Maternal and Paternal Cognitive Engagement and Children’s Literacy Skills in 25 African Countries

Abstract

There is increasing interest in promoting positive parenting to improve childhood development in low- and middle-income countries. Following propositions in parenting and cultural-ecological frameworks about the importance of early parent-child engagement in fostering children’s literacy skills, we used the UNICEF Multiple Indicator Surveys to assess the associations between maternal and paternal book reading, storytelling, and naming/counting and early literacy skills in 90,397 families and their preschool-aged children in 25 low- and middle-income African countries. Individual participant data meta-analysis indicated strong associations between maternal and paternal engagement in reading, storytelling, and naming/counting and children’s letter recognition, reading simple words, and recognizing symbols. Preschool enrollment and maternal education consistently moderated the associations between maternal and paternal engagement activities and children’s literacy skills. These findings are in agreement with those found in the high-income countries and have implications for the transferability of parenting practices from high-income to low- and middle-income countries in efforts to improve the early academic skills of young children.

Keywords

meta-analysis children’s literacy African countries paternal cognitive engagement individual participant data

Positive parenting is increasingly being emphasized by prominent world bodies (e.g., UNICEF, 2016; World Health Organization, 2010) and researchers (Engle et al., 2011; Levtov, van der Gaag, Greene, Kaufman, & Barker, 2015) as an effective way of promoting optimal childhood development in low- and middle-income countries. That said, much of what is known about levels of mother-child and father-child engagement and their associations with children’s cognitive skills is from the high-income countries (see Baker, 2015; McWayne, Downer, Campos, & Harris, 2013), and it is not clear whether these findings generalize to families in other parts of the world. Data on parental engagement and early childhood development from outside the high-income countries are not only necessary for the development of more inclusive theories on parenting and literacy skills development, but are vitally important for guiding parenting interventions and the construction of policy initiatives aimed at setting children on a good path to school success in low- and middle-income countries. To help remedy the deficiency in studies on parental cognitive engagement activities and childhood development in low- and middle-income countries, we determined associations between maternal and paternal engagement in book reading, storytelling, and naming/counting objects and preschoolers’ literacy skills, and the role of sociodemographic factors and preschool attendance in moderating these associations across 25 African countries.

Research conducted on different ethnic and cultural groups in North America and Western Europe have showed positive associations between levels and quality of maternal and paternal engagement in different cognitively laden activities (e.g., book reading, storytelling) and children’s early language skills (Baker, 2015; Baker & Vernon-Feagans, 2015; Duursma, Pan, & Raikes, 2008; Fagan, Iglesias, & Kaufman, 2016; McWayne et al., 2013; Varghese & Wachen, 2016). However, in a number of these studies, associations were tempered by material resources, literacy materials in the home environment, ethnicity, child-rearing beliefs, paternal residence, and neighborhood quality (Adamsons & Johnson, 2013; Duncan, Magnuson, & Votruba-Drzal, 2014; Sastry & Pebley, 2010), suggesting that the relative benefits children accrue from maternal and paternal cognitive engagement may vary across socioeconomic and ethnic/cultural groups in the rich democratic countries.

Relations between positive parenting and preschool-aged children’s language skills in low- and middle-income countries, such as those in Africa, remain sketchy. What research exists on early child-rearing practices in African families has mostly focused on differences in rates of maternal and paternal caregiving activities (verbal interactions, holding, displays of warmth and affection) in South Africa (Makusha & Richter, 2015), different groups in Kenya, such as the Kalenjins, Abagusii, Gikuyu, Luhyia, Luo, and Kamba (Mwoma, 2015; Putnick et al., 2012), and in foraging groups in the Central African Republic (i.e., Aka, Ngandu, Bofi), the Democratic Republic of the Congo (Efe), and Tanzania (Hadza; Fouts, 2008, 2013; Hewlett & MacFarlan, 2010; Marlowe, 2005). Not surprisingly, given fully entrenched traditional beliefs about child care and household labor in African countries, these studies indicated that mothers engaged in far more caregiving activities (e.g., cleaning, holding, feeding) than did fathers (Harkness & Super, 1992; Makusha & Richter, 2015; Mwoma, 2015). The UNICEF Multiple Indicator Surveys (MICS; UNICEF, 2015), one of the first global attempts to consistently document rates of cognitive engagement with children across low- and middle-income countries, revealed stark gender disparities in paternal and maternal cognitive engagement with children across 14 sub-Saharan countries (Bornstein et al., 2017).

Because of poor economic conditions in African countries, quite a bit of attention has been directed toward determining the impact of poverty on children’s physical health, nutritional status, and well-being (Bornstein et al., 2017) than on parental cognitive engagement and the acquisition of early language skills during the preschool years. Two small-scale, correlational studies found weak positive associations between paternal engagement in academic work at home and children’s number, word, and language activities (Wanjiku, 2016) and between paternal involvement in education (e.g., buying books, attending children’s school functions) and children’s academic performance in Kenyan children (Mwoma, 2009). An analysis of 44 studies on parenting practices and health and well-being in older children also found associations between parental connection to children, parental monitoring, and modeling of desirable parental behaviors and good educational outcomes in adolescents (Devlin, Wight, & Fenton, 2018). By contrast, a study on Ghanaian families found negative associations between parental cognitive stimulation and children’s literacy skills (Wolf & McCoy, 2019). From this small body of work, it is difficult to make broad statements about the associations between maternal and paternal cognitive engagement activities and young children’s literacy skills in African families across countries and in varying economic circumstances.

With rich civilizations, child-rearing beliefs (e.g., about extended breast feeding, cosleeping) and practices (e.g., alloparenting in which multiple individuals extend care to young children) that can differ markedly from those in high-income countries (Fouts, 2008, 2013; Marlowe, 2005), histories rooted in long periods of colonization, and impressive linguistic and ethnic/cultural diversity, the low- and middle-income African countries provide a good opportunity to further test the potential benefits of early maternal and paternal cognitive engagement for children’s early literacy skills development. Employing an individual participant data (IPD) meta-analytic technique, we used the UNICEF MICS to determine associations between maternal and paternal engagement in book reading, storytelling, and naming/counting objects and preschool-aged children’s literacy skills in the low- and middle-income African countries of Algeria, Benin, Cameroon, Chad, Central African Republic, Congo, Cote d’Ivoire, Democratic Republic of Congo, Gambia, Ghana, Guinea, Guinea Bissau, Madagascar, Malawi, Mali, Mauritania, Nigeria, Sao Tome and Principe, Senegal, Sierra Leone, Somalia, Swaziland, Togo, Tunisia, and Zimbabwe. An additional aim was to examine whether specific sociodemographic factors, such as maternal education, children’s age and gender, fathers’ residential status, and preschool attendance, moderated the associations between maternal and paternal engagement and children’s literacy skills.

Parenting and literacy frameworks (Baumrind, 1996; Morrison & Cooney, 2002; Shonkoff, 2010; Shonkoff & Phillips, 2000) and cultural-ecological models (Nsamenang, 2008, 2011; Ogbu, 1981; Whiting & Whiting, 1975) that situate child-rearing and parent-child relationships in cultural and socioeconomic contexts guided the study. These frameworks and models acknowledge the family setting and its associated cultural beliefs and practices as central to learning cognitive and social skills during the formative early childhood years. As such, mother-child and father-child sensitively calibrated cognitive activities (e.g., book reading and labelling) within the home learning environment (HLE) are seen as critical to laying the foundation for the development of language skills during the preschool years (Hart & Risley, 1995; Mol & Bus, 2011; Pong & Hao, 2007). Moreover, literacy frameworks (Morrison & Cooney, 2002) emphasize the significance of the availability of print-rich materials and developmentally appropriate objects within the HLE for early language development in preschool-aged children.

Based on principles in the aforementioned frameworks and model and related research (Baker, 2015; Isbell, Sobol, Lindauer, Lindauer, & Lowrance, 2004; McWayne et al., 2013; Mol & Bus, 2011), we selected maternal and paternal engagement in reading, telling stories, and naming/counting as predictors, and children’s ability to name/identify letters in the alphabet, read simple words, and recognizing symbols as outcome measures. In view of their heavy linguistic properties and the joint engagement opportunities that they provide for children to learn language skills, we labelled book reading, storytelling, and naming/counting objects as cognitive engagement activities that are subsumed under positive parenting. The cognitive benefits of these parent-child activities are outlined in several reviews (Bus, van Ijzendoorn, & Pellegrini, 1995; Isbell et al., 2004; Mol & Bus, 2011). The childhood outcome indicators were chosen because they are important markers of early academic skills in different cultural communities (Mistry, Biesanz, Taylor, Burchinal, & Cox, 2004).

