Protective indigenous collective value of Ubuntu and child Protective indigenous collective value of Ubuntu and child neglect: Implications for rural child protection practice neglect: Implications for rural child protection practice

Theories on collective efficacy and social support suggest that indigenous values that support collective practices and sanction community obligations to childcare would be protective against child neglect. Likewise, new qualitative findings show that collective values are stronger in rural areas than in urban. This study tested the claims that the value of Ubuntu, which is a symbolic cultural value of ‘ being for others ’ , will be protective against the likelihood of neglect; this relationship will be stronger in rural compared with urban communities in Ghana. Using data obtained from a nationally representative sample of 1100 mothers (from 22 communities) in Ghana, we tested the claims using fixed effects logistic regression. The Ubuntu norms were significantly endorsed in rural communities compared with the urban. The overall model showed that higher levels of Ubuntu are associated with lower odds of child neglect ( OR .47, [.29, .76] p < 0.05), and the relationship remained significant only in the rural sample ( OR .13, [.06, .31] p < 0.001). Similar evidence was recorded for the Ubuntu norms of community care and compassion. The results suggest that child protection in rural Ghana can be fruitful when interventions are developed to boost the value of Ubuntu and the norms of collective childcare


| INTRODUCTION
Growing evidence within the field of community-based child protection and child maltreatment prevention highlights the function of collective values that propel community actions against child maltreatment (Coulton et al., 2007;McDonell et al., 2015;Molnar et al., 2021).Where there are legitimate and culturally sanctioned collective practices to support families and care for children, child neglect (and other kinds of maltreatment) would be expected to be less prevalent.This logic, which is based on culture and value-driven community social support, has informed most community-based child maltreatment interventions (Gross-Manos & Cohen, 2022) such as the Strong Communities for Children (Melton, 2014), Community Partnerships for Protecting Children (CPPC) and the Durham Family Initiative (cf.Daro & Dodge, 2009).Indeed, empirical findings show that child maltreatment is less prevalent in communities with high collective practices (Nadan et al., 2015).Researchers and advocates for indigenous approaches to social work and community-based child protection in Africa espouse the need to centre the value of Ubuntu within social work and community-based child protection practice (Mayaka & Truell, 2021;Mugumbate & Nyanguru, 2013).Ubuntu demonstrates a communal belief often expressed as 'I am because who we all are' and in Zulu language as 'ubuntu ngumuntu ngabantu' (Mugumbate & Chereni, 2019).The norms underpinning Ubuntu: mutual support, respect, collective care, collective well-being and compassion (empathy) (Mayaka & Truell, 2021) signify a collective commitment to the well-being of members in the community.Mugumbate and Chereni (2019) argued that the norms of Ubuntu could significantly impact the rates of child maltreatment in communities when they are translated into positive practices that support families in caring for children.This is because beliefs in Ubuntu are embedded within the relationship and community support practices (Muwanga-Zake, 2009).Their proposition is underpinned by the core maxim of Ubuntu, which says 'it takes a village to raise a child' (Mugumbate & Chereni, 2019).Although there has been a significant increase in research on Ubuntu and the benefits to community well-being (cf.Chilwalo, 2020;Mayaka & Truell, 2021;Mugumbate & Chereni, 2019;Mugumbate & Nyanguru, 2013), to the best of our knowledge, no empirical research has examined the protective effects of Ubuntu (including its associated norm of collective care) on child neglect.This study sought to establish this relationship for the first time and offer practical solutions to boost Ubuntu and indigenous community social support framework in child maltreatment prevention.

