike information criterion (AIC). AIC provides a penalty based on the complexity of the model, multiplying the number of parameters (p) by two and adding this result to the error term also called deviance statistic (Heck et al., 2014). If the sample size is above 100 individuals AIC has been noted to work well to select the best-fit model (Heck et al., 2014). All calculations were done using SPSS 2.6 GENLIN MIXED. When selecting the test of fixed effects and coefficients the setting “use robust estimation to handle violations of model assumptions” was used. The robust standard errors model is generally more conservative regarding possible variations from normality (Heck et al., 2012). Since
underestimating the standard errors could lead to more findings of significance, the robust model uses more caution when calculating the standard errors (Heck et al., 2012). Sensitivity tests were done to evaluate this choice.
4.4 Ethics
Data that were used in this study were publicly available online, and for that reason it was not necessary to apply for ethical approval.
4.5 Descriptive statistics
was also similar, resulting in a mean of 48.51 % males in the final sample. The mean for the variable “Went hungry” ranged from 1.3 to 2.93, Zambia having the highest rate at 2.93 (corresponding to sometimes went hungry) and people in Chile being the least hungry (1.3
corresponding to never went hungry). GDP/ppp had a sample mean of 13.01 thousand dollars, with a range of 0.94 thousand dollars to 30.8 thousand dollars. Emancipative values scores ranged from 29 to 54, with a mean of 40.63. GII scores ranged from 30 to 62, with a mean of 41.55.
TABLE 1 HERE
In order to answer the research questions, the analyses use two dependent variables. One measuring sexual activity and one measuring condom use. Table 2 presents the frequency of intercourse for each country and the frequency of condom use among those who had intercourse for each country. The proportion of adolescents who had experienced intercourse within each country ranged from 0.6 % to 43.4 %, with a mean of 21.3 % in the final sample. Argentina II and Zambia had the highest scores and Indonesia, Thailand I and Malaysia had the lowest scores. The adolescents who had experienced intercourse and used condom ranged from 30.9 % to 81.6 % within each country, with a mean of 59.6 % in the final sample. Uruguay II and North Macedonia had the highest frequency of condom use and Malaysia, Zimbabwe and Chile I the lowest. The substantial differences between countries indicate systematic differences that motivate further analysis. TABLE 2 HERE
5.0 Result
5.1 Analysing research questions
Tables 3 and 4 present the analysis of the cross-country differences. Table 3 presents the multilevel logistic regression models for research question 1 and 2 where sexual activity is investigated and Table 4 presents the multilevel logistic regressions for research question 3 and 4 where condom use is investigated. Both tables share the same structure. For each research question there are 5 models. The first model is empty in order to observe if there is indications of a contextual phenomenon (Merlo et al., 2006). In other words, whether it seems to be a significant difference between countries based on the dependent variable. Thereafter the second model includes all level 1 variables. The second model is utilized in order to observe the differences between countries explained by the composition of individuals within each country (Merlo et al., 2006). The third, fourth and fifth models include level 2 variables where GDP/ppp (which is treated as a confounder) is constant and GII and emancipatual values are calculated one by one and then together. The level 2 variables were utilized one by one and then together to handle potential multicollinearity. In each table the result is presented in 3 parts. Table 3A and 4A presents the results for both genders
together, Tables 3B and 4B only males and Tables 3C and 4C only females.
Starting with sexual activity in Table 3, there seems to be some difference between countries due to contextual factors since model 1 were significant. Turning to model 2 where the individual level variables are entered, the strongest individual level factors affecting sexual activity were age and gender. Sexual activity significantly increased with age and females were significantly less likely to be sexually active. Age were a categorical scale and age 12 years or younger were used as a
increase on the scale increased the odds of having intercourse by approximately 12.7 % (b = 0.12). Turning to model 3-5 and the analyse of level 2 variables, much of the individual level variables did not change. The effect of age was almost the same and only minute changes could be observed for the age categories 14 years old and 13 years old. Gender and hunger had the same effect.
Continuing with model 3-5 and looking at the second level variables neither GDP/ppp or GII was significantly associated with sexual activity. Emancipative values, on the other hand, were found to have a significant positive association with sexual activity in both model 4 and 5 and increased the odds of sexual activity with approximately 7.4 % for each step of the scale (model 4A b = 0.07; model 5A b = 0.08). Analysing boys and girls separately in Tables 3B and 3C there was almost no difference in the effect of the second level variables.