Noting strong associations between children’s early cognitive skills and socioeconomic conditions, area of residence, and material resources (e.g., books and toys; see Dede Yildirim & Roopnarine, 2017; Sastry & Pebley, 2010; Votruba-Drzal, Miller, Coley, 2015), we entered household wealth, number of books in the home, and area of residence as control variables in our analyses. Because it has also been demonstrated that children’s age and gender, maternal education, and preschool attendance are associated with children’s early intellectual functioning (Burger, 2010; Engle & Black, 2008; Dede Yildirim & Roopnarine, 2017; Kramer, Allen, & Gergen, 1995; Liaw & Brooks-Gunn, 1994; Lima, Caughy, Nettles, & O’Campo, 2010; Loeb, Bridges, Bassok, Fuller, & Rumberger, 2007; Magnuson & Waldfogel, 2008; Reardon, 2011; Reardon & Robinson, 2008; Votruba-Drzal et al., 2015), these variables were also entered as covariates in our analyses. Given the diverse mating/marital systems across African countries, and the impact of fathers’ presence on childhood well-being and cognitive development (Adamsons & Johnson, 2013; Roopnarine, 2015; Roopnarine & Dede Yildirim, 2019a), we controlled for fathers’ residential status in our analyses.

Based on gendered beliefs and patterns of parental involvement with children in African families (Makusha & Richter, 2015, Marlowe, 2005; Mwoma, 2015) and findings on the associations between sociodemographic factors and childhood outcomes across cultural communities (e.g., Burger, 2010; Dede Yildirim & Roopnarine, 2017; Loeb et al., 2007; Roopnarine, Krishnakumar, Logie, Narine, & Davidson, 2017; Samms-Vaughan, 2005; Sastry & Pebley, 2010), we proposed two hypotheses. We predicted that the associations between book reading, storytelling, and naming/counting and indicators of childhood literacy skills would be stronger for mothers and children than for fathers and children across countries, after controlling for key sociodemographic variables and preschool attendance. We further predicted that fathers’ residential status, children’s age and gender, maternal education, and preschool enrollment would moderate the associations between maternal and paternal book reading, storytelling, and naming/counting and indicators of childhood literacy skills, after controlling for number of children’s books in the home environment and household wealth. As such, it was expected that the associations between maternal and paternal engagement activities and children’s literacy skills would be stronger for those who lived with resident fathers, who attended preschool, and whose mothers had at least a high school education than those who did not have resident fathers, did not attend preschool, and whose mothers had lower than a high school education. It was also expected that boys and older children would benefit more from paternal and maternal engagement than girls and younger children.

Method

Procedure

For this study, we drew on publicly available data from the UNICEF’s Multiple Indicator Cluster Surveys rounds 4, 5, and 6 (UNICEF, 2015) conducted between 2009 and 2017 in 25 African countries: Algeria, Benin, Cameroon, Chad, Central African Republic, Congo, Cote d’Ivoire, Democratic Republic of Congo, Gambia, Ghana, Guinea, Guinea Bissau, Madagascar, Malawi, Mali, Mauritania, Nigeria, Sao Tome and Principe, Senegal, Sierra Leone, Somalia, Swaziland, Togo, Tunisia, and Zimbabwe. The MICS surveys gather nationally representative data on women and children using face-to-face interviews in nationally representative samples of households in low- and middle-income countries by utilizing a three-stage stratified cluster sampling design.

Participants

For these analyses, participants were families who had children younger than 5 years and limited to those who completed the MICS early child development surveys in the 25 countries. This sample was composed of 95,742 preschool-aged children and their families. However, we used a subsample of participants selecting only one child per household to avoid within-family dependency due to the nested feature of data and excluding children whose biological mother and father were not alive at the time of data collection. The restricted sample consisted of 90,397 preschool-aged children (ranged from 36 to 59 months) and their families. The sociodemographic characteristics of caregivers and children included in our analyses are shown in Table 1.

Table 1.

Sample Characteristics.

Country	n	ECE attendance, %	Urban, %	Child, male, %	Secondary education, %	Father in household, %	Wealth index quintiles					Child age (in months)
Country	n	ECE attendance, %	Urban, %	Child, male, %	Secondary education, %	Father in household, %	Poorest, %	Second, %	Middle, %	Fourth, %	Richest, %	M	SD
Algeria	5,275	15.42	65.84	52.06	29.41	97.63	23.07	19.60	21.23	19.39	16.70	47.47	6.85
Benin	3,365	16.27	49.87	49.51	13.50	80.72	20.08	18.63	18.16	18.54	24.59	47.47	6.80
Cameroon	2,530	31.54	45.81	49.33	33.23	72.09	18.18	22.92	22.09	20.36	16.44	46.58	6.78
Central African Republic	3,365	3.62	33.08	48.83	11.41	73.79	20.89	25.14	24.28	18.66	11.03	46.16	6.62
Chad	6,340	4.96	35.88	49.72	5.40	85.73	16.89	17.63	21.06	24.15	20.27	48.05	6.96
Congo	3,340	23.72	31.02	51.77	54.18	67.49	44.85	25.33	12.87	9.31	7.63	48.03	6.90
Cote d’Ivoire	3,402	10.49	25.69	50.94	9.58	73.99	30.16	26.81	20.93	13.55	8.55	46.67	6.65
Democratic Republic of Congo	3,651	5.72	37.72	49.90	36.09	81.43	19.25	17.94	21.34	23.45	18.02	45.97	6.85
Gambia	3,772	14.88	25.72	51.06	10.84	75.37	30.54	21.66	19.70	16.65	11.45	46.00	6.64
Ghana	2,877	57.43	28.61	50.43	25.58	78.28	46.96	20.06	14.01	10.67	8.31	46.92	6.82
Guinea	2,810	8.50	29.93	51.49	11.75	78.51	23.95	23.13	20.50	18.15	14.27	47.08	6.70
Guinea Bissau	2,702	9.93	27.54	49.11	10.73	65.77	29.90	24.20	22.69	14.62	8.59	47.08	6.75
Madagascar	1,085	7.86	4.06	49.59	9.06	68.76	25.25	19.82	22.58	20.18	12.17	46.05	6.49
Malawi	7,081	38.93	11.21	49.95	15.07	72.87	21.00	21.80	21.75	19.57	15.87	47.42	6.83
Mali	6,016	5.52	21.56	51.40	10.11	88.80	21.66	19.90	19.76	20.69	17.99	46.18	6.74
Mauritania	3,988	13.20	42.50	49.45	47.29	61.31	21.29	23.27	21.09	17.98	16.37	46.54	6.75
Nigeria	10,566	38.36	26.41	50.75	34.39	90.45	23.47	22.20	18.81	17.96	17.56	46.91	6.73
Sao Tome and Principe	806	37.58	59.31	49.75	26.80	59.06	25.56	22.95	20.47	19.98	11.04	47.76	7.20
Senegal	1,653	42.43	NA	50.03	23.32	66.12	21.72	21.96	21.90	19.48	14.94	46.83	6.83
Sierra Leone	4,276	NA	28.62	50.19	20.21	64.29	28.74	25.02	22.31	12.91	11.01	47.23	6.84
Somalia	3,268	2.05	57.28	50.18	5.63	78.46	21.36	20.90	20.26	19.40	18.08	44.99	6.64
Swaziland	941	28.94	15.20	51.33	53.52	32.62	28.16	23.70	20.72	14.67	12.75	47.18	6.80
Togo	1,706	26.07	23.80	51.00	13.02	82.24	26.96	27.37	20.22	13.25	12.19	46.63	6.89
Tunisia	1,072	39.42	58.21	54.10	44.96	95.99	27.71	22.29	17.07	19.68	13.25	47.09	6.66
Zimbabwe	3,393	22.77	28.09	49.72	59.15	58.09	21.49	21.54	18.72	20.25	18.01	47.13	6.59

Note: N = 90,347. ECE = early childhood education.

Measures

The MICS surveys include a questionnaire on household factors, questionnaires for individual women and men between the ages 15 and 49 years, a questionnaire for children younger than 5 years, and a questionnaire for vaccination records at health facilities (UNICEF, 2015). The surveys cover key indicators of the well-being of children and their families including nutrition, early parental involvement, early childhood development, child protection and child labor, maternal and child health, prenatal and postnatal health care, low birth weight, infant and child mortality, water and sanitation, and HIV/AIDS. Household questionnaires were administered to all participants, and selected households completed questionnaires for children younger than 5 years and for individual men between 15 and 49 years of age.

Mothers’ and fathers’ cognitive engagement

Primary caregivers reported on mothers’ and fathers’ engagement in cognitive activities separately utilizing the questionnaire for children younger than 5 years. Based on different perspectives on parenting (Isbell et al., 2004; Mol & Bus, 2011; Morrow, 1985; Morrison & Cooney, 2002; Shonkoff, 2010) and research findings on the importance of early parental cognitive engagement for children’s early language skills (Bus et al., 1995; Mol & Bus, 2011), three items were chosen from the child questionnaire as parental engagement indicators. Caregivers reported whether mothers and fathers “read books to or looked at picture books with child,” “told stories to child,” and “named, counted, or drew things to or with child” during the past 3 days. In prior work, the internal consistency of the scores of composite measures of parental engagement in play and learning activities (Cronbach’s α = .77; Jeong, McCoy, Yousafzai, Salhi, & Fink, 2016; McCoy et al., 2016) and paternal and maternal cognitive engagement in these activities with children were good (Guttman’s Lambda ranged from .80 to .90; Roopnarine & Dede Yildirim, 2019b). In this study, each item was entered into the models separately, and item-factor reliability estimates were calculated using confirmatory factor analysis (CFA; Raykov, Dimitrov, & Asparouhov, 2010) using single dichotomous item factor indicators. Item-factor loadings were λ = .86 for maternal and λ = .85 for paternal engagement in book reading, λ = .77 for maternal and λ = .78 for paternal storytelling, and λ = .71 for maternal and λ = .77 for paternal naming/counting objects. Table 2 displays the percentages of mothers and fathers who engaged in book reading, storytelling, and naming/counting for each country.

