Abstract
In the Gambia, three out of four women of reproductive age have undergone Female Genital Cutting (FGC). Many studies and policy advocates suggest that for such a practice that is deeply rooted in culture, a more holistic approach focusing on educating the population will have sustainable impact. This research examined whether educational level of women has an association with their attitude towards the practice of FGC.
Data from the 2013 Gambia Demographic Health Survey (GDHS) were analyzed. The sample included 6217 households: 10,233 females aged between 15 to 49 years and 3831 males between 15–59 years. This study focused only on women participants. The outcome variable was the attitude of women toward the practice of FGC.
In multivariate regression model, women who were circumcised are found to have 80 times higher odds of supporting FGC [Odds Ratio = 80 (95% CI 50.93–124.4)] compared to uncircumcised women. Women with primary and secondary level education have lower odds of supporting FGC [OR = 0.73 (95% CI 0.915–0.007)) and those with higher education had the lowest odds [OR = 0.28 (95% CI 0.147–0.543)) of supporting FGC relative to women with no education at all.
Education and awareness programs targeting women who are married and older, those with less education and those who are already circumcised can help change attitudes towards the practice of FGC.
Introduction
Female Genital Cutting (FGC) is a spectrum of practices that describes the complete or partial cutting of a woman’s genitalia for non-medical reasons.1–5 The exact origin of the practice of FGC is unknown or lost6 but many believe it originated in Egypt or Northern Sudan.2,4,7 FGC is carried out in Africa, the Middle East, some ethnic communities in South America and some immigrant groups in Europe and North America.1,2,4,5,8–11 About 200 million women are estimated to have undergone this practice in 30 countries over the past two decades alone.12 In the Gambia, three out of four women of reproductive age (15–49 years) have experienced FGC.12 In some rural regions, the prevalence is as high as 90%.13
The classification of FGC varies by culture and depends on the nature and severity of the cutting. For this study, the original classification of FGC by The World Health Organization (WHO) is followed. Type I is the most commonly practiced in The Gambia which is clitoridectomy where the clitoral tip is removed with or without removal of the entire clitoris.14–17
The Gambia has ratified many international and regional treaties to stop FGC.17 However, enforcement of these agreements remains to be a big challenge.8,18 In November 2015, FGC was criminalized in the Gambia through legislation.19 Nevertheless, criminalizing this practice on paper without serious enforcement measures and without due acknowledgment of existing social norms and values could lead to new challenges for a practice that is so deeply rooted.8 There have been attempts to medicalize FGC in other places as well; however, The Gambia Medical and Dental Association and Medical Research Council in the Gambia both strongly opposed the medicalization of FGC as there are no associated health benefits.17,20 Nevertheless, about 8% of healthcare professionals in the Gambia are reported to have carried out FGC in their medical practice.21
Many studies and policy advocates propose that for such a practice so deeply rooted in culture, a more holistic approach focusing on educating the population on the negative health effects and developing decision making authority among more women will have lasting impact. With higher education should come higher empowerment of the women in family and social level which may then help in obstructing such a practice.
This study utilized an ecological model of behavior change in explaining the attitude of women regarding FGC practice. This framework presents a multifaceted approach that looks at the individual, environmental and sociocultural factors within the broader community and beyond.22 It has been used in many previous studies on child abuse,22 domestic violence and other social issues.22,23 The key independent variable used in this study is womens’ education. Education was considered to be both a process and outcome of women’s advancement.
This study examined the support for the continuation of FGC practice by considering a key determinant: educational level of women. This study is unique from other studies as it used a national country level dataset that is comprehensive enough to assess this issue in a country where FGC is deeply rooted and practiced. It used education as important determinant and at the same time included income, race, region, marital status, etc. in the model. No other studies have examined FGC in the Gambia where all these important variables have been analyzed in a comprehensive way from the GDHS. This study also reported the difference in attitude between married and never married women towards this practice.
Methods
Data from the 2013 Gambia Demographic Health Survey (GDHS) conducted by the Gambia Bureau of Statistics were analyzed. The sample included 6217 households: 10 233 females aged 15–49 and 3831 males aged 15–59. This study focused only on women participants.
The sample is representative of both urban and rural Gambia and was designed with two-stage probability stratification technique with appropriate sampling weights. All analyses conducted in this study included these sample design features. Variables were re-coded and crossed-checked against the original variables as part of data cleaning and quality control before conducting actual analyses.
