Religiosity as a Factor Supporting Parenting and Its Perceived Effectiveness in Hungarian School Children’s Families

: The impact of a family’s socio-economic status on parenting activities has been a thoroughly researched topic. Parental involvement, which refers to parenting that supports school achievement, may be home-based and school-based, as typiﬁed by Epstein and Sanders. Earlier ﬁndings suggest that a more favorable socioeconomic status results in more active and effective parental involvement. Only a few studies show low-status parents can be more effective, whereby some factor compensates for the negative impact of low status. Our hypothesis is that parents’ self-identiﬁed religiosity can reduce the disadvantages arising from the family’s unfavorable socio-economic status. We use data from a survey conducted in 2020 among the parents of fourth-grade Hungarian primary school children, which is geographically and socially representative with a sample size of 1156. Our results show that religiosity has a signiﬁcant positive inﬂuence on parenting activities and their perceived effectiveness, even after controlling for other variables. In addition, our indicator of home-based parental attention in itself is also strongly supportive of parents’ perceived effectiveness in parenting. Our ﬁndings suggest that parental religiosity and home-based parental attention are important factors that can make parenting effective despite social and cultural disadvantages.


Introduction
Remote learning during the COVID-19 pandemic has drawn attention to social differences in parenting skills and practices. This has been particularly apparent in countries where there are large social differences between children and schools, including modern European countries such as Hungary. Parenting practices are determined by how parents interpret their own role and what efforts they perceive to be effective (Hoover-Dempsey and Sandler 1995). Parenting requires substantial investment by families over a long period of time, and a diverse set of parenting activities is necessary to raise a child. Examining parental attention and interactions is more common in early childhood (Ulfers 2020); in contrast, the perspective of learning is often studied among school-aged children, in particular how their parents support their academic success and get involved in their academic advancement. The research clearly shows that parental involvement is a key factor for children's academic success, but the fact that parental involvement is measured by a wide range of indicators makes it difficult to examine the concept and compare results. Thus, there is no clear answer to the question as to which indicator of parental involvement has the greatest impact on children's academic success and future. In addition, few longitudinal studies have been carried out, whereas in cross-sectional studies, the positive impact of parental involvement on academic achievement is not clear, since in many cases it is precisely children with problems who receive higher levels of parental involvement. In the latter case, the relationship is inverted, with low achievement implying high involvement (Boonk et al. 2018).
Researchers have distinguished between school-based and home-based dimensions of parental involvement (Epstein and Sanders 2002). School-based parental involvement relates to parents' activities at school and their interactions with teachers, while home-based involvement includes parents' support for, expectations of, and attitudes towards child learning. However, little is known about the impact of everyday parental practices which are not directly aimed at learning and the factors which influence them. When examining home-based parental activities which are not closely related to learning, studies often focus on joint leisure activities, which could support skill development and subsequent positive academic advancement. Joint activities can be considered in terms of the subject of the parent-child interaction, the variety of activities undertaken together, or the amount of parental time spent with the child. Studies on parental time spent with children find that highly educated parents, especially mothers, spend more time with their children, despite their longer working hours (Sayer et al. 2004;Guryan et al. 2008). As for the relationship between parental time investment and children's academic success, highly educated parents spend more time with their children on intellectual activities (Lareau 2011). The usefulness of playing or reading together and engaging in play-based activities has also been demonstrated in several studies, sometimes even if parental socio-economic status taken into account as well (Wood 2002;Dodici et al. 2003;Wingard 2007). The stability of relationships within the family and the support network surrounding the family can also make children more productive. Coleman interprets the family as a social network, emphasizes the effectiveness of closed and strong relationships, and points out that the effectiveness of home-based parenting depends on the relationship with and connection to the child, because a closed network of relationships enables the transferring of norms and expectations (Coleman 1988;Schneider and Coleman 1993). Social capital within the family is a resource that can enable even low educated parents to raise resilient children (Slates et al. 2012). The relationship between parent and child is strengthened by the structural integrity of the family, parental attention, time spent together, and joint activities (Coleman 1988;Parcel and Dufur 2001;Parcel et al. 2010). The source of social capital within the family is a unified set of values and norms. One dimension of this includes agreement on parenting values and consensus on worldviews, implying that the parents' similar religiosity may also contribute to social capital.