Table 2.

Maternal and Paternal Engagement in Cognitive Activities.

Country	Mothers				Fathers
Country	n	Read books, %	Tell stories, %	Name objects, %	n	Read books, %	Tell stories, %	Name objects, %
Algeria	5,275	28.87	51.62	62.05	5,275	13.38	18.39	31.50
Benin	4,482	11.49	20.30	10.31	4,482	7.48	10.73	7.70
Cameroon	2,530	8.38	24.22	14.59	2,530	3.35	9.62	5.80
Central African Republic	3,365	3.84	27.52	42.41	3,365	4.97	19.51	22.17
Chad	6,340	5.38	30.48	59.60	6,340	4.26	14.43	13.97
Congo	3,340	17.85	30.83	25.32	3,340	8.00	13.12	11.44
Cote d’Ivoire	3,402	5.78	16.94	10.04	3,402	4.03	8.95	6.99
Democratic Republic of Congo	3,651	3.84	27.33	22.68	3,651	3.98	15.01	11.49
Gambia	3,772	1.78	13.18	8.42	3,772	1.22	2.80	5.45
Ghana	2,877	6.71	10.25	9.28	2,877	5.18	5.25	5.67
Guinea	2,810	8.77	20.49	8.89	2,810	3.44	8.77	4.92
Guinea Bissau	2,702	2.52	9.33	15.77	2,702	1.59	6.92	3.66
Madagascar	1,085	2.40	5.00	9.70	1,085	1.48	3.24	6.28
Malawi	7,081	7.98	14.13	11.27	7,081	4.88	6.37	6.13
Mali	6,016	14.27	35.25	18.24	6,016	6.91	14.74	9.96
Mauritania	3,988	25.10	36.84	31.85	3,988	7.34	12.65	11.83
Nigeria	10,566	30.76	40.47	26.83	10,566	15.79	19.23	14.39
Sao Tome and Principe	806	16.19	20.15	17.68	806	6.10	9.08	6.35
Senegal	1,653	12.75	28.87	18.70	1,653	4.44	10.32	6.87
Sierra Leone	4,276	11.27	24.00	11.06	4,276	7.66	12.22	7.43
Somalia	3,268	17.18	56.82	37.58	3,268	6.17	13.61	12.15
Swaziland	941	11.38	13.19	24.17	940	2.77	2.34	4.79
Togo	1,706	7.47	29.29	38.84	1,706	6.29	13.24	15.88
Tunisia	1,072	33.61	50.23	53.03	1,072	17.09	21.66	30.16
Zimbabwe	3,393	12.26	26.99	21.97	3,393	4.52	11.65	6.53
Chi-square (df = 24)		6994.43**	6964.30**	13193.63**		2557.51**	1995.03**	4203.41**

Note: df = degrees of freedom; n_mother = 90,397; n_father = 90,396.

p < .001.

Children’s literacy skills

Drawing from studies that delineated the importance of knowing basic literacy skills, such as recognizing letters, words, symbols, and the ability to read simple words for early schooling and for setting children on a good path for early school success (LoCasale-Crouch, Mashburn, Downer, & Pianta, 2008), we used three binary items as measures of children’s early literacy skills. Primary caregivers indicated whether their “child can identify or name at least 10 letters of the alphabet,” “child can read at least four simple popular words,” and “child knows the name and recognizes the symbol of all numbers from 1 to 10.” These items are similar to those used in standardized instruments to assess children’s early language skills (e.g., Kaufman & Kaufman, 1993) in low- and middle-income countries (e.g., Caribbean). Items for children’s literacy skills were also modeled separately, and item-factor reliability estimates were calculated using CFA (Raykov et al., 2010) with single-item factor indicators. Item-factor loadings were λ = .80 for identifying letters, λ = .94 for reading words, and λ = .86 for recognizing symbols.

Covariates

Sociodemographic data provided by caregivers were used to create several covariates. These variables were the number of children’s books in the household (ranged from 0 to 10), mothers’ highest level of education (1 = at least secondary, 0 = less than secondary), area of residence (1 = urban, 0 = rural), children’s age (in months), and sex (1 = male, 0 = female), preschool enrollment (1 = enrolled, 0 = not enrolled), and fathers’ residential status (1 = father lives in the household, 0 = father is alive but lives elsewhere). The household wealth index was created by using unrestricted principal component analysis (PCA) utilizing 234 binary items from country-specific household surveys based on all available information to indicate the main source of drinking water, the main sanitation facility, number of rooms for sleeping, the main floor, roof, and wall materials, type of fuel for cooking, and household assets. In the final index, the type of drinking water consisted of 25 items (e.g., piped into dwelling, compound, yard or plot, or to neighbor, public tap/standpipe, tube well or borehole, protected well, unprotected well, rainwater collection, surface water, and bottled water), the main sanitation facility had 17 items (e.g., flush to pit, pit latrine with slab, flush to unknown place, composting toilet, bucket, hanging toilet, hanging latrine, no facility, bush or field, and shared public sanitation facility). The main floor material consisted of 19 items (e.g., bricks or stones, mat, reeds, wood planks, plywood, ceramic tiles, natural floor, and earth/sand), the main roof material had 23 items (e.g., no roof, mud with dry hard straw, sod, rustic mat, cloth/canvas/tent, tin cans, roofing shingles, and cardboard/carton), and the main wall material had 33 items (e.g., no walls, mud/dirt, grass wall, cane, straw, mud, uncovered adobe, plywood, cloth/canvas/tent, and prefabricated). Twelve indicators were used to assess the main fuel for cooking (e.g., electricity, natural gas, biogas, kerosene, charcoal, and animal dung) and household assets consisted of 95 items (e.g., radio, TV, refrigerator, vacuum cleaner, table, bed, sofa, water tank, tractor, domestic animals, and agricultural land; see Martel, 2016, for a detailed description of household wealth index). Based on item loadings in the final PCA, the index scores were calculated for each household, and then households were assigned to five equally distributed wealth quintiles. Across countries, the household wealth index score ranged from −4.97 to 7.29, and household wealth index quintiles (poorest, second, middle, fourth, and richest) were included as covariates.

Meta-Analytic Procedure

Two-stage meta-analysis

The R program (R Core Team, 2018) was used to perform sets of multilevel binary logistic regression to assess country-specific associations between maternal and paternal engagement in book reading, storytelling, and counting and naming objects, and the three individual binary items representing children’s literacy skills (identifying letters, reading words, recognizing symbols), controlling for preschool enrollment, children’s sex and age, number of children’s books in the household, mothers’ educational level, area of residence, fathers’ residential status, and household wealth index. Maternal and paternal engagement and the outcome variables were each modeled separately in order to avoid potential multicollinearity between them. Children’s sample weights for those younger than 5 years along with the household cluster numbers, which represent number of households in the primary geographic areas, were entered into the models to account for the complex sampling design and hierarchical nesting of data.

Next, a conventional meta-analysis was conducted using odds ratios (ORs) and confidence intervals (CIs) as the effect size measure using the “metafor” package (Viechtbauer, 2010). A random effects model was used because of expected heterogeneity of maternal and paternal engagement in cognitive activities across the 25 African countries. To test the homogeneity of effect sizes, we calculated I² statistics in percentages as indicators of heterogeneity. I² with a value of 0% to 25% indicates unimportant heterogeneity, with larger values indicating increasing heterogeneity (25% to 50% as moderate, 50% to 75% substantial, and 75% to 100% as considerable). The 95% CIs of I² were calculated using the noncentral chi-square approach. Funnel plots were produced to visually assess the small sample bias (available on request from the corresponding author), and the Egger’s test (Egger, Smith, Schneider, & Minder, 1997) and trim-and-fill procedure (Duval & Tweedie, 2000) were evaluated. We created forest plots for each child outcome measure using the “forestplot” package (Gordon & Lumley, 2017).

One-stage meta-analysis

To assess whether children’s sex and age, preschool enrollment, fathers’ residence, and maternal education moderated the associations between maternal and paternal engagement in cognitive activities and children’s early literacy skills, we conducted an IPD meta-analysis pooling all individual-level data into one data set. IPD meta-analyses were conducted with the “lme4” package (Bates, Maechler, Bolker, & Walker, 2015). Mixed-effect logistic regression models were run using the country as the random effect. Maternal and paternal engagement in book reading, storytelling, and naming/counting objects were entered as predictors and the three individual binary items of children’s early literacy skills were entered as outcome variables. Interaction terms were created between the three maternal and paternal engagement activities and children’s sex and age, preschool enrollment, fathers’ residential status, and maternal education. All models included number of children’s books in the home environment and household wealth index as covariates.