TABLE 3 HERE
In the next table, table 4A, the probability of using condoms among those youths who had sexual intercourse is investigated. The overall result is that females have a significantly lower probability to use condoms in comparison with males, hunger significantly decrease the probability to use condoms and emancipatual values significantly increased the probability of a condom being used. Beginning with model 1 there seems to be some difference between countries due to contextual factors since the log odds coefficient were significant. Turning to model 2 where the individual level variables are entered, the factors significantly affecting condom use were gender and level of hunger. To analyse gender females were compared with males. Females had significantly decreased odds of using condom, with approximately 21.8 % (b = -0.25). Hunger were significant and a 1point increase on the scale decreased the odds of using condoms by approximately 8.7 % (b = -0.09). Turning to model 3-5 and the analyse of level 2 variables, the individual level variables did not change.
Continuing with model 3-5 and looking at the second level variables neither GDP/ppp or GII was significantly associated with condom use. Emancipatual values, on the other hand, were found to have a significant association with condom use and increased the odds to use condoms with
approximately 4.3 % for each step of the scale in model 4 and 5 (b = 0.04). Analysing boys and girls separately in Tables 4B and 4C there was almost no difference in the effect of the second level variables.
TABLE 4 HERE
5.2 Sensitivity test
than a slight underestimation of effects. Therefore, the robust model was used for the presented main analysis.
6.0 Discussion
This article has investigated whether egalitarian values focusing on structural gender inequality af-fect sexual behaviours in terms of sexual activity and if, during intercourse, a condom was used. Sexual intercourse was significantly associated with age, hunger, gender and emancipatual values index. Higher age, hunger and emancipatual values index increased the probability of sexual inter-course. Females were less likely to be sexually active, compared to males. Condom use was signifi-cantly associated with hunger, gender and emancipatual values index. Hunger decreased the proba-bility to use condom and females had a lower probaproba-bility than males to use a condom. Emancipatual values increased the probability of a condom being used. GII was never significant in any model or analyses for either of the sexual behaviour variables and the effects were similar for males and fe-males when the genders were analysed separately. The similarity between fe-males and fefe-males could possibly be explained by considering the context. Both genders are part of the same social structure and thus are affected by that structure to the same extent even though the consequences of the social structure differ. One consequence is that females have a smaller probability to use condoms in com-parison with males. But both genders would probably use condom to a larger extent if egalitarian values were strengthened. The effect of egalitarian values at country level seem to be the same for females’ and males’ activity and condom use.
GII and emancipatual values index measure different dimensions of equality and egalitarian values, thus the conclusion holds even though GII never was significantly associated with sexual behav-iours. Maybe this could be partly explained by considering institutional theory. Emancipatual values seem to be more closely linked to sexual freedom, social norms and attitudes. GII measures struc-tural differences further from the everyday life. GII are usually used to compare bigger social con-texts, as a macro variable, since it measures seats in parliament for example. Social norms and atti-tudes exist on all levels: micro, meso and macrolevel. Emancipatual values could capture the move-ments within the social institution. Returning to the aim of the article the overall conclusion is that egalitarian values are associated with more sexual activity among youth as well as a greater use of condoms when being sexually active, increasing the overall sexual health.
These results are in line with previous research where several studies (Marston & King, 2006; UNESCO, 2017; Vasilenko et al., 2015) illustrate how gender inequalities in a society could limit the access and use of contraceptives. It is possible that the lower rates of females carrying, using and suggesting the use of condoms (Leung & MacDonald, 2018; Marston & King, 2006) could partly be explained by lack of equality and egalitarian values. It has been noted by multiple studies (Berrocal et al., 2019; Leung & MacDonald, 2018; Marston & King, 2006; Pearson, 2018;
UNESCO, 2017; Vasilenko et al., 2015) that signs of females sexual activity, carrying or use of contraceptives, could lead to stigma and/or embarrassment. UNESCO reinforce the importance of gender equality and discussion of power within relationships as a crucial components of effective comprehensive sexuality education CSE (UNESCO, 2017). Furthermore, including gender in CSE renders it up to five times more effective in terms of its ability to significantly decrease sexual transmitted diseases and pregnancy (Haberland, 2015).
2012). This reflection should, however, not be taken as an indication that change is impossible. In-stitutional factors evolve and change, and change is also possible under inIn-stitutional constraints. One such example is education. Education have the potential to enhance emancipatual values (Welzel, 2012). How would education designed to enhance emancipatual values affect adolescents in a social institution not favouring emancipatual values? Research from other fields suggest that the emancipatual values would increase (Welzel, 2012). For future research, an important aspect would be to investigate that effect by asking how it is possible to live according to egalitarian val-ues in a social institution that does not favour egalitarian valval-ues. Furthermore, it is of importance to understand whether an individual favouring values that is not favoured by the social institution will have a harder time to act upon those values. The first focal point would be to which extent living in a way not favoured by the social institution an individual is part of is possible and the second focal point would be how hard that would be.