As institutional religiosity has loosened, the proportion of parents regularly practicing religion in a community has declined, but the proportion of those who still consider themselves religious in some way remains significant (Davie 2002;Tomka 1995;Casanova 2017;Pusztai and Demeter-Karászi 2019). Global indicators of religiosity are closely related to the number of children, family structure, and the division of roles within the family (Petro et al. 2018;Wilcox 2004;Fényes et al. 2020). The impact of parents' self-identified religiosity on home-based parenting activities and their effectiveness, which is the focus of this study, is rarely examined.

Parenting Activities within the Family
The evolution of parenting activities is usually explained by the social status of the family. Most theories interpreting the parents' role in influencing their children's progress point to the transmission of behavioral patterns, which are not even conscious due to structural social constraints and are therefore unavoidable (Bourdieu and Passeron 1977;DiMaggio 1982;Lareau 2011;Li et al. 2020). Other theories suggest rational choices based on individual value attachments behind parental investment (Boudon 2003;Thompson 2018). Theories that also consider social network structures see parental involvement as a source of social capital within the family (Coleman 1988). However, over-individualized psychological theories on parenting do not examine social and cultural differences in the effectiveness of parenting (Baumrind 2012).
Investment strategies in terms of parenting show great variation historically, culturally, and socially. This is determined, among other things, by the view of children in society, parents' role perceptions, cultural and material resources, and cost-benefit calculations for parenting. In his famous 1988 paper, Coleman classifies the resources invested in parenting into three categories, distinguishing between financial or economic inputs, human capital investments rooted in the parents' cognitive abilities and skills, and the time, energy, presence, and attention devoted to parenting, which he terms social capital investments (Coleman 1988). Having children and caring for newborns is a major parental investment, but daily parenting practices, which require a great deal of effort over a long period of time, are not discussed as much as they should be. We do not have a reliable overview of the diversity of everyday parenting activities in society.
The literature uses different terms in relation to parenting, depending on whether researchers focus on the elements and nature of the activity (parental involvement, parental support) or on the beliefs about parenting (parental engagement) (Epstein and Sanders 2002;Li et al. 2020). The time parents devote to their children and parental attention can be interpreted as an investment in the children's future (Coleman 1988). Parental engagement, in contrast, refers to the recognition of the importance of schooling and learning as well as the understanding and support of academic aspirations (Goodall and Montgomery 2014).
Studies on parental participation and involvement focus on how schools can compensate for the decline in parental functions and what factors enable parents to make more effective educational investments in the human capital of their children's generation. The effectiveness of parenting is most often examined from the perspective of the children's school success and learning at home. In this sense, the parent is a kind of external supporter of the school and a facilitator of the processes which take place at the educational institution (Epstein and Sanders 2002). In examining the activities of parents and schools, Epstein and Sanders point out that the home and the school are intertwined, with the communities surrounding the children constituting a third sphere.
Parental involvement at school includes parents' visibility at school, attendance of events for parents organized by the school as well as of other school programs and activities, and volunteering at school. The other aspect of dedicated parenting relates not to schoolbased but to home-based activity (Kim 2009), sometimes referred to as parental support, which is distinguished from activities related to the parents' visibility at school (Schneider and Coleman 1993). Home-based parental involvement includes helping with homework and discussing what has happened at school, where the parent clearly acts as an ally of the school. However, home-based parental involvement may strive not only to develop a specific skill in the child but also to transmit values, norms, and expectations which lay the foundation for good academic performance later on (Sy et al. 2013). In addition to monitoring academic performance and compliance with the behavioral standards of the school, the aim of parental support is also to convey the entire family culture (goals, values, norms), which parents view as a way of ensuring their children's long-term and widely defined success and development into competent adults (Kim 2009). This broad and not always conscious process of cultural transmission involves a range of activities in the family context, which are independent of school tasks and events, and could include discussions with the child or joint activities and tasks.
The school-centered evaluation of parenting is changing, as reflected by the emerging concept of family-school partnership (Epstein and Sheldon 2006). Goodall and Montgomery (2014) see parenting activities on an axis, with the parent responsible for everything at one end and the school at the other, while shared concerns are in the space in between. Epstein and Sanders (2002) distinguish six areas on this continuum, two of which focus on processes at home. Learning at home can be delineated from parenting, which involves parents working on their children's physical, psychological, and spiritual development. This takes place in a largely informal, unplanned setting, mostly in the family home or where shared leisure time is spent. Learning at home, although done by parents and children, is organized and controlled by the school (Epstein and Sanders 2002). Good performance in a certain area is correlated with performance in other fields and with academic achievement. In this study, we focus on the time spent with children and joint activities. These parenting activities are used to construct a complex indicator of parental involvement, which we analyze as both a dependent and an explanatory variable.