Results

Preliminary Analyses

Missing data on predictors, covariates and outcome variables were identified, and descriptive statistics of study variables were analyzed. Overall, most variables had low levels of missing data (ranged from 0.5% to 1.5%) and there were no outliers. Missing data were handled using pairwise deletion in each analysis. The analytic sample varied per analysis and ranged from 89,048 to 90,397 families. Overall, 13.4% of mothers and 6.9% of fathers read books, 28.4% of mothers and 12.3% of fathers told stories, and 25.2% of mothers and 11.2% of fathers named, counted, or drew things to or with children across countries. Among all children, 18.3% of them identified or named at least 10 letters of the alphabet, 13.6% read at least four simple, popular words, and 19.2% were judged by caregivers to recognize symbols.

Descriptive Statistics

Table 2 presents the weighted chi-square tests on the percentages of mothers and fathers who engaged in the three cognitive activities with children in each country. As can be seen in Table 2, there were significant variations and differences (using an α of .05) among countries across all three paternal and maternal engagement activities. For instance, 1.78% of mothers and 1.22% of fathers in Gambia engaged in book reading, whereas more than 30% of mothers and 15% of fathers in Nigeria and Tunisia engaged in book reading with children. Mothers (62.5%) and fathers (31.5%) in Algeria were more likely to name/count objects with children than were those in all other countries (e.g., Central African Republic, Democratic Republic of Congo, Mauritania, Nigeria). Madagascar (5%), Guinea Bissau (9.33%), and Ghana (10.25%) had the lowest numbers of mothers, and Swaziland (2.34%), Gambia (2.80%), and Madagascar (3.24%) had the lowest numbers of fathers who told stories to children.

Two-Stage Meta-Analysis

The random effects two-stage meta-analysis revealed that mothers’ engagement in book reading, storytelling, and counting/naming objects were positively associated with children’s ability to identify letters (µ [random effect OR] = 1.51, µ = 1.19, and µ = 1.30, respectively), read words (µ = 1.56, µ = 1.21, and µ = 1.26, respectively), and recognize symbols (µ = 1.48, µ = 1.15, and µ = 1.34, respectively). Similarly, fathers’ engagement in book reading, storytelling, and counting/naming objects were positively associated with children’s ability to identify letters (µ = 1.49, µ = 1.25, and µ = 1.46, respectively), read words (µ = 1.47, µ = 1.14, and µ = 1.43, respectively), and recognize symbols (µ = 1.50, µ = 1.24, and µ = 1.38, respectively). Table 3 represents effect sizes for associations between maternal and paternal engagement and children’s literacy skills. Forest plots (Figures 1 -9) represent the links between maternal and paternal engagement in the three cognitive activities and indicators of children’s literacy skills for each country. Overall, models estimated zero heterogeneity, indicating no variation in associations between countries, and the standard estimates of heterogeneity ranged from Q = 9.99, p = .99, I² = .01 to Q = 17.23, p = .84, I² = .01, suggesting minimal variation in associations between countries.

Table 3.

Effect Sizes for Associations Between Maternal and Paternal Engagement in Cognitive Activities and Children’s Literacy Skills.

	Letter identification		Reading simple words		Recognizing symbols
	OR	95% CI	OR	95% CI	OR	95% CI
Mothers
Reading	1.51***	[1.28, 1.77]	1.56***	[1.32, 1.84]	1.48***	[1.26, 1.74]
Storytelling	1.19*	[1.02, 1.37]	1.21*	[1.04, 1.40]	1.15*	[1.00, 1.33]
Naming/counting objects	1.30***	[1.12, 1.52]	1.26**	[1.08, 1.47]	1.34***	[1.16, 1.55]
Fathers
Reading	1.49***	[1.24, 1.79]	1.47***	[1.22, 1.76]	1.50***	[1.26, 1.80]
Storytelling	1.25***	[1.06, 1.48]	1.14**	[0.96, 1.35]	1.24***	[1.05, 1.47]
Naming/counting objects	1.46***	[1.24, 1.73]	1.43***	[1.21, 1.69]	1.38***	[1.17, 1.63]

Note: OR = odds ratio; CI = confidence interval; n_mother = 90,397; n_father = 90,396.

p < .05. **p < .01. ***p < .001.

Figure 1.

The links between maternal and paternal engagement in book reading activities and identifying letters.

Figure 2.

The links between maternal and paternal engagement in story telling activities and identifying letters.

Figure 3.

The links between maternal and paternal engagement in naming objects and identifying letters.

Figure 4.

The links between maternal and paternal engagement in book reading activities and reading words.

Figure 5.

The links between maternal and paternal engagement in story telling activities and reading words.

Figure 6.

The links between maternal and paternal engagement in naming objects and reading words.

Figure 7.

The links between maternal and paternal engagement in book reading activities and recognizing symbols.

Figure 8.

The links between maternal and paternal engagement in story telling activities and recognizing symbols.

Figure 9.

The links between maternal and paternal engagement in naming objects and recognizing symbols.

Individual Participant Data Meta-Analysis

Almost identical patterns of associations were obtained from the one-stage meta-analysis using mixed-effect logistic regression estimates. Mothers’ engagement in book reading, storytelling, and counting/naming objects were positively associated with children’s success at identifying letters (OR = 1.57, OR = 1.13, and OR = 1.38, respectively), reading words (OR = 1.69, OR = 1.15, and OR = 1.26, respectively), and recognizing symbols (OR = 1.57, OR = 1.11, and OR = 1.35, respectively; see Table 4 for full results). Likewise, fathers’ book reading, storytelling, and naming/counting objects were positively associated with children’s success at identifying letters (OR = 1.42, OR = 1.26, and OR = 1.46, respectively), reading words (OR = 1.41, OR = 1.23, and OR = 1.35, respectively), and recognizing symbols (OR = 1.52, OR = 1.24, and OR = 1.30, respectively; see Table 5).

Table 4.

IPD Meta-Analysis Estimates for Associations Between Maternal Engagement in Cognitive Activities and Children’s Literacy Skills.

	Letter identification				Reading simple words				Recognizing symbols
	B	SE	OR	95% CI	B	SE	OR	95% CI	B	SE	OR	95% CI
Reading	0.45***	0.03	1.57	[1.48, 1.67]	0.53***	0.03	1.70	[1.59, 1.80]	0.45***	0.03	1.57	[1.48, 1.66]
Storytelling	0.13***	0.03	1.13	[1.08, 1.19]	0.14***	0.03	1.15	[1.09, 1.21]	0.11***	0.03	1.11	[1.06, 1.17]
Naming/counting objects	0.32***	0.03	1.38	[1.31, 1.46]	0.23***	0.03	1.26	[1.19, 1.33]	0.30***	0.03	1.35	[1.29, 1.42]
Preschool attendance	1.81***	0.03	6.11	[5.80, 6.44]	1.32***	0.03	3.73	[3.53, 3.94]	1.64***	0.03	5.14	[4.89, 5.40]
Number of books	0.15***	0.01	1.16	[1.14, 1.18]	0.12***	0.01	1.13	[1.11, 1.14]	0.16***	0.01	1.18	[1.16, 1.20]
Child age	0.56***	0.02	1.74	[1.67, 1.82]	0.47***	0.02	1.61	[1.53, 1.68]	0.69***	0.02	2.00	[1.92, 2.09]
Wealth score	0.35***	0.01	1.42	[1.38, 1.45]	0.25***	0.01	1.29	[1.26, 1.32]	0.35***	0.01	1.41	[1.38, 1.45]
Child sex (male)	−0.09***	0.02	0.92	[0.88, 0.96]	−0.06*	0.02	0.95	[0.91, 0.99]	−0.12***	0.02	0.89	[0.86, 0.93]
Father residence (yes)	0.001	0.03	1.00	[0.95, 1.06]	0.04	0.03	1.04	[0.98, 1.10]	0.01	0.03	1.01	[0.96, 1.06]
M. education (secondary)	.025***	0.03	1.28	[1.22, 1.35]	0.16***	0.03	1.18	[1.11, 1.25]	0.30***	0.03	1.36	[1.29, 1.43]
Interactions
Reading × Preschool A.	−0.36***	0.06	0.70	[0.62, 0.79]	−0.36***	0.06	0.70	[0.62, 0.79]	−0.38***	0.06	0.68	[0.61, 0.77]
Storytelling × Preschool A.	−0.04	0.05	0.96	[0.87, 1.07]	0.05	0.05	1.05	[0.94, 1.17]	0.08	0.05	1.08	[0.98, 1.20]
Naming/C. objects × Preschool A.	−0.47***	0.05	0.63	[0.56, 0.69]	−0.40***	0.06	0.67	[0.60, 0.75]	−0.36***	0.05	0.70	[0.63, 0.78]
Reading × Child age	.019**	0.06	1.21	[1.08, 1.36]	0.13*	0.06	1.14	[1.01, 1.28]	0.14*	0.06	1.16	[1.03, 1.30]
Storytelling × Child age	−0.07	0.05	0.94	[0.85, 1.03]	−0.07	0.05	0.93	[0.84, 1.03]	−0.06	0.05	0.94	[0.85, 1.03]
Naming/C. objects × Child age	0.02	0.05	1.02	[0.92, 1.13]	0.03	0.05	1.03	[0.93, 1.15]	0.01	0.05	1.01	[0.92, 1.11]
Reading × Male	−0.09	0.06	0.92	[0.82, 1.03]	0.001	0.06	1.00	[0.89, 1.13]	−0.18**	0.06	0.83	[0.74, 0.93]
Storytelling × Male	0.08	0.05	1.08	[0.98, 1.20]	0.04	0.05	1.04	[0.94, 1.16]	0.13**	0.05	1.14	[1.03, 1.25]
Naming/C. objects × Male	0.07	0.05	1.07	[0.97, 1.18]	0.001	0.05	1.00	[0.90, 1.11]	−0.05	0.05	0.96	[0.87, 1.05]
Reading × Fathers R.	0.13	0.08	1.13	[0.98, 1.32]	0.04	0.08	1.04	[0.90, 1.21]	0.08	0.07	1.08	[0.93, 1.25]
Storytelling × Fathers R.	0.04	0.07	1.04	[0.91, 1.18]	−0.03	0.07	0.97	[0.85, 1.11]	−0.04	0.06	0.96	[0.85, 1.09]
Naming/C. objects × Fathers R.	0.15*	0.07	1.17	[1.02, 1.33]	0.03	0.07	1.03	[0.90, 1.18]	0.01	0.06	1.01	[0.89, 1.14]
Reading × M. education	−0.18**	0.06	0.83	[0.74, 0.94]	−0.27***	0.06	0.76	[0.67, 0.86]	−0.20**	0.06	0.82	[0.73, 0.92]
Storytelling × M. education	−0.13*	0.05	0.88	[0.79, 0.98]	0.10	0.05	1.11	[0.99, 1.23]	−0.04	0.05	0.96	[0.87, 1.06]
Naming/C. × M. education	−0.28***	0.05	0.75	[0.68, 0.84]	−0.20***	0.06	0.82	[0.73, 0.91]	−0.16**	0.05	0.85	[0.77, 0.94]

