Internet use, exposure to digital family planning messages, and sexual agency among partnered women in Northern Nigeria: implications for digital family planning intervention

Abstract

Digital health interventions are gaining ground in conflict-affected countries, but studies on their reproductive health benefits for women are scanty. Focusing on conflict-affected northern Nigeria, this study examined the relationships between Internet use, exposure to digital family planning messages via text messages or social media, and sexual agency – measured as the ability to refuse sex and ask a male partner to use a condom – among partnered women including the rural-urban differentials. Partnered women's data (n= 18,205) from the 2018 Nigeria Demographic and Health Survey were analysed using descriptive and multinomial logistic regression analyses. 44.6% of women are able to refuse sex, and 31.4% to ask a male partner to use a condom. Internet use was positively associated with women's ability to refuse sex in the northern region and urban areas, and across the region to ask a male partner to use a condom. It was also positively associated with women's uncertainty about asking a male partner to use a condom. Exposure to digital family planning messages was positively associated with women's ability to ask a male partner to use a condom across the region, in both urban and rural areas. However, exposure to digital family planning messages was negatively associated with women's uncertainty in urban areas about their ability to refuse sex. Implications of these findings for digital family planning interventions are discussed.

Résumé

Les interventions de santé numérique gagnent du terrain dans les pays touchés par des conflits, mais les études sur leurs avantages en matière de santé reproductive des femmes sont rares. Cette étude, concentrée sur le Nord du Nigéria touché par un conflit, a examiné les relations entre l'utilisation d'Internet, l'exposition aux messages numériques de planification familiale par le biais de textos ou des médias sociaux, et la maîtrise de sa propre sexualité (mesurée comme la capacité de refuser un rapport sexuel et de demander à un partenaire masculin d'utiliser un préservatif) chez les femmes en couple, y compris les différences entre zones rurales et urbaines. Les données sur les femmes en couple (n=18 205) tirées de l'enquête démographique et de santé du Nigéria de 2018 ont fait l'objet d'analyses de régression logistique descriptive et multinomiale. 44,6% des femmes peuvent refuser un rapport sexuel et 31,4% sont en mesure de demander à leur partenaire masculin d'utiliser un préservatif. L'utilisation d'Internet était positivement associée à la capacité d'une femme de refuser un rapport sexuel dans la région septentrionale et les zones urbaines, et dans l'ensemble de la région à la capacité de demander à un partenaire d'utiliser un préservatif. L'utilisation s'Internet était aussi positivement associée à l'incertitude des femmes concernant leur capacité de demander à un partenaire d'utiliser un préservatif. Dans l'ensemble de la région, aussi bien dans les zones urbaines que rurales, l'exposition aux messages numériques de planification familiale était positivement associée à la capacité des femmes de demander à leur partenaire d'utiliser un préservatif. Néanmoins, l'exposition aux messages numériques de planification familiale était négativement associée à l'incertitude des femmes dans les zones urbaines concernant leur capacité de refuser un rapport sexuel. Les conséquences de ces conclusions pour les interventions numériques de planification familiale ont fait l'objet d'une discussion.

Resumen

Las intervenciones sanitarias digitales están ganando terreno en países afectados por conflictos, pero escasean los estudios sobre sus beneficios para la salud reproductiva de las mujeres. Enfocado en la región septentrional de Nigeria afectada por conflictos, este estudio examinó las relaciones entre el uso de internet, la exposición a mensajes digitales sobre planificación familiar vía mensajes de texto o redes sociales y la agencia sexual, medida como la capacidad de mujeres con pareja para negar sexo y la capacidad para pedir a su pareja masculina que use el condón, teniendo en cuenta las diferencias entre zonas rurales y urbanas. Utilizando análisis de regresión logística multinomial y descriptiva, se analizaron los datos de mujeres con pareja (n= 18,205) de la Encuesta Demográfica y de Salud Nacional de 2018. El 44.6 % de las mujeres eran capaces de negar sexo y el 31.4 % de pedir a su pareja masculina que usara un condón. El uso de internet se asoció positivamente con la capacidad de las mujeres para negar sexo en la región septentrional y en zonas urbanas; y en toda la región con su capacidad para pedir a su pareja masculina que usara un condón. También se asoció positivamente con la incertidumbre de las mujeres en cuanto a pedir a su pareja masculina que usara un condón. La exposición a mensajes digitales sobre planificación familiar se asoció positivamente con la capacidad de las mujeres para pedir a su pareja masculina que usara un condón en toda la región, tanto en zonas urbanas como rurales. Sin embargo, la exposición a mensajes digitales sobre planificación familiar se asoció negativamente con la incertidumbre de las mujeres en zonas urbanas respecto a su capacidad para negar sexo. Se analizan las implicaciones de estos hallazgos para las intervenciones digitales de planificación familiar.

Keywords:

• digital connectedness
• family planning
• sexual agency
• Nigeria

Introduction

Women's ability to negotiate safer sex is a key indicator of sexual autonomy and a component of sexual agency.^1,2 Sexual agency is a multidimensional concept that includes sexual assertiveness, feelings of entitlement to pleasure, and sexual satisfaction while also aiming to protect oneself from adverse sexual and reproductive health outcomes from sexual activities.³ It is also an important determinant of women's susceptibility to infection and resilience.^4–7 Women's ability to negotiate safer sex, especially in marriage in sub-Saharan Africa (SSA), is shaped by interspousal power relations.^5–8 In highly patriarchal climes like Nigeria, female sexuality is socially constructed and controlled by powerful and repressive traditions that hinder women's autonomy.⁹ Women’s limited autonomy prevails in the form of a lack of sexual decision-making power, economic independence, self-efficacy, and access to sexual and reproductive health information, and fear of the consequences of negotiating for safer sex.^7,10,11

Northern Nigeria is bedevilled by conflicts and insurgency, resulting in the death and displacement of thousands of people, while health systems have been badly affected).^12,13 Therefore, it is reasonable to expect pregnancy or fertility desire among women to reduce as well as fertility levels.¹⁴ However, fertility remains high in the region, including poor reproductive health indices.¹⁵ The key reasons are male partners’ pronatalism and dominance in reproductive decisions, fuelling their desire for more children.¹⁵ Such men can prevent their female partners from using contraceptives and jeopardise their sexual autonomy; these are tantamount to reproductive coercion.¹⁶ Unplanned pregnancies may occur in violent relationships.¹⁷ Therefore, it is crucial to understand the extent of women's sexual agency and associated factors in Northern Nigeria.

Globalisation, telecommunication, and the Internet have improved interconnectivity among people.¹⁸ They drive millions of people's exposure to information through the mass media of television, radio, newspapers, and magazines. Besides, we are currently in a digital age representing the most information-laden era in human history. The role of mass media in every area of human life, including public health, cannot be overemphasised; it has been explored by scholars globally.^19–23 Aboagye²⁴ submitted that mass media exposure was associated with safer sex negotiations among women in SSA. Concerning digital media, some conflict-affected and hard-to-reach regions in SSA have experienced growth in deploying eHealth technologies to cut programme costs and enhance health equity.²⁰ However, disparities exist in the availability of these eHealth tools.²⁰ Also, there is a paucity of studies on the health impacts of such interventions in the conflict regions like northern Nigeria, where only very few of the nineteen northern states – Kaduna and Kano states – have high Internet use.²¹ However, understanding whether sexual agency is higher among women who use the Internet can inform digital interventions in the region.

A systematic review of SSA found that digital health interventions were associated with women's autonomy and decision-making abilities.²² In the Kaduna state in northern Nigeria, women's contraceptive ideation and use improved due to their exposure to digital reproductive health programmes.²³ These findings emphasise the framework of family planning programmes and expected positive outcomes, such as the ability to exercise reproductive rights and enjoy reproductive health.^24,25 However, little is known about whether exposure to digital family planning messages is positively associated with sexual agency among women in northern Nigeria. Therefore, this study examined the: (a) extent of sexual agency, (a) relationship between Internet use and sexual agency; (b) the association between exposure to family planning messages (via text messages or social media) and sexual agency, and (d) rural–urban differentials in these relationships among partnered women in northern Nigeria.