Factors Influencing Parenting Activities and Their Effectiveness
Various studies investigate the determinants that lead parents to engage in active and effective parenting activities. One group of these factors is subjective in nature, in that they explain differences in parental effort by the parents' views on the child's future and the role of the school (Ðurišić and Bunijevac 2017). However, the analysis of objective indicators reveals significant social inequalities in the parameters of investment in homebased parenting. The socio-economic status or minority position of the family is closely related to the types of parenting activities and is in fact part of a coherent status culture (DiMaggio 1982;Lee and Bowen 2006;Lareau 2011;Li et al. 2020). This reinforces the transmission of inequalities to the next generations (Bourdieu and Passeron 1977;Thompson 2018). Some authors have found cross-cultural differences in parenting practices at home (Whiteside-Mansell et al. 2001), but others have argued that the impact of socio-economic status overrides cross-cultural differences (Lareau 2011;Tan et al. 2020).
There are different explanations for the lower parenting activity and worse outcomes of those with low social status. Some explain the low intensity of parental involvement by economic disadvantage, low income, inflexible and longer working hours, the need to supplement income, and time constraints due to overtime (Dyson et al. 2007;Ho 2020). Financial hardships may also act through another factor, namely the instability of relationships and daily routines within the family and the lack of recreational cultural consumption (Roubinov and Boyce 2017). Other interpretations suggest that unfavorable social status influences parenting through low educational attainment, lack of positive school experiences, lack of information, uncertainty, and insufficient confidence in the educational process (Morawska et al. 2009;Baeck 2010).
While there is a fairly broad consensus on the disadvantages of low status in schoolrelated activities, the findings on home-based involvement are not clear. Some find that this is the only area where low-status parents can catch up with higher-status parents (Coleman 1988;Stacer and Perrucci 2013), while others argue that low-status parents' insufficient knowledge and poor verbal skills mean that they cannot help with learning at home (Clinton and Hattie 2013;Ho 2020). Results from Pusztai et al. (Forthcoming) show that having one parent with and another without a tertiary degree increases parental involvement at home, compared to homogeneous graduate or non-graduate parental backgrounds.
The lack of social capital associated with low status may also be another cause for underinformed parenting (Lee and Bowen 2006). From a social network approach, multiplex relationships within the family support the quality and effectiveness of parenting, and parental attention and time investment are very effective indicators of social capital in the family (Coleman 1988;Schneider and Coleman 1993;Meier 1999;Parcel and Dufur 2001;Pusztai 2015). The concept of family-school partnership and Coleman's social capital theory also attach considerable importance to the communities and social networks surrounding the immediate family, such as cooperation with fellow parents, which Epstein has incorporated into his model as a community factor (Epstein and Sanders 2002).

Religiosity as a Factor Influencing Parenting and Its Effectiveness
Research exploring the impact of religious education on academic achievement has not examined the mechanisms through which religious education exerts its impact (Gutman and Midgley 2000;Jeynes 2003;Park and Bonner 2008). There is also no clear evidence on whether religiosity has an indirect or direct effect on parental attention, or which dimension of religiosity matters in this respect. Parental involvement may be indirectly supported in that parental religiosity, especially the structural trait of religious homogamy, supports lasting romantic relationships, reduces the risk of divorce, leads to favorable child-parent ratios, and enhances parental attention (Cohen-Zada and Sander 2008; Bartkowski et al. 2019). Religious parents are more likely to develop a close relationship with their children in primary school, and the parents' religious homogamy strengthens family stability and parental attention (Pusztai 2015).
According to the integrative model of parenting in the family, the processes and effectiveness of parenting are determined by the parents' corresponding values, which influence decisions about everyday parenting activities the most (Darling and Steinberg 1993). The relationship with religiosity clearly indicates ends values and means values, which in turn determine parenting values (Zulehner et al. 2008;Pusztai 2011). Members of different religions and denominations may show great variation in terms of preferred parenting values but undoubtedly share some core values, which supports children's achievement in the school system and in society (by strengthening self-discipline and through the prohibition of aggressivity and self-destructive deviance) (Bartkowski et al. 2019).