Note: N = 89,048 - 90,397. OR = odds ratio; CI = confidence interval; SE = standard error; Preschool A. = preschool attendance; Fathers R. = fathers residence; M. Education = maternal education.

p < .05. **p < .01. ***p < .001.

Table 5.

IPD Meta-Analysis Estimates for Associations Between Paternal Engagement in Cognitive Activities and Children’s Literacy Skills.

	Letter identification				Reading simple words				Recognizing symbols
	B	SE	OR	95% CI	B	SE	OR	95% CI	B	SE	OR	95% CI
Reading	0.35***	0.04	1.42	[1.31, 1.53]	0.34***	0.04	1.41	[1.31, 1.52]	0.42***	0.04	1.52	[1.41, 1.64]
Storytelling	0.23***	0.03	1.26	[1.18, 1.35]	0.21***	0.03	1.23	[1.15, 1.32]	0.21***	0.03	1.24	[1.16, 1.32]
Naming/counting objects	0.38***	0.03	1.46	[1.37, 1.56]	0.30***	0.03	1.35	[1.26, 1.45]	0.27***	0.03	1.30	[1.22, 1.39]
Preschool attendance	1.83***	0.03	6.23	[5.92, 6.57]	1.34***	0.03	3.83	[3.63, 4.05]	1.65***	0.03	5.23	[4.98, 5.49]
Number of books	0.16***	0.01	1.17	[1.15, 1.19]	0.13***	0.01	1.14	[1.12, 1.15]	0.17***	0.01	1.19	[1.17, 1.21]
Child age	0.55***	0.02	1.73	[1.66, 1.80]	0.46***	0.02	1.59	[1.52, 1.66]	0.69***	0.02	1.99	[1.91, 2.07]
Wealth score	0.36***	0.01	1.44	[1.40, 1.47]	0.27***	0.01	1.31	[1.28, 1.34]	0.36***	0.01	1.43	[1.40, 1.47]
Child sex (male)	−0.10***	0.02	0.90	[0.86, 0.94]	−0.07**	0.02	0.93	[0.89, 0.97]	−0.13***	0.02	0.88	[0.84, 0.91]
Father residence	−0.08**	0.03	0.92	[0.87, 0.97]	−0.04	0.03	0.96	[0.91, 1.02]	−0.07***	0.03	0.93	[0.89, 0.98]
Maternal education	0.34***	0.03	1.40	[1.33, 1.48]	0.26***	0.03	1.30	[1.23, 1.38]	0.39***	0.03	1.48	[1.40, 1.55]
Interactions
Reading × Preschool A.	−0.37***	0.08	0.69	[0.60, 0.81]	−0.36***	0.08	0.69	[0.60, 0.81]	−0.43***	0.08	0.65	[0.56, 0.75]
Storytelling × Preschool A.	−0.09	0.07	0.92	[0.80, 1.05]	−0.05	0.07	0.95	[0.83, 1.09]	0.09	0.07	1.10	[0.96, 1.25]
Naming/C. objects × Preschool A.	−0.44***	0.07	0.65	[0.57, 0.74]	−0.27***	0.07	0.77	[0.67, 0.87]	−0.29***	0.07	0.75	[0.65, 0.85]
Reading × Child age	0.19*	0.08	1.21	[1.04, 1.41]	0.15*	0.08	1.17	[1.00, 1.36]	0.09	0.08	1.09	[0.94, 1.27]
Storytelling × Child age	−0.01	0.07	0.99	[0.87, 1.13]	−0.09	0.07	0.91	[0.80, 1.04]	−0.02	0.06	0.98	[0.86, 1.11]
Naming/C. objects × Child age	0.05	0.07	1.05	[0.92, 1.19]	−0.04	0.07	0.97	[0.85, 1.10]	0.03	0.06	1.03	[0.91, 1.17]
Reading × Male	−0.07	0.08	0.93	[0.80, 1.09]	0.05	0.08	1.06	[0.91, 1.23]	−0.09	0.08	0.91	[0.78, 1.06]
Storytelling × Male	0.02	0.07	1.02	[0.90, 1.16]	−0.04	0.07	0.96	[0.84, 1.09]	0.001	0.06	1.00	[0.88, 1.13]
Naming/C. objects × Male	0.001	0.07	1.00	[0.88, 1.14]	−0.02	0.07	0.98	[0.86, 1.12]	0.03	0.06	1.03	[0.91, 1.17]
Reading × Fathers R.	0.12	0.18	1.12	[0.80, 1.58]	−0.19	0.17	0.83	[0.60, 1.16]	−0.19	0.17	0.82	[0.59, 1.15]
Storytelling × Fathers R.	0.23	0.15	1.26	[0.93, 1.69]	0.16	0.15	1.17	[0.87, 1.57]	−0.05	0.14	0.96	[0.72, 1.26]
Naming/C. objects × Fathers R.	−0.26	0.14	0.77	[0.59, 1.01]	−0.06	0.14	0.95	[0.72, 1.25]	0.22	0.14	1.25	[0.94, 1.66]
Reading × M. education	−0.14	0.08	0.87	[0.75, 1.02]	−0.27***	0.08	0.76	[0.66, 0.89]	−0.28***	0.08	0.75	[0.65, 0.88]
Storytelling × M. education	−0.03	0.07	0.97	[0.85, 1.11]	0.05	0.07	1.05	[0.92, 1.21]	−0.04	0.07	0.96	[0.84, 1.09]
Naming/C. × M. education	−0.27***	0.07	0.76	[0.67, 0.87]	−0.14*	0.07	0.87	[0.76, 0.99]	−0.25***	0.07	0.78	[0.69, 0.89]

Note: N = 89,048 - 90,397. OR = odds ratio; CI = confidence interval; SE = standard error; Preschool A. = preschool attendance; Fathers R. = fathers residence; M. education = maternal education.

p < .05. **p < .01. ***p < .001.

Moderation Analysis

A set of mixed-effect logistic regression models were tested to determine whether preschool attendance, children’s age and sex, fathers’ residential status, and maternal education moderated the associations between maternal and paternal engagement in book reading, story-telling, and naming/counting and children’s early literacy skills. The associations between mothers’ engagement in book reading and identifying letters (OR = 0.70), reading words (OR = 0.70), and recognizing symbols (OR = 0.68), and the associations between fathers’ engagement in book reading and identifying letters (OR = 0.70), reading words (OR = 0.69), and recognizing symbols (OR = 0.65) were stronger for children who did not attend preschool relative to those who did. In addition, the links between mothers’ engagement in counting/naming objects and identifying letters (OR = 0.63), reading words (OR = 0.67), and recognizing symbols (OR = 0.70), and the links between fathers’ engagement in counting and naming objects and identifying letters (OR = 0.65), reading words (OR = 0.77), and recognizing symbols (OR = 0.74) were stronger for children who did not attend preschool versus those who did.

Turning to child characteristics, sex and age moderated the associations between mothers’ engagement in cognitive activities and children’s literacy skills inconsistently. The link between mothers’ engagement in book reading and recognizing symbols (OR = 0.83) was stronger for girls than for boys, but the link between mothers’ engagement in storytelling and recognizing symbols (OR = 1.14) was stronger for boys than for girls. Associations between mothers’ engagement in book reading and identifying letters (OR = 1.21), reading words (OR = 1.14), and recognizing symbols (OR = 1.16), and the associations between fathers’ engagement in book reading and identifying letters (OR = 1.21), and reading words (OR = 1.17) were stronger for 4-year-old than for 3-year-old children.