Dependent variable
The outcome variable is the attitude of women in the Gambia toward the practice of FGC. The participants were asked whether the practice of cutting girls in their community should continue or end. The response categories are: (1) Continue, (2) Come to an end (3) Not sure/Undecided. These categories were re-coded into a dichotomous variable (Continue = 1 and Stop = 0); the response Not sure/Undecided was dropped because of few responses.
Since the outcome variable is binary, logistic regression model was chosen for analysis.24 Logistic regression estimates an odds ratio to understand the relationships between two events (supporting and not supporting FGC in this case). Odds ratio in logistic regression can be interpreted as the effects of a one-unit change in the input variable in predicting change in odds ratio of another variable whereas all other variables are held constant in the model.25
Independent variables
For Educational level (no education, primary, secondary and higher education), primary and secondary education were combined for this study. Education categories refer to the highest level of schooling attended whether that level was completed or not. This was done to help explain the relationship between educational level and the attitude towards FGC practice.
Other independent variables
Several socio-demographic and cultural variables were included in the regression model including age, literacy level, circumcision status, whether the flesh was removed (cut), religion, place of residence, ethnicity and administrative regions. Age was dichotomized: young were defined as 15–30 years based on the mean age of the married women and older was 30–49 years. The mean age was used as the cut-off point to help understand whether age is a critical factor in determining support for FGC continuation in the Gambia. Perhaps younger Gambian women (who are better educated) will be less likely to support the continuation of FGC.
For literacy level, three categories were utilized: cannot read at all, able to read parts of a sentence and able to read the whole sentence. Literacy status of the survey respondents was assessed based on a survey respondent’s ability to read all or parts of a simple sentence from a card. Those with secondary education or higher were assumed to be literate. Whether the woman is circumcised was coded as binary (1 ‘yes = woman was circumcised’ and 0 ‘no = woman was not circumcised’) and cut (meaning flesh was removed) during circumcision was also coded as (1 ‘yes’ and 0 ‘no’). Comparisons were made between married and never married women regarding their educational levels and their support for FGC practice to gain insights on whether future generations of women in The Gambia will be supportive of continued FGC practice.
Religion was classified as either Muslim or Christian; Wealth status as -poorest, poor, middle, rich and richest. The classification of the household members in those categories is based on the wealth index. ‘Wealth index is a composite measure of household’s living standard.’26 ‘Wealth status is defined to mean possession of durable consumer goods. These items include a radio, television, refrigerator and other durable household goods; these are indicators of a household’s socioeconomic status.’13 ‘Radio could be utilized to access valuable information and ideas; a refrigerator could help in storing food and contribute to food security; and a bicycle could provide access the market and health facilities that are not near.’13 ‘This is a very useful indicator of household socio-economic standing especially in countries where data are difficult to obtain on individual employment and income.’26
Also, other variables included type and place of residence (rural or urban) in the Gambia. In addition, the ethnic composition of survey respondents (Mandinka/Jahanka, Wollof, Jola/Koroninka, Fula/Tukulor/Lorobo, Serere, Serahuleh, Creole/Aku Marabout, Manjago, Bambara) were reported. The Mandinka ethnicity which is the largest ethnic group as well as a strong supporter of this practice was used as a reference group against other ethnicities. Local Government Areas such as Banjul and Kanifing Municipalities, Brikama, Mansakonko, Kerewan, Kuntaur and Janjabureh regions were also included in the analysis. Banjul was used as a reference category to make urban and rural comparisons.
Statistical analysis
The analysis included descriptive statistics, bivariate analysis and binomial logistic regression. Binomial logistic regression analysis was used to analyze how women’s educational levels might explain their attitude towards FGC while controlling for other independent variables. STATA version 14 (StataCorp LP College Station, TX) was used.
Results
Figure 1 shows the regions of the Gambia and its administrative regions. Majority of the study respondents were from Brikama region. The Mandinka ethnic group is the largest ethnic group in the Gambia and constitutes 33% of the population; they were followed by the Fula ethnic group at 23%. Muslims represent 97% of the Gambian population.
Administrative Regions of the Gambia and Ethnic groups and the Study respondents by Ethnicity and Regions. Map source: Google.