| Child neglect in Ghana
Over the past three decades, global research has identified child neglect as the most common, most reported and substantiated form of child maltreatment (Radford et al., 2013;Solem et al., 2020;Stoltenborgh et al., 2013).Over three out of every four children in Ghana was reported to have experienced some form of neglect during the peak of the COVID-19 pandemic (UNICEF, 2020).Estimates from a recent nationally representative survey in Ghana showed 56% overall lifetime prevalence of neglect (Abdullah et al., 2023), with the majority of them happening in rural settlements.For instance, over 19% of parents in rural areas reported to have left their children in an unsafe environment more than three times in the past year, compared with only 4% for parents in urban areas (cf.Abdullah et al., 2023).The prevalence of neglect in rural areas may be motivated by structural factors, such as poverty, which is most prevalent in rural Ghana (Dzanku, 2015).Though neglect has been found to be common in rural Ghana, in contrast, collective values (including the value of Ubuntu) and collective practices are known to be stronger in rural areas compared with urban areas (Abdullah & Frederico et al., 2020;Nukunya, 2003).
Findings from a recent study on the conceptualization of neglect revealed that most parents in Ghana understand child neglect as caregiver actions or inactions, including omission in care and failure to provide basic needs for children (Manful & Abdullah, 2021), which agreed with widely accepted definitions of neglect (Dubowitz et al., 2005).
However, some studies identified nuances, including findings on contextual and cultural elements, such as concentrated poverty, inheritance expectations (Manful & Abdullah, 2021) and gendered norms (see Awortwe et al., 2020;Ayim et al., 2023, for further reading), as key variables that influence neglect in Ghana.Though the cultural component (cf.Awortwe et al., 2020;Ayim et al., 2023) concurs with suggestions about context and the influence of culture on child neglect definition (Lonne, 2015), evidence on the thematic constructs identified in the only Ghanaian study that focused on parents' conceptualization of neglect (Manful & Abdullah, 2021) suggested that standard practices that support parents to meet the basic needs of children may contribute to reducing the amount of neglect in Ghanaian communities.Moreover, if these social support practices were ingrained in the culture and normative structure of communities, the effect could be greater.

| Theoretical link between Ubuntu and child neglect
Ubuntu is a philosophy that espouses the values of 'being human' and sanctions traditional normative practices that keep the community together.The word 'Ubuntu' is derived from the Bantu and Nguni languages of people in Southern Africa, which symbolizes a collectivist value orientation (Mugumbate & Nyanguru, 2013).Researchers (Lutz, 2009;Mayaka & Truell, 2021) have identified Ubuntu as the primary philosophy and value behind collectivism in Africa.The core values of Ubuntu are expressed with languages such as bumuntu in Tanzania, unhu/botho in Zimbabwe, gimuntu in Angola, ubuntu in Bostwana, bato in Cameroon, biako ye/abrewatia in Ghana, maaya in Burkina Faso, mutunchi/ogwa in Nigeria, maaya in Gambia and so forth (Mugumbate & Chereni, 2019).Thereby suggesting that although Ubuntu belief is widespread across Africa, the philosophy is represented using different nouns in different societies.The common meaning of these words is a confirmation that values and normative practices espoused in Ubuntu are upheld across different communities in Africa.For instance, community solidarity practices such as caring for orphans and vulnerable children in communities are established in Ghana (Goody, 1966(Goody, , 1973) ) and Zimbabwe communities (Mugumbate & Chereni, 2019).A key proverbial statement of Ubuntu 'mwana wa mnzako ngwako yemwe, ukachenjera manja udya naye' meaning your neighbour's child is your own child, represents collectivist and key child protection element in Ubuntu (Mugumbate & Nyanguru, 2013).Within Ubuntu culture, individuals are deemed to satisfy their common good through the process of pursuing good for the community (Lutz, 2009).Molose (2019) identified community solidarity, mutual respect, collective survival, collective responsibility to care for children and compassion (empathy) among the core normative underpinnings (tenets) of Ubuntu.These normative underpinnings were validated in his measurement model for Ubuntu.Guided by the normative expectations of Ubuntu, a right and acceptable action within society is one that is deemed to promote harmony, mutual support, community togetherness, solidarity, empathy and benevolence, kindness and caring for others in the community (Hailey, 2008;Metz, 2007).Confrontation, vengeance and retribution are not accepted in Ubuntu-driven communities.The link between the practices of Ubuntu and child neglect can be explained via social support and collective efficacy theory.Literature on social support theory (Lakey & Cohen, 2000) and collective efficacy theory (Sampson et al., 1997) suggests that practices that promote the norms of Ubuntu-community care, mutual respect/support, community solidarity and compassion-may help to reduce the incidence of neglect in communities.