The links between mothers’ engagement in book reading and counting/naming objects and identifying letters (OR = 0.83, and OR = 0.75, respectively), reading words (OR = 0.76, and OR = 0.82, respectively), and recognizing symbols (OR = 0.82, and OR = 0.85, respectively) were stronger for children whose mothers had less than a secondary school education compared with those whose mothers had at least a secondary school education. The same was true for the links between mothers’ engagement in storytelling and identifying letters (OR = 0.88). For fathers, the links between engagement in book reading and naming/counting objects and reading words (OR = 0.76, and OR = 0.87, respectively), and recognizing symbols (OR = 0.75, and OR = 0.78, respectively), and between engagement in counting/naming objects and identifying letters (OR = 0.76) were stronger for children whose mothers had less than a secondary school education as opposed to those whose mothers had at least a secondary school education. In terms of paternal residential status, there was a stronger association between mothers’ engagement in counting/naming objects and children’s ability to identify letters for children who lived with their fathers in comparison with those who did not (OR = 1.17).

Discussion

Joining attempts to establish the importance of early parental investment and childhood outcomes more broadly across cultural communities (see Baker, 2015; Baker & Vernon-Feagans, 2015; McWayne et al., 2013), we used data from the UNICEF MICS to explore relations between early parent-child cognitive engagement activities and children’s early literacy skills in 25 low- and middle-income countries in Africa. More specifically, drawing on propositions within frameworks on parenting and parent-child stimulation activities and childhood development (Isbell et al., 2004; Mol & Bus, 2011; Morrison & Cooney, 2002; Shonkoff, 2010), we determined associations between maternal and paternal engagement in reading, telling stories, and naming/counting and preschool-aged children’s literacy skills and whether certain sociodemographic factors and preschool enrollment moderated these associations.

As in other studies on African families (e.g., Bornstein et al., 2017), there was quite a bit of variation in the levels of maternal and paternal engagement in reading, storytelling, and naming/counting objects across the 25 countries. For instance, parents in Algeria and Tunisia were far more likely to engage in cognitive activities with children than were those in other parts of Africa, such as Gambia, Madagascar, and Swaziland (see Table 2). What stood out were the low levels of engagement in the three cognitive activities by both parents across most countries, and in no country did paternal engagement exceed maternal engagement. In fact, in most countries a small percentage of fathers (10% to 15%) engaged in reading, telling stories, and naming/counting objects with children. These rates of engagement are way below those of parents in the United States and England (Duursma, 2016; McMunn, Martin, Kelly, & Sacker, 2015) and in the developing countries of the Caribbean (e.g., Barbados, Guyana, and Jamaica; Dede Yildirim & Roopnarine, 2017). Whether the discrepancies in rates of maternal and paternal engagement between the African countries and those in the United States and the Caribbean are due to a lack of awareness of the benefits of and beliefs in early parent-child stimulation activities, economic conditions, or scarce literacy-type materials in the home environment is not clear from the data.

Our main hypothesis that the associations between maternal and paternal book reading, storytelling, and naming/counting objects and children’s letter identification, reading simple words, and recognizing symbols would be stronger for mothers and children than for fathers and children was not confirmed. Instead, there were similar patterns of positive associations between the three indicators of cognitive engagement and children’s literacy skills for both mothers and fathers. These associations are in agreement with those obtained in two small-scale studies conducted in Kenya on paternal investment in children’s educational activities (Mwoma, 2009; Wanjiku, 2016), and further underline the importance of both maternal and paternal cognitive engagement activities for childhood development during the preschool years in challenging ecological niches where poor economic conditions, a shortage of literacy materials in the home, and low parental educational attainment can be a hindrance to investment in childhood stimulation activities (see Dede Yildirim & Roopnarine, 2017).

Along with initiatives to improve the health and nutritional status of children, there have been calls for greater emphasis in advancing early childhood development through quality early parent-child positive engagement activities in low- and middle-income countries (Engle et al., 2011; Roopnarine, Johnson, Quinn, & Patte, 2018; UNICEF, 2016; World Health Organization, 2010). Additionally, persuasive arguments have been made to frame childhood development in culturally situated conceptions of parenting and early education (Ng’asike, 2018; Nsamenang, 2008, 2011; Serpell, 1996). In this vein, concerns have been raised about the transferability of parenting interventions from the high- to the low-income countries (see Devlin et al., 2018, for discussion). With these two concerns in mind, we were able to show that the patterns and degree of associations between parental engagement activities and children’s literacy skills during the preschool years in the 25 African countries were similar to those determined for families in high-income countries (McWayne et al., 2013; Mol & Bus, 2011). These findings offer support for the implementation of the positive parenting model in low- and middle-income countries.

The hypothesis on the moderating role of sociodemographic factors, such as maternal education, children’s age and gender, enrollment in preschool, and fathers’ residential status, on the associations between parental engagement and childhood outcomes received partial support. Maternal education did have a significant moderating role on the associations between maternal and paternal cognitive engagement activities and children’s literacy skills. Children whose mothers had at least a secondary school education were more likely to demonstrate basic literacy skills than those whose mothers did not. With respect to child characteristics, there were stronger associations between maternal and paternal engagement activities and children’s literacy skills for older than younger children. The age trends make intuitive sense given that parents tend to increase their levels of joint activities with children as their cognitive maps become increasingly more mature. Contrary to assertions about gendered patterns of socialization in African countries (see Mwoma, 2015), the child’s sex assumed a minor role in tempering the associations between parental engagement activities and children’s literacy skills. More research is needed to further delineate the role of child sex in parental cognitive engagement.

A factor that consistently modified the associations between maternal and paternal cognitive engagement activities and children’s literacy skills was preschool education. That is, children who attended a preschool program seem to profit more from maternal and paternal cognitive engagements compared with those who did not attend a preschool program. It is reasonable to assume that exposure to linguistic activities in preschool may work in collaboration with parental engagement activities at home to enhance children’s early literacy skills development. Preschool education remains a viable mechanism for boosting children’s early cognitive skills in low- and middle-income countries (see Burger, 2010; Loeb et al., 2007), and in some, such as those in the Caribbean region, it contributed uniquely to children’s literacy skills above and beyond maternal cognitive engagement (Dede Yildirim & Roopnarine, 2017). Unfortunately, preschool enrollment is quite low in a number of African countries (see Table 1).

With one exception, father’s residential status did not seem to temper the associations between maternal or paternal engagement activities and children’s literacy skills in the African countries. In the high-income countries such as the United States, living in a nonresident father household was not consistently associated with lower academic achievement and social adjustment difficulties (Adamsons & Johnson, 2013; Carlson & McLanahan, 2010; Castillo, Welch, & Sarver, 2011). As indicated earlier, alloparenting is common in a number of cultural groups in Africa (Marlowe, 2005), and it is possible that children in households without resident fathers may have benefitted from linguistic exchanges with other familial members (e.g., siblings, grandparents, community fathers). At the same time, some nonresident fathers could have remained connected to their children through distal involvement. Several studies have cataloged different ways in which nonresident fathers in high-income countries stay in touch with and offer resources to their children (e.g., Cheadle, Amato, & King, 2010; Nepomnyaschy & Garfinkel, 2010). Distal interactions between African fathers and children could have also lessened the impact of not having a resident father on children’s literacy skills.

Limitations and Implications

As far as we know, this study is one of the first attempts to explore the associations between three major early cognitive stimulation activities and children’s early literacy skills across a range of low- and middle-income African countries that are characterized by diverse family structural and social organization patterns for bearing and rearing children. In spite of its broad scope in attempting to examine links between parental cognitive engagement activities and children’s early literacy skills in countries across the African continent, this study has several limitations. Key among them are the correlational nature of the study and the fact that the UNICEF MICS recorded whether parents engaged in book reading, storytelling, and naming/counting without any attention to the quality and complexity of these activities with children. The same can be said for the childhood literacy skills indicators. Undoubtedly, detailed information on the nature of the parental engagement activities and standardized assessments of children’s early language skills in more controlled experiments would be more desirable for assessing causal links between the two sets of constructs. A related concern entails the total reliance on single-source reports for all data. The drawbacks of this method of collecting data are well-known and include method-variance issues and reporting bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). It would have also been helpful had we assessed and included the relationship quality between spouses/partners and support networks in our analyses as these family factors have been shown to influence paternal engagement with children and to mediate associations between parenting and childhood outcomes (Roopnarine & Dede Yildirim, 2018).

These limitations aside, our findings provide additional support regarding the importance of maternal and paternal cognitive engagement for advancing children’s literacy skills during the preschool years when impressive brain development occurs. That there were consistent associations between three major early parental engagement activities and children’s literacy skills for both mothers and fathers, and across so many low- and middle-income African countries should strengthen confidence in these findings. As indicated above, emphasis has been on culturally situated efforts to strengthen parent-child engagement and, relatedly, the transferability of intervention modalities from high- to low-income countries. A message here is that parent-child engagement activities that are commonly held as important for childhood development in high-income countries may have the same benefit in fostering early childhood development in the less developed nations of the world, and some activities such as storytelling and oral narratives that are endemic to African cultural communities may have more currency in shaping children’s literacy skills. It appears that when booking reading, storytelling, and naming/counting are coupled with preschool enrollment, they may have an added advantage in facilitating children’s early literacy skills.

