INTRODUCTION

School homework refers to activities assigned by teachers to be completed outside of class time (^{Cooper et al., 2000}). Over the years, evidence has emerged supporting the importance of homework as a pedagogic tool to enhance academic achievement, as well as an alternative way to encourage parental involvement in children’s school lives (^{Bas et al., 2017}; ^{Cooper et al., 2000}; ^{Valdés-Cuervo et al., 2020}).

In Brazil, homework is a widespread activity that gained popularity after the 1990s due to educational policies and investments to strengthen family-school relationships (^{Carvalho, 2006}; ^{Gomes & Cunha, 2019}; ^{Resende et al., 2018}). It is estimated that 55.6% of 9th-grade Brazilian students have their homework checked by their parents, comprising 56.00% of students aged between 13 and 15 years old and 41.30% between 16 and 17 years old (^{Instituto Brasileiro de Geografia e Estatística [IBGE], 2016}).

In recent decades, the effects of parental involvement with homework time (PIH) on children and family outcomes have been the focus of various studies. Evidence suggests that PIH can impact children’s school achievement, behavior, motivation, and family relationships. However, findings regarding the effects of PIH still remain mixed and controversial (^{Barger et al., 2019}; ^{Fernández-Alonso et al., 2022}; ^{Kim, 2022}). Studies indicate that PIH increases children’s academic success and contributes to their development of socioemotional skills, fostering responsibility and self-efficacy (^{Xu et al., 2019}). In contrast, some studies have shown that parental involvement in children’s education can increase family stress, complicate children’s behavioral problems, and reduce academic achievement (^{Fernández-Alonso et al., 2022}; ^{Moè et al., 2020}). More recent studies have also shown that the context of the COVID-19 pandemic aggravated these difficulties, increasing problems related to PIH (^{Laguna et al., 2021}; ^{Touloupis, 2021}).

Additionally, studies encountered factors that could mediate PIH’s effects and its manifestation. It is already a consensus in literature that the quality of PIH is more related to positive outcomes than the frequency of involvement (^{Flunger et al., 2021}). Furthermore, PIH is also impacted by a diversity of children’s factors, for example, age, presence of diagnosis, learning problems, and sex and others. Parents of young children (elementary school) usually spend more time supervising homework compared with parents of children in middle and high school (^{Barger et al., 2019}; ^{Dettmers et al., 2019}; ^{Wei et al., 2019}). The same way, studies showed that the quality of PIH varies also according to the presence of previous diagnosis or learning difficulties in children, because in these cases, parents tend to struggle more to establish rules and to deal with behavior problems and learning difficulties during homework supervision (^{Sipila-Thomas et al., 2020}; ^{Xiao et al., 2022}). Finally, some studies demonstrated that parents showed different patterns of involvement depending on children sex (^{Silinskas & Kikas 2019}). Parents usually report more behavior problems in boys during homework time (^{Lee et al., 2007}; ^{Russell et al., 2014}), however, they tend to use more supportive practices for boys than for girls (^{Braza et al., 2015}; ^{Lee et al., 2007}).

An explanation that could account for the mixed findings regarding PIH impacts, and manifestation may be the complexity of the construct, which involves a range of parental behaviors that might have contrasting impacts depending on the context (^{Fernández-Alonso et al., 2022}). The controversial results on PIH can also be explained by how the construct is assessed and defined (^{Dumont et al., 2014}; ^{Trautwein, 2007}, ^{Xiao et al., 2022}). There are different theoretical models in literature explaining parental involvement in school as a construct (^{Núñez et al., 2017}; ^{Pomerantz et al., 2007}; ^{Trautwein, 2007}; ^{Watkins, 1997}). Next, three evidence-based models underlying homebased parental involvement assessment are presented. These models were chosen because they offer a valuable perception of parental involvement, and they present extensive evidence regarding PIH’s impacts on children’s and family outcomes.

The Grolnick and Ryan (1994) model proposes three parenting styles that could predict children’s socioemotional competencies and academic performance: (1) Autonomy support: parental practices that encourage children’s autonomy in problem-solving, help with collaborative decision-making, and reinforce success and personal choices, (2) direct parental involvement which includes parental interest and knowledge about children’s school life and (3) providing structure which encompasses parental provision of clear and consistent guidelines and physical study space to help children establish an academic routine.