The impact of the ideological, emotional, and intellectual dimensions of religiosity on parenting is rarely discussed, with the exception of the relationship between the subjective perception of God and parenting style, but it is not associated with a different parenting style after controlling for background variables (De Roos et al. 2004;Mahoney 2010).
Religiousness is often operationalized in terms of ritual dimension, which is the most easily measured and involves organizational membership and participation in ceremonies, thus adopting the perspective of churches and religious organizations. However, with the declining importance of institutional religious practice, a third category has emerged alongside the traditional dichotomy of religious and non-religious categories, characterized by religiosity with little or no connection to religious organizations. Tomka (1995) calls such people religious "in their own way" and Davie (2002) uses the term "believing without belonging". This individual religiosity, which lacks community cohesion, is a commitment based on individual choice and differs from the following categories on Tomka's scale, namely the "neither religious nor non-religious" uncertainty and the "nonreligious" category.
When examining the impact of religiosity on academic careers, we have found that religiosity without organizational ties represents a strong value commitment, with a similar, albeit weaker, effect compared to religiosity supported by organizational cohesion (Pusztai 2015). From the perspective of parenting, the advantage of parents' community-based religious practice may be that the gatherings of religious communities sometimes feature lessons on parenting, and parents can also learn several good practices from other parents in the same community. The religious affiliation of parents, especially those from disadvantaged backgrounds, gives them access to transitive relationships which act as a particularly effective information network and a factor to assist norm-consciousness in parenting (Coleman and Hoffer 1987;Pusztai 2005). The networking of families in religious communities increases the range of social capital that can be mobilized in parenting (Wilcox 2004;Mahoney 2010). In a crisis or in a single-parent family, this resource can be drawn upon, highlighting the protective potential of parental religiosity (Petts 2012). At the same time, of course, religious parents who are not embedded in an organization can also get inspired from their religiosity to independently acquire principles, knowledge, and skills for parenting.

Research Questions and Hypotheses
In our main research question, we ask whether parental religiosity has an effect on home-based parental attention and on the effectiveness of parenting activities. To examine the effect of parental religiosity, we take into account the effect of other explanatory variables, such as the parents' educational attainment, financial situation, social resources, family structure, and number of children, all of which may have an impact on both homebased parental attention and parenting outcomes perceived by the parents, measured mainly by the child's current academic performance.
Our first hypothesis suggests that parental religiosity is positively correlated with high social status, intact family structure, a large number of children, and a wide network that can be mobilized to assist parenting. We base this hypothesis on research findings showing that after the Millennium, religiosity is associated with higher status in Hungary (Tomka 2011;Hámori and Rosta 2013). In addition, religious parents tend to have a more stable relationship and richer social resources (Pusztai 2015).
We also examine the factors that influence parental attention. We measure home-based parental attention by various parenting activities and the time spent parenting.
According to our second hypothesis, the parents' high social status and religiosity support parental attention independently of each other, even after controlling for other variables.
According to our third hypothesis, high status increases the perceived effectiveness of parenting, but the positive effect of religiosity can be detected also, even if high status is taken into account.
The last two hypotheses are based on the experience of our previous research, in which we pointed out activities in the pedagogical practices of church-run schools that make low-status students high achievers (Pusztai 2005(Pusztai , 2015. In addition, we have observed that low-status young people who belong to religious communities have more successful academic careers (Pusztai 2015).

The Database
For the empirical analysis, we used the database of parents (N = 1156) collected for the "Value-Creating Child Rearing 2020" research project supported by the Mária Kopp Institute for Demography and Families. This study was approved by the Research Ethics Committee of the affiliated institution. The sample is representative of fourth-grade primary school children in Hungary by region, settlement type of the place of residence, and the family's social background. Multistage clustered sampling was applied, based on the student and site database of the Hungarian National Assessment of Basic Competencies. In the first stage of sampling, schools were selected, and in the second stage, the classes were selected to be included in the survey. Among the selected classes, a survey of all parents was attempted. Parents responded using a paper-based self-administered questionnaire.