Footnotes

Acknowledgements

The second author was supported by funds made available through a Pearls S. Falk Professorship in the David B. Falk College of Sport and Human Dynamics at Syracuse University. This support is gratefully acknowledged.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The second author was supported by funds made available through a Pearls S. Falk Professorship in the David B. Falk College of Sport and Human Dynamics at Syracuse University.

ORCID iD

Elif Dede Yildirim

References

Adamsons

Johnson

S. K.

(2013). An updated and expanded meta-analysis of nonresident fathering and child well-being. Journal of Family Psychology, 27, 589-599. doi:10.1037/a0033786

Baker

C. E.

(2015). African American and Hispanic fathers’ work characteristics and preschool children’s cognitive development. Journal of Family Issues, 37, 1514-1534.

Baker

C. E.

Vernon-Feagans

(2015). Fathers’ language input during shared book activities: Links to children’s kindergarten achievement. Journal of Applied Developmental Psychology, 36, 53-59. doi:10.1016/j.appdev.2014.11.009

Bates

Maechler

Bolker

Walker

(2015). Fitting linear mixed-effect models using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01

Baumrind

(1996). Parenting: The discipline controversy revisited. Family Relations, 45, 405-414. doi:10.2307/585170

Bornstein

M. H.

Putnick

D. L.

Oburu

Lansford

J. E.

Deater-Deckard

Bradley

R. H.

. . .Britto

P. R.

(2017). Parenting, environment, and early child development in sub-Saharan Africa. In Abubakar

van de Vijver

F. J. R.

(Eds.), Handbook of applied developmental science in sub-Saharan Africa (pp. 15-33). New York, NY: Springer.

Burger

(2010). How does early care and education affect cognitive development? An international review of the effects of early interventions for children from different social backgrounds. Early Child Research Quarterly, 25, 140-165.

Bus

A. G.

van Ijzendoorn

M. H.

Pellegrini

A. D.

(1995). Joint book reading makes for success in learning to read: A meta-analysis on intergenerational transmission of literacy. Review of Educational Research, 64, 1-21.

Carlson

M. J.

McLanahan

S. S.

(2010). Fathers in fragile families. In Lamb

M. E.

(Ed.), The role of the father in child development (5th ed., pp. 241-269). Hoboken, NJ: Wiley.

10.

Castillo

Welch

Sarver

(2011). Fathering: The relationship between fathers’ residence, fathers’ sociodemographic characteristics, and father involvement. Maternal and Child Health Journal, 15, 1342-1349.

11.

Cheadle

Amato

P. R.

King

(2010). Patterns of nonresident father contact. Demography, 47, 205-225.

12.

Dede Yildirim

Roopnarine

J. L

. (2017). Paternal and maternal engagement across six Caribbean countries and childhood outcomes. Journal of Applied Developmental Psychology, 53, 64-73. doi:10.1016/j.appdev.2017.08.007

13.

Devlin

A. M.

Wight

Fenton

(2018). Are parenting practices associated with the same child outcomes in sub-Saharan African countries as in high-income countries? A review and synthesis. BMJ Global Health, 3(6). doi:10.1136/bmjgh-2018-000912

14.

Duncan

Magnuson

Votruba-Drzal

(2014). Boosting family income to promote child development. Future of Children, 24, 99-120.

15.

Duursma

(2016). Who does the reading, who the talking? Low-income fathers and mothers in the US interacting with their young children around a picture book. First Language, 36, 465-484. doi:10.1177/0142723716648849

16.

Duursma

Pan

B. A.

Raikes

(2008). Predictors and outcomes of low-income fathers’ reading with their toddlers. Early Childhood Research Quarterly, 23, 351-365. doi:10.1016/j.ecresq.2008.06.001

17.

Duval

Tweedie

(2000). Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics, 56, 455-463.

18.

Egger

Smith

G. D.

Schneider

Minder

(1997). Bias in meta-analysis detected by a simple, graphical test. British Medical Journal, 315, 629-634.

19.

Engle

P. L.

Black

M. M.

(2008). The effect of poverty on child development and educational outcomes. Annals of the New York Academy of Sciences, 1136, 243-256. doi:10.1196/annals.1425.023.

20.

Engle

P. L.

Fernald

L. C. H.

Alderman

Behrman

O’Gara

Yousafzai

. . . Iltus

(2011). Strategies for reducing inequalities and improving developmental outcomes for young children in low-income and middle-income countries. Lancet, 378, 1339-1353. doi:10.1016/S0140-6736(11)60889-1

21.

Fagan

Iglesias

Kaufman

(2016). Associations among Head Start fathers’ involvement with their preschoolers and child language skills. Early Child Development and Care, 186, 1342-1356. doi:10.1080/03004430.2015.1094654

22.

Fouts

(2008). Father involvement with young children among the Aka and Bofi foragers. Cross-Cultural Research, 42, 290-312.

23.

Fouts

(2013). Fathering in Central and East Africa: Cultural and adaptationist perspectives in small-scale societies. In Shwalb

Shwalb

Lamb

M. E.

(Eds.), Fathers in cultural context (pp. 151-172). New York, NY: Routledge.

24.

Gordon

Lumley

(2017). Forestplot: Advanced forest plot using “grid” graphics (R package version 1.7.2) [Computer software]. Retrieved from https://rdrr.io/cran/forestplot/

25.

Harkness

Super

C. M.

(1992). From parents’ cultural belief systems to behavior. In Eldering

Leseman

(Eds.), Effective early intervention: Cross-cultural perspectives (pp. 67-90). New York, NY: Falmer Press.

26.

Hart

Risley

T. R.

(1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Paul H. Brookes.

27.

Hewlett

B. S.

MacFarlan

S. J.

(2010). Fathers’ roles in hunter-gatherer and other small-scale societies. In Lamb

M. E.

(Ed.), The role of the father in child development (pp. 413-434). Hoboken, NJ: Wiley.

28.

Isbell

Sobol

Lindauer

Lowrance

(2004). The effects of storytelling and story reading on the oral language complexity and story comprehension of young children. Early Childhood Education Journal, 32, 157-163.

29.

Jeong

McCoy

D. C.

Yousafzai

A. K.

Salhi

Fink

(2016). Paternal stimulation and early child development in low- and middle-income countries. Pediatrics, 138(4), e20161357. doi:10.1542/peds.2016-1357

30.

Kaufman

A. S.

Kaufman

N. L.

(1993). K-SEALS: Kaufman survey of early academic and language skills. Circle Pines, MN: American Guidance Service.

31.

Kramer

R. A.

Allen

Gergen

P. J.

(1995). Health and social characteristics and children’s cognitive functioning: Results from a national cohort. American Journal of Public Health, 85, 312-318.

32.

Levtov

van der Gaag

Greene

Kaufman

Barker

(2015). State of the world’s fathers. Retrieved from https://promundoglobal.org/resources/state-of-the-worlds-fathers-2015/

33.

Liaw

Brooks-Gunn

(1994). Cumulative familial risks and low birth weight children’s cognitive and behavioral development. Journal of Clinical Child Psychology, 23, 360-372. doi:10.1207/s15374424jccp2304_2

34.

Lima

Caughy

Nettles

S. M.

O’Campo

P. J.

(2010). Effects of cumulative risk on behavioral and psychological well-being in first grade: Moderation by neighborhood context. Social Science & Medicine, 71, 1447-1454. doi:10.1016/j.socscimed.2010.06.022

35.

LoCasale-Crouch

Mashburn

A. J.

Downer

J. T.

Pianta

R. C.

(2008). Pre-kindergarten teachers’ use of transition practices and children’s adjustment to kindergarten. Early Childhood Research Quarterly, 23, 124-139. doi:10.1016/j.ecresq.2007.06.001

36.

Loeb

Bridges

Bassok

Fuller

Rumberger

(2007). How much is too much? The influence of preschool centers on children’s social and cognitive development. Economics of Education Review, 26, 52-66. doi:10.1016/j.econedurev.2005.11.005

37.

Magnuson

Waldfogel

(2008). Steady gains and stalled progress: Inequality and the Black-White test score gap. New York, NY: Russell Sage Foundation.

38.

Makusha

Richter

(2015). Black fathers in South Africa. In Roopnarine

J. L.

Roopnarine

J. L.

(Eds.), Fathers across cultures: The importance, roles, and diverse practices of dads (pp. 391-409). Santa Barbara, CA: Praeger/ABC-CLIO.

39.

Marlowe

(2005). Who tends Hadza children? In Hewlett

B. S.

Lamb

M. E.

(Eds.), Hunter-gatherer childhoods: Evolutionary, developmental, and cultural perspectives (pp. 175-213). New Brunswick, NJ: Aldine Transaction.

40.

Martel

(2016). Review of options for reporting water, sanitation and hygiene coverage by wealth quintile (MICS Methodological Paper No. 4). New York, NY: Data and Analytics Section Division of Data, Research and Policy, UNICEF.

41.

