Bahamian Fathers’ Involvement with their Child’s Schooling: To What Extent Does Family Structure Matter?

This study examined characteristics of Bahamian fathers’ involvement with their child’s schooling using data from a comprehensive study on parent involvement within one Bahamian community. We triangulated data from a parent survey ( n = 91 males, n = 278 females) and community interview data ( n = 33 community members) to compare fathers to mothers, examine whether fathers living in the same home as their child had an impact on their involvement with schooling, and pinpoint beliefs regarding family structure and gender norms. Chi-square analyses demonstrate that Bahamian fathers engaged with schools in very similar ways to mothers on over half the involvement indicators, with slight differences on the remaining indicators. Mean difference analyses of factor scores showed slight differences between mothers and fathers in academic involvement and more pronounced differences in involvement with basic needs. Interestingly, despite beliefs about family structure voiced by some participants, living in the same home as their child did not play a role in fathers’ involvement. We discuss our findings within the social context of The Bahamas, raise questions about the real impact of family structure on paternal involvement, and call for more exploration of the impact of class and socioeconomic status on involvement with schools.


Introduction: Context of Bahamian Parent Involvement inside Schools
Schools within The Bahamas are constantly calling for more parent involvement. There is a perception within Bahamian society that many parents place a low value on their child's education, especially parents whose children attend government (public) schools (Bowe & Johnson, 2021). Given this backdrop, we conducted a comprehensive study on parent involvement with schooling within one Bahamian community to understand patterns of involvement, barriers parents face, and practical solutions that schools could immediately implement. One recent finding is that single-parent homes are perceived by Bahamian school staff as one of the barriers to involvement (Bowe & Johnson, 2021). It is questionable, however, whether this belief has any supporting evidence. The perception that parents place a low value on education is possibly increased for Bahamian fathers because Caribbean fathers are often stereotyped as deficient or absent (Green et al., 2019;Reynolds, 2009;Roopnarine et al., 1995). Our study examines the extent to which fathers living in the same home as their child has an impact on their involvement with schooling. Our findings help clarify the extent to which family structure matters to involvement within Bahamian society.

Parent Involvement with Schooling: The Need for Culturally Responsive Approaches
The literature on parent involvement with schooling from industrialized nations demonstrates that involvement looks different among communities from diverse ethnic, cultural, and sociodemographic backgrounds (Calzada et al., 2015;Desforges & Abouchaar, 2003;Feuerstein, 2000). Traditional parent involvement models (and initiatives) often do not take into account the roles that ethnic, cultural, and sociodemographic backgrounds play regarding parent interactions with school environments (Auerbach, 2001;Baquedano-López et al., 2013). Notably, the call for culturally responsive approaches for examining parent involvement is longstanding (Daniel-White, 2002;Fine, 1993;Souto-Manning & Swick, 2006). This study addresses the call within a Caribbean context that parent involvement should be operationalized and measured using findings in Caribbean (rather than international) literature as well as by the opinions of local Bahamian informants.

Caribbean Literature on Parent Involvement with Schooling
Caribbean literature demonstrates that Caribbean parents engage in more homebased than school-based involvement (Munroe, 2009;Murphy, 2002;Troupe, 2017). Examples of home-based involvement are helping children with homework and taking them to community centres. Examples of school-based involvement are attending PTA meetings and volunteering at school events. Home-based involvement that Caribbean parents tend to employ includes providing structure at home, helping with homework, hiring tutors or arranging for extra classes, and giving emotional support and encouragement (Abdirahman et al., 2012;Anderson et al., 2009;Collie-Patterson, 2008;Murphy, 2002;Munroe, 2009). Demographic factors that influence parent involvement include the sex of the child (Marshall & Jackman, 2015), the parents' educational level (Cole, 2020;Murphy, 2002), the performance level of the school (Bevill, 2007), and the grade level of the child (Marshall & Jackman, 2015;O'Neil-Kerr, 2014).
The literature demonstrates that the relationship between parent involvement and academic achievement in the Caribbean is mixed. For example, some scholars have found a positive relationship (Anderson et al., 2009;Bevill, 2007) whilst others have found no relationship (Collie-Patterson, 2003). There are also a number of social and schoollevel barriers parents face that hinder their involvement. Societal barriers include work hours, limited finances, and household responsibilities (Blackman & Mahon, 2016;Cole, 2020;Munroe, 2009;Murphy, 2002). Some school-level barriers include teachers not wanting parents inside their classrooms (Munroe, 2009), parents not having voice in curricula decision-making for their child (Blackman & Mahon, 2016;Murphy, 2002), parents not having voice in school governance (Murphy, 2002;Walker, 2012), and school staff holding negative perceptions of parents (Watson-Williams et al., 2011). While the Caribbean literature cites more school-level than societal-level barriers, more research is needed to clarify which barriers pose more problems.