This study had some limitations that need to be considered. All the data used in this article was pooled from many sources. This means that the sampling process, wording and the structure of the survey were definite. This created some challenges during the cleaning of the data and some coun-tries were disqualified due to difficulties interpreting what the variables measured. Condom use is related to sexual health, even though some of the respondents might be using additional forms of contraceptives and some of the students who do not use condom might use other forms of contra-ceptives. Much of the literature review was done with this limitation in mind. The definite sampling process chosen by GSHS, WVS and GII also limited witch countries could be included within the timeframe of this study. It proved to be a challenge to find enough countries that could be matched (GSHS, WVS and GII) in a satisfactory way. The analyses would probably be more robust if more countries could have been matched. The countries selected for this study were the countries availa-ble, no random selection was possible. This leads to one important note; this result is not representa-tive for countries within the western world or middle eastern context since none of these contexts where represented. On the other hand, both GII score and emancipatual index scores varied in a sat-isfactory way and the representation of countries outside those two contexts was robust and repre-sent underreprerepre-sented context in the generally western centric research within the field. This article through this provide evidence expanding the contextual reach of previous research.
On additional limitation is important to have in mind: GSHS is a school-based survey. Conse-quently, the results presented in this article are valid for adolescents going to school, not all adoles-cents. In some countries where fewer adolescents go to school, this may have a larger impact. since critical thinking, using logic and asking questions generally are part of the education systems educa-tion have been showed to have the potential of increasing emancipatual values (Welzel, 2012). It is possible that the associations presented in this article would be stronger if adolescents not attending a school would be included.
On the other hand, the data used in this article are regularly used by established researches. WVS have been used in more than 400 publications and GSHS is well established and controlled by sev-eral international organs including WHO (Centers for Disease Control and Prevention, 2013; Inglehart, R., C. et al, 2014). The other indexes (GII, GDP/ppp) used in this article are also widely used by others and generally regarded as accurate measurements (Klugman, 2010). Thus, this study contributes to the understanding of egalitarian values effect on sexual behaviours, the importance of context and understanding of egalitarian values outside the western world.
6.1 Conclusion
between countries were limited within the current research. Furthermore, the majority of the countries included in this article were from outside the western world, were research investigating sexual behaviours are lacking (Boislard et al., 2016). Finally, there are some paths future research could explore. Firstly, since egalitarian values contribute to sexual health and thus health overall, it would be interesting to know how to enhance egalitarian values. Secondly, it would be interesting to research movements within a social institution, since they are known to evolve and change. That would require data that makes analyses of the connections between individual values and
behaviours on one hand and macro structures and groups behaviours on the other hand possible. Thirdly, the difference between structural equality and emancipative values would be of interest for future research. GII were never significantly associated with sexual activity or condom use whereas emancipative values were. It would seem like they are related but some crucial difference sets them apart. It would be interesting to understand those differences and how those could be used to
Referential
Andersson, Ingemar. (2006) Epidemiologi för hälsovetare – en introduktion. Studentlitteratur AB, Lund.
Assarsson, R., Petersen, S., Högberg, B., Strandh, M., & Johansson, K. (2018). Gender inequality and adolescent suicide ideation across Africa, Asia, the South Pacific and Latin America–a cross-sectional study based on the Global School Health Survey (GSHS). Global Health
Action, 11(sup3). https://doi.org/10.1080/16549716.2019.1663619
Bedree, H., Moller-Mullen, M., Rose, E., & Flanders, C. E. (2020). Sexual Well-Being Among College Students: A Qualitative Study. Sexuality and Culture, 24(1), 140–156.
https://doi.org/10.1007/s12119-019-09631-5
Bell, J. (2009). Why embarrassment inhibits the acquisition and use of condoms: A qualitative approach to understanding risky sexual behaviour. Journal of Adolescence, 32(2), 379–391. https://doi.org/10.1016/j.adolescence.2008.01.002
Berrocal, M. D. C. G., Vallejo-Medina, P., Moyano, N., & Sierra, J. C. (2019). Sexual double standard: A psychometric study from a macropsychological perspective among the spanish heterosexual population. Frontiers in Psychology, 10(AUG), 1–14.
https://doi.org/10.3389/fpsyg.2019.01869
Boislard, M. A., Bongardt, D. Van De, & Blais, M. (2016). Sexuality (and lack thereof) in adolescence and early adulthood: A review of the literature. In Behavioral Sciences (Vol. 6, Issue 1). https://doi.org/10.3390/bs6010008
Cambridge international dictionary of English. (1995). EGALITARIAN. Cambridge: Cambridge University Press. https://dictionary.cambridge.org/dictionary/english/egalitarian
Casique, I. (2019). Gender Differences in the Sexual Well-Being of Mexican Adolescents.
