Adolescent Girls’ Safety In and Out of School: Evidence on Physical and Sexual Violence from Across Sub-Saharan Africa

Figures & data

Table 1. Proportion of girls aged 15–19 years old who experienced physical or sexual violence previously (DHS data)

	All girls		Did not attend school the past year		Attended school the past year
	Experienced physical or sexual violence (%)	Obs	Experienced physical or sexual violence (%)	Obs	Experienced physical or sexual violence (%)	Obs	Simple difference (%)	Adjusted difference (%)
Cross-country average	28.8	21,545	30.6	12,222	27.0	9,268	−3.6***	−1.0
Angola	23.7	1,980	22.4	868	24.3	1,112	1.8	2.5
Burkina Faso	20.6	1,815	18.8	1,396	25.8	413	7.1**	1.6
Cameroon	31.7	1,312	33.6	568	30.5	744	−3.1	0.0
Chad	20.5	777	23.1	601	15.1	165	−7.9**	−12.4**
Congo Dem Rep	42.5	1,143	46.4	589	39.8	544	−6.6	−4.3
Cote d‘Ivoire	32.8	990	33.6	720	31.1	254	−2.5	−6.3
Ethiopia	14.0	976	17.7	501	10.9	475	−6.9*	−3.7
Ghana	35.6	380	35.1	194	35.8	184	0.7	3.0
Kenya	34.7	772	40.9	287	31.8	479	−9.1*	−8.0
Malawi	29.4	1,080	33.6	565	25.7	515	−7.9**	−3.6
Mali	36.6	635	40.5	478	28.3	157	−12.2**	−10.2
Mozambique	24.5	1,212	24.0	744	25.2	468	1.2	3.4
Nigeria	34.4	1,529	31.4	880	37.7	649	6.3*	7.0*
Rwanda	32.6	378	41.2	174	25.3	201	−15.9***	−11.8**
Senegal	20.1	364	18.4	182	21.9	182	3.6	−0.6
South Africa	14.4	233	14.3	76	14.5	157	0.2	0.1
Tanzania	27.3	1,533	31.2	1,005	19.7	527	−11.4***	−6.1*
Uganda	44.6	1,638	47.5	935	41.6	703	−6.0*	−1.5
Zambia	23.9	1,612	26.9	847	21.2	765	−5.7**	−2.9
Zimbabwe	31.3	1,186	38.0	612	25.4	574	−12.6***	−9.9**

Notes: We used the most recent DHS survey for each country for which the domestic violence section is available. We used the indicators for various physical abuse and forced unwanted sexual acts by partners or others. We control for the country, age, type of residence (urban/rural), parents’ education, and a wealth index from household assets to report the adjusted difference. Each country’s tabulation uses domestic violence module sample weights. The all-country tabulation use denormalised weights. Standard errors are clustered at the household level. ***p < 0.01; **p < 0.05; *p < 0.1.

Table 2. Distribution of perpetrators of first sexual violence among girls aged 15–19 years old who report having experienced sexual violence previously (DHS data)

	Total (%)	Did not attend school the past year (%)	Attended school the past year (%)
Former or current husband	4.1	6.0	3.3
Former or current boyfriend	19.5	16.1	21.8
Family members	8.5	6.0	14.5
Friends	14.4	9.6	22.5
Teacher	2.1	1.4	2.0
Employer	0.9	1.0	0.3
Stranger	13.7	8.3	23.3
Others	5.6	6.2	6.3
Did not report perpetrator	31.3	45.4	5.9
Observations	1,666	1,142	516

Notes: We use the DHS surveys for the same set of countries in Table 1. We use denormalised weights for this tabulation. The sum of the observations for those who did not attend school (1,142) and those who attended school (516) are not equal to the total observations (1,666) because some of the observations do not have school attendance recorded.