Table 1 presents the key variables by marital status. The mean age of married women (and those living with a partner) at the time of the survey (n = 6905) was 30 years compared to a mean age of 19 years for never-married women (Table 1). Half of the married women (50%) were employed and about half resided in the urban Gambia.
Description of the Study respondents: Married to Never Married Women in The Gambia
| Characteristics | Married | Never married |
|---|---|---|
| Mean age*** | 30 years old | 19 years old |
| Percent who have been circumcised | 77% | 73% |
| Woman’s Educational Status*** | ||
| No education | 61% | 14% |
| Primary or secondary education | 35% | 77% |
| Higher education | 4% | 9% |
| Percent who cannot read at all*** | 70% | 21% |
| Wealth status*** | ||
| Poorest | 19% | 12% |
| Poor | 21% | 14% |
| Middle | 20% | 16% |
| Rich | 20% | 23% |
| Richest | 20% | 34% |
| Percent employed*** | 49.6% | 25% |
| Percent who reside in urban areas*** | 49% | 69% |
| Percent who are Muslim*** | 97% | 92% |
| Ethnicity*** | ||
| Mandinka/Jahanka | 34% | 35% |
| Fula/Tukulor/Lorobo | 23% | 20% |
| Wollof | 13% | 12% |
| Jola/Karoninka | 9% | 14% |
| Sarahule | 7% | 5% |
| Percent residing in different administrative regions*** | ||
| Brikama | 34% | 37% |
| Kanifing | 19% | 31% |
| Basse | 15% | 7% |
| Kerewan | 12% | 9% |
| Banjul | 2% | 3% |
| Characteristics | Married | Never married |
|---|---|---|
| Mean age*** | 30 years old | 19 years old |
| Percent who have been circumcised | 77% | 73% |
| Woman’s Educational Status*** | ||
| No education | 61% | 14% |
| Primary or secondary education | 35% | 77% |
| Higher education | 4% | 9% |
| Percent who cannot read at all*** | 70% | 21% |
| Wealth status*** | ||
| Poorest | 19% | 12% |
| Poor | 21% | 14% |
| Middle | 20% | 16% |
| Rich | 20% | 23% |
| Richest | 20% | 34% |
| Percent employed*** | 49.6% | 25% |
| Percent who reside in urban areas*** | 49% | 69% |
| Percent who are Muslim*** | 97% | 92% |
| Ethnicity*** | ||
| Mandinka/Jahanka | 34% | 35% |
| Fula/Tukulor/Lorobo | 23% | 20% |
| Wollof | 13% | 12% |
| Jola/Karoninka | 9% | 14% |
| Sarahule | 7% | 5% |
| Percent residing in different administrative regions*** | ||
| Brikama | 34% | 37% |
| Kanifing | 19% | 31% |
| Basse | 15% | 7% |
| Kerewan | 12% | 9% |
| Banjul | 2% | 3% |
Note: Married women include women living with a partner. n = 6839–10 182 depending on the categories.
Differences between married and never married women are statistically significant as follows: *p < 0.05, **P < 0.01, ***P < 0.001.
Other Ethnicities: Serere, Creole/Aku marabout, Manjago, Bambara, Other, & Non-Gambians.