International Journal of Sexual Health, 31(1), 1–16.
https://doi.org/10.1080/19317611.2018.1561587
Centers for Disease Control and Prevention, W. H. O. (2013). Global Student Health Survey
(GSHS) Data User’s Guide. January, 1–16.
http://www.who.int/chp/gshs/methodology/en/index.html%5Cnhttp://www.cdc.gov/gshs/questi onnaire/index.ht
Curtin, N., Ward, L. M., Merriwether, A., & Caruthers, A. (2011). Femininity ideology and sexual health in young women: A focus on sexual knowledge, embodiment, and agency. International
Journal of Sexual Health, 23(1), 48–62. https://doi.org/10.1080/19317611.2010.524694
Foshay, J. E., & O’Sullivan, L. F. (2020). Home-based sex communication, school coverage of sex, and problems in sexual functioning among adolescents. The Canadian Journal of Human
Sexuality, 29(1), e20190025. https://doi.org/10.3138/cjhs.2019-0025
Grose, R. G., Grabe, S., & Kohfeldt, D. (2014). Sexual education, gender ideology, and youth sexual empowerment. Journal of Sex Research, 51(7), 742–753.
https://doi.org/10.1080/00224499.2013.809511
Gunkelman, J. (2002). Technical notes. Journal of Neurotherapy, 6(2), 45. https://doi.org/10.1300/J184v06n02_06
Haberland, N. A. (2015). The case for addressing gender and power in sexuality and hiv education: A comprehensive review of evaluation studies. International Perspectives on Sexual and
Reproductive Health, 41(1), 31–42. https://doi.org/10.1363/4103115
Harden, K. P. (2014). A Sex-Positive Framework for Research on Adolescent Sexuality.
Perspectives on Psychological Science, 9(5), 455–469.
https://doi.org/10.1177/1745691614535934
Havaei, F., Doull, M., & Saewyc, E. (2019). A trend analysis of sexual health behaviours of a national sample of Canadian adolescents using HBSC data from 2002-2014. Canadian Journal
of Human Sexuality, 28(1), 17–25. https://doi.org/10.3138/cjhs.2018-0035
Comparative Sociology, 60(4), 269–291. https://doi.org/10.1177/0020715219874386
Hull Terence, Trojan Ondrej ,Corona Esther. (2002). Defining sexual health Sexual health document series. Report of a Technical Consultation on Sexual Health, January, 1–35.
Inglehart, R., C. Haerpfer, A. Moreno, C. Welzel, K. Kizilova, J. Diez-Medrano, M. Lagos, P. Norris, E. P. & B. P. et al. (eds. . (2014). WVS Database. World Values Survey: Round Six - Country-Pooled Datafile. http://www.worldvaluessurvey.org/WVSContents.jsp
Jellema, I. J., Abraham, C., Schaalma, H. P., Gebhardt, W. A., & Van Empelen, P. (2013). Predicting having condoms available among adolescents: The role of personal norm and enjoyment.
British Journal of Health Psychology, 18(2), 453–468.
https://doi.org/10.1111/j.2044-8287.2012.02088.x
Jovanovic, J., & Williams, J. C. (2018). Gender, Sexual Agency, and Friends with Benefits Relationships. Sexuality and Culture, 22(2), 555–576. https://doi.org/10.1007/s12119-017-9483-1
Klugman, J. (2010). Human Development Report 2010. In Population and Development Review (Vol. 21, Issue 4). https://doi.org/10.2307/2137795
Klugman, J., & Choi, H. (2011). Human Development Research Paper 2011/01 The HDI 2010: New Controversies, Old Critiques. Human Development.
Leung, P. C., & MacDonald, T. K. (2018). Attitudes toward condom carriers: The role of gender.
Canadian Journal of Human Sexuality, 27(3), 215–225.
https://doi.org/10.3138/cjhs.2018-0003
Manlove, J., Ryan, S., & Franzetta, K. (2004). Contraceptive Use and Consistency in U.S. Teenagers’ Most Recent Sexual Relationships. Perspectives on Sexual and Reproductive
Health, 36(6), 265–275. https://doi.org/10.1363/3626504
Marston, C., & King, E. (2006). Factors that shape young people’s sexual behaviour: a systematic review. Lancet, 368(9547), 1581–1586. https://doi.org/10.1016/S0140-6736(06)69662-1 Merlo, J., Chaix, B., Ohlsson, H., Beckman, A., Johnell, K., Hjerpe, P., Råstam, L., & Larsen, K.