The Examined Variables
The database we used allows for a nuanced study of parenting within the family. The dependent variable in the first model of our regression analysis, which was also included as an explanatory variable in the second model, was a complex indicator of home-based parental attention. Its first 12 items comprised the frequency of joint activities with the child in the past month (measured on a four-point Likert scale). The items were based on a revised version of a previous questionnaire developed to investigate parental attention (Pusztai 2015), adapted for primary school children. Based on these questions, we examined the multiplexity and diversity of the parental attention. In addition, we included parental time spent with the child on weekends as item 13, which indicated the amount of time spent with the child by the responding parent, also on a four-point scale. Principal component analysis was conducted using 13 variables, and the resulting principal component explained 31.6% of the variance of all variables included, with an eigenvalue of 4.1, a minimum of −4.64, a maximum of 2.38, a mean of 0, and a standard deviation of 1. Descriptive statistics for the 13 items and principal component factor weights are presented in Table A1 of the Appendix A.
The dependent variable in our second regression model related to parenting outcomes as perceived by the parent. The first two components reflected data on the child's academic performance: whether the child has the best or second-best grades (42% of the sample, factor weight 0.827) and the parent reported the child to participate in academic competitions (49% of the sample, factor weight 0.68) (called objective indicators). A six-degree relative indicator showed the parent's estimate of the child's abilities compared to the classmates (called relative indicator) with a mean of 4.11, standard deviation of 1.04, and factor weight of 0.809. Based on the three components, we performed principal component analysis using standardized variables to eliminate the effect of different scales. The resulting principal component accounted for 59.9% of the variance of the three variables, with an eigenvalue of 1.8, a minimum of −2.45, a maximum of 1.44, a mean of 0, and a standard deviation of 1.
In both models, the most important explanatory variable was parental religiosity. Religiosity was measured on a single dimension using Tomka's five-point scale of parents' religious self-identification (Tomka 1995). 1 Those who explicitly defined themselves as religious, either institutionally or individually, were separated from the other respondents, resulting in two groups of roughly equal size. Parents were grouped into three categories, according to whether both parents, one parent, or neither parent could be defined as religious. On this basis, we included in our regression models two dummy variables, one for homogeneous religious background and another for heterogeneous religious background, with neither parent being religious as the reference category. The explanatory variables also included parental educational attainment, measured by a dummy variable for homogeneous background of low educational attainment (indicating when neither parent had a secondary school leaving certificate). The family's financial background was measured by an objective indicator, the parent's income (the four possible income brackets were coded to the mean). Finally, the parent's social resources were measured by the strength of the parental network, which included people on whom the parent could rely in parenting decisions and when busy. We also measured the certainty with which the parent could rely on these people, in a similar manner to Widmer's (2007) method. From this, we created an index by summing the four-point intensity indicators for the ten individuals listed. The regression analyses also included the place of residence (1: urban, 0: rural) and the child's gender (1: boy, 0: girl). The effect of family structure was also examined, and measured by two variables. First, we examined whether the child was raised in a nuclear (intact) family, a stepfamily, or a single-parent family, from which two dummy variables were constructed (single-parent family, stepfamily, with intact nuclear family as the reference category). Second, the final explanatory variable measured the number of children in the family. Descriptive statistics of the explanatory variables are reported in Table A2 of the Appendix A.

Methods
In addition to bivariate contingency tables and analyses of variance, multistage linear regression analyses were also produced. The dependent variables in our regression models were first created using principal component analysis. The regression analyses were carried out with the stepwise inclusion of the explanatory variables. Parenting activities at home (parental attention principal component) were included as the dependent variable in the first model and as an explanatory variable in the second model.

The Correlation of Parental Religiosity with Social Status, Family Structure, and Available Network (First Hypothesis)
In line with the religiosity of the adult Hungarian population (Hámori and Rosta 2013), 16.2% of the parents who completed the questionnaire identified as religious according to the teachings of a church, and 38% reported to be religious in their own way, with a combined 54.2% identifying as religious. Since in the modern world religiosity can take many different forms, we have no reason to doubt the respondents' choice. Of those who could be defined as non-religious, 11.3% were undecided, 19.2% reported not to be religious, 6.2% had other beliefs, and 9.1% did not answer. About one third of the parents who completed the survey had a tertiary degree, another third possessed a secondary school leaving certificate, and a final third had primary or secondary vocational qualification without a secondary school leaving certificate. Most families lived in villages (40%) and small or medium-sized towns (35%), with 11% each living in county seats or the capital.