Theoretical perspectives

According to the sexual script theory, cultures are responsible for the emergence of social norms, known as social scripts, about humans’ sexuality, which govern their attitudes, perceptions, and behaviours in sexual relationships.²⁶ The sexual script conditions men to be less emotional, sex-craving, and to dominate women.²⁷ Conversely, women are expected to be sex limiters (by placing limits on men's access to sex), to attract men but to be submissive to them.²⁷ According to Seabrook et al.²⁸ mass media are important agents of socialisation used for spreading gender roles. Programmes on mass media, especially television, are replete with erotic scenes sexualising women and reinforcing sexual roles, and encouraging them to accept, internalise and demonstrate such roles.²⁹

Conversely, this digital era has ignited social changes challenging the dominant gendered sexual scripts. Digital connectedness is positively associated with women's decision-making autonomy and negatively associated with gender equality.³⁰ Increased rates of digital connectedness have enabled women to secure jobs (especially remote ones) and contribute financially in their households rather than just taking care of their children and homes while their male partners provide. With increased access to mobile phones and the Internet, women can learn and copy socially desirable reproductive behaviours of other women online and decide whether to have children and use contraceptives.³⁰ Studies have documented the reproductive health impacts of mobile-phone-based healthcare interventions, including increased contraceptive use and acceptability.^31–33 We expected reproductive empowerment to be higher for women who are digitally connected. Hence, we hypothesised that Internet use is positively associated with women's sexual agency in northern Nigeria.

Furthermore, according to social cognitive theory, people's knowledge, attitude, self-efficacy, and skills are important determinants of health promotion, which includes prevention and control of human immunodeficiency syndrome (HIV) and acquired immunodeficiency syndrome (AIDS).^34–37 This theory underlies various public health interventions using digital media to improve participants’ knowledge, attitude, self-efficacy, and skills in adopting positive behaviours, such as smoking cessation and physical activity.^38,39 A study revealed that digital public health interventions in SSA enhanced women's self-efficacy, autonomy, and involvement in decision-making, apart from fostering spousal communication and support.²² Exposure to digital health interventions caused contraceptive ideation and its use in a cluster-randomised control trial in Kaduna, a northern state in Nigeria.²³ Thus, exposure to digital family planning messages can improve women's reproductive knowledge, attitudes, and self-efficacy, leading to sexual agency. Hence, we hypothesised that exposure to digital family planning messages is positively associated with sexual agency among partnered women in northern Nigeria.

Methods

Data source

This study used a cross-sectional design. The Individual Recode (IR) file, containing women's data from the 2018 Nigeria Demographic and Health Survey (NDHS), was analysed in this study. NDHS is a quinquennial survey conducted by the National Population Commission (NPC). The overarching aim of the survey is to produce demographic and health indicators useful for policymakers and programme managers to monitor and develop public health interventions to improve population health in the country.¹³ The 2018 NDHS, while leveraging the technical support of ICF through the DHS Programme,¹³ targets women aged 15–49 years and men aged 15–59 years. The survey involved a two-stage design that combines stratified (urban and rural strata) and cluster sampling (using enumeration areas) to randomly select eligible households from each state (from the 36 states, including the Federal Capital Territory) of the federation. Information about the study, alongside the sample size computation and weighting approach, is available in the 2018 NDHS report. They are accessible at www.measuredhs.com. In the NDHS, a total of 41,821 women were interviewed.

Target population

The target population in this study were women of reproductive age (15–49 years) who were currently married or living with a man (i.e. partnered women) and resided in the north region (i.e. North Central, North East, and North West regions) of Nigeria at the time of the survey. Hence, a weighted sample of 18,753 women met the criteria for inclusion in this study.

Measures

The outcome variable in this study is sexual agency, measured with two questions about women's ability to: (a) refuse sex from their male partners and (b) ask them to use condoms. This operationalisation aligns with previous studies.^40–42 The main explanatory variables are Internet use and exposure to family planning messages via text or social media. The covariates are the socio-demographic and economic characteristics of women and of their male partners. Table 1 shows these variables alongside the data manipulation performed before the statistical analysis. Table 1 describes the control variables in this study. Women's socio-demographic and economic characteristics were age, education, employment status, phone ownership, residence, ethnicity, type of marriage, religion, region, and health facility visits. The male partner's characteristics were age, education, employment status, and desire for children. Household wealth index was also considered. The DHS Programme computed the index using principal components analysis for variables indicating household assets such as television, bicycle, car, source of drinking water, toilet facilities, and flooring materials.¹³ The socio-demographic and economic characteristics were considered control variables – whose effects were statistically controlled to estimate the unique effects of Internet use and exposure to family planning messages (via texts or social media) on sexual agency – based on contextual knowledge and recent studies identifying them as important correlates of sexual autonomy and contraceptive decision-making ability of partnered women in Nigeria and Africa.^43–45

Table 1. Description of the outcome, explanatory and control variables, variable and value labels

Types of Variables	Questions or Variable Names in the Dataset	Variable Labels and Values in the Dataset	Variable Labels and Value Labels
Outcome variables: Sexual agency
a. Ability to refuse sex	Can you say no to your (husband/partner) if you do not want to have sexual intercourse?	Yes (1), No (2) Depends/ not sure (8)
b. Ability to ask male partner to use condom	Could you ask your (husband/partner) to use a condom if you wanted him to?	Yes (1), No (2) Depends/ not sure (8)
Main explanatory variables
i. Internet use	In the last 12 months, have you used the Internet? (v171a)	Never (1), yes, last 12 months (2), and yes, before last 12 months (3)	Non-use (0); and use (1);
ii. Exposure to family planning by text messages on mobile phone	In the last few months have you: received a voice or text messages about family planning on a mobile phone? (v384d)	Yes (1), No (2)	Summed both and dichotomised as unexposed (0); and exposed (1) to family planning messages
iii. Exposure to family planning messages on social media	In the last few months have you: read/heard (about family planning) from social media? (s815e)	Yes (1), No (2)
Covariates
Personal socio-demographic and economic characteristics
Ownership of mobile phone	Do you own a mobile phone? (v169a)	Yes (1), No (2)
Age in 5-year groups	How old were you at your last birthday? (v013)	15–19 years (1), 20–24 years (2), 25–29 years (3), 30–34 years (4), 35–39 years (5), 40–44 years (6) and 45–49 years (7)	15–24 years (1), 25–29 years (2), 30–34 years (3), 35–39 years (4), 40–44 years (5), and 45–49 years (6)
Type of marriage	Including yourself, in total, how many wives or live-in partners does he (male partner) have? (v505)	No other wives (0), … , 12	Monogamous (0) and polygamous (1-12)
Highest educational level	What is the highest level of school you attended: primary, secondary, or higher? (v106)	No education (0), primary (1), secondary (2) and higher (3)
Employment status	Have you done any work in the last 12 months? (v714)	Yes (1) and No (2)	Unemployed (1) and Employed (2)
Wealth index	v190	Poorest (1), Poorer (2), Middle (3),Richer (4), and Richest (5)
Ethnicity	What is your ethnic group? (v131)	Ekoi (1), Fulani (2), Hausa (3), … , Igbo (6), Kanuri/Beri Beri (8), … , Yoruba (10), Others (96) and Don't Know (98)	Hausa/Fulani/Kanuri/Beri Beri (1), Yoruba (2), Igbo (3) and Others (4)
Religion	What is your religion? (v130)	Catholic (1), Other Christian (2), Islam (3), Traditionalist (4) and other (96).	Christianity (1, Catholic and other Christians) and Islam/other (2) (i.e. traditionalist and other)
Type of place of residence		Urban (1), rural (2)
Region	v024	North Central (1), North East (2), North West (3), South East (4), South South (5) and South West (6)	North Central (1), North East (2), and North-West (3)
Decision-maker for respondent's health care	Who usually makes the decision about health care for yourself: you. (v743a)	Respondent (1), Husband/partner (2), Joint Decision (3), Someone else (4) and Other (specify) (6)	Respondent (1), and others (0)
Health facility visit	In the last 12 months, have you visited a health facility for care for yourself? (v394)	Yes (1) and No (2)
Male partners’ socio-demographic and economic characteristics
Age	How old was your (husband/partner) on his last birthday? (v730)	15, 16. 17, … , 95	Young adult (15-35 years), middle-age (36-55 years) and older adult (56 & above)
Education level	What was the highest level of school he attended: primary, secondary, or higher? (v701)	No education (0), Primary (1), Secondary (2) and Higher (3)
Employment status	What is your (husband's/partner's) occupation? That is, what kind of work does he mainly do? (v705)	Did not work (0), professional/technical/managerial (1), clerical (2), … , unskilled manual (9), other (96) and do not know (98).	Employed (1) (any of professional/technical/managerial; clerical, … , unskilled manual and other) and Unemployed (0) (i.e. did not work). Don't know was coded as missing.
Partner's desire for children	Does your (husband/partner) want the same number of children that you want, or does he want more or fewer than you want? (v621)	Both want the same (1), husband wants more (2), husband wants fewer (3) and Don't know (8).	Desire same as partner (1), Partner desires more children (2), Partner desire fewer (3) and Don't know (8)

Given that maternal healthcare visits are associated with contraceptive use in Nassarawa, a northen state in Nigeria,⁴⁶ women’s health facility was considered a control variable which may predict women’s exposure to family planning information from health facility workers who may influence their decision to learn more about family planning from programmes on digital and non-digital media. Such exposure can shape women’s sexual decision-making and behaviours. Socio-demographic and economic variables, such as education, wealth, and urbanity, correlate with female Internet use in Nigeria.²¹ They were, alongside employment status and phone ownership, considered control variables that can predict Internet use which, in turn, can predict exposure to digital family planning messages.

Statistical analyses

Descriptive statistics (frequencies and percentages) were used to describe all the variables. Multiple multinomial logistic regressions were used to model the two polytomous (and nominal) outcome variables separately as functions of Internet use and exposure to family planning messages (via texts or social media). There was no serious collinearity among the explanatory and control variables, given that the average variance inflation factor (VIF) was 1.64. The first two rows in Table 1 present the regression of sexual agency indicators on Internet use and exposure to family planning messages via text or social media. The base outcome for both outcome variables was “unable”, coded as 0. The multinomial logistic regression is mathematically expressed as. (1) ${\displaystyle \frac{P(y=1)}{P(y=2)}=e^{X\beta (1)}}$(1) (2) ${\displaystyle \frac{P(y=3)}{P(y=2)}=e^{X\beta (3)}}$(2) p (y = 1) = The probability of ability to refuse sex or ask a male partner to use condom relative to the base outcome: inability (p (y = 2)). p (y = 3) = The probability of being uncertain about ability to refuse sex or ask a male partner to use condom relative to the base outcome. e = Approximately 2.7183. X = Vector of main explanatory variables (e.g. Internet use) and covariates. β = Vector of coefficients of main explanatory variables and covariates.

The ratios of the probabilities in equations 1 and 2 are the relative risk ratios (RRR).⁴⁷ We fitted different multiple multinomial logistic regressions. The first and second models examined the relationships between Internet use and (a) the ability to refuse sex and (b) ask a male partner to use condom. The third and fourth models examined the relationship between exposure to family planning messages (via texts or social media), and the ability to refuse sex and ask a male partner to use condom. Subsequently, we stratified each regression model by urban and rural residences. In total, 12 regressions were fitted. Before the analyses, we accounted for the complex nature of the data by using the “svyset” command (in Stata 14 package) to specify the variables for the survey weights, strata and clusters. The “svy, subpop” command was applied to restrict all analyses to the target sample (i.e. partnered women in northern Nigeria) and obtain valid standard errors.⁴⁷ The significance of the main explanatory variables, using their associated RRRs, was assessed against 5% level of significance and 95% confidence intervals. We analysed only complete cases (18,205 women, 97.08%).

Results

Descriptive analyses

Table 2 shows the percentage and frequency distributions of women by sexual agency (ability to refuse sex, ability to ask a male partner to use condoms), their socio-demographic and economic other characteristics, including their male partners’. More than one-third of women (44.6%) stated that they could refuse sex from their male partners and ask their male partners to use condoms (31.4%), respectively. Less than one-tenth of the women used the Internet (4.3%) and were exposed to family planning messages via texts or social media (1.8%). Most women were less than 35 years old (65.8%) and uneducated (64.9%). Slightly more than half of the women resided in the North-West region (52.2%) and had middle-aged (52.3%) and uneducated male partners (51.3%).

Table 2. Descriptive statistics of women and their male partners in northern Nigeria

Variables	Frequency (%)
Ability to refuse sex
No	9,776 (53.70)
Yes	8,114 (44.60)
DK/NS/Depends	316 (1.70)
Ability to ask male partner to use condom
No	12,012 (66.00)
Yes	5,708 (31.40)
DK/Not Sure/Depends	485 (2.70)
Use of Internet
Non-use	17,427 (95.70)
Use	778 (4.30)
Exposure to FP messages via sms/social media
Unexposed	17,877 (98.20)
Exposed	328 (1.80)
Ownership of mobile phone
Don't own mobile phone	11,232 (61.70)
Owned mobile phone	6,974 (38.30)
Age
15-24 years	4,986 (27.40)
25-29 years	3,852 (21.20)
30-34 years	3,131 (17.20)
35-39 years	2,612 (14.30)
40-44 years	1,935 (10.60)
45-49 years	1,690 (9.30)
Highest educational level
No education	11,824 (64.90)
Primary	2,437 (13.40)
Secondary	3,035 (16.70)
Higher	909 (5.00)
Health facility visit^$
No visit	9,485 (52.10)
Visit	8,720 (47.90)
Employment status
Unemployed	6,951 (38.18)
Employed	11,254 (61.82)
Wealth index
Poorest	5,534 (30.40)
Poorer	5,222 (28.69)
Middle	3,621 (19.89)
Richer	2,380 (13.07)
Richest	1,447 (7.95)
Ethnicity
Hausa/Fulani/Beri Beri	12,365 (67.90)
Yoruba	418 (2.30)
Igbo	170 (0.90)
Others	5,252 (28.90)
Religion
Christianity	3,220 (17.70)
Islam/ others	14,985 (82.30)
Region
North Central	3,976 (21.80)
North East	4,733 (26.00)
North West	9,496 (52.20)
Decision-maker on respondent's healthcare
Other	16,963 (93.20)
Respondent	1,242 (6.80)
Type of marriage
Monogamous	10,879 (59.75)
Polygynous	7,327 (40.25)
Place of residence
Urban	4,738 (26.00)
Rural	13,467 (74.00)
Male partner's characteristics
Partner's age
Young adult (15-35 years)	6,338 (34.80)
Middle-age (36-55 years)	9,529 (52.30)
Older (56+years)	2,338 (12.80)
Partner's level of education
No education	9,341 (51.30)
Primary	2,290 (12.60)
Secondary	4,225 (23.20)
Higher	2,349 (12.90)
Partner's employment status
Unemployed	815 (4.50)
Employed	17,390 (95.50)
Partner's desire for children
Both want same	5,295 (29.10)
Partner wants more	10,257 (56.30)
Partner wants fewer	765 (4.20)
Don't know	1,888 (10.40)

Sample size: 18,205. DK = Don't know. NS = Not sure. Depends = It depends. FP = Family planning. $ = over the previous 12 months.

Multivariate analyses

Table 3 presents the covariate-adjusted relationships between Internet use, exposure to family planning messages (via texts or social media) and ability to refuse sex and ask a male partner to use condoms. The probability of being able to refuse sex, compared to that of inability, was 52% higher for women who used the Internet than those who did not (RRR = 1.52; 95%CI = 1.150-1.999). There was 77% increase in the probability of being able to ask a male partner to use condoms, compared to that of inability, among women who used the Internet than those who did not (RRR = 1.77; 95%CI = 1.388–2.266). The chances of being able to ask a male partner to use condoms, compared to that of inability, was 79% higher for women exposed to family planning messages (via texts or social media) than the unexposed women (RRR = 1.79; 95%CI = 1.333–2.413).

Table 3. Relationships between Internet use, exposure to FP messages and sexual agency

Explanatory variables	Ability to refuse sex		Ability to ask partner to use condom
	Able vs Unable	DK/Not Sure/Depends vs Unable	Able vs Unable	DK/Not Sure/Depends vs Unable
	RRR [95% CI]	RRR [95% CI]	RRR [95% CI]	RRR [95% CI]
Use of Internet (ref = Non-user)^a	1.000	1.000	1.000	1.000
Used	1.516* [1.150,1.999]	0.822 [0.329,2.056]	1.774* [1.388,2.266]	1.001 [0.481,2.083]
Exposure to FP messages via texts/social media (ref = No exposure)^b	1.000	1.000	1.000	1.000
Exposed	1.051 [0.789,1.400]	0.291 [0.062,1.361]	1.793* [1.333,2.413]	0.838 [0.330,2.131]

a = covariates are phone ownership, age, education, employment status, wealth, residence, ethnicity, type of marriage, religion, region, health facility visit, desire for children, male partner's age, education and employment status.

b = covariates are phone ownership, Internet use, age, education, employment status, wealth, residence, ethnicity, type of marriage, religion, region, health facility visit, desire for children, male partner's age, education and employment status.

DK = Don't know. 95% confidence interval in bracket. *p<0.05. ref = Reference category. Sample size: 18,205

Urban–rural stratified multivariate analyses

According to the results in Table 4, in urban areas, the probability of being able to refuse sex, compared to that of inability, was 57% higher for women who used the Internet than those who did not (RRR = 1.57; 95%CI = 1.108–2.228). The probability of being able to ask a male partner to use condoms (relative to that of inability) was 64% higher for women who used the Internet than those who did not (RRR = 1.64; 95%CI = 1.222–2.207). The probability of being able to ask a male partner to use condoms, compared to that of inability, was 96% higher for women who were exposed to family planning messages (via texts or social media) than the unexposed women (RRR = 1.96; 95%CI = 1.321–2.897). Conversely, the probability of being uncertain (i.e. saying don’t know, not sure or that it depends) about the ability to refuse sex (relative to that of inability) was 92% lower for women exposed to family planning messages (via texts or social media) than their unexposed counterparts (RRR = 0.08; 95%CI = 0.011–0.615).

Table 4. Relationships between Internet use, exposure to FP messages and sexual agency by residence

Explanatory variables	Ability to refuse sex		Ability to ask partner to use condom
	Able Vs Unable	DK/Not Sure/Depends Vs Unable	Able Vs Unable	DK/Not Sure/Depends Vs Unable
	RRR [95% CI]	RRR [95% CI]	RRR [95% CI]	RRR [95% CI]
Urban
Use of Internet (ref = Non-user)^a	1.000	1.000	1.000	1.000
Used	1.572* [1.108,2.228]	0.535 [0.163,1.760]	1.643* [1.222,2.207]	0.604 [0.238,1.531]
Exposure to FP messages via texts/social media (ref = No exposure)^b	1.000	1.000	1.000	1.000
Exposed	1.184 [0.833,1.682]	0.081* [0.011,0.615]	1.957* [1.321,2.897]	0.401 [0.119,1.346]
Rural
Use of Internet (ref = Non-user)^a	1.000	1.000	1.000	1.000
Used	1.212 [0.758,1.938]	1.499 [0.433,5.188]	1.940* [1.235,3.047]	3.633* [1.152,11.453]
Exposure to FP messages via texts/social media (ref = No exposure)^b	1.000	1.000	1.000	1.000
Exposed	0.922 [0.567,1.500]	0.448 [0.063,3.194]	1.581* [1.004,2.491]	1.617 [0.447,5.842]

a = covariates are phone ownership, age, education, employment status, wealth, residence, ethnicity, type of marriage, religion, region, health facility visit, desire for children, male partner's age, education and employment status.

b = covariates are phone ownership, Internet use, age, education, employment status, wealth, residence, ethnicity, type of marriage, religion, region, health facility visit, desire for children, male partner's age, education and employment status.

DK = Don't know. 95% confidence interval in bracket. *p < 0.05. ref = Reference category. Sample size: 4,738 (urban) and 13,467 (rural).

In rural areas, among women who used the Internet, compared to those who did not, the probability of being able to ask a male partner to use condoms was 94% higher (RRR = 1.94; 95%CI = 1.235–3.047) compared to the probability of inability to do so. The probability of being uncertain (i.e. saying don’t know, not sure, or that it depends) about asking a male partner to use condoms, compared to that of inability, was at least three times higher for women who used the Internet than those who did not (RRR = 3.63; 95%CI = 1.152–11.453). Similarly, the probability of being able to ask a male partner to use condoms, compared to that of inability, was 58% higher for women who were exposed to family planning messages than the unexposed women (RRR = 1.58; 95%CI = 1.004–2.491).

Discussion

This study examined the extent of sexual agency, investigated the associations of Internet use and exposure to family planning messages (via texts or social media) with sexual agency, and examined these associations at the residential levels among partnered women in northern Nigeria. Less than half of the women could refuse sex (44.5%) and ask their male partners to use condoms (31.4%). These estimates are lower than those reported by Sano et al.⁴⁸ for women who could refuse sex (64%) and ask their male partners to use a condom (64% and 41%), respectively.

Internet use was positively associated with women's ability to refuse sex (rather than inability to do so) and ask their male partners to use condoms (rather than their inability to ask them to do so). These relationships may be due to their exposure to vital sexual and reproductive health information and messages on the Internet, thus empowering them to navigate unwanted sexual advances from their male partners and convince/persuade them to use condoms. Similarly, we found that women who were exposed to family planning messages (via text messages or social media) were more likely to be able to ask their male partners to use condoms (than being unable to do so). A study in India supports these explanations, and other previous studies,^24,25 that women's use of the Internet could provide them with the information needed to claim and exercise their reproductive health and rights.⁴⁹

Concerning the associations between Internet use, exposure to family planning messages (via texts or social media) and sexual agency in urban areas, this study found that women who used the Internet were more likely to be able to refuse sex and ask their male partners to use condoms. Women who were exposed to family planning messages (via texts or social media) were more likely to be able to ask their male partners to use condoms. In line with this explanation, a study in Ghana revealed that in urban areas, access to and use of mobile phones enabled young people to overcome cultural barriers to exercise their sexual rights.⁵⁵ Similar results were observed for rural areas where women who used the Internet and those exposed to family planning messages (via texts or social media) were more likely to be able to ask their male partners to use condoms. This further echoes the importance of Internet use as a gateway to a trove of sexual and reproductive health information (including information on women's reproductive empowerment) and digital family planning messages for women's sexual agency.^17,19

Nevertheless, among urban women exposed to family planning messages, compared to unexposed women, the likelihood of being uncertain about their ability to refuse sex was lower than the likelihood of their inability to refuse sex. In contrast, among women in rural areas, the likelihood of being uncertain about their ability to ask their male partners to use condoms was higher for women who used the Internet. This finding is attributable to the patriarchal culture⁵⁶ men's pronatalism and dominance in fertility decisions, and women’s pronatalism in the north.¹³ These sociocultural factors can hinder women’s involvement in sexual and reproduction-related decisions,⁵⁷ which include refusing sex with their male partners and asking their male partners to use condoms. Effective digital family planning interventions, driven by the social cognitive theory, are needed to challenge the formidable sociocultural inhibitors of women's sexual agency in urban and rural areas of northern Nigeria.

The findings from this study support the potential of digital family planning in improving sexual agency among partnered women in northern Nigeria, which can improve their sexual and reproductive health and ability to exercise their sexual and reproductive rights. More importantly, the third Sustainable Development Goal (SDG) targets universal access to sexual and reproductive health, information, and education among women of reproductive age.⁵⁸ The ninth SDG targets increased access to information and communications technology and interventions to provide universal and affordable access to the Internet in least-developed countries by 2030.⁵⁸ In addition, the fifth SDG includes ensuring universal access to sexual and reproductive health and rights.⁵⁸ Based on findings in this study, it appears that the likelihood of women's sexual agency, a factor of reproductive rights in SDG 5, may depend on achieving the targets of SDGs 3 and 9. Nevertheless, Internet use and exposure to digital family planning messages can be instrumental in ensuring partnered women's sexual agency in northern Nigeria and, by extension, their sexual and reproductive health and rights.

Limitations and Strengths

The study fills existing knowledge gaps on the roles of Internet use and exposure to family planning messages (via text or mass media) in women's sexual agency in northern Nigeria. This region is affected by insurgency, violence against women, and poor reproductive health. This study analysed the Demographic and Health Survey, which conforms to international standards and ensures that inferences are generalisable to the population of partnered women in northern Nigeria. However, the study has some limitations. The data used were cross-sectional, making it difficult to prove causal relationships between Internet use, exposure to family planning messages, and sexual agency. We did not control for contextual factors, such as Internet availability and telecommunication coverage in the women's communities, because the survey did not capture this information. These factors can affect women's Internet use and exposure to digital family planning messages. Despite the limitations, our study filled the existing knowledge gap on the roles of Internet use and digital family planning messages in partnered women’s sexual agency in northern Nigeria – a region ridden with crisis/conflicts and where most partnered women are characterised by low socio-economic empowerment, poor sexual and reproductive health, and inability to exercise their sexual and reproductive rights.

Conclusion

This study concludes that Internet use and exposure to family planning messages (via text or social media) are positively associated with sexual agency among partnered women in northern Nigeria. This result is consistent in urban and rural areas of northern Nigeria. While digital connectedness and digital family planning messages should be promoted and incorporated into women’s reproductive empowerment programmes in northern Nigeria to improve their sexual agency, such programmes should specifically address partnered women in urban areas who are exposed to family planning messages (via text or social media) but unable to refuse sex and those in rural areas who use the Internet but are uncertain about their ability to ask their male partners to use a condom.

Ethics approval and consent to participate

The 2018 Nigerian Demographic and Health Survey data were used in this study. Ethical approval was obtained from the ICF Review Board in the United States (ICF IRB FWA00000845) and the National Health Research Ethics Committee of Nigeria for the NPC to conduct the survey (NHREC/01/01/2007). Written and verbal consents were obtained from participants during the survey. The authors received the DHS Program's approval to download and use the data.

Acknowledgements

The authors thank the MEASURE DHS and the National Population Commission (Nigeria) for permitting us to download and analyse the datasets.

Disclosure statement

No potential conflict of interests was reported by the authors.

Data availability statement

The data are accessible via https://dhsprogram.com/data/dataset/Nigeria_Standard-DHS_2018.cfm?flag = 1.

Appendix

Full tables

Relationship between Internet use and sexual agency (full results).

	Ability to Refuse Sex						Ability to Ask Partner to Use Condom
	Able Vs Unable			DK/Not Sure/Depends Vs Unable			Able Vs Unable			DK/Not Sure/Depends Vs Unable
	RRR	95% CI		RRR	95% CI		RRR	95% CI		RRR	95% CI
Use of Internet (ref = Non-user)
Used	1.516	1.150	1.999	0.822	0.329	2.056	1.774	1.388	2.266	1.001	0.481	2.083
Mobile phone ownership (ref = No)
Owned mobile phone	1.261	1.134	1.402	1.624	1.233	2.139	1.348	1.185	1.533	1.701	1.329	2.177
Age (ref = 15–24 years)
25–29 years	0.925	0.811	1.053	1.136	0.827	1.560	1.055	0.926	1.202	1.044	0.781	1.396
30–34 years	1.026	0.894	1.177	1.209	0.857	1.705	1.155	0.995	1.339	1.140	0.821	1.583
35–39 years	1.014	0.862	1.192	1.505	0.798	2.840	1.032	0.853	1.249	1.343	0.902	2.002
40–44 years	1.034	0.860	1.242	1.284	0.803	2.053	1.028	0.848	1.248	0.987	0.664	1.467
45–49 years	0.927	0.755	1.137	1.221	0.701	2.128	0.732	0.580	0.924	1.308	0.830	2.061
Highest level of education (ref = No education)
Primary	1.165	1.025	1.324	0.620	0.394	0.976	1.223	1.082	1.382	0.868	0.656	1.148
Secondary	1.281	1.111	1.479	0.626	0.371	1.058	1.381	1.206	1.580	0.896	0.599	1.340
Higher	1.404	1.061	1.857	1.047	0.477	2.299	1.751	1.344	2.281	1.144	0.607	2.154
Employment status (ref = Unemployed)
Employed	1.018	0.911	1.137	1.249	0.929	1.679	0.976	0.868	1.096	1.271	0.975	1.657
Wealth quintile (ref = Poorest)
Poorer	1.039	0.909	1.189	1.200	0.716	2.012	1.170	1.005	1.362	0.819	0.569	1.180
Middle	1.311	1.065	1.613	1.164	0.614	2.206	1.553	1.235	1.952	0.797	0.529	1.202
Richer	1.299	1.041	1.621	1.302	0.717	2.367	1.958	1.495	2.565	1.184	0.720	1.948
Richest	1.495	1.116	2.003	0.902	0.381	2.135	1.829	1.325	2.524	0.631	0.338	1.178
Residence (ref = Rural)
Urban	1.143	0.971	1.345	0.731	0.450	1.188	1.127	0.942	1.349	1.086	0.748	1.578
Ethnicity (ref = Hausa/Fulani/Beri Beri)
Yoruba	4.088	2.912	5.738	2.105	0.760	5.827	6.270	4.534	8.672	4.976	2.071	11.957
Igbo	1.105	0.687	1.778	0.327	0.037	2.878	1.220	0.674	2.209	2.939	0.936	9.232
Other	1.132	0.948	1.352	1.237	0.708	2.160	1.215	1.007	1.466	1.586	1.035	2.430
Religion (ref = Islam and others)
Christian	2.539	2.123	3.036	0.928	0.561	1.534	2.719	2.218	3.333	1.327	0.823	2.141
Region (ref = North East)
North Central	0.483	0.411	0.568	0.308	0.175	0.543	0.558	0.466	0.668	0.210	0.133	0.331
North West	0.564	0.484	0.657	0.501	0.290	0.866	0.793	0.672	0.937	0.505	0.357	0.714
Decision-maker on respondent's healthcare (ref = Other)
Respondent	0.964	0.785	1.184	0.170	0.062	0.466	1.090	0.873	1.362	0.511	0.262	0.996
Health Facility Visit (ref = No visit)
Visit	1.051	0.952	1.159	1.021	0.663	1.571	0.942	0.846	1.050	0.888	0.699	1.127
Partner's age (ref = Young adult (15–35 years))
Middle-age (36–55 years)	1.029	0.915	1.158	0.768	0.490	1.203	0.950	0.841	1.074	0.934	0.703	1.242
Older (56 + years)	1.134	0.954	1.348	0.702	0.360	1.372	0.844	0.684	1.043	0.692	0.455	1.052
Partner's level of education (ref = No education)
Primary	1.176	1.032	1.340	1.561	0.882	2.762	1.043	0.883	1.233	0.850	0.591	1.222
Secondary	1.260	1.098	1.447	0.927	0.603	1.425	1.178	1.002	1.385	0.706	0.498	1.001
Higher	1.324	1.097	1.599	1.581	0.912	2.741	1.202	0.992	1.456	1.135	0.797	1.616
Partner's employment status (ref = Unemployed)
Employed	0.518	0.397	0.677	4.925	1.614	15.022	0.356	0.274	0.462	4.009	1.734	9.270
Type of marriage (ref = Monogamous)
Polygynous	0.963	0.883	1.050	0.846	0.591	1.212	0.890	0.790	1.003	0.860	0.665	1.112
Partner's desire for children (ref = both want same)
Partner wants more	0.740	0.661	0.828	1.480	0.951	2.302	0.672	0.602	0.750	1.337	0.982	1.821
Partner wants fewer	1.865	1.454	2.393	5.382	2.169	13.357	1.835	1.423	2.367	3.626	1.956	6.721
Don't know	0.801	0.657	0.977	8.837	5.567	14.029	0.855	0.692	1.056	10.935	7.615	15.705

DK = Don’t know. CI = Confidence intervals. Sample = 18,205. ref = Reference category. RRR = Relative Risk Ratio.

Relationship between exposure to digital family planning messages and sexual agency (full results).

	Ability to Refuse Sex						Ability to Ask Partner to Use Condom
	Able Vs Unable			DK/Not Sure/Depends Vs Unable			Able Vs Unable			DK/Not Sure/Depends Vs Unable
	RRR	95% CI		RRR	95% CI		RRR	95% CI		RRR	95% CI
Exposure to FP messages via texts/social media (ref = No exposure)
More Exposed	1.051	0.789	1.400	0.291	0.062	1.361	1.793	1.333	2.413	0.838	0.330	2.132
Use of Internet (ref = Non-user)
Used	1.510	1.146	1.990	0.936	0.367	2.385	1.700	1.328	2.177	1.052	0.498	2.223
Mobile phone ownership (ref = No)
Owned mobile phone	1.260	1.133	1.401	1.638	1.243	2.157	1.339	1.177	1.522	1.702	1.329	2.178
Age (ref = 15–24 years)
25–29 years	0.924	0.811	1.053	1.140	0.830	1.565	1.054	0.925	1.200	1.044	0.781	1.396
30–34 years	1.025	0.894	1.176	1.212	0.860	1.709	1.152	0.994	1.337	1.138	0.820	1.580
35–39 years	1.013	0.862	1.192	1.509	0.798	2.853	1.030	0.851	1.246	1.341	0.900	1.997
40–44 years	1.033	0.860	1.242	1.294	0.810	2.066	1.023	0.843	1.241	0.990	0.666	1.471
45–49 years	0.927	0.755	1.137	1.222	0.701	2.130	0.731	0.580	0.922	1.306	0.829	2.058
Highest level of education (ref = No education)
Primary	1.165	1.025	1.325	0.618	0.393	0.973	1.227	1.086	1.387	0.867	0.655	1.148
Secondary	1.282	1.111	1.479	0.629	0.372	1.062	1.382	1.208	1.581	0.899	0.602	1.343
Higher	1.399	1.055	1.855	1.086	0.500	2.361	1.702	1.304	2.221	1.148	0.612	2.154
Employment status (ref = Unemployed)
Employed	1.018	0.911	1.137	1.250	0.929	1.681	0.977	0.870	1.098	1.272	0.976	1.658
Wealth quintile (ref = Poorest)
Poorer	1.039	0.909	1.189	1.201	0.717	2.013	1.169	1.004	1.360	0.819	0.568	1.179
Middle	1.311	1.065	1.613	1.162	0.613	2.206	1.553	1.235	1.953	0.795	0.527	1.199
Richer	1.298	1.040	1.621	1.308	0.720	2.375	1.949	1.487	2.554	1.187	0.723	1.948
Richest	1.494	1.115	2.002	0.921	0.390	2.176	1.817	1.317	2.507	0.631	0.339	1.175
Residence (ref = Rural)
Urban	1.142	0.970	1.345	0.732	0.450	1.191	1.124	0.939	1.346	1.087	0.749	1.580
Ethnicity (ref = Hausa/Fulani/Beri Beri)
Yoruba	4.093	2.913	5.751	2.070	0.753	5.694	6.385	4.611	8.841	4.956	2.056	11.943
Igbo	1.104	0.687	1.775	0.355	0.041	3.071	1.207	0.657	2.216	3.035	0.960	9.595
Other	1.132	0.948	1.353	1.230	0.704	2.149	1.220	1.012	1.472	1.586	1.035	2.429
Religion (ref = Islam and others)
Christian	2.539	2.124	3.036	0.924	0.559	1.526	2.729	2.225	3.346	1.328	0.823	2.142
Region (ref = North East)
North Central	0.483	0.411	0.568	0.309	0.175	0.545	0.556	0.465	0.664	0.210	0.133	0.330
North West	0.564	0.484	0.657	0.498	0.288	0.861	0.800	0.678	0.945	0.504	0.355	0.713
Decision-maker on respondent's healthcare (ref = Other)
Respondent	0.965	0.786	1.185	0.169	0.062	0.464	1.100	0.881	1.373	0.513	0.264	0.999
Health Facility Visit (ref = No visit)
Visit	1.050	0.952	1.159	1.027	0.667	1.582	0.939	0.843	1.046	0.891	0.702	1.130
Partner's age (ref = Young adult (15–35 years))
Middle-age (36–55 years)	1.030	0.916	1.158	0.765	0.488	1.200	0.953	0.843	1.077	0.935	0.703	1.243
Older (56 + years)	1.134	0.954	1.348	0.699	0.358	1.367	0.847	0.685	1.047	0.691	0.454	1.051
Partner's level of education (ref = No education)
Primary	1.176	1.032	1.340	1.562	0.883	2.765	1.042	0.882	1.232	0.850	0.591	1.223
Secondary	1.261	1.098	1.448	0.928	0.604	1.426	1.179	1.003	1.386	0.704	0.496	0.998
Higher	1.323	1.095	1.598	1.595	0.923	2.757	1.187	0.979	1.439	1.136	0.799	1.617
Partner's employment status (ref = Unemployed)
Employed	0.518	0.397	0.677	4.929	1.613	15.060	0.356	0.274	0.462	4.014	1.735	9.288
Type of marriage (ref = Monogamous)
Polygynous	0.963	0.883	1.050	0.845	0.590	1.211	0.890	0.790	1.003	0.860	0.664	1.112
Partner's desire for children (ref = both want same)
Partner wants more	0.740	0.662	0.828	1.475	0.947	2.297	0.674	0.604	0.752	1.337	0.981	1.821
Partner wants fewer	1.866	1.454	2.395	5.337	2.154	13.223	1.844	1.428	2.379	3.615	1.950	6.702
Don't know	0.801	0.657	0.977	8.826	5.555	14.022	0.854	0.691	1.056	10.962	7.628	15.753

DK = Don’t know. CI = Confidence intervals. Sample = 18,205. ref = Reference category. RRR = Relative Risk Ratio.

Relationship between Internet use and sexual agency in urban areas (full results).

	Ability to Refuse Sex						Ability to Ask Partner to Use Condom
	Able Vs Unable			DK/Not Sure/Depends Vs Unable			Able Vs Unable			DK/Not Sure/Depends Vs Unable
	RRR	95% CI		RRR	95% CI		RRR	95% CI		RRR	95% CI
Use of Internet (ref = Non-user)
Used	1.572	1.108	2.228	0.535	0.163	1.760	1.643	1.222	2.207	0.604	0.238	1.531
Mobile phone ownership (ref = No)
Owned mobile phone	1.072	0.896	1.283	2.898	1.605	5.231	1.408	1.157	1.713	3.107	2.034	4.747
Age (ref = 15–24 years)
25–29 years	1.216	0.931	1.588	1.899	0.773	4.668	1.246	0.965	1.609	1.507	0.748	3.035
30–34 years	1.254	0.900	1.746	2.235	0.965	5.176	1.411	1.017	1.956	2.245	1.060	4.755
35–39 years	1.346	0.944	1.919	3.182	1.345	7.531	1.198	0.829	1.730	2.051	1.022	4.116
40–44 years	1.390	0.941	2.054	3.146	1.016	9.743	1.234	0.830	1.834	1.674	0.717	3.910
45–49 years	1.283	0.874	1.881	2.861	0.921	8.888	0.898	0.591	1.366	3.063	1.361	6.892
Highest level of education (ref = No education)
Primary	1.142	0.909	1.434	0.995	0.504	1.965	1.126	0.892	1.422	0.645	0.357	1.165
Secondary	1.495	1.201	1.861	0.637	0.252	1.614	1.364	1.088	1.710	0.659	0.329	1.321
Higher	1.745	1.182	2.574	0.782	0.303	2.017	1.760	1.211	2.557	1.046	0.521	2.101
Employment status (ref = Unemployed)
Employed	1.121	0.910	1.380	1.197	0.672	2.135	1.130	0.941	1.357	1.327	0.851	2.072
Wealth quintile (ref = Poorest)
Poorer	1.356	0.903	2.036	0.821	0.167	4.034	0.974	0.591	1.604	0.190	0.048	0.762
Middle	1.548	1.032	2.322	0.479	0.128	1.801	1.346	0.802	2.257	0.642	0.237	1.743
Richer	1.602	1.079	2.378	0.714	0.191	2.667	1.540	0.913	2.596	0.649	0.241	1.747
Richest	1.817	1.165	2.834	0.442	0.108	1.810	1.448	0.825	2.542	0.338	0.119	0.957
Ethnicity (ref = Hausa/Fulani/Beri Beri)
Yoruba	4.466	2.993	6.663	1.342	0.345	5.213	6.519	4.332	9.811	3.197	1.057	9.671
Igbo	1.571	0.891	2.767	0.352	0.030	4.180	2.156	0.924	5.032	2.325	0.533	10.137
Other	1.437	1.032	2.001	1.807	0.674	4.842	1.495	1.085	2.061	1.778	0.740	4.271
Religion (ref = Islam and others)
Christian	1.677	1.200	2.342	1.134	0.465	2.766	1.789	1.254	2.553	1.285	0.597	2.767
Region (ref = North East)
North Central	0.494	0.370	0.661	0.386	0.171	0.872	0.602	0.440	0.823	0.369	0.170	0.802
North West	0.640	0.458	0.894	0.419	0.192	0.915	0.804	0.581	1.112	0.545	0.308	0.963
Decision-maker on respondent's healthcare (ref = Other)
Respondent	1.064	0.689	1.643	0.173	0.022	1.345	1.113	0.690	1.796	1.021	0.306	3.412
Health Facility Visit (ref = No visit)
Visit	1.090	0.869	1.369	1.154	0.609	2.189	0.940	0.725	1.219	0.925	0.540	1.583
Partner's age (ref = Young adult (15–35 years))
Middle-age (36–55 years)	1.067	0.825	1.379	0.576	0.278	1.193	0.925	0.724	1.181	0.861	0.503	1.474
Older (56 + years)	1.329	0.945	1.870	0.307	0.101	0.933	0.861	0.590	1.257	0.430	0.187	0.991
Partner's level of education (ref = No education)
Primary	0.983	0.717	1.346	1.137	0.395	3.276	0.863	0.635	1.171	0.895	0.435	1.841
Secondary	1.093	0.832	1.434	1.508	0.612	3.712	1.078	0.824	1.412	1.216	0.583	2.537
Higher	1.184	0.855	1.640	1.597	0.587	4.344	1.305	0.985	1.729	1.416	0.687	2.920
Partner's employment status (ref = Unemployed)
Employed	0.599	0.341	1.052	5.650	0.597	53.492	0.512	0.325	0.807	12.453	1.442	107.568
Type of marriage (ref = Monogamous)
Polygynous	0.797	0.663	0.958	1.010	0.548	1.864	0.815	0.666	0.997	0.928	0.576	1.494
Partner's desire for children (ref = both want same)
Partner wants more	0.811	0.649	1.015	0.722	0.380	1.373	0.653	0.527	0.810	1.099	0.610	1.980
Partner wants fewer	2.119	1.259	3.567	1.487	0.407	5.424	2.190	1.350	3.552	3.302	1.062	10.261
Don't know	0.674	0.499	0.910	5.735	2.521	13.048	0.745	0.561	0.987	9.053	4.482	18.286

DK = Don’t know. CI = Confidence intervals. Sample = 4,738. ref = Reference category. RRR = Relative Risk Ratio.

Relationship between Internet use and sexual agency in rural areas (full results).

	Ability to Refuse Sex						Ability to Ask Partner to Use Condom
	Able Vs Unable			DK/Not Sure/Depends Vs Unable			Able Vs Unable			DK/Not Sure/Depends Vs Unable
	RRR	95% CI		RRR	95% CI		RRR	95% CI		RRR	95% CI
Use of Internet (ref = Non-user)
Used	1.212	0.758	1.938	1.499	0.433	5.188	1.940	1.235	3.048	3.633	1.152	11.453
Mobile phone ownership (ref = No)
Owned mobile phone	1.344	1.181	1.530	1.430	1.017	2.009	1.333	1.134	1.567	1.427	1.046	1.948
Age (ref = 15–24 years)
25–29 years	0.865	0.744	1.006	1.004	0.724	1.392	1.010	0.867	1.176	0.959	0.702	1.309
30–34 years	0.982	0.849	1.135	1.036	0.711	1.510	1.089	0.924	1.284	0.942	0.655	1.355
35–39 years	0.933	0.779	1.119	1.249	0.568	2.748	0.985	0.784	1.237	1.186	0.730	1.926
40–44 years	0.952	0.774	1.171	1.022	0.609	1.715	0.966	0.776	1.203	0.839	0.532	1.323
45–49 years	0.850	0.667	1.083	0.974	0.525	1.807	0.686	0.518	0.908	0.942	0.553	1.607
Highest level of education (ref = No education)
Primary	1.173	1.004	1.369	0.525	0.304	0.906	1.264	1.093	1.462	0.955	0.697	1.310
Secondary	1.201	0.997	1.447	0.601	0.318	1.136	1.410	1.193	1.668	0.978	0.599	1.596
Higher	0.975	0.654	1.454	1.247	0.356	4.361	1.457	0.994	2.136	0.585	0.162	2.118
Employment status (ref = Unemployed)
Employed	0.970	0.848	1.109	1.271	0.911	1.775	0.914	0.789	1.060	1.311	0.954	1.801
Wealth quintile (ref = Poorest)
Poorer	1.004	0.869	1.162	1.258	0.742	2.135	1.192	1.013	1.403	0.946	0.650	1.376
Middle	1.302	1.012	1.674	1.372	0.700	2.692	1.590	1.215	2.081	0.746	0.454	1.225
Richer	1.315	1.002	1.725	1.021	0.411	2.539	2.303	1.648	3.217	1.365	0.661	2.819
Richest	1.303	0.849	2.001	0.909	0.163	5.068	1.814	1.125	2.925	0.504	0.164	1.552
Ethnicity (ref = Hausa/Fulani/Beri Beri)
Yoruba	4.077	2.187	7.599	4.579	0.708	29.610	4.651	2.480	8.723	6.685	1.437	31.110
Igbo	1.069	0.459	2.488	0.000	0.000	0.000	0.732	0.410	1.308	3.765	0.328	43.264
Other	1.066	0.865	1.313	1.159	0.608	2.211	1.136	0.907	1.422	1.545	0.959	2.486
Religion (ref = Islam and others)
Christian	2.826	2.302	3.469	0.879	0.481	1.604	3.108	2.429	3.976	1.469	0.823	2.621
Region (ref = North East)
North Central	0.491	0.404	0.596	0.272	0.128	0.575	0.538	0.431	0.673	0.160	0.088	0.290
North West	0.536	0.452	0.636	0.517	0.278	0.961	0.774	0.638	0.938	0.499	0.327	0.761
Decision-maker on respondent's healthcare (ref = Other)
Respondent	0.902	0.716	1.137	0.167	0.052	0.535	1.077	0.853	1.360	0.356	0.178	0.713
Health Facility Visit (ref = No visit)
Visit	1.037	0.932	1.152	1.005	0.610	1.658	0.942	0.840	1.056	0.880	0.672	1.152
Partner's age (ref = Young adult (15–35 years))
Middle-age (36-55 years)	1.029	0.903	1.173	0.839	0.493	1.426	0.969	0.841	1.116	0.976	0.701	1.360
Older (56 + years)	1.063	0.866	1.305	0.882	0.409	1.901	0.843	0.649	1.096	0.866	0.535	1.402
Partner's level of education (ref = No education)
Primary	1.211	1.048	1.398	1.689	0.924	3.087	1.077	0.882	1.316	0.843	0.563	1.261
Secondary	1.305	1.109	1.535	0.736	0.433	1.252	1.210	0.991	1.477	0.529	0.358	0.782
Higher	1.398	1.104	1.770	1.766	0.895	3.484	1.097	0.843	1.427	1.235	0.815	1.870
Partner's employment status (ref = Unemployed)
Employed	0.481	0.359	0.646	4.868	1.341	17.673	0.300	0.222	0.406	3.001	1.205	7.471
Type of marriage (ref = Monogamous)
Polygynous	1.022	0.928	1.125	0.809	0.535	1.224	0.916	0.791	1.060	0.839	0.621	1.133
Partner's desire for children (ref = both want same)
Partner wants more	0.720	0.632	0.820	1.847	1.058	3.226	0.678	0.596	0.771	1.442	0.995	2.089
Partner wants fewer	1.772	1.355	2.317	8.088	2.808	23.301	1.677	1.251	2.250	3.956	1.875	8.348
Don't know	0.861	0.672	1.102	10.360	5.867	18.295	0.903	0.687	1.186	11.846	7.770	18.059

DK = Don’t know. CI = Confidence intervals. Sample = 4,738. ref = Reference category. RRR = Relative Risk Ratio.

Relationship between exposure to digital family planning messages and sexual agency in urban areas (full results).

	Ability to Refuse Sex						Ability to Ask Partner to Use Condom
	Able Vs Unable			DK/Not Sure/Depends Vs Unable			Able Vs Unable			DK/Not Sure/Depends Vs Unable
	RRR	95% CI		RRR	95% CI		RRR	95% CI		RRR	95% CI
Exposure to FP messages via texts/social media (ref = No exposure)
More Exposed	1.184	0.833	1.682	0.081	0.011	0.615	1.957	1.321	2.897	0.401	0.119	1.346
Use of Internet (ref = Non-user)
Used	1.550	1.095	2.194	0.603	0.180	2.022	1.560	1.160	2.097	0.664	0.258	1.708
Mobile phone ownership (ref = No)
Owned mobile phone	1.071	0.895	1.281	2.894	1.608	5.208	1.401	1.152	1.703	3.106	2.037	4.737
Age (ref = 15-24 years)
25–29 years	1.214	0.929	1.586	1.913	0.775	4.723	1.240	0.959	1.604	1.491	0.742	2.995
30–34 years	1.253	0.900	1.745	2.203	0.948	5.122	1.410	1.016	1.956	2.198	1.039	4.651
35–39 years	1.345	0.943	1.917	3.157	1.338	7.446	1.193	0.825	1.724	2.009	1.000	4.037
40–44 years	1.388	0.939	2.051	3.212	1.036	9.957	1.226	0.825	1.822	1.692	0.734	3.901
45–49 years	1.283	0.875	1.882	2.824	0.902	8.849	0.901	0.593	1.369	2.993	1.334	6.713
Highest level of education (ref = No education)
Primary	1.145	0.911	1.438	0.986	0.498	1.953	1.137	0.900	1.435	0.640	0.354	1.158
Secondary	1.496	1.202	1.862	0.656	0.259	1.661	1.366	1.090	1.711	0.669	0.334	1.343
Higher	1.733	1.172	2.563	0.834	0.334	2.081	1.724	1.183	2.513	1.055	0.530	2.100
Employment status (ref = Unemployed)
Employed	1.120	0.910	1.379	1.212	0.675	2.173	1.128	0.941	1.352	1.335	0.855	2.085
Wealth quintile (ref = Poorest)
Poorer	1.355	0.903	2.035	0.810	0.164	4.000	0.971	0.590	1.599	0.188	0.047	0.756
Middle	1.546	1.031	2.318	0.475	0.126	1.789	1.339	0.798	2.245	0.638	0.235	1.732
Richer	1.598	1.076	2.372	0.725	0.194	2.706	1.521	0.902	2.563	0.656	0.244	1.762
Richest	1.812	1.162	2.826	0.458	0.112	1.880	1.431	0.815	2.510	0.342	0.121	0.965
Ethnicity (ref = Hausa/Fulani/Beri Beri)
Yoruba	4.486	2.999	6.711	1.321	0.341	5.118	6.660	4.415	####	3.162	1.039	9.622
Igbo	1.570	0.890	2.768	0.359	0.030	4.308	2.158	0.912	5.109	2.370	0.516	10.881
Other	1.438	1.032	2.003	1.827	0.688	4.849	1.499	1.084	2.072	1.782	0.743	4.274
Religion (ref = Islam and others)
Christian	1.676	1.199	2.343	1.141	0.470	2.773	1.791	1.250	2.565	1.304	0.608	2.797
Region (ref = North East)
North Central	0.494	0.370	0.661	0.376	0.166	0.850	0.602	0.442	0.822	0.360	0.165	0.785
North West	0.643	0.460	0.899	0.407	0.187	0.884	0.822	0.594	1.136	0.532	0.301	0.941
Decision-maker on respondent's healthcare (ref = Other)
Respondent	1.068	0.692	1.650	0.177	0.023	1.374	1.133	0.703	1.825	1.047	0.312	3.508
Health Facility Visit (ref = No visit)
Visit	1.088	0.867	1.366	1.182	0.627	2.226	0.932	0.718	1.209	0.940	0.552	1.601
Partner's age (ref = Young adult (15–35 years))
Middle-age (36–55 years)	1.068	0.827	1.381	0.584	0.281	1.216	0.929	0.727	1.187	0.872	0.510	1.491
Older (56 + years)	1.328	0.944	1.870	0.309	0.101	0.948	0.856	0.585	1.253	0.432	0.187	1.001
Partner's level of education (ref = No education)
Primary	0.981	0.716	1.345	1.155	0.397	3.363	0.859	0.632	1.166	0.901	0.436	1.864
Secondary	1.095	0.834	1.437	1.469	0.599	3.600	1.087	0.831	1.422	1.192	0.571	2.489
Higher	1.179	0.850	1.635	1.596	0.586	4.345	1.282	0.967	1.700	1.414	0.686	2.912
Partner's employment status (ref = Unemployed)
Employed	0.599	0.341	1.054	5.379	0.560	51.660	0.515	0.325	0.816	12.288	1.415	106.681
Type of marriage (ref = Monogamous)
Polygynous	0.797	0.663	0.959	0.991	0.533	1.844	0.817	0.668	0.998	0.923	0.570	1.494
Partner's desire for children (ref = both want same)
Partner wants more	0.814	0.651	1.017	0.709	0.374	1.343	0.659	0.532	0.817	1.085	0.603	1.954
Partner wants fewer	2.127	1.263	3.581	1.438	0.394	5.249	2.219	1.366	3.604	3.219	1.035	10.017
Don't know	0.673	0.497	0.910	5.707	2.521	12.919	0.739	0.555	0.984	9.089	4.480	18.441

DK = Don’t know. CI = Confidence intervals. Sample = 13,467. ref = Reference category. RRR = Relative Risk Ratio.

Relationship between exposure to digital family planning messages and sexual agency in rural areas (full results).

	Ability to Refuse Sex						Ability to Ask Partner to Use Condom
	Able Vs Unable			DK/Not Sure/Depends Vs Unable			Able Vs Unable			DK/Not Sure/Depends Vs Unable
	RRR	95% CI		RRR	95% CI		RRR	95% CI		RRR	95% CI
Exposure to FP messages via texts/social media (ref = No exposure)
More Exposed	0.922	0.567	1.500	0.448	0.063	3.194	1.581	1.004	2.491	1.617	0.447	5.842
Use of Internet (ref = Non-user)
Used	1.220	0.762	1.952	1.663	0.451	6.129	1.889	1.205	2.962	3.497	1.059	11.549
Mobile phone ownership (ref = No)
Owned mobile phone	1.346	1.182	1.532	1.441	1.024	2.028	1.325	1.127	1.558	1.418	1.038	1.939
Age (ref = 15–24 years)
25–29 years	0.865	0.744	1.006	1.006	0.725	1.394	1.010	0.868	1.177	0.957	0.700	1.308
30–34 years	0.982	0.849	1.135	1.040	0.714	1.516	1.087	0.923	1.281	0.940	0.654	1.353
35–39 years	0.933	0.779	1.119	1.250	0.568	2.754	0.984	0.784	1.236	1.184	0.729	1.923
40–44 years	0.953	0.774	1.172	1.028	0.614	1.723	0.962	0.772	1.198	0.835	0.529	1.318
45–49 years	0.850	0.667	1.083	0.978	0.527	1.815	0.684	0.517	0.906	0.940	0.551	1.604
Highest level of education (ref = No education)
Primary	1.172	1.004	1.369	0.524	0.303	0.905	1.266	1.095	1.464	0.956	0.697	1.311
Secondary	1.201	0.997	1.447	0.599	0.316	1.135	1.411	1.194	1.667	0.978	0.599	1.596
Higher	0.982	0.656	1.469	1.311	0.378	4.543	1.407	0.958	2.066	0.564	0.156	2.035
Employment status (ref = Unemployed)
Employed	0.969	0.848	1.108	1.271	0.910	1.775	0.916	0.791	1.062	1.314	0.956	1.804
Wealth quintile (ref = Poorest)
Poorer	1.005	0.869	1.162	1.259	0.742	2.134	1.191	1.012	1.402	0.945	0.650	1.376
Middle	1.302	1.012	1.674	1.371	0.699	2.692	1.592	1.216	2.083	0.747	0.455	1.228
Richer	1.315	1.002	1.725	1.020	0.410	2.537	2.302	1.648	3.217	1.368	0.659	2.837
Richest	1.306	0.850	2.008	0.940	0.167	5.285	1.791	1.109	2.891	0.496	0.159	1.551
Ethnicity (ref = Hausa/Fulani/Beri Beri)
Yoruba	4.065	2.180	7.579	4.425	0.675	28.994	4.729	2.523	8.861	6.808	1.471	31.508
Igbo	1.076	0.459	2.524	0.000	0.000	0.000	0.709	0.392	1.284	3.601	0.301	43.035
Other	1.065	0.865	1.312	1.156	0.606	2.206	1.139	0.911	1.425	1.547	0.961	2.488
Religion (ref = Islam and others)
Christian	2.824	2.300	3.467	0.875	0.479	1.597	3.120	2.439	3.991	1.476	0.827	2.635
Region (ref = North East)
North Central	0.491	0.404	0.596	0.272	0.128	0.575	0.536	0.430	0.669	0.159	0.088	0.289
North West	0.536	0.452	0.636	0.516	0.277	0.959	0.776	0.640	0.940	0.500	0.328	0.763
Decision-maker on respondent's healthcare (ref = Other)
Respondent	0.902	0.716	1.136	0.165	0.051	0.532	1.081	0.856	1.365	0.357	0.178	0.715
Health Facility Visit (ref = No visit)
Visit	1.037	0.932	1.153	1.007	0.611	1.660	0.941	0.839	1.054	0.878	0.671	1.149
Partner's age (ref = Young adult (15–35 years))
Middle-age (36–55 years)	1.029	0.903	1.172	0.837	0.492	1.423	0.970	0.842	1.118	0.978	0.701	1.364
Older (56 + years)	1.062	0.866	1.304	0.877	0.406	1.893	0.847	0.652	1.101	0.870	0.536	1.410
Partner's level of education (ref = No education)
Primary	1.211	1.048	1.398	1.690	0.924	3.090	1.077	0.882	1.315	0.842	0.563	1.261
Secondary	1.305	1.109	1.536	0.740	0.435	1.258	1.208	0.989	1.475	0.527	0.356	0.780
Higher	1.399	1.105	1.771	1.767	0.896	3.487	1.090	0.837	1.421	1.228	0.811	1.861
Partner's employment status (ref = Unemployed)
Employed	0.481	0.359	0.646	4.896	1.346	17.813	0.300	0.221	0.405	2.994	1.203	7.454
Type of marriage (ref = Monogamous)
Polygynous	1.022	0.928	1.125	0.810	0.536	1.225	0.915	0.791	1.060	0.838	0.621	1.131
Partner's desire for children (ref = both want same)
Partner wants more	0.720	0.632	0.820	1.847	1.057	3.230	0.678	0.596	0.772	1.442	0.996	2.089
Partner wants fewer	1.772	1.355	2.316	8.076	2.811	23.202	1.678	1.251	2.250	3.957	1.874	8.353
Don't know	0.861	0.672	1.102	10.366	5.862	18.333	0.904	0.688	1.188	11.856	7.779	18.069

DK = Don’t know. CI = Confidence intervals. Sample = 13,467. ref = Reference category. RRR = Relative Risk Ratio.