Caribbean Fathers, Parent Involvement in Child-rearing
While the literature on maternal parent involvement in the Caribbean is growing, less attention has been paid to the specific roles of fathers (Devonish & Anderson, 2017;Yildirim & Roopnarine, 2017). Anderson's (2021) study on masculinity and fathering in Jamaica is one of the most comprehensive and it summarizes much of the literature on Caribbean fathering over the past seven decades. She highlights the two main schools of thought within Caribbean literature that have examined Afro-Caribbean families since the 1950s (Anderson, 2021;Anderson & Daley, 2015). The first school of thought endorses more pluralist approaches to the Caribbean family and seeks to understand the differences between various social groups. The second school of thought seeks to uncover common value systems across Caribbean families. A third and more inconspicuous line of research within the 1960s was grounded in social anthropology and focused more on male peer groups within smaller communities (Anderson, 2021).
Much of the research on the Caribbean family focuses on single-parent homes and child development since single-parent homes are prevalent in many Caribbean societies (Anderson & Daley, 2015). Similar to critiques in the parent involvement literature from industrialized societies, Anderson and Daley (2015) point out that many parenting models used to explain Caribbean families are based upon traditional family arrangements and assumptions that oftentimes do not align with the Caribbean context. Their findings echoed Barrow's (2001) call for scholars and societies to reject deficit views of non-traditional family arrangements, and instead create models that embrace the diversity of family arrangements in the Caribbean.
There is a growing body of research focused on Caribbean parent involvement and early childhood development led by Roopnarine and colleagues. These researchers employ more positivist frames (that is, more quantitative approaches that employ causal modelling) to understand Caribbean family dynamics, and their findings demonstrate varying parenting styles and varying effects of parenting styles on children's outcomes (Roopnarine et al., 2013;Roopnarine & Yildirim, 2018. Here we provide three examples of their findings. First, the typology of paternal involvement differs among fathers from Belize compared to Jamaica, Guyana, Suriname, and the Dominican Republic (Roopnarine & Yildirim, 2018). Second, whilst Caribbean families do exhibit different parenting styles, they only impact children's academic outcomes if the child is from a particular ethnic group (Roopnarine et al., 2013). Third, while certain types of paternal linguistic involvement do positively impact children's cognitive skills, this is only true for certain ethnic groups (Roopnarine & Yildirim, 2019). Overall, the findings of Roopnarine and colleagues support both pluralistic and common value systems among Caribbean families. That is, they support pluralistic theorizing due to differential effects of parenting styles and types of involvement. They also support common value systems because their instruments demonstrate a reasonable psychometric fit for multiple Caribbean communities.
One important factor to consider in parent involvement is whether or not the father lives with the child. The summary of the literature provided by Anderson (2021) shows that whether or not the father lives with their child(ren) is linked to the economic system of that Caribbean community. For example, Caribbean economies built around subsistence farming versus seasonal labour show different family structures and patterns, with subsistence farming being more associated with stable mating patterns. In line with this, Yildirim and Roopnarine (2017) found that socioeconomic status was related to whether or not fathers lived with their children. They found that fathers who were more financially stable tended to live in homes with a partner and children while those who were not tended to live by themselves. The growing research on fathers also shows that whether or not they lived with their children only matters for certain ethnic groups and, further, has different effects (Roopnarine & Yildirim, 2019). For example, Roopnarine and Yildirim (2019) found that children from Afro-Caribbean families whose fathers lived with them displayed higher literacy skills whereas children from Indo-Caribbean families whose fathers lived with them displayed lower literacy skills. In contrast, whether or not fathers lived with their children had no impact on the literacy skills of children from mixed-ethnic families.
Furthermore, the Caribbean literature demonstrates that younger fathers report more engagement than older fathers, though there is no social class (e.g., middle and upper class, working class, lower class) difference in their levels of engagement (Devonish & Anderson, 2017). Also, fathers who only had outside children (that is, children who were not part of a marriage union and who did not reside with their fathers) tended to be more engaged with their outside children than fathers who had both outside and inside children (that is, children who were a part of a marriage union) but who lived with their inside children (Devonish & Anderson, 2017).
In short, the Caribbean literature points to several factors that may influence father involvement with schooling. These include whether or not the child resides with the father, the father's age, and socioeconomic status. Given these findings, our survey asked parents to indicate with whom their child lived (mother, father, both parents, grandparents, other) because we wanted to know the extent to which this impacted their involvement. We also asked parents to indicate the age band they belonged to so we could examine the role age played in parental involvement. While the Caribbean literature demonstrates that fathers' socioeconomic status (typically proxied by social class status) has a relationship with their living arrangements, we acknowledge that socioeconomic status is challenging to define and measure within The Bahamas. We proxied parents' socioeconomic status based on whether their child attended a government (lower-income) versus a private (higherincome) school, though we acknowledge that this is a very crude and tentative estimate within the Bahamian context.