| Social support
The social constructionist preposition of social support theory (Lakey & Cohen, 2000) argues that social support from network members can impact well-being outcomes via self-regulation and selfesteem media, irrespective of the influence of risk factors.This means that in communities where there is strong mutual support among community members, parents may be less likely to neglect their children, regardless of the presence of risk factors, such as poverty.
Empirical evidence from a review of research on the relationship between social support and child maltreatment affirmed that social support ameliorates the risk of child neglect (Thompson, 2015).Parents with documented histories of childhood neglect reported lower levels of social support (Sperry & Widom, 2015).Sampson et al. (1997) theorized collective efficacy as a combination of neighbourhood social cohesion (mutual trust, bond and support) and informal social control (expectations to intervene in undesirable behaviours).They argued that crime and maltreatment would be less in communities that are high in social cohesion (also called solidarity), and members have higher expectations that people will intervene to correct wrongs (high in informal social control).Sampson et al.'s (1997) theory has informed the majority of crime and child maltreatment research in the USA.A review of a decade of research on the association between collective efficacy and child maltreatment largely supported Sampson et al. (1997), with evidence showing that neighbourhood collective efficacy (social cohesion and informal social control) predicted fewer instances of neglect and abuse (Abdullah, Emery & Jordan, 2020).Maguire-Jack and colleagues found robust evidence from the Fragile Family and Child Well-being data in the USA, and other community surveys to support the claim that neighbourhood social cohesion is negatively associated with child neglect and abuse (cf.Barnhart & Maguire-Jack, 2016;Kim & Maguire-Jack, 2015;Maguire-Jack et al., 2022;Maguire-Jack & Showalter, 2016).The relationship between Ubuntu and child neglect may follow a similar trend because Ubuntu propagates norms of community solidarity, mutual support and collective childcare duties.If confirmed, the findings will provide significant evidence to consider Ubuntu (and associated norms) as the foci for community intervention to prevent child maltreatment.Such approaches can build on traditional practices such as zunderamambo; a practice where village members come together to grow food, which is used to cater to the needy in the community (Mugumbate & Nyanguru, 2013).

| Collective efficacy
Whether the relationship follows the paths of social support theory or collective efficacy theory, the outcome will be desirable to communities and community-based child maltreatment prevention.Some evidence suggested that in cases where neighbourhood social cohesion (collective efficacy) does not directly predict neglect, the effect can be indirectly mediated by social support (Xu et al., 2020).
An indication that the two pathways may cohere to explain the effects of Ubuntu on child neglect.

| Objective and hypothesis
This study sought to contribute to the community approach to child maltreatment prevention by examining the relationship between Ubuntu and child neglect.It hoped to achieve this objective through testing the following hypotheses: H1. Ubuntu will be associated with lower odds of child neglect.
H2.The normative underpinnings of Ubuntu (respect, community care, collective solidarity, community survival and compassion) will each predict lower odds of child neglect.

H3. Association between Ubuntu, norms of Ubuntu
and child neglect will be stronger in rural communities compared with urban communities in Ghana.

| Data and procedure
We obtained data from a nationally representative sample of 1100 mothers in Ghana following a stratified four-stage random probability proportional to size (PPS) sampling approach.The four-stage stratification entailed rural (59%) and urban (41%) stratification.Although Ghana is 57.3% urban, according to the 2010 population census, we oversampled from rural communities because collective norms and traditional practices are deemed to be common in rural areas (Nukunya, 2003).Hence, we conducted random PPS sampling to select seven districts (four rural and three urban) using data from the 2010 Ghana census.The PPS cluster sampling technique has the added advantage of increasing the probability of random selection based on the proportion of the cluster to the total population.Again, using the PPS approach, we randomly selected 22 communities from the seven districts (at least three communities per district).We randomly selected 50 mothers from each community through a nextdoor-neighbour sampling, facilitated by Google Maps.We developed community maps from Google Maps.On each community map, we carefully demarcated the settlement into equal regions (neighbourhoods/suburbs) using latitude and longitude and assigned numbers to each region.We used a random draw from a uniform distribution to select five regions from the total regions on the settlement map.These regions were used as start points to select a neighbourhood cluster of 10 households.
Six research assistants from a University in Ghana were trained to locate 10 households close to each start-point and visit the households to recruit mothers with primary childcare duties.Because the study had a sole focus on child neglect incidence perpetrated by primary caregiving mothers, only mothers with current caregiving duties were considered.Where there were two or more caregiving mothers in the household, mothers with the most recent birthday were selected.This yielded 50 mothers per community (10 each from the five neighbourhood cluster in the community) and a total sample of 1100 mothers.Each research participant provided written or thumbprint consent prior to their participation in the study.The researchers provided information about the project, including the objectives and information about the participants' rights to participate and how to withdraw from the study.A 95% response rate was recorded, an indication that the project was welcomed by the mothers.An honorarium of 10 Ghana cedis (USD 2.00 at the time of data collection) was provided to each mother as compensation for their time.
To ensure the questionnaires fitted the local contexts, the final questionnaire was translated into Twi (the common lingua franca in Ghana) by an academic from Ghana and back-translated into English.
Measures, including the de-identification of identifiable information from the survey, were carried out to maintain the confidentiality and privacy of the participants.As such, the telephone details of the participants were separated and stored in separate files before the analysis of the data.Completed surveys were also stored in a secured cabinet in the office of the PI, and data were stored on the passwordprotected computer of the PI.Formal ethics approval for the study was obtained from the University of Hong Kong.