We found strong religious homogamy among married couples. Almost half of children were raised by homogeneously religious parents (45.6%), and more than a third of children had non-religious parents (35.2%). Heterogeneously religious parents were in the minority (19.2%).
We also examined the relationship between religiosity and social status, family structure, and the network at the parents' disposal. The proportion of parents with low educational attainment was the lowest among parents with heterogeneous religious background (14.3%), followed by homogeneously religious parents (20.2%), and non-religious parents (29.3%) (p = 0.000). The patterns of income were in line with parents' education. Heterogeneously religious parents had the highest income, followed by homogeneously religious parents, while non-religious parents had the lowest income (p = 0.018). The settlement type of the place of residence (urban or rural) did not differ significantly across the three groups defined by the parents' religiosity.
We also observed that religious parents lived in families with a more stable structure, while single-parent families and stepfamilies were significantly more common among nonreligious parents and those with heterogeneous religious background. The latter group had an exceptionally high incidence (16.8%) of the single-parent family type compared to other two groups, but the stepfamily structure was also the most common (11%) among them compared to homogenously religious and non-religious parents. Interestingly, however, the results of the analysis of variance did not show significant differences in the number of children across the three groups of parental religiosity.
In addition to economic and cultural capital, the social network outside the family is an important resource for parents. This includes the number and reliability of those who are involved in parenting by providing counsel and assistance. The size of the network is important as it is a good indicator of the size of the community from which the child receives attention. However, as the likelihood of unexpected events (illness, minor or major accidents) is relatively high in families with young children, the size of the network of people who can help with the child is as important as the ability to make use of it. The value of the index showing the strength of the parental network reflects this: the higher the value, the more people the network involves and the more reliable assistance is. However, the bivariate analysis of variance did not reveal significant differences in the magnitude of the index across the three groups of parental religiosity.

Effects on Home-Based Parental Attention (Second Hypothesis)
In our analysis, we investigate the home-based dimension of parental attention. The examined data were based on retrospective self-reports and the respondents' estimates and are therefore obviously not objective, but they provide an excellent comparative measure for differences between groups of parents. The first indicator of parental attention consisted of the parent-child joint activity scale, which we developed to measure the diversity of activities the parent had engaged in with the child in the month prior to data collection. An important analytical aspect is the multiplexity of the relationships, which increases their strength and stability (Coleman 1988). This also applies to parent-child relationships, and even supports children's success in subsequent stages of their academic career (Pusztai 2019). The more diverse activities the parent does together with the child, the richer and more productive their relationship can be.
In the bivariate analysis of variance, we found strong and significant (p = 0.000) differences in the frequency of each activity (see Table A1 in the Appendix A for the 12 activities tested) across the groups of parental religiosity, which favored children growing up in a religious environment. Homogeneously religious families led in terms of engagement with children in 10 of the 12 examined activities. The only activities that parents with heterogeneous religious background did more often with their child than homogeneously religious parents were the discussion of spiritual matters and world affairs. Homogeneously nonreligious families, in contrast, engaged in all activities the least frequently compared to the other two groups. The other indicator in the principal component of parental attention measured the time the parent spent with the child on weekends, which was also a retrospective self-reported estimate, but the bivariate analysis showed no significant difference in this across the groups by parental religiosity.
In the following, we present the results of our multivariate regression model for the factors affecting the principal component of parental attention, whereby the explanatory variables were included in several steps; in the first step, we included just the variables of religiosity, in the second, the socioeconomic status and social network indicators, and finally family structure and sex of the children (Table 1). Step 1 Step 2 Step 3 We found that parental attention was higher when both parents or even only one parent was religious. This effect was significant even after controlling for other variables. However, after accounting for the positive effect of parental network, the effect of religiosity weakened slightly, suggesting that religious parents could have a wider support network, although this was not observed in the bivariate analysis discussed above. Contrary to expectations, the parents' educational attainment, financial status, and place of residence had no discernible effect on parental attention. However, the effect of parental religiosity was slightly stronger when taking into account the negative effect of the number of children, which may suggest that the number of children among them could be slightly higher, although this also differed from the results obtained in the bivariate analysis. Parental attention was independent of the child's gender, and interestingly, also of family structure. In other words, parental attention was unaffected by whether the child lived in a singleparent family or stepfamily.