(2006). A brief conceptual tutorial of multilevel analysis in social epidemiology: Using measures of clustering in multilevel logistic regression to investigate contextual phenomena.
Journal of Epidemiology and Community Health, 60(4), 290–297.
https://doi.org/10.1136/jech.2004.029454
Pearson, J. (2018). High School Context, Heterosexual Scripts, and Young Women’s Sexual Development. Journal of Youth and Adolescence, 47(7), 1469–1485.
https://doi.org/10.1007/s10964-018-0863-0
Ryan, S., Franzetta, K., & Manlove, J. (2007). Knowledge, perceptions, and motivations for contraception: Influence on teens’ contraceptive consistency. Youth and Society, 39(2), 182– 208. https://doi.org/10.1177/0044118X06296907
Shannon, K., Leiter, K., Phaladze, N., Hlanze, Z., Tsai, A. C., Heisler, M., Iacopino, V., & Weiser, S. D. (2012). Gender inequity norms are associated with increased male-perpetrated rape and sexual risks for HIV infection in Botswana and Swaziland. PLoS ONE, 7(1).
https://doi.org/10.1371/journal.pone.0028739
Sun, W. H., Miu, H. Y. H., Wong, C. K. H., Tucker, J. D., & Wong, W. C. W. (2018). Assessing Participation and Effectiveness of the Peer-Led Approach in Youth Sexual Health Education: Systematic Review and Meta-Analysis in More Developed Countries. Journal of Sex Research,
55(1), 31–44. https://doi.org/10.1080/00224499.2016.1247779
UNESCO. (2017). Early and Unintended Pregnancy- Recommendations for the education secto.
UNESCO, 12.
van der Geugten, J., van Meijel, B., den Uyl, M. H. G., & de Vries, N. K. (2017). Protected or Unprotected Sex: The Conceptions and Attitudes of the Youth in Bolgatanga Municipality, Ghana. Sexuality and Culture, 21(4), 1040–1061. https://doi.org/10.1007/s12119-017-9432-z Vasilenko, S. A., Kreager, D. A., & Lefkowitz, E. S. (2015). Gender, contraceptive attitudes, and
condom use in adolescent romantic relationships: A dyadic approach. Journal of Research on
Vestheim, G. & Kangas, A. (2010). Institutionalism, cultural institutions and cultural policy in the Nordic countries. Nordisk Kulturpolitisk Tidsskrift, 13, 267–284.
Vetenskapsrådet. (2002). Forskningsetiska principer. 1–17. http://www.codex.vr.se/texts/HSFR.pdf Welzel. (2013). Freedom Rising Online appendix (pp. 1–146).
https://www.cambridge.org/files/8613/8054/8416/FreedomRising_OA.pdf
Welzel, C. (2012). Freedom rising: Human empowerment and the quest for emancipation. In
Freedom Rising: Human Empowerment and the Quest for Emancipation (Issue December
2013). https://doi.org/10.1017/CBO9781139540919
Appendix
Praktikrelevans
Att möta människor som befinner sig i olika skeden i livet är en vardag för många socialarbetare. Det är en praktik där socialarbetare kommer i kontakt med stora delar av en människas livspektrum. En betydande del av en människas livsspektrum är sexualitet (Knutagård, 2016). Ofta krävs en viss nivå av tillit och samarbete innan brukaren är bekväm och vill dela med sig av erfarenheter kring sexualitet eller berätta om områden som kan vara problematiska i relation till sexualitet (Knutagård, 2016). Detta kan ske inom ramarna för skolverksamhet, utredningar, fritidsverksamhet, sjukvård och ungdomsmottagningar för att nämna några. Det händer att människor inte vågar dela med sig av sina erfarenheter för att den som lyssnar bedöms vara för okunnig eller otillräcklig för att orka lyssna på den berättelse som brukaren bär på (Knutagård, 2016). Detta gör att ökad kunskap om sexualitet och sexuell hälsa är av vikt för de flesta socialarbetare. Vi kan inte veta i förväg om det är i samtal med kuratorn på skolan eller fritidsledaren på ungdomsgården som socialarbetaren i
Argentina kan möta en människa som vill eller behöver berätta om olika aspekter av sin sexualitet. Således är att förstå vad som kan främja sexuell hälsa relevant för de allra flesta socialarbetare. Viljan att undersöka social kontext väcktes av tanken att alla kan göra sin del för att främja hälsa då alla människor befinner sig i en social kontext. Och just frigörelse, emancipation, kan interageras i de flesta av socialarbetarens kontexter. Följaktligen är det av intresse för det sociala arbetets praktik huruvida att främja emancipationella värderingar kan öka den sexuella hälsan. Utfallet talar även sitt tydliga språk: vill vi främja sexuell hälsa är att stärka emancipationella värderingar en god idé. Framtida forskning borde utreda hur detta kan anpassas till det sociala arbetets praktik i Sverige. Hur kan socialarbetaren på försäkringskassa, sjukhuset, grundskolan eller fritidsgården arbeta för att stärka emancipationella värderingar?