Other Regions: Mansakonko, Kuntaur, and Janjabureh,
Description of the Study respondents: Married to Never Married Women in The Gambia
| Characteristics | Married | Never married |
|---|---|---|
| Mean age*** | 30 years old | 19 years old |
| Percent who have been circumcised | 77% | 73% |
| Woman’s Educational Status*** | ||
| No education | 61% | 14% |
| Primary or secondary education | 35% | 77% |
| Higher education | 4% | 9% |
| Percent who cannot read at all*** | 70% | 21% |
| Wealth status*** | ||
| Poorest | 19% | 12% |
| Poor | 21% | 14% |
| Middle | 20% | 16% |
| Rich | 20% | 23% |
| Richest | 20% | 34% |
| Percent employed*** | 49.6% | 25% |
| Percent who reside in urban areas*** | 49% | 69% |
| Percent who are Muslim*** | 97% | 92% |
| Ethnicity*** | ||
| Mandinka/Jahanka | 34% | 35% |
| Fula/Tukulor/Lorobo | 23% | 20% |
| Wollof | 13% | 12% |
| Jola/Karoninka | 9% | 14% |
| Sarahule | 7% | 5% |
| Percent residing in different administrative regions*** | ||
| Brikama | 34% | 37% |
| Kanifing | 19% | 31% |
| Basse | 15% | 7% |
| Kerewan | 12% | 9% |
| Banjul | 2% | 3% |
| Characteristics | Married | Never married |
|---|---|---|
| Mean age*** | 30 years old | 19 years old |
| Percent who have been circumcised | 77% | 73% |
| Woman’s Educational Status*** | ||
| No education | 61% | 14% |
| Primary or secondary education | 35% | 77% |
| Higher education | 4% | 9% |
| Percent who cannot read at all*** | 70% | 21% |
| Wealth status*** | ||
| Poorest | 19% | 12% |
| Poor | 21% | 14% |
| Middle | 20% | 16% |
| Rich | 20% | 23% |
| Richest | 20% | 34% |
| Percent employed*** | 49.6% | 25% |
| Percent who reside in urban areas*** | 49% | 69% |
| Percent who are Muslim*** | 97% | 92% |
| Ethnicity*** | ||
| Mandinka/Jahanka | 34% | 35% |
| Fula/Tukulor/Lorobo | 23% | 20% |
| Wollof | 13% | 12% |
| Jola/Karoninka | 9% | 14% |
| Sarahule | 7% | 5% |
| Percent residing in different administrative regions*** | ||
| Brikama | 34% | 37% |
| Kanifing | 19% | 31% |
| Basse | 15% | 7% |
| Kerewan | 12% | 9% |
| Banjul | 2% | 3% |
Note: Married women include women living with a partner. n = 6839–10 182 depending on the categories.
Differences between married and never married women are statistically significant as follows: *p < 0.05, **P < 0.01, ***P < 0.001.
Other Ethnicities: Serere, Creole/Aku marabout, Manjago, Bambara, Other, & Non-Gambians.
Other Regions: Mansakonko, Kuntaur, and Janjabureh,
The associations between attitude towards FGC and the covariates were explored. In the bivariate comparison, most covariates were found to be associated with the view towards FGC. It shows employment, urban, Muslim, house ownership, literacy, ethnicity, wealth status, region and women’s level of education were all statistically significant regardless of the women’s marital status. About two-third of married women support the continuation of FGC.
Table 2 provides the summarized adjusted model of selected odds ratio and corresponding 95% CI of independent variables. Women who were circumcised have 80 times higher odds of supporting FGC [OR = 80 (95% CI 50.93–124.4)] compared to the uncircumcised group. Women with primary and secondary education have lower odds of supporting FGC [OR = 0.73 (95% CI 0.915–0.007)) and those with higher education [OR = 0.28 (95% CI 0.147–0.543)) had the lowest odds of supporting this practice compared to women with no education at all.
Should FGC be continued? Logistic regression results of the association of education, age and other co-variates
| Characteristics | Adjusted OR | 95% CI | P-value |
|---|---|---|---|
| Woman’s level of education | |||
| Reference category: (no education) | |||
| Primary + secondary education | 0.719 | 0.569–0.909 | 0.006** |
| Higher education | 0.271 | 0.132–0.555 | <0.001*** |
| Age | |||
| Reference category: (1 = young) | |||
| Age | 0.660 | 0.527–0.828 | <0.001*** |
| Ethnicity | |||
| Reference category: (Mandinka) | |||
| Fula/Tukulor/Lorobo | 0.440 | 0.205–0.679 | <0.001*** |
| Wollof | 0.355 | 0.209–0.604 | <0.001*** |
| Region | |||
| Reference category: (Banjul) | |||
| Basse | 2.47 | 1.035–5.830 | 0.0042* |
| Brikama | 1.62 | 1.076–2.536 | 0.022* |
| Kuntaur | 0.402 | 0.223–0.765 | 0.005** |
| Respondent was circumcised | 79.4 | 51.3–122.7 | <0.001** |
| Islam | 2.16 | 1.01–4.617 | 0.045 |
| Characteristics | Adjusted OR | 95% CI | P-value |
|---|---|---|---|
| Woman’s level of education | |||
| Reference category: (no education) | |||
| Primary + secondary education | 0.719 | 0.569–0.909 | 0.006** |
| Higher education | 0.271 | 0.132–0.555 | <0.001*** |
| Age | |||
| Reference category: (1 = young) | |||
| Age | 0.660 | 0.527–0.828 | <0.001*** |
| Ethnicity | |||
| Reference category: (Mandinka) | |||
| Fula/Tukulor/Lorobo | 0.440 | 0.205–0.679 | <0.001*** |
| Wollof | 0.355 | 0.209–0.604 | <0.001*** |
| Region | |||
| Reference category: (Banjul) | |||
| Basse | 2.47 | 1.035–5.830 | 0.0042* |
| Brikama | 1.62 | 1.076–2.536 | 0.022* |
| Kuntaur | 0.402 | 0.223–0.765 | 0.005** |
| Respondent was circumcised | 79.4 | 51.3–122.7 | <0.001** |
| Islam | 2.16 | 1.01–4.617 | 0.045 |
Note: FGC = female genital cutting, OR = odds ratio, CI = confidence interval, n = 6906.