Factors Affecting the Effectiveness of Parenting (Third Hypothesis)
As the effectiveness of parenting can only be reliably tested after a sufficiently long period of time, we used indicators that could make parents perceive their parenting as effective based on the child's current outcomes. Parents' perception of the effectiveness of parenting was split into two dimensions, with one relative and two objective indicators. Parents generally associate objective success with academic achievement, such as having the best or second-best grades and participating in academic competitions. Relative success reflects the parents' views whether their child has outstanding or above-average achievement, at least in selected fields. Based on these indicators, we constructed an aggregate principal component of the effectiveness of parenting and examined the effects of explanatory variables on it in our regression model. We included the principal component of parental attention as an explanatory factor in the last step, as well (Table 2). Table 2. Factors influencing the principal component of the effectiveness of parenting, multistage linear regression results (regression betas and the significance level of coefficients).
Parental religiosity had a positive effect in the first step, although the effect of heterogeneous religious background was only weakly significant in the first step and disappeared in subsequent steps. However, the positive effect of homogeneous religious background on perceived effectiveness remained significant throughout and declined in magnitude only slightly when the effect of the parents' cultural, financial, and social resources was taken into account. This suggests that parents with homogeneous religious background had more resources, which, when considered in the analysis, decreased the independent effect of religiosity slightly. As expected, low educational attainment had a negative effect on perceived effectiveness, while a favorable financial status and a wide support network had a positive impact. In contrast, the place of residence exerted no effect. The positive effect of a homogeneous religious background was slightly further weakened when controlling for girls' higher achievement and the negative effect of a stepfamily structure. The latter may be explained by the fact that, consistent with our bivariate results, intact nuclear families were more common among homogeneously religious parents than stepfamilies. Finally, the positive effect of homogeneous religious background marginally declined further with the inclusion of the principal component of parental attention; in other words, religiosity also increased the effectiveness of parenting through greater associated parental attention. The fact that the effect of homogeneous religious background remained positive and significant throughout, however, implied that religiosity had an influence not only through greater parental attention but also in itself.

Discussion
We found that parental religiosity had an independent, significant, positive effect on both home-based parental attention and parenting effectiveness as perceived by the parents, regardless of social status. Our analysis did not address the dimensions and potential mechanisms of the effect of religiosity, but it may be rooted in a transcendence-based view of parenting, whereby caring for the child is not a barrier to self-actualization but a mission, and belonging to a community with similar values also strengthens the commitment to parenting. The effectiveness of religious parents is especially noteworthy because the beneficial effects of religiosity were observed independently, even after accounting for family structure, educational attainment, and financial status.
We measured the parents' social status using indicators of financial, cultural, and social capital. While home-based parental attention was unaffected by the parents' education or financial status, the perceived effectiveness of parenting increased if the parent possessed a secondary school leaving certificate and had high income. However, for both home-based attention and the effectiveness of parenting, a strong positive effect was exerted by the ability to rely on a trusted network of people outside the family with whom parenting tasks could be shared. Urban or rural residence had no significant influence on either parental attention or its' perceived effectiveness. The latter may be explained by the generally lowstatus composition of small rural schools, where competition for grades and recognition is not as fierce as in urban schools. Contrary to expectations, family structure (singleparent family, stepfamily, or intact nuclear family) had no effect on parental attention, but parenting effectiveness was lower among stepfamilies. In contrast, parental attention was reduced by a high number of children, while parenting effectiveness remained unaffected. Furthermore, we found that the child's gender did not influence home-based parental attention, but the perceived academic achievement was higher for girls than boys.
Finally, our measure of parental attention, which included a variety of joint activities and time spent with the child on weekends, significantly improved the effectiveness of parenting, independently of social background variables, family structure, and religiosity. This suggests that home-based parental attention, which does not necessarily relate to learning and schoolwork, can have a significant positive impact even among those of lower status.

Conclusions
In this study, we focused on home-based parenting activities. We were primarily interested in the impact of the parents' religiosity on both parenting itself and its effectiveness, while accounting for the effects of the parents' social status and family structure, among other factors.