Child neglect
We measured child neglect using the neglect subscale of the Conflict Tactics Scale (CTS) (Straus et al., 1998).Mothers reported instances where they had neglected their children by endorsing the items: (1) were not able to make sure their child got the food he/she needed, (2) had to leave their child alone, even when they thought some adult should be with him/her, (3) were not able to make sure their child got to the doctor when he/she needed, (4) were so stressed that they had a problem taking care of their child and (5) were so caught up with their own problems that they were unable to tell their child they love him/her.Two items measuring educational neglect and neglect due to parent-child conflict were added to create a seven-item neglect scale.
The two items were: (6) My child ran away, so I did not know where he/she was, and (7) my child skipped school after a fight with me.Responses were: (1) once in the past 12 months, (2) twice in the past 12 months, (3) 3-5 times in the past 12 months, (4) 6-10 times in the past 12 months, (5) 11-20 times in the past 12 months, (6) more than 20 times in the past 12 months, (7) neglect occurred but not in the past 12 months and (0) no neglect perpetration.Following suggestions by Straus et al. (1998), we created a dichotomous 'ever neglect' scale by recoding the actual neglect cases (either past year or lifetime) as 1 and no neglect as 0. The mother's responses to the neglect items focused on only one child (below 18 years), called the focal child.
Where mothers had more than one child, the child with the most recent birthday was selected.

Ubuntu
We measured Ubuntu using a modified version of the Ubuntu scale by Molose (2019).Because this was the first time we were applying this scale in Ghana, we conducted principal factor analysis with varimax rotation to examine the factor structure of the Ubuntu scale.In addition to assessing the uni-dimensionality of the scale, we also explored the subscales (factors) within the scale, herein normative underpinnings.We used the conventional criterion of eigenvalue >1 (18) It is important to me that I respect the decisions made by my neighbour, (α = 0.79).Two items made up the compassion subscale: (19) I feel that sharing my difficulties (e.g., childcare problems) with other community members makes me strong; (20) my neighbours share his/her burden during hard times (e.g., caregiving problems) as part of a member of the community, (α = 0.76).We created a fourpoint Likert scale (from strongly disagree to strongly agree) to capture mothers' endorsement of the items.Cronbach's alpha for the entire 20-item Ubuntu scale was excellent (α = 0.92).Responses were summed to create the unidimensional Ubuntu scale and the individual subscales.

Legitimacy
We controlled for the legitimacy of collectivism within the neighbourhood using Emery et al.'s (2022) collective legitimacy scale.Using a response dimension of strongly disagree to strongly agree, mothers rated the level of legitimacy of collectivism within their neighbourhoods by endorsing the following items: (1) help each other in times of need, (2) encourage each other not to act in ways that are disruptive to the community, (3) stop each other from doing things that might be harmful to themselves and (4) stop each other from doing things that might be harmful others.The reliability of the legitimacy scale was excellent (Cronbach's α = 0.92).

Severity of IPV
Injury items from the revised CTS 2 scale by Straus and Douglas (2004) were used to capture the severity of intimate partner violence experienced by the mothers in the study.Mothers reported the number of times they had experienced the following items in the past 12 months: (1) I went to see a doctor (M.D.) or needed to see a doctor because of a fight with my partner; (2) I had a sprain, bruise, small cut or pain the next day because of a fight with my partner.We calculated midpoint values for the frequencies reported, as Straus and Douglas (2004) suggested, and summed the midpoint values to create a continuous severity of the IPV scale.Cronbach's alpha of the two items was great, α = 0.79.