Gender Socialization of Males in the Caribbean
We now turn attention to gender-role stereotypes because, arguably, these shape how males perceive their fathering roles. Gender-role stereotypes about maleness and femaleness are a set of beliefs that are culture-specific and endorsed by members of that community (Walters & Carpenter, 2017). Gender-role stereotypes are prevalent in Caribbean societies (Walters & Carpenter, 2017) and some of these stereotypes position Caribbean fathers in deficit or absentee terms (Bethell-Bennet, 2014;Green et al., 2019;Reynolds, 2009;Roopanarine et al., 1995). Recent findings though on parenting styles from Trinidad and Guyana demonstrate that fathers show high warmth and active engagement toward their young children (Roopnarine et al., 2013) which counters negative stereotypes of Caribbean fathers. Anderson (2021) reminds us that stereotypes and gender-role norms are defined individually, collectively and institutionally,, and that schools play a large role in shaping and perpetuating gender identities via their structure and operations. There are many studies focused on male academic underachievement and male marginalization within Caribbean school systems and a summary of them can be found in Bowe (2015). Gendered experiences within Caribbean schools and the perception within many Caribbean societies that education is feminine (Bethell-Bennett, 2014;Miller, 1994;Reddock, 2003Reddock, , 2010, may drive parents to relegate school involvement activities to mothers (Munroe, 2009). Further, Bahamian parents tend to make more financial investments in their daughters' education rather than their sons (Fielding & Gibson, 2015) which arguably may contribute to boys being more marginalized within schooling contexts. In particular, Bethell-Bennett (2014) offers a solid foundation to guide our understanding of Bahamian masculinities and structural inequities that have led to the marginalization of Bahamian Black males. His findings concur with Caribbean scholars in that education in The Bahamas is perceived as feminine. Further, the belief that educational attainment leads to social advancement is disappearing. Given the gender-role socialization of males within Bahamian society, we anticipate different patterns of involvement between mothers and fathers. Overall, our present study examined whether Bahamian fathers enact different forms of involvement than mothers and the extent to which age, whether they lived with their child, and socioeconomic status mattered (proxied by the type of school their child attended).

Research questions
1. Are there differences in parent involvement between Bahamian male and female parents/caregivers? 2. To what extent does (a) age, (b) socioeconomic status, and (c) whether fathers lived with their child impact fathers' level of involvement?

