Parent-Adolescent Child Concordance in Social Norms related to Gender Equity in Marriage- Findings from Rural India

Abstract

The purpose of this exploratory study was to assess parent-adolescent child concordance on social norms related to gender equity in marriage in rural Maharashtra, India. Survey data on marital norms related to girl’s marital age and choice, contraception, and marital violence (MV) were collected from unmarried adolescents (n=113 girls, 116 boys) and their parents (n=227 mothers, 203 fathers). Concordance was assessed using a Cohen’s unweighted Kappa statistic, with analyses stratified by sex of parent and child. Analyses revealed fair (K=.25-.27) mother-daughter concordance on girls’ right to choose when to marry, contraception use, and acceptability of MV. Father-son concordance was seen on girls’ right to choose when (K=.22, slight) and who (K=.20, fair) to marry and MV acceptability (K=.53, moderate). No opposite sex parent-child concordance was revealed. Results indicate same but not opposite sex parent-child concordance on gender equity social norms related to marriage, suggesting same sex transfer of these norms.

Introduction

Girl child marriage, non-use of contraception by young couples, and violence in marital relationships are intersecting issues that have been implicated in driving ongoing high rates of child mortality in India (Raj, McDougal, & Rusch, 2014; Raj, Saggurti, Lawrence, et al., 2010; Raj, Saggurti, Balaiah, & Silverman, 2009; Silverman, Decker, Cheng, et al., 2011; IIPS & Macro International, 2007). Vulnerabilities for all of these concerns are greater in rural relative to urban India (IIPS & Macro International, 2007). Altering adolescent social norms regarding acceptability of these practices is an important mechanism through which change could occur on these important health issues (Santillan, 2013; McCleary-Sills, 2013). Prior research documents the role of intergenerational transmission of social norms related to marriage and family, including those regarding spousal violence, age at first birth, and family size, and how these related to gender roles and expectations (Farre & Vella, 2009; Axinn, Clarkberg & Thorton, 1994; Barber, 2001). Much of this work has focused on transfer of norms from mother to child, documenting the role of mothers in affecting normative practices of both daughters and sons (Barber, 2001). Less of this work has focused on age at marriage, girl marital choice, and marital contraception use, or on the intergenerational transfer of gender and marital norms from fathers to children. Almost all of this research has focused on normative practices, rather than the normative beliefs that may be a better guide for understanding unmarried adolescents in low and middle income countries, where there can be stronger traditional normative beliefs regarding marriage, sex and fertility. This study seeks to understand parent and adolescent child social normative beliefs on these issues and their concordance to guide consideration of social norm change approaches regarding girl child marriage, marital contraception use, and marital violence in rural India.

Methods

Survey data were collected via mobile devices with non-sibling unmarried girls (n=113) and boys (n=116) not of the same household, and their parents (n=227 mothers and 203 fathers). At least one parent was included for all youth; 87.8% included both parents. Participants were recruited via convenience sampling from January to March 2013 from thirteen villages in rural Maharashtra. These villages were selected based on their attachment to two public health sub-centers that served to support a pilot study for a larger randomized controlled trial of a family planning intervention for adults; as part of pilot study efforts, relationships with village leaders were established, which facilitated the undertaking of this supplemental study, as well. The pilot villages were selected based on their high rates of adolescent childbirth and non-use of contraception. Convenience sampling involved referrals and introductions from community leaders to eligible youth and families. Eligible youth were age 14-16 for girls and 16-21 for boys (selected for age prior to typical age at marriage) and parental consent for youth participation.

Participants were recruited from households with identified eligible youth. Notably, parent-provided ages of youth were not always consistent with ages reported by youth on the survey, resulting in girl self-reported ages ranging from 13-16 years and boy self-reported ages ranging from 15-20 years. Lack of birth registry data for many youth in the area impedes reliable age data Additionally, because it is not culturally normative to celebrate birthdays in region, there is not a focus on age that supports its accurate recall by youth or parents. As there was no incentive for study participation, it is unlikely parents misrepresented youth data with the goal of study inclusion. No participants identified as eligible refused participation.