Neglect secrecy
We captured the social desirability of self-reporting neglect using Emery et al. (2018)  Responses were summed to create a neglect secrecy scale.

Demographic controls
We controlled for key demographic variables, including the age of the mother (in years), marital status of the mother (1 for married and cohabiting), sex of the focal child (1 for female child) and age of the focal child (in years).We divided the mother's self-reported monthly income (USD equivalent) by the number of children under their direct care to obtain the value for income per child.We created a dichotomous measure for kinship ties by accounting for the presence of family members and close friends within the neighbourhood.

| Analysis approach
The cluster design of the sample demanded that analysis should cater to clustering at the community level.Hence, we conducted fixed effects logistic regressions to analyse the data.Fixed effects sufficiently handle row dependence issues emanating from clustering within communities and associated biases in statistical inferences.
Fixed effects regression models concurrently address biassed statistical inferences that occur due to correlation within the error terms and eliminate bias that may occur due to unobserved settlement variables.
We performed diagnostics to assess the robustness of the models and to ascertain any model violations.The highest VIF value is 1.79 (Ubuntu-solidarity), an indication that multicollinearity is not a concern in the model.Link tests suggest that the independent variables are linearly related to the log odds of the outcome variable (hat squared = 0.39, p = 0.20).A full interaction test showed that models for rural and urban are significantly different ( p = 0.0001).