Etiska överväganden
Det finns 4 grundläggande krav för forskning inom humaniora. Dessa är samtyckeskravet, informationskravet, nyttjandekravet och konfidentialitetskravet (Vetenskapsrådet, 2002).
Samtyckeskravet innebär att informanten ska samtycka till och har rätt att påverka sin medverkan i undersökningen. Om personen är minderårig kan samtycke behöva hämtas från vårdnadshavare (Vetenskapsrådet, 2002). Under arbetet med denna artikel användes 2 huvudsakliga datakällor: GSHS och WVS. Båda dessa har liknande upplägg: att samla in data som delas med
forskningsgrupper i världen. Båda dessa kräver samtycke från sina informanter och båda är öppna med att det är en källa som andra forskare använder för att hämta data. Detta innebär följande: samtycker dom som deltar till att ge sin information till någon av dessa samtycker informanten också till att tredje part använder data. Samma princip gäller de index som har använts. Detta gör även att informationskravet täcks per samma resonemang. Informationskravet innebär att
informanten ska informeras om forskningens syfte, upplägg och framtida användningsområden på ett sätt som är begripligt (Vetenskapsrådet, 2002).
Nyttjandekravet innebär att uppgifter från informanter enbart får användas för forskningens ändamål (Vetenskapsrådet, 2002). Detta är något som betonas på de hemsidor som använd data finns. Jag gör bedömningen att detta krav har uppfyllts då denna undersökning endast kommer användas i forskningssyfte. Konfidentialitetskravet innebär att informanten ska erbjudas största möjliga konfidentialitet och personuppgifter förvaras så att obehöriga inte har tillgång till dessa (Vetenskapsrådet, 2002). Detta innebär även att ingen enskild deltagare ska kunna identifiera sig själv eller andra när resultatet redovisas. Detta uppfylls då all data har avidentifierats av
omöjligt för utomstående att spåra vem som kan ha svarat vad.
Validitet och reliabilitet
Validitet är det begrepp som används för att undersöka huruvida en forskare mäter det hen avser att mäta (Andersson, 2006). Exempel på detta kan vara huruvida enkäten som används innehåller syftningsfel, huruvida beräkningar har gjorts på ett korrekt sätt eller huruvida rätt personer har inkluderats i studien. Tidigt var det ett strategisk val att endast använda källor som har visat sig tillförlitliga över tid. WVS har samlat in ett flertal vågor data och granskats vid flera tillfällen under samtliga vågor. GSHS är ett välanvänt verktyg för att undersöka olika aspekter av ungdomars hälsa världen över och anses även det vara ett tillförlitligt instrument. Den fråga som återkom under arbetet med denna artikel var: hur kommer det sig att det verkar finnas ett mönster kring vilka länder som har valt att inkludera sexuella beteenden i sin enkät? Slutsatsen blev att det snarast talar för att frågorna mäter det dom avser att mäta eftersom dessa länder samtidigt hade relativt låga poäng på jämställdhetsindexar och emancipatual values index. En annan fråga som tidigt blev aktuell och som var en del av att rensa data var att ta reda på vilka översättningar av ordet kondom som syftade på kondom och inte preventivmedel över lag. Lexikon kollades och vissa länder uteslöts. Möjligen kan valet av mått för egalitära värderingar ifrågasättas. Egalitära värderingar är ett brett begrepp som skulle kunna inkludera mer än det som denna artikel mäter, exempelvis rasism. Den övervägning som gjordes i detta fall var medvetet restriktiv, då typ 2 fel (falsk positiv) ansågs vara värre än alternativet. Detta är en ansatts att mäta en effekt av egalitära värderingar, nämligen jämställhet mellan kön. Därav valdes ett index som mäter strukturella förutsättningar och ett index som mäter värderingar som skulle kunna bidra till dessa eller potentiellt ha en egen effekt för att främja jämställdhet. Det är känt att emancipation är en av de starkaste faktorerna för att främja kvinnors rättigheter (Welzel, 2013). Det tydliga fokuset på jämställdhet var avgörande för att valet föll på just dessa sätt att mäta egalitära värderingar. I framtiden kanske ett index för att mäta egalitära värderingar kan skapas som väger in fler faktorer, vilket kan vara användbart om en forskare skulle vilja undersöka egalitära värderingar med ett bredare fokus.