Trained research staff obtained written consent from parents for both youth and parent participation prior to survey implementation. Brief (30 minutes for youth, 20 minutes for adults) one-on-one interviewer-administered surveys were conducted in Marathi by sex-matched research staff in a private setting. No identifiers were placed on surveys. Procedures were approved by the Institutional Review Boards of University of California at San Diego and the National Institute for Research in Reproductive Health, Indian Council of Medical Research.

Survey measures included assessments of demographic characteristics and beliefs regarding marriage and family life. Single item measures on attitudes toward girl child marriage and marital choice, as well as attitude toward contraception use in marriage, were created for this exploratory study. Due to lack of validated measures on these issues, these items were developed by the co-investigator team using guidance from their prior qualitative research in the region and testing created items via cognitive interviewing with 20 participants from the target population. We also included the 5-item DHS measure on attitudes toward acceptability of marital violence (MV) (IIPS & Macro International, 2007). This measure demonstrated good internal reliability across samples, as indicated by Cronbach alphas of .71 for youth, .77 for fathers, and .79 for mothers.

Identical measures were utilized across all participant groups to allow for parent-child concordance analyses. Concordance was assessed using a Cohen’s unweighted Kappa statistic, with analyses stratified by sex of parent and sex of child. Unweighted statistics were used in light of the exploratory nature of this study. To assess strength of agreement, Kappa statistics were interpreted utilizing the guidelines proposed by Landis and Koch (1977): Poor<0.00, Slight=0.00-0.20, Fair=0.21–0.40, Moderate=0.41–0.60, Substantial=0.61–0.80, Almost Perfect=0.81–1.00.

Results

Mean age of youth participants was 17.8 years (SD=1.5) and 14.4 years (SD=0.9), for males and females respectively; all except one boy had received formal schooling. Mothers were aged 26-55 years (mean=39.4, SD=5.2); 77.5% had received any formal schooling, and 47.1% were employed in the past year. Fathers were aged 28-62 years (mean=46.6; SD=6.0); 91.6% had received any formal schooling, and 88.7% were employed in the past year. Few participants endorsed marriage of girls before age 18 years; some endorsed girls’ not having the right to choose who and when to marry, with fathers being most likely and girls being least likely to endorse these claims (Table 1.) With regard to marital contraception use, girls predominantly opposed the practice (85.8%); boys were least likely to oppose it (41.4%). Acceptability of MV was not uncommon and similar across fathers (40.4%) and mothers (41.0%), and male and female youth (33.6% and 30.1%, respectively). Concordant analyses revealed fair concordance between daughters and mothers on girl rights to choose when to marry, attitudes toward marital contraception use, and endorsement of justification for MV (see Table 2). Slight concordance between sons and fathers was seen on rights of girl to choose when to marry, with fair concordance on rights of girl to choose who to marry, and a moderate concordance was seen with father and son attitudes toward marital contraception use.Poor concordance in attitudes toward marital contraception use was seen for fathers and daughters, and for mothers and sons.

Table 1.

Prevalence of child and parent attitudes toward early and forced marriage of girls, contraception use in marriage, and domestic violence against women (n=659)

	Girls (n=113) % (n)	Boys (n=116) % (n)	Mothers (n=227) % (n)	Fathers (n=203) % (n)
Girls shouldn’t choose who to marry	19.5 (22)	21.6 (25)	33.9 (77)	48.8 (99)
Girls shouldn’t choose when to marry	9.7 (11)	22.4 (26)	18.9 (43)	43.4 (88)
Contraception shouldn’t be used in marriage	85.8 (97)	41.4 (48)	63.4 (144)	46.8 (95)
Endorse justification for marital violence	30.1 (34)	33.6 (39)	41.0 (93)	40.4 (82)
Positive effects of girl child marriage	2.7 (3)	1.7 (2)	4.4 (10)	1.5 (3)

Table 2.