McCoy

D. C.

Peet

E. D.

Ezzati

Goodarz

Black

M. M.

Sudfeld

C. R.

. . . Fink

(2016). Early childhood developmental status in low- and middle-income countries: National, regional, and global prevalence estimates using predictive modeling. PLoS Medicine, 14(1), e1002233. doi:10.1371/journal.pmed.1002034

42.

McMunn

Martin

Kelly

Sacker

(2015). Fathers’ involvement: Correlates and consequences for child socioemotional behavior in the United Kingdom. Journal of Family Issues, 38, 1109-1131. doi:10.1177/0192513X15622415

43.

McWayne

Downer

J. T.

Campos

Harris

R. D.

(2013). Father involvement during early childhood and its association with children’s early learning: A meta-analysis. Early Education and Development, 24, 898-922. doi:10.1080/10409289.2013.746932

44.

Mistry

R. S.

Biesanz

J. C.

Taylor

L. C.

Burchinal

Cox

M. J.

(2004). Family income and its relation to preschool children’s adjustment for families in the NICHD study of early child care. Developmental Psychology, 40, 727-745. doi:10.1037/0012-1649.40.5.727

45.

Mol

S. E.

Bus

A. G.

(2011). To read or not to read: A meta-analysis of print exposure from infancy to early adulthood. Psychological Bulletin, 137, 267-296. doi:10.1037/a0021890

46.

Morrison

F. J.

Cooney

R. R.

(2002). Parenting and academic achievement: Multiple paths to early literacy. In Borkowski

J. G.

Ramey

S. L.

Bristol-Power

(Eds.), Parenting and the child’s world: Influences on academic, intellectual, and social-emotional development (pp. 141-160). Mahwah, NJ: Lawrence Erlbaum.

47.

Morrow

(1985). Retelling stories: A strategy for improving young children’s comprehension, concept of story structure, and oral language complexity. Elementary School Journal, 85, 647-661. doi:10.1086/461427

48.

Mwoma

T. B.

(2009). Paternal involvement in children’s education: An implication of children’s performance at preschool in Gucha district Kenya (Unpublished doctoral dissertation). Kenyatta University, Nairobi, Kenya.

49.

Mwoma

T. B.

(2015). Fathering in Kenya. In Roopnarine

J. L.

(Ed.), Fathers across cultures: The importance, roles, and diverse practices of dads (pp. 410-428). Santa Barbara, CA: Praeger/ABC-CLIO.

50.

Nepomnyaschy

Garfinkel

(2010). Child support enforcement and fathers’ contributions to their nonmarital children. Social Service Review, 84, 341-380. doi:10.1086/655392

51.

Ng’asike

(2018). Early childhood education development in Kenya. In Roopnarine

J. L.

Johnson

J. E.

Quinn, S.

S. F.

Patte

(Eds.), Handbook of international perspectives on early childhood education (pp. 263-281). London, England: Routledge.

52.

Nsamenang

A. B.

(2008). (Mis)Understanding ECD in Africa: The force of local and global motives. In Garcia

Pence

Evans

J. L.

(Eds.), Africa’s future, Africa’s challenge: Early childhood care and development in sub-Saharan Africa (pp. 135-149). Washington, DC: World Bank.

53.

Nsamenang

A. B.

(2011). Developmental learning in African cultural circumstances. In Nsamenang

A. B.

Tchombe

T. M. S.

(Eds.), Handbook of African educational theories and practices: A generative teacher education curriculum (pp. 233-244). Bamenda, Cameroon: Human Development Resource Centre.

54.

Ogbu

(1981). Origins of human competence: A cultural ecological perspective. Child Development, 52, 413-429. doi:10.2307/1129158

55.

Podsakoff

P. M.

MacKenzie

Lee

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. doi:10.1037/0021-9010.88.5.879

56.

Pong

Hao

(2007). Neighborhood and school factors in the school performance of immigrants’ children. International Migration Review, 41, 206-241.

57.

Putnick

Bornstein

M. H.

Lansford

J. E.

Chang

Deater-Deckard

Di Giunta

. . . Bombi

A. S.

(2012). Agreement in mother and father acceptance-rejection, warmth, and hostility/rejection/neglect of children across nine cultures. Cross-Cultural Research, 46, 191-223. doi:10.1177/1069397112440931

58.

R Core Team. (2018). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.

59.

Raykov

Dimitrov

D. M.

Asparouhov

. (2010). Evaluation of scale reliability with binary measures using latent variable modeling. Structural Equation Modeling: A Multidisciplinary Journal, 17(2), 265-279. doi:10.1080/10705511003659417

60.

Reardon

S. F.

(2011). The widening academic achievement gap between rich and poor: New evidence and possible explanations. In Duncan

G. J.

Murnane

R. J.

(Eds.), Whither opportunity? Rising inequality, schools, and children’s life chances (pp. 91-115). New York, NY: Russell Sage Foundation.

61.

Reardon

S. F.

Robinson

J. P.

(2008). Patterns and trends in racial/ethnic and socioeconomic academic achievement gaps. In Ladd

H. F.

Fiske

E. B.

(Eds.), Handbook of research in education finance and policy (pp. 497-516). Mahwah, NJ: Lawrence Erlbaum.

62.

Roopnarine

J. L.

(Ed.). (2015). Fathers across cultures: The importance, roles, and diverse practices of dads. New York, NY: Praeger Press.

63.

Roopnarine

J. L.

Dede Yildirim

(2018). Influence of relationship skills education on pathways of associations between paternal depressive symptoms and IPV and childhood behaviors. Psychology of Men and Masculinity, 19, 223-233. doi:10.1037/men0000100

64.

Roopnarine

J. L.

Dede Yildirim

(2019a). Fathers across cultures: Developmental and clinical issues. London, England: Routledge.

65.

Roopnarine

J. L.

Dede Yildirim

(2019b). Fathers’ cognitive engagement and children’s literacy skills in three ethnic groups in Trinidad and Tobago. Manuscript submitted for publication.

66.

Roopnarine

J. L.

Johnson

J. E.

Quinn

Patte

(Eds.). (2018). Handbook of international perspectives on early childhood education. London, England: Routledge.

67.

Roopnarine

J. L.

Krishnakumar

Logie

Narine

Davidson

K. L.

(2017). Moderating role of neighborhood characteristics on the associations between parenting practices and children’s early language skills in Trinidad and Tobago. Caribbean Journal of Psychology, 9, 21-35.

68.

Samms-Vaughan

(2005). The Jamaican pre-school child: The status of early childhood development in Jamaica. Kingston, Jamaica: Planning Institute of Jamaica.

69.

Sastry

Pebley

A. R.

(2010). Family and neighborhood sources of socioeconomic inequality in children’s achievement. Demography, 47, 777-800. doi:10.1353/dem.0.0114

70.

Serpell

(1996). Cultural models of childhood in indigenous socialization and formal schooling in Zambia. In Philip Hwang

Lamb

M. E.

Sigel

I. E.

(Eds.). Models of childhood (pp. 129-142). Mahwah, NJ: Lawrence Erlbaum.

71.

Shonkoff

J. P.

(2010). Building a new biodevelopmental framework to guide the future of early childhood policy. Child Development, 81, 357-367. doi:10.1111/j.1467-8624.2009.01399.x

72.

Shonkoff

J. P.

Phillips

D. A.

(2000). From neurons to neighborhoods: The science of early childhood development. Washington, DC: National Academies Press.

73.

UNICEF. (2015). Monitoring the situation of children and women for 20 years: The Multiple Indicator Cluster Surveys (MICS) 1995–2015. New York, NY: United Nations Children’s Fund.

74.

UNICEF. (2016). State of the world’s children 2016: A fair chance for every child. New York, NY: United Nations Children’s Fund.

75.

Varghese

Wachen

(2016). The determinants of father involvement and connections to children’s literacy and language outcomes: Review of the literature. Marriage & Family Review, 52, 331-359. doi:10.1080/01494929.2015.1099587

76.

Viechtbauer

(2010). Conducting meta-Analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1-48. doi:10.18637/jss.v036.i03

77.

Votruba-Drzal

Miller

Coley

R. L.

(2015). Poverty, urbanicity, and children’s development of early academic skills. Child Development Perspectives, 10, 3-9. doi:10.1111/cdep.12152

78.

Wanjiku

K. N.

(2016). Influence of paternal involvement on children’s performance in number work, and language activities in public preschools in Starehe subcounty, Nairobi County (Master’s thesis). Retrieved from http://erepository.uonbi.ac.ke/bitstream/handle/11295/99593/Karanja_Influence%20of%20Paternal%20Involvement%20on%20Children%E2%80%99s%20Performance%20in%20Number%20Work%20and%20Language%20Activities%20in%20Public%20Preschools%20in%20Starehe%20Subcounty%2C%20Nairobi%20County.pdf?sequence=1&isAllowed=y

79.

Whiting

B. B.

Whiting

J. W. M.

(1975). Children of six cultures: A psycho-cultural analysis. Cambridge, MA: Harvard University Press.

80.

Wolf

McCoy

D. C.

(2019). Household socioeconomic status and parental investments: Direct and indirect relations with school readiness in Ghana. Child Development, 90, 268-270. doi:10.1111/cdev.12899

81.

World Health Organization. (2010). Parenting for lifelong health (PLH). Geneva, Switzerland: Author.