Methods
We examined parent involvement within the Bahamian community via a mixed methods case-study approach (Yin, 2003). We collected data using surveys, interviews and a focus group. Interview and focus group data allowed us to deepen our understanding of survey trends. Participants included parents, teachers, faith leaders, employers, health care workers, and social service providers. We used local radio and TV stations to inform the public of the study and to invite them to participate. Principals invited us to use their schools as data collection sites. Hospitals, clinics, government workplaces, the public library, churches and police/fire stations were also data collection sites. Our participants were community members who willingly participated, thus our study represents a convenience sample. To address the above research questions, we report on parent survey data and community interview data.

Survey
Three hundred and seventy-seven parents (males = 91, females = 278, unknown = 8) completed the parent survey (see Table 1 for parent demographics). Parents with more than one child were asked to think about one child and respond based on that child. They indicated which child they were thinking of in the demographic section.
We proxied parents' socioeconomic status based on whether their child attended a government (lower-income) versus a private (higher-income) school (see Table 1). We acknowledge this is a very crude and tentative estimate within the Bahamian context. We included this variable as a crude proxy because findings in the Caribbean literature demonstrates that socioeconomic status has a relationship with whether or not the father lived with the child.

Interviews
We conducted telephone and face-to-face semi-structured interviews with parents (n = 10), pastors (n = 8), employers (n = 8), social service workers (n = 2), and health workers (n = 5) asking them their views on parent involvement. Interviews with parents were participant initiated, that is, parents called the researcher to share their views. Interviews with the other community members occurred by the researcher calling or visiting various companies, churches, and social service offices to inform them of the study and to invite participation. Examples of interview questions were, "What does parent involvement within schooling mean to you?" and "What types of barriers do parents face to being more involved with schooling?"

Limitations
Neither the survey nor the interview protocol asked parents to indicate whether any other adult in their family completed the survey, and this has implications for the appropriateness of the type of analyses used. This study assumed one parent per household completed a survey or interview since informal conversation with the participants indicated that was the case for the majority of participants (about 90%). However, uncertainty in this area is a limitation of this study. Another limitation is that the sample sizes for men were small. We do not know the power needed to detect an effect if one is present. It is possible that the analyses may not have had enough power to detect an effect if one is actually present.

Preliminary Validation Study on Survey Instrument
We conducted a preliminary cross-validation study (see Brown, 2015) for guidelines and procedures) on the first 16 items to determine the factor structure. Our exploratory findings suggested a one or two-factor structure. We adopted the two-factor structure (basic needs involvement scale, Cronbach's α = .739; academic involvement scale, Cronbach's α = .569) because the two-factor structure yielded slightly better fit indices than the onefactor structure (χ 2 model fit (76) = 181.01, χ 2 baseline model (91) = 1687.56, RMSEA = .06, CI90 [.049, .072], CFI = .93, TLI = .92). We note that the α for the academic involvement scale is low meaning that research-level decisions should not be made based on this scale. During the confirmatory stage of the analysis, we removed items 1 and 2 as suggested by MPLUS software to improve fit. Table 2 outlines the items on each scale. Note. α = Cronbach's α estimates. Factor scores were calculated for these scales and used in the independent samples t-test and multivariate regression analyses.

Quantitative analysis
We compared males to females on their survey responses. We used Chi-square tests on the first 16 closed-ended item to compare involvement patterns between males and females. Next we compared fathers to mothers on their involvement using independent samples t-test on the factor scores (basic needs involvement scores and academic involvement scores). Third, we used multivariate regression to regress the factor scores for basic needs involvement and academic involvement onto age, socioeconomic status, and whether the father lived with their child. We ran a separate model for fathers and mothers to make possible interaction effects based on sex more interpretable. We included the grade level of their child as a covariate as well, given that that variable consistently moderates involvement within Caribbean and international literature (Marshall & Jackman, 2015;O'Neil-Kerr, 2014).
To clarify beliefs about the impact of whether the father lives with the child or not, we analysed responses to the question about barriers to parent involvement. We used the content analysis via direct coding method (Hsieh & Shannon, 2005) and specifically coded responses that referenced beliefs about (a) the impact of family structure on the child's schooling, (b) the roles fathers played in involvement, and (c) fathers' experiences within schools. We triangulated open-ended responses and interview data with survey data to explain quantitative trends.