Reliabilitet är ett begrepp som används för att uttrycka precisionen i en studie och förutsätter att dom slumpmässiga felen är små så att om studien skulle upprepas skulle den ge samma resultat (Andersson, 2006). Samtliga datakällor använder sig av flera steg för at kontrollera att de
slumpmässiga felen i datainsamlingen är små. Vad gäller min bearbetning och beräkning av data finns det flera saker som talar för en hög reliabilitet. När sensitivitetstester utfördes var resultaten i princip oförändrade när olika länder av olika anledningar uteslöts. Vidare fördes noggranna
Tables
Table 1. Deskreptiv data
Countries N GSHS WVS BNP/ppp mean GII mean Emancipative index Age mean Went hungry Males (%)
Argentina (I) 3958 2007 2006 16,2 37,1 48 14,5 1,5 49,0 Argentina (II) 56736 2012 2013 20,1 36,5 52 14,5 1,6 47,8 Chile (I) 16262 2004 2000 10,7 40,8 46 13,7 1,3 49,2 Chile (II) 4096 2013 2012 22,0 34,5 51 14,4 1,4 50,1 Ghana 12472 2007 2007 2,5 57,5 29 14,5 2,7 52,4 Guatemala 11184 2009 2004 5,9 55,7 41 14,3 1,5 45,6 Indonesia 6232 2007 2006 6,8 53,9 34 13,8 2,1 47,7 Malaysia 51014 2012 2012 23,1 30,2 36 14,7 2,0 50,0 North Macedonia 4228 2007 2001 7,6 X 41 14,5 1,2 49,5 Peru 5764 2010 2012 10,6 41,8 44 14,5 1,8 49,0 Thailand (I) 5534 2008 2007 12,1 33,4 37 13,6 1,9 49,4 Thailand (II) 11788 2015 2013 15,8 34,6 37 14,3 1,9 43,1
Trinidad & Tobago (I) 5938 2007 2006 28,3 35,3 40 14,1 1,7 48,5
Trinidad & Tobago (II) 5622 2011 2010 30,8 35,4 39 14,1 1,9 54,6
Uganda 6430 2003 2001 0,9 62,2 30 14,9 2,1 51,4
Uruguay (II) 7048 2012 2011 18,5 32,1 54 14,4 1,3 46,5
Venezuela 8830 2003 2000 10,9 47,5 43 13,4 1,4 44,7
Zambia 4514 2004 2007 2,4 60,4 40 14,6 2,9 48,9
Zimbabwe 11330 2003 2001 2,2 60,8 30 14,7 1,8 44,6
*BNP/ppp have been divided by 1000 for increased clarity and comparability. It should be read as GDP per capita, PPP ($1000) **GII 100 = Equality, GII 0 = inequality
Table 2. Frequncies of sexual behaviours in the studied countries
Ever had intercourse
Used condom last intercourse
Country Yes (%) Country Yes (%)
Argentina (II, n=52676 ) 43,4 Uruguay (II, n=1978) 81,6
Zambia (n=1482) 43 North Macedonia (n=486) 78,6
Uruguay (II, n=6768) 33,7 Argentina (I, n=904) 77,66
Argentina (I, n=3588) 32,7 Argentina (II, n=19102) 75,57
Chile (II, n=3932) 32,3 Peru (n=1030) 64,85
Trinidad & Tobago (II, n=5118) 29,7 Uganda (n=1168) 63,36
Uganda (n=4230) 28,6 Thailand (II, n=1042) 60,27
Trinidad & Tobago (I, n=5244) 24,7 Thailand (I, n=280) 59,29
Peru (n=5638) 20,1 Guatemala (n=1146) 58,29
Thailand (II, n=10776) 16,4 Trinidad & Tobago (I, n=1246) 57,95
Venezuela (n=6952) 15,9 Trinidad & Tobago (II, n=1110) 57,48
Guatemala (n=10538) 15,3 Ghana (n=1286) 54,9
Chile (I, n=14100) 14,7 Venezuela (n=1094) 54,66
Zimbabwe (n=7474) 14,2 Chile (II, n=1020) 53,14