Table 3. Proportion of 15–19-year-old girls who experienced sexual violence in the past 12 months (VACS data)

Country	Year of survey	Obs	At least one incident of sexual violence (%)	Types of sexual violence				% of incidents that happened in school (among those who experienced violence) (%)
				Unwanted touch (%)	Attempted unwanted sex (%)	Physically forced sex (%)	Pressured sex (%)
Cross-country average			16.7	9.0	9.3	2.1	1.7	10.2
Kenya	2010	462	13.6	10.4	3.8	0.9	1.8	18.1
Malawi	2013	401	27.7	11.9	17.6	1.9	2.2	9.7
Nigeria	2014	665	20.8	12.7	12.7	3.7	1.7	13.7
Tanzania	2009	882	15.7	7.5	7.3	2.2	2.0	10.6
Zambia	2014	348	16.7	9.2	11.0	2.4	1.6	7.1
Zimbabwe	2017	3,486	5.9	2.1	3.2	1.6	0.9	1.7

Notes: We used the most recent publicly available Violence Against Children Survey (VACS) from the Together for Girls partnership. The survey asks where the incident of sexual violence occurred, and respondents may say school regardless of enrollment status. Each country’s tabulation uses sample weights. The cross-country average is the simple average of the countries in the table.

Table 4. Proportion of 15–19-year-old girls who reported experiencing sexual violence in the past 12 months according to school attendance (VACS data)

		Not currently attending school			Currently attending school			Simple difference		Adjusted difference
Country	Year of survey	Obs	At least one incident of sexual violence (%)	% of violence that happened at school (%)	Obs	At least one incident of sexual violence (%)	% of violence that happened at school (%)	At least one incident of sexual violence (%)	% of violence that happened at school (%)	At least one incident of sexual violence (%)	% of violence that happened at school (%)
Cross-country average			15.8	4.6		19.3	14.4	3.5	9.8	2.3	11.0
Kenya	2010	166	16.5	4.4	284	12.2	28.3	−4.3	23.9**	−3.2	24.4*
Malawi	2013	181	21.5	3.6	208	34.8	13.1	13.4	9.5	8.7	8.6
Nigeria	2014	185	20.2	8.9	344	27.2	16.6	7.0*	7.7	8.0*	17.7**
Tanzania	2009	359	13.8	9.6	449	19.1	13.1	5.3	3.5	6.0	2.0
Zambia	2014	171	15.4	0.0	164	17.6	12.9	2.2	12.9*	−3.5	12.1
Zimbabwe	2017	1,656	7.3	1.1	1,804	4.6	2.6	−2.6***	1.5	−2.2**	1.0

Notes: We use the most recent publicly available VACS surveys. We control for age, household assets (electricity, radio, television, telephone, and bicycle) and report the adjusted difference. For Kenya: additional controls such as distance to nearest clinic and highest schooling of adult in household did not change the sign and significance of the coefficient of the adjusted difference of ‘At least one incident of sexual violence’ but reduced the significance level of ‘% of violence that happened at school’ so that the adjusted difference is not statistically significant. For Tanzania: adding a control for the highest schooling attained by the head of household did not change the sign and significance of the coefficients. Country-level tabulations and regressions use sample weights. ***p < 0.01; **p < 0.05; *p < 0.1.

Table 5. Comparison between the adjusted difference (enrolled vs not enrolled) among the proportions of 15–19-year-old girls who reported experiencing physical or sexual violence in the past 12 months (coefficients by survey used)

	DHS		VACS
Country	Physical (%)	Sexual (%)	Physical	Sexual (%)	Forced sex only (%)
Kenya	−0.3	−2.0	16.4**	−3.2	−2.9
Malawi	−3.7	−5.3	9.4	8.7	−1.2
Nigeria	−4.7	−1.9**	12.6**	8.0*	−1.6
Tanzania	0.0	−3.3**	13.8**	6.0	−1.4
Zambia	−4.2*	−2.2**	7.6	−3.5	1.3
Zimbabwe	−0.8	−7.2***	2.4*	−2.2**	−2.5***

Notes: Differences with a negative sign indicate that levels are lower for girls who are enrolled in school. DHS controls include age, wealth index, urban/rural, parents’ educational level. VACS controls include age, ownership of television, telephone, bicycle and radio and access to electricity. Country-level tabulations and regressions use sample weights. These coefficients are from Table 4 (sexual violence as reported in VACS) and Appendix Tables A6 (physical violence reported in DHS), A7 (sexual violence reported in DHS), A17 (physical violence as reported in VACS), and A20 (being forced to have sex as reported in VACS). ***p < 0.01; **p < 0.05; * p < 0.1.

Data availability statement

The Demographic and Health Surveys (DHS) and the Violence against Children Surveys (VACS) are both publicly available. The necessary do-files for replicating all the results in this article using Stata, together with guidance as to which data sets to download, are available at the Harvard Dataverse: https://doi.org/10.7910/DVN/06IBGB.