RQ1. Are there differences in parent involvement between Bahamian male and female parents/caregivers? Quantitative Survey Data
Chi-squared analysis revealed males differed from females on seven of the 16 items, and a visual inspection of the cells suggested that females tended to respond most of the time and always more often than males. Only the items males and females differed on are reported in Table 3. Effect size estimates revealed that in all cases these differences were small. Note: Effect size are Kendall's τ-b correlation estimates: .10 represents a small effect, .30 is a medium effect, and .50 is a large effect (Cohen, 2013).
We ran independent samples t-test on the factor scores for the two scales, basic needs (11 items) and academic involvement (three items) with sex as the independent variable. Findings revealed that fathers were less involved than mothers with differences reaching a small effect size for academic involvement (Academic Involvement, t(365) = -2.84, p = .005, Cohen's d = .33) and close to a medium effect for basic needs involvement (Basic Needs: t(365) = -3.88, p < .001, Cohen's d = .46). Table 4 demonstrates that the father's age, socioeconomic status, and whether or not he lived with his child did not predict his involvement. In fact, none of the covariates were useful in explaining fathers' involvement as indicated by the adjusted R 2 values. In contrast, the models predicting mother involvement showed a number of significant predictors. First, we see that mothers tended to be less involved if the father was not living in the home with them. Age was also a predictor of mothers' involvement but only for providing basic needs. As the child advanced through schooling, mothers' involvement decreased, which is similar to findings in Caribbean literature (Marshall & Jackman, 2015;O'Neil-Kerr, 2014). Partial η 2 estimates reveal that the child's grade level was the most important predictor of involvement and explained the majority of variation in the mothers' models. The approximate adjusted R 2 values for mother models showed that the model was almost twice as useful in explaining academic involvement (15%) as meeting basic needs (8%). Note: These are estimates for unstandardized coefficients: SE. * = p < .05; ** = p <.01; *** = p < .001

RQ2. To what extent does (a) age, (b) socioeconomic status, and (c) whether fathers live with their child, impact fathers' level of parent involvement?
To further understand participants' beliefs about the impact of fathers living in the home of their child, we report on our qualitative findings that referenced the family structure, gender roles, and the experiences fathers had with schools. Of the 33 interviewees, 10 made reference to the family structure, gender roles, or the experiences fathers had with schools. For example, one female employer explained, "Back in the day, mom wasn't working. But now when children are out of school, the mom is still at work". This person was referring to a lack of parental presence at home after school hours. This statement also reflects an underlying belief that parent involvement is the responsibility of mothers. Another female employer commented that since there are too few adults in the home, children are now performing the roles of adults by taking care of other siblings. One male faith leader stated, "the problem we have [in the Bahamas] is single parents" and that boys needed their fathers. A female faith leader remarked that barriers to involvement were actually compounded for some single parents who were working hard trying to "make ends meet". Similarly, a female healthcare worker commented, "most of the times, it's only the mom. And if you have like three children … there are a lot of moms that have to take on two jobs". Together, while these comments referred to family structure and gender roles, they also illustrated that work schedules were also perceived as a barrier to involvement.
Another female health worker also alluded to the unavailability of parents when she explained caretaker policies for the hospital and schools. She commented that according to Bahamian law, any family member aged 18 years and above could represent the child in the hospital and at school. So, one solution to the unavailability of parents was having other adult relatives step in. Finally, one parent and an employer (both females) commented on the current age of parents. For example, the parent said, "now we have a younger generation who are still kids themselves and they are now raising kids". Interestingly, the parent survey data shows that only five of the 274 who responded to the question about barriers to involvement commented on single family homes as barriers. Furthermore, only one parent commented that some parents were too young and immature to parent effectively.
In reference to gender roles, the faith leader who said that boys needed their fathers further elaborated upon the role of fathers in the home. He commented that boys were often misunderstood as bad, but as the father has the same biological makeup as the son, a father understands boys better than the mother. He lamented that many boys in single-parent homes were deprived of the nurturing and development that only fathers could provide.
Last, there is some evidence that school staff has lower expectations of the involvement of fathers. For example, one father commented that he was chided for advocating for ways he wanted to be involved in the school. He wanted access to the PTA policies but a school administrator asked him, "Why do you care?" He explained that he was shocked at their response because the person insinuated that they expected him to not care as much as he did. Another father also commented on the lower expectations school staff had of his involvement. He recalled an instance when his daughter was emotionally depressed at school and the guidance counsellor called his wife rather than him. When he inquired about it, the counsellor explained that she assumed he was too busy to respond to their calls. This was despite the fact that both parents were full-time workers within the healthcare system. Likewise, a third father commented that schools should make an effort to deal with both parents, not just the mother.
Taken together, qualitative findings demonstrate that some participants believed family structure has an impact on parent involvement in schools. Quantitative findings help clarify this relationship and demonstrate that while fathers living in the home with their children has no impact on their involvement in school, it does have an impact on the mother's involvement. Also, some fathers may receive subliminal messages from school staff regarding the less involved role they are expected to play with their children.