Should FGC be continued? Logistic regression results of the association of education, age and other co-variates
| Characteristics | Adjusted OR | 95% CI | P-value |
|---|---|---|---|
| Woman’s level of education | |||
| Reference category: (no education) | |||
| Primary + secondary education | 0.719 | 0.569–0.909 | 0.006** |
| Higher education | 0.271 | 0.132–0.555 | <0.001*** |
| Age | |||
| Reference category: (1 = young) | |||
| Age | 0.660 | 0.527–0.828 | <0.001*** |
| Ethnicity | |||
| Reference category: (Mandinka) | |||
| Fula/Tukulor/Lorobo | 0.440 | 0.205–0.679 | <0.001*** |
| Wollof | 0.355 | 0.209–0.604 | <0.001*** |
| Region | |||
| Reference category: (Banjul) | |||
| Basse | 2.47 | 1.035–5.830 | 0.0042* |
| Brikama | 1.62 | 1.076–2.536 | 0.022* |
| Kuntaur | 0.402 | 0.223–0.765 | 0.005** |
| Respondent was circumcised | 79.4 | 51.3–122.7 | <0.001** |
| Islam | 2.16 | 1.01–4.617 | 0.045 |
| Characteristics | Adjusted OR | 95% CI | P-value |
|---|---|---|---|
| Woman’s level of education | |||
| Reference category: (no education) | |||
| Primary + secondary education | 0.719 | 0.569–0.909 | 0.006** |
| Higher education | 0.271 | 0.132–0.555 | <0.001*** |
| Age | |||
| Reference category: (1 = young) | |||
| Age | 0.660 | 0.527–0.828 | <0.001*** |
| Ethnicity | |||
| Reference category: (Mandinka) | |||
| Fula/Tukulor/Lorobo | 0.440 | 0.205–0.679 | <0.001*** |
| Wollof | 0.355 | 0.209–0.604 | <0.001*** |
| Region | |||
| Reference category: (Banjul) | |||
| Basse | 2.47 | 1.035–5.830 | 0.0042* |
| Brikama | 1.62 | 1.076–2.536 | 0.022* |
| Kuntaur | 0.402 | 0.223–0.765 | 0.005** |
| Respondent was circumcised | 79.4 | 51.3–122.7 | <0.001** |
| Islam | 2.16 | 1.01–4.617 | 0.045 |
Note: FGC = female genital cutting, OR = odds ratio, CI = confidence interval, n = 6906.
Discussion
The main findings of this study
The findings of this study show that women with some level of education are less likely to support the continuation of FGC compared to women with no education.
Age is an important factor in understanding attitudes towards FGC in The Gambia. The findings show that young married women between the ages of 15–30 years have 34% lower odds of supporting FGC relative to married women who are older than 30 years. This is statistically significant at P > 0.001. A previous study in 2012 also found that older women in West Africa are more likely to support the continuation of this practice.27
Our study confirms that FGC is practiced in all regions of The Gambia and this practice is prevalent across ethnic, regional, religious and economic background. Previous study also supported this finding FGC occurs in all social-linguistic groups in practicing communities.6
The strongest predictor in our model is the FGC status of married women. A woman who has undergone FGC has 80 times higher odds of supporting the continuation of FGC of young women relative to the uncircumcised group. This is in line with other studies7 that also reported that women who have undergone FGC are more likely to support continuation of this practice.