Ghana (n=9892) 13,7 Zambia (n=582) 52,23
North Macedonia (n=3916) 12,6 Chile (I, n=2040) 46,67
Malaysia (n=47290) 8 Zimbabwe (n=1022) 45,79
Thailand (I, n=5100) 5,7 Malaysia (n=1056) 30,87
Indonesia (n=6134) 0,6 Indonesia x
Table 3. Multilevel logistic regression: Sexual activity
BOTH GENDER, A MALE, B FEMALE, C
b b b b b b b b b b b b b b b Level 1 variabels Age 12 Years or younger 0b 0b 0b 0b 0b 0b 0b 0b 0b 0b 0b 0b 13 Years 0,33* 0,34* 0,33* 0,34* 0,37* 0,38* 0,37* 0,37* 0,29 0,29 0,29 0,29 14 Years 0,79*** 0,80*** 0,79*** 0,80*** 0,81*** 0,81*** 0,81*** 0,81*** 0,80*** 0,80*** 0,80*** 0,80*** 15 Years 1,31*** 1,31*** 1,31*** 1,31*** 1,29*** 1,29*** 1,29*** 1,29*** 1,35*** 1,35*** 1,35*** 1,35*** 16 Years or older 1,88*** 1,88*** 1,88*** 1,88*** 1,83*** 1,83*** 1,83*** 1,83*** 1,96*** 1,97*** 1,97*** 1,97*** Gender Female -0,72*** -0,70*** -0,72*** -0,70*** M ale 0b 0b 0b 0b
Went hungry last 30 days 0,12*** 0,12*** 0,12*** 0,12*** 0,11*** 0,12*** 0,11*** 0,12*** 0,12*** 0,12*** 0,13*** 0,12***
Level 2 variabels BNP/ppp 0,02 0,01 0,04 0,03 0,00 0,05 0,05* 0,02 0,06** GII 0,00 0,04 0,01 0,05 0,01 0,04 Secular values Emancipatual values 0,07* 0,08** 0,08** 0,09*** 0,07* 0,09*** SSE N countries 19 18 17 18 17 19 18 17 18 17 19 18 17 18 17 Intercept -1,56*** -2,49*** -2,69 -5,53*** -7,89** -1,17*** -2,42*** -3,20 -5,64*** -9*** -2,05*** -3,34*** -4,39** -6,54*** -9,51***
Akaike information criterion 1030810 1025468 1005233 1025410 1005036 466810 464672 455965 464651 455829 568480 562700 550520 562682 550326
Table 4 multilevel logistic regression: Condom use
BOTH GENDERS, A M ALE, B FEM ALE, C
b b b b b b b b b b b b b b b Level 1 variabels Age 12 Years or y ounger 0b 0b 0b 0b 0b 0b 0b 0b 0b 0b 0b 0b 13 Years -0,13 -0,13 -0,13 -0,13 -0,14 -0,13 -0,15 -0,13 -0,06 -0,10 -0,06 -0,10 14 Years 0,01 0,02 0,01 0,02 0,03 0,05 0,03 0,05 0,00 -0,04 -0,01 -0,04 15 Years -0,06 0,00 0,01 0,00 0,14 0,15 0,13 0,15 -0,17 -0,20 -0,17 -0,21 16 Years or older -0,06 -0,06 -0,06 -0,06 0,06 0,07 0,05 0,07 -0,22 -0,25 -0,22 -0,26 Gender Female -0,25*** -0,25*** -0,25*** -0,25*** M ale 0b 0b 0b 0b
Went hungry last 30 day s -0,09** -0,09** -0,09** -0,09** -0,10** -0,10** -0,10** -0,10** -0,08* -0,08* -0,08** -0,08*
Level 2 variabels
BNP/p p p -0,01 -0,02 -0,01 -0,02 -0,01 -0,01 -0,01 -0,03 0,00
GII -0,01 0,00 -0,02 -0,01 0,00 0,02
Secular values
Emancip atual values 0,04* 0,04* 0,04* 0,04* 0,04* 0,05*
SSE
N countries 18 17 16 17 16 18 17 16 17 16 18 17 16 17 16
Intercep t 0,42*** 0,74** 1,4272 -0,78665 -1,04584 0,49*** 0,67** 1,8358 -0,971 -0,509 0,32* 0,65* 0,577 -0,734 -2,18
Akaike information criterion 167864 158829 156680 158849 156689 99506 94958 93245 94979 93255 66321 64051 63627 64070 63635
*p-value < 0,05 **p-value < 0,01