Discussion
Overall, our findings for Bahamian father involvement yielded mixed results. First, we see slight differences between mothers and fathers on academic involvement indicators, but more pronounced on the basic needs indicators. On one hand, this counters deficit stereotypes regarding fathers for the academic involvement domain, but reinforces stereotypes about their lack of contributions to the family's welfare. Findings suggest that some fathers may be pushed out by school systems because school staff may send subliminal messages about the roles they are expected to play. This leans towards Bethell-Bennett's (2014) findings on ways in which Black males are marginalized within Bahamian society. Further research is needed to determine the extent to which this is pervasive since few males within the sample mentioned these types of interactions. Further research is also needed to clarify which type of involvement, whether it is meeting basic needs or providing academic support, has a stronger relationship with achievement.
Fathers living in the home with their child had no impact on their involvement with schooling. There are a number of ways we can interpret these findings. For example, we can argue that fathers who are "absent" are similarly involved to fathers who reside with their children. In a different vein, we can argue that where the father lives actually does matterbut only for mothers. That is, we can infer that the presence of fathers at home frees up time for mothers to attend to school matters. Thus, findings also underscore the gender norm that parent involvement with schooling is more likely to be the responsibility of mothers than fathers. Overall, our findings support the growing body of literature on the differential impact of Caribbean fathers living with their children (Roopnarine & Yildirim, 2019;Yildirim & Roopnarine, 2021). They also call for Bahamian society to reject positioning single-parent headed homes as deficient when it comes to parent involvement with school. Whilst Caribbean scholars have been rejecting deficit frames for some time now, their findings are seemingly not being relayed to the general public. The disparity between scholarly findings and popular beliefs underlines the need for scholars to communicate their findings to the general public.
However, our findings demonstrate that parental age does make a difference for providing basic needs, which is similar to findings within Caribbean literature. The Bahamas' National Development Plan (2016) has shown that poverty amongst young adults is twice that of the national average. This offers a different narrative to the immature and irresponsible young parent stereotype because it reveals that younger parents simply may not have the means to adequately provide for their children. As well, our findings demonstrate that younger mothers are similar to older mothers for academic types of involvement, which again discounts the stereotype that younger parents are too irresponsible and immature to care about their child's education.
Last, similar to findings in Caribbean literature, socioeconomic status did not play a role in involvement. The Bahamas has a high cost of living and many families struggle to make ends meet (Bethell-Bennett, 2014;National Development Plan, 2016). Thus, regardless of whether the child attends a public or private school, the high cost of living in general may require parents to work long hours and limit their availability to be more involved with schooling. These findings leave room for future research on the roles class and social capital play with involvement and the threshold income that families may need to make in order for parents to be more involved.