| RESULTS
Table 1 reports the mean, percentages and standard deviations of the variables in the models, classified into rural and urban.The sample included 650 mothers from rural and 450 from urban settlements in Ghana.Seventy-nine percent (79%) of focal children in rural areas compared with 38% in urban areas are reported to be neglected, according to self-report from mothers in the survey.Compared with urban settlements, endorsement of Ubuntu and the normative underpinnings were significantly higher in the rural communities: compassion (3.50 vs. 3.31, p < 0.001), collective solidarity (3.11vs.2.90, p < 0.001), community care (3.44 vs. 3.14, p < 0.001), respect (3.67 vs. 3.47, p < 0.001) and survival (3.22 vs. 2.74, p < 0.001).Similarly, there were significant differences between rural and urban areas in terms of mothers' self-report of IPV experience (0.16 vs. 0.10, p < 0.05) and a number of kinship ties (94% vs. 75%, p < 0.001), with rural areas having the highest frequency.About 75% of mothers in rural communities were either married legally or living in a commonlaw marriage, compared with 71% of mothers in urban communities.
The difference between rural vs. urban was statistically significant (p < 0.05).However, mothers' average income per child was higher for those in the urban settlements (US $42.39 vs. 47.44), but the difference was not statistically significant.
Table 2 shows odds ratios and coefficient standard errors for the ever neglect fixed effects logistic regression models, separated by settlement type.The table shows the effects of the overall Ubuntu scale on child neglect within the full sample (columns 2 to 4) and the effects within the rural and urban sub-samples (columns 5 to 10).In the full sample model, the strength of Ubuntu is associated with lower odds of neglect (OR = 0.47, p < 0.05 CI [0.29, 0.76]).Also, a stronger endorsement of the legitimacy of collectivism within the neighbourhoods is negatively associated with child neglect (OR = .62,p < 0.01 CI [0.44,0.87]).Mothers' IPV severity (OR = 1.50, p < 0.05 CI [1.00, 2.25]) and their secrecy in reporting neglect (OR = 1.50, p < 0.01 CI [1.19, 1.89]) predicted higher odds of neglect.Indeed, parents who are more likely to conceal their neglectful acts would probably neglect their children more.Kinship ties did not significantly predict neglect in the full sample model.
The final two models in Table 2 show that the relationship between Ubuntu and child neglect is robust and significant in the rural model (OR = 0.13, p < 0.001 CI [0.06, 0.31]) but not significant for the urban model (OR = 1.05, p = 0.89 CI [0.55, 2.00]), and the direction may reverse for rural vs. urban.Similarly, the relationship between the legitimacy of collectivism and neglect was only significant for the rural sample (OR = 0.37, p < .001CI [0.23, 0.60]), and the coefficient was in the opposite direction for the urban sample (OR = 1.14, p = 0.65 CI [0.65, 1.97]).The severity of IPV was not Hypotheses 2 and 3 predicted that the normative underpinnings of Ubuntu would predict lower odds of neglect, which will be stronger in rural communities compared with urban ones.Both hypotheses were supported to a large degree.Generally, these findings suggest that activities that strengthen Ubuntu norms of community care, collective solidarity and compassion will yield positive outcomes by helping to reduce the incidence of neglect in rural communities in Ghana.
Although previous research (largely argumentative or commentary articles) have documented the potential benefits of Ubuntu to child protection (Chilwalo, 2020;Mayaka & Truell, 2021;van Breda, 2019), this is the first empirical paper to examine the relationship between Ubuntu (including norms of Ubuntu) and child neglect.Also, the first empirical paper to test claims about the strengths of Ubuntu and collective practices in rural areas in Ghana.Abdullah and Emery (2023) argued that the architectural structure of rural communities, especially the dominance of family compound housing structures (Danso-Wiredu & Poku, 2020), influences collective activities and facilitates community members' intervention to support families and remedy neglect (Abdullah et al., 2022).Such supportive activities may involve community arrangements to support families caring for orphan children (Mugumbate & Chereni, 2019) and community informal kinship care practices (Abdullah & Frederico et al., 2020).Our findings on the relationship between Ubuntu (Ubuntu norms) and child neglect in rural communities validate these claims empirically.Especially as the norms of community care in Ubuntu may influence collective child care practices, such as the community care arrangements for orphan children in Zimbabwe (Mugumbate & Chereni, 2019), and the traditional kinship care practices in Africa (cf.Cudjoe et al., 2019).Similarly, community members are more likely to support vulnerable families in communities when the mutual bond and trust between members are strong (Maguire-Jack & Showalter, 2016).Robust evidence on collective efficacy and child maltreatment substantiate this claim (cf.Abdullah, Emery, & Jordan, 2020;Maguire-Jack et al., 2022;Maguire-Jack & Showalter, 2016).Findings on the relationship between the norm of compassion and child maltreatment also confirm that empathetic mothers are less likely to neglect their children (Rodriguez, 2013).
The finding of a non-significant relationship between the Ubuntu norms of respect, solidarity and child neglect may be an issue of low statistical power, given that coefficients were consistent in the right direction.Studies in different contexts using a large sample with high statistical power could be useful to validate this finding.However, the counter-predictive findings on the relationship between the Ubuntu norm of survival and neglect are worth discussing.Empirically, notions and issues of survival depict a high level of deprivation.Deprivation (including those resulting from poverty) was a strong predictor of neglect (Maguire-Jack & Font, 2017;Sun & Chen, 2022).This suggests that beliefs of survival and neglect could correlate in the same direction, as evident in our findings.That said, we cannot fully validate the findings with this claim, given that normative interpretations of values/norms and concrete practices that accompany norms may differ (Parsons, 1937).As such, qualitative findings that seek to unpack these findings will be useful in providing better clarity about the determinants of this relationship.
The statistically non-significant findings on the relationship between Ubuntu and neglect in the urban sample have some implications.At face value, these findings suggest that the strength of Ubuntu and collective values, in general, have diminished in urban areas in Ghana.Although some researchers surmised that collective values (including values on kinship) in urban areas have declined and life in the city is shifting towards an individualistic orientation (Antwi, 2017;Nukunya, 2003), no study has substantiated this claim.Our findings could serve as a wake-up call for policymakers and community leaders to strengthen collective values in urban areas.

| Implications for child protection practice in rural Ghana
Our findings suggest that the prevalence of child neglect in rural  (Wessells, 2015).Whereas the above practice recommendations and intervention measures are informed by the current findings and findings from existing studies (Wessells, 2015), their efficacy and effectiveness should be tested using rigorous experimental designs.