^{Watkins (1997)} proposed an example of a bifactorial structure for PIH. In his model there were two styles of involvement: mastery and performance orientation. Mastery orientation is the style when parents drive their attention to the children’s learning process, valuing small academic achievements. In contrast, performance oriented parents tend to value only when children achieve great grades.

^{Pomerantz et al. (2007)} also proposed four dimensions of PIH: (1) Parental autonomy/control: autonomy support allows children to independently explore their environment, initiate actions, and engage in proactive problem-solving, while controlling involves parental pressure to steer children towards specific outcomes (e.g., excelling in school), often through commands, directives, or love withdrawal; (2) Parental person-focus and process-focus: parents with a person-focused orientation typically place greater emphasis on their children’s accomplishments, whereas process-focused parents tend to prioritize their children’s learning and mastery process. (3) Emotions cultivated and (4) Parental beliefs: the authors also acknowledge the potential for parents to develop positive or negative beliefs and emotions regarding their involvement.

Evidently, the diversity of theories and the PIH concept has led to numerous ways of assessing this construct. Parental involvement has primarily been measured and qualitatively analyzed through interviews, often consisting of open-ended questions (e.g., as used in ^{Cooper et al., 2000}). Instruments featuring yes/no questions or Likert scales have also been used, although less frequently. Scales designed explicitly to assess PIH are relatively scarce in the literature. More commonly, PIH is assessed indirectly through instruments that address the broader construct of involvement in school. For instance, parental perceptions of the level of parent-child conflict (see ^{Power et al., 2015}), children’s perceptions of the quality of involvement they receive from parents (e.g., ^{Núñez et al., 2019}; ^{Silinskas & Kikas, 2019}), and parental involvement in school life in general, combining home-based and school-based involvement (^{Puccioni, 2018}), and children behavior problems during homework (^{Power et al., 2015}). Other scales have been proposed to assess PIH, however, their psychometric properties have not been thoroughly evaluated. Additionally, these scales are often tailored for use within specific research contexts, such as scales designed to assess children’s perceived parental help with math homework (^{Dumont et al., 2014}; ^{Silinskas & Kikas, 2019}), and to evaluate children’s perceptions of parental control and parental support of homework (^{Núñez et al., 2017}; ^{Touloupis, 2021}; ^{Valdés-Cuervo et al., 2020}).

An example of a scale that attempted to assess the quality of parental involvement in homework was developed by ^{Watkins (1997)}. Based on his theory of parental types of involvement orientation, he created a 27-item scale that measures two involvement styles: mastery and performance-oriented. The Parental Involvement Inventory also included questions about general parental involvement, often framed as inquiries beginning with “how often” and questions assessing parents’ perceptions of teachers’ behavior towards homework.

^{Cunha et al. (2018)} introduced the Parental Homework Management Scale (PHMS) based on ^{Xu (2008)}. The items in the scale were modified to indicate parental involvement in homework assistance, resulting in eight items using a 5-point response scale. These items represented the three homework management strategies: (1. Arranging the environment, 2. Time management, 3. Monitoring children’s motivation and emotions). The PHMS has been employed in various studies to examine parental involvement in homework and its impact on student outcomes (^{Suárez Fernández et al., 2022}; ^{Xu et al., 2017}; ^{Xu & Wu, 2013}).

In Brazil, parental involvement is often assessed qualitatively, predominantly using interviews and focus groups (^{Carvalho, 2006}; ^{Resende et al., 2018}). When scales are serviced, the construct is commonly evaluated through indirect measures such as parent-child relationships, social skills, and parenting styles (e.g., ^{Glidden & Weber, 2020}; ^{Gomes & Cunha, 2019}; ^{Silveira et al., 2021}). Nevertheless, there are no validated scales in Brazilian Portuguese designed to assess parental involvement in homework. Therefore, in this study, we aimed to develop and evaluate preliminary validation evidence of a 15-item scale named the School Homework Involvement of Parents (SHIP) that assesses parental involvement during children’s homework. It is expected to find a multifactorial structure and positive correlations with variables that measure similar constructs and no correlations with variables that measure different constructs. Moreover, it is expected to find differences between children’s age, sex and diagnosis groups, according to what was previously demonstrated in literature.

METHODS

RESULTS

DISCUSSION

The current study aimed to assess the psychometric properties of the SHIP scale, developed by the authors to evaluate PIH. The scale originated from a semi-structured survey produced by ^{Cooper et al. (2000)} and incorporates five frequent aspects founded on evidence-based theoretical models of home-based parