Page 107 - IJPS-7-1

Rawat and Tirkey

measured by three dichotomous questions: had any STIs in past 12 months, bad smelling/abnormal genital discharge, and
genital sore or ulcer. If answering “Yes” to any of three questions, the woman was considered having STIs/STSs, coded 1
and 0 otherwise. Sixth, whether the respondent who had STIs/STSs sought advice/treatment in the last 12 months? (Yes
“1” vs. No “0”).

2.2.2. Independent variables
The study selected some key socio-economic and demographic variables which were considered as potential characteristics
that could affect the outcome of the interest. In a nutshell, age group of the women (15 – 24, 25 – 34, 35 – 44, and 45 –
49 years), place of residence (rural and urban), marital status of the women (never married, married, and others), religion
of the respondents (Muslim and non-Muslim), ethnic groups (Poular, Wolof, Serer and others), levels of education (no
education, primary education, secondary education, and tertiary education), the wealth index (poorest, poor, middle, rich,
and richest), and media exposure (i.e., reading magazines/newspapers, listening to radio, and watching television [yes vs.
no]) were used as the independent or predictors in the study. All these variables have been identified as the key factors that
are associated with our dependent variables (e.g., Sougou et al., 2021; Ahinkorah et al., 2021).

2.3. Analytical methods
In addition to bivariate analyses, binary logit (logistic) regression models were carried out to estimate the prevalence of
FGM and its associated factors, factors associated with attitudes toward termination of FGM, factors associated with STIs
knowledge, the prevalence of STIs, and factors associated with seek treatments among who had STI symptoms. Multinomial
logit regression models were used to examine factors associated with women’s attitude toward the continuation of FGM
practice. The application of multinomial logit regression model is because the outcome variable has three categories
(continuing, stopped, and do not know/depends). In all analyses, appropriate sampling weights were used to adjust for the
complex survey design of the DHS. All analyses of this study were carried out using SPSS v20 and STATA v16.

3. Results
3.1. Sample characteristics

The weighted percentage distributions of sample by selected background characteristics are shown in the left panel of
Table 1. Women aged 15 – 24, 25 – 34, and 35 – 44 years accounted for 41%, 31%, and 21%, respectively; and women
of aged 45 – 49 accounted for 7%. Urban women accounted for nearly 49%. About 65% of women in reproductive ages
15 – 49 were married and approximately 97% were Muslims. Poular and Wolof ethnicities accounted for 28% and 40.2%,
respectively, and Serer ethnicity accounted for 15.3%. Around half (47.2%) of Senegal women in reproductive ages
had no education, and 21.1%, 27.3%, and 4.4% of them received primary education, secondary education, and tertiary
education, respectively. More than 90% of Senegal women in reproductive ages had ever exposed to the mass media.
Overall, about 25% of reproductive women experienced FGM practice (Column B in Table 1). Bivariate analysis shows
that. Bivariate analysis shows that there was no age difference in weighted prevalence of FGM practice, although the youngest
women (aged 15 – 24 years) and the oldest women (aged 45 – 49) had lower FGM prevalence (23 – 24%) than the two other
age groups (ages 25 – 34 and 35 – 44) (about 28%). Results further show that rural women reported a higher prevalence of
FGM practices (29.1%) in comparison to their urban counterparts (21.1%). About 28% of married women reported a practice
of FGM in comparison with only 20% among never married women and 21% in those divorced or widowed. Muslim
women reported much higher prevalence of FGM (25.7%) than non-Muslim (7.5%). The higher prevalence of FGM found
in Poular (54%) and others (57%) in comparison with only 1 – 1.5% among Wolof and Serer ethnicities. Women with tertiary
education reported a lower prevalence rate of having FGM in comparison with those in other educational categories: 27% in
women with no education, 23 – 24% in women with primary and secondary education. The prevalence rate of having FGM
is linked to wealth, from nearly 48% in the poorest down to <15% in the richest. Women with media exposure reported a
prevalence rate of having FGM 23%, lower than 56% among those who had non-exposure to media.
Among the women who had a FGM practice, about 96.4% of FGM cases were performed by traditional circumcisers
and the remaining 3.6% of cases were performed by other practitioners. This pattern is almost universal for all subgroups
of the population with few exceptions. This indicates that even today almost all FGM cases were performed by traditional
practitioners in Senegal. The distributional difference was only found by education, wealth, and urban-rural residence.
The lowest prevalence of FGM performed by a traditional practitioner was found in those with tertiary (90.3%) and
primary education (91.6%), the richest (94.1%), and rich (95.0%) women, and urban women (95.4%).

International Journal of Population Studies | 2021, Volume 7, Issue 1 101