| Limitations and recommendations for further research
This study is not without limitations.We relied on the mother's selfreports of their neglectful behaviour, which may be subjected to social desirability bias.Although we accounted for the risk of social desirability through the neglect secrecy measure, that cannot eliminate the risk of social desirability in our models.As this is a cross-sectional study, associations do not suggest causality.Further, the CTS scale for (an informal farming support activity) and byelaws on mutual childcare duties.
and cut of point of factor loading >0.40 to extract the factors.The modified items loaded strongly onto five factors, which was confirmed in the scree plot analysis-showed a strong elbow after the fifth factor.The five factors were labelled: collective solidarity, community care, survival, respect and compassion.The collective solidarity subscale was composed of items: (1) My neighbour treats each community member as if he/she was a member of a family; (2) I have a genuine backing (support) of my neighbours, such that they are willing to help me when I need it; (3) I actively contribute to community goals that benefit a wider group particularly, where they are worse off than me; (4) I generally do trust my neighbours in matters of support or extending a helping hand; (5) I do helpful things that will benefit the community members I know and me; (6) I see myself as part of a diverse community rather than as individual from a different cultural background or nationality, had a Cronbach's alpha of (α = 0.86).The community-care subscale had the following items: (7) I feel it is my duty to take care of my neighbour, even if I have to sacrifice what I want; (8) I believe each member of this community should be willing to share (the little) they have with others as a way of brotherly care; (9) my neighbour expects me to respect his/her decisions; (10) I feel that all community members should stick together as a family no matter what sacrifices are required; (11) being a valuable community member is very important to me than my personal identity; (12) the wellbeing of my neighbour is important to me, Cronbach's alpha of (α = 0.90).Items for the survival subscale included: (13) My neighbour and family members are usually present (emotionally) to share my pain during difficult times; (14) my neighbour and family members are usually available (physically) to suffer with me during difficult times; (15) my neighbour and family members encourage me to remain polite even when I disagree with what the guest says, (α = 0.83).The respect subscale was composed of items: (16) I feel that my neighbour treats me with utmost respect and dignity; (17) my neighbour greets me whenever he/she sees me; neglect secrecy scale.Mothers agreed to the following four-point Likert scale items: If I could not take good enough care of my child, I would try to keep it secret from (1) my friends, (2) my family, (3) my neighbours, (4) my co-workers, (5) my boss and (6) everyone.The reliability of the scale was very good (α = .83).
Ghana could reduce if the Ubuntu norms were effectively utilized and translated into positive practices.Child protection workers may adopt a community-focused approach and work with leaders in rural communities to collaboratively enhance traditional practices that strengthen these collective norms.Part of the measures could entail the formation and recognition of informal parental associations.Periodic training on (1) parenting support, (2) building and sustaining positive relations with neighbours, (3) addressing childcare and parenting difficulties and (4) sustaining partnerships with local leaders could be useful in strengthening the Ubuntu norms, especially norms of social solidarity, community care and compassion.It is recommended that local leaders, including the village or area Chiefs, should be empowered to develop local bylaws that strengthen collective activities.Bylaws that promote mutual childcare duties and participation in community activities are desired.This proposed community-focused and Ubuntu-oriented child protection practice should be spearheaded by the Department of Social Welfare in Ghana in collaboration with local communities to ensure effective implementation neglect might not capture the unique cultural contexts in Ghana as well as the influence of neighbourhood structural factors, such as poverty, particularly in rural Ghana.As such, future studies should consider more contextual measures for neglect in Ghana, especially in rural communities.We propose the need to develop a community neglect scale to capture neglect in collectivist societies where acts of neglect may be associated with community negligence due to norms of collective childcare and motivated by structural factors.Studies should be conducted in other African countries to validate the findings.Ethnographic studies with community members to unravel innovative pathways to strengthen the values of Ubuntu in rural and urban areas would be desired.4.4 | ConclusionChildren in rural areas are known to be at high risk of neglect due to the presence of poverty and other structural predisposing factors.Also, evidence suggests that child protection measures are underdeveloped in rural areas in Africa.But research has identified collective values as important strengths that can be tapped to help protect children against the risk of neglect in rural areas.Our findings have tested this last claim and found significant support for the claim.The findings suggest that despite the effects of poverty in rural areas, effective use of Ubuntu and the norms that underpin it may help to protect against the risk of child neglect in rural Ghana.Collaboration with chiefs and stakeholders within child protection should be reshaped to focus on innovative indigenous practices that promote collective values.Such interventions may focus on promoting community durbars, informal parent associations, the traditional collective practice of 'ndoboa' T A B L E 1 Sample descriptive statistics.Fixed effects logistic regression models on ever neglect (n = 996).