What is already known on this topic
Previous research reported that a woman’s level of education is a central determinant of her position on FGC. There is a negative association between a woman’s level of education and her attitude towards FGC.6 Similar research in this area has pointed out that access to education could contribute to reducing the prevalence of FGC.28 Women with the higher level of educational attainment are more likely to oppose FGC.29
A study in West Africa on FGC shows that women with no formal education who were currently or previously in a relationship and Muslims are at higher odds of experiencing FGC.27 Skramstad, 2008, cited in Allen, Hurn, Loveday, Sivan, (2015), acknowledged the importance of education related to the Mandinka circumcision that is thought to be connected to female behavioral attributes. They conveyed that the Mandinka ‘Nyaka’ (circumcision) is connected to respect (horomo), secrecy (suturo) and endurance (sabati). The continuance of this form of social education in Mandinka communities has been advocated without promoting the actual cutting of the clitoris. However, the institution of ‘Nyaka’ reinforces gender inequality. Hence, education could target to shatter this deeply rooted belief.
Social determinants of health30 are fundamental drivers that influence the attitudes of women towards FGC continuation in The Gambia. For example, access to education and jobs have shown to have incrementally positive relationship to better health.30,31 Other determinants include gender factor, political representation, and participation in the labor market.32
Most of the women in The Gambia have a limited voice in society and their views and needs are often ignored in the political process.33 Therefore, an in-depth understanding of the socioeconomic status of women is needed to shape their attitudes regarding FGC. When more women are engaged in the decision-making process, their requirements and priorities are more likely to be given importance.34 Therefore, education can help improve participation in decision making for these women in their households and in the wider communities. Human rights education for girls and women could help change their roles and expectation in a traditional society.35,36
The role of education in preventing child abuse37 and domestic violence38 was investigated in previous studies.
In the Gambia, FGC is widely practiced and still considered to be a contentious issue. Decisions leading to practicing FGC are complex39 and it requires a deeper understanding of the social and cultural norms associated with this practice.
What this study adds
In the Gambia, over 60% of men and women have no education at all and 69% of married women cannot read at all. Education is considered as an enabler that could help women climb the social mobility ladder and hence increase their autonomy and empowerment.40 This study shows that women with some level of education are less likely to support the continuation of FGC compared to women with no education. Therefore, the introduction and incorporation of targeted human rights education in the school system, rural communities and an overall strengthening of civic education programs could all contribute towards raising awareness and hence mobilize opinion and actions against FGC practice.
Most of the FGC prevention programs have failed to explain their program effects in changing attitudes and behaviors.41 The focus of FGC related education should be on creating change in knowledge, attitude and behavior of individuals and communities.42
Education is critical in communities where it is pervasive to hear offensive and insulting terms like ‘solema’ denoting uncircumcised (Hernlund 2003; cited in Hernlund & Shell-Duncan39). This is very similar to the concept of ‘ghalfa’ in Sudan which is also a derogatory term with a suggestion of prostitution. For fear of this social stigma, some women in many circumstances are often pressured into accepting FGC so that they feel accepted within their ethnolinguistic community.41 The social determinants of status such as age and gender are also important in determining the position of individuals in society.43
Strengths and limitations of this study
The strengths of this study are many including the large sample size and the scientific sampling design. This study has several limitations as well which are mostly data related. Variables were coded in a certain way that limited exploration of important measures like empowerment. Therefore, the study results should be interpreted with caution. The secondary data used for analysis is cross-sectional in nature. The information on attitudes towards FGC was self-reported. There was no follow-up medical examination to confirm or disapprove FGC status or the type of FGC carried out on these women. Age was used as a binary variable to help understand the attitude of the ‘young’ and the ‘old’ to distinguish between two generations and this limited interpretation of study findings by narrower age brackets.
Conclusion
The present study provides valuable new information about the prevailing attitudes of women in The Gambia towards FGC practice. The study offers benchmark numbers against which future researchers can compare their findings when analyzing similar national-level datasets inside and outside The Gambia. It will also help to monitor the prevalence trends of FGC over the next few cycles of GDHS. These results may help program implementers and evaluators working toward measuring progress in the Gambia on obstructing FGC practice and promoting other women empowerment issues.
In conclusion, more educational opportunities in general and targeted education in certain demographic groups in the Gambia—older and married women and those already circumcised may help obstruct support for FGC practice continuation.