In the first hypothesis, we hypothesized that in families of high status and intact families the parents were more likely to be religious, and that religiosity was associated with a wide support network and a large number of children. However, contrary to our hypothesis, parents with heterogeneous religious background had the most favorable educational attainment and financial status in the sample, followed by homogeneously religious parents and then non-religious parents. We also observed, in line with our hypothesis, that religious parents had a more stable family structure, while heterogeneously religious and non-religious parents were more likely to form a single-parent family or a stepfamily. However, the bivariate analyses revealed no significant difference in the number of children and the available parental network in favor of religious parents.
In relation to our second hypothesis, we combined the multiplexity of parent-child joint activities and the amount of time parents devoted to the child on weekends into the principal component of parental attention. We examined how certain factors affected this indicator. In the regression analysis, contrary to our hypothesis, social status had no effect, but parental religiosity as well as the size and intensity of the network assisting the parent had a positive effect. The effect of religiosity somewhat weakened upon the inclusion of the parental network, implying a wider parental network associated with religiosity. The number of children in the family, in contrast, reduced home-based parental attention, and the effect of religiosity was slightly stronger after accounting for the number of children, which may suggest that religious families had more children. Overall, the second hypothesis was partly corroborated as religiosity increased the intensity of homebased parental attention, however, it was unaffected by the parents' educational attainment and financial capital.
As for the third hypothesis, we found that a homogeneous religious background had a positive effect on the perceived effectiveness of parenting, even after the positive effect of higher parental status was taken into account. In addition, complex and intensive parental attention had a separate positive effect, which slightly decreased the effect of religiosity, implying that parental religiosity also influenced children's achievement through this factor. Our results also suggest that the parenting activities preferred by religious parents could also be used effectively without religiosity, and that the positive effect of religiosity on parenting outcomes is not limited to the parenting patterns examined. Multiplex activities with the child and above-average parenting time on weekends are only part of the parenting repertoire of religious parents. In addition, other effective parenting norms and practices may also serve as a mechanism through which religious parents can be more effective than non-religious ones. However, exploring these requires further investigation.
Overall, religiosity had a stable positive influence on parenting activities and their effectiveness. In addition, the indicator of parental attention we used was also a strong determinant of effectiveness independently. Our study suggests that parental religiosity and parental attention are valuable factors that can make parents effective regardless of adverse socio-cultural backgrounds. If low-status parents, despite their difficulties, can invest more time, engage in a more diverse set of joint activities, and operate a wider support network, their children's achievement is likely to increase. The awareness and dissemination of related research findings is highly important for professionals and parents alike and could offer a way out of the inertia of social determinism.
A limitation of our research is that it is difficult to measure parenting activities and their effectiveness. Parenting mostly involves the presentation of a model of behavioral patterns, which are constantly monitored by the child and comprise many spontaneous moments. These can best be assessed by qualitative methods. In addition, our data on parenting and its effectiveness, collected based on retrospective self-reporting and the respondents' estimation, are not objective but still provide a comparative reflection of differences between parent groups. An interesting result is that the discussion of spiritual matters and world affairs is also more common among heterogeneously religious parents. Maybe the diversity of parents' religiosity provides an opportunity for this discussion. However, this seems to contradict our second and third hypotheses. This activity, which develops the child's critical thinking, does not necessarily result in academic success during primary education, but outside of school and later in adult life.
Our analysis confirms the view that parenting skills are not innate, and parenting is a challenging task that can be done better or worse. However, its importance is underlined by the long-term consequences. In this research, both parents and teachers indicated commitment of parents to parenting, but roughly half of the respondents felt they required help with a variety of developmental and recreational activities with their children because they did not feel adequately prepared. Our research findings could have practical benefits through the development of a network of locally available educational staff to assist with everyday parenting tasks and improve parenting skills and practices, which could significantly enhance the quality of parenting.
Author Contributions: Conceptualization, G.P.; methodology, H.F.; formal analysis, H.F.; investigation, G.P.; resources, G.P.; data curation, G.P., H.F.; writing-original draft preparation, G.P.; writing-review and editing, G.P., H.F.; supervision, G.P.; project administration, G.P.; funding acquisition, G.P. All authors have read and agreed to the published version of the manuscript.  The scale lists those who are religious according to the teachings of a church, those who are religious in their own way, those who are undecided, the non-religious, and those of a decidedly different worldview (Tomka 1995).