Concordance between child and parent on attitudes toward early and forced marriage of girls, contraception use in marriage, and domestic violence against women (n=659)

	Mother – Child			Father - Child
	# Total Pairs	% Agreement (# pairs)	Kappa* (95% CI)	#Total Pairs	%Agreement (# pairs)	Kappa* (95% CI)
Female Child
Girls choose who to marry	112	63.4 (71)	0.17 (<0.001, 0.34)	97	54.6 (53)	0.13 (-0.002, 0.27)
Girls choose when to marry	112	81.3 (91)	0.27 (0.05, 0.49)	97	66.0 (64)	0.18 (0.05, 0.31)
Contraception use in marriage	112	70.5 (79)	0.25 (0.08, 0.42)	97	40.2 (39)	-0.20 (-0.35, -0.06)
Endorsement of marital violence	112	65.2 (73)	0.26 (0.08, 0.43)	97	57.7 (56)	0.06 (-0.14, 0.25)
Male Child
Girls choose who to marry	115	64.3 (74)	0.05 (-0.14, 0.23)	106	63.2 (67)	0.22 (0.05, 0.38)
Girls choose when to marry	115	67.8 (78)	0.01 (-0.17, 0.20)	106	61.3 (65)	0.20 (0.03, 0.36)
Contraception use in marriage	115	35.6 (41)	-0.23 (-0.40, -0.07)	106	77.4 (82)	0.53 (0.37, 0.70)
Endorsement of marital violence	115	48.7 (56)	-0.10 (-0.27, 0.08)	106	58.5 (62)	0.12 (-0.07, 0.31)

^*Cohen’s unweighted Kappa. Strength of agreement was defined as: Poor < 0.00, Slight = 0.00 - 0.20, Fair = 0.21 – 0.40, Moderate = 0.41 – 0.60, Substantial = 0.61 – 0.80, Almost Perfect = 0.81 – 1.00 (9).

Discussion

Current findings from this exploratory study indicate that many parents and youth within participating Indian villages continue to hold norms inhibiting girl choice in marriage and marital contraception, and supporting acceptability of MV, factors linked to poor maternal and child health and survival in India (Raj et al., 2014; Raj et al., 2010; Raj et al., 2009; Silverman et al., 2011; IIPS & Macro International, 2007). The very high proportion of adolescent girls who oppose contraception use in marriage requires particular attention. The need for intervention with youth across all these issues is substantial, particularly given that rural Maharashtra fairs better than many other Indian states on these health concerns (IIPS & Macro International, 2007). There may be some benefit to engaging parents in social norm approaches designed to reach youth, with an emphasis on gender concordance in programming, given findings of concordance in norms with regard to mothers and daughters, and for fathers and sons. The utility of engaging parents to delay marriage and increase marital contraception use has been demonstrated within an effective community intervention in rural India (Pathfinder, 2013), and current findings suggest that such an approach can be expanded to also include prevention of MV. Additionally, this work gives support more broadly to the importance of considering intergenerational transfer of norms related to marriage and family, expanding prior work in this area by documenting its importance in understanding norms of child marriage and contraception as well as father influence on those norms among boys.

Findings must be considered in light of certain limitations, including reliance on self-report and consequent social desirability, limited generalizability due to use of a convenience sample of respondents from two villages, and inability to compare same-age youth responses by sex due differences in age criteria at study entry. Further research with larger samples is also needed to help clarify whether observed findings vary by demographic characteristics of youth (e.g., age, birth order), parent (e.g., education level), and community (e.g., tribal)Nonetheless, these findings suggest that there is concordance between same-sex parents and children, indicating same sex transfer of normative beliefs related to girl child marital choice, family planning and marital violence. Programming targeting mother-daughter and father-son dyads may offer a promising approach to promoting gender equity marital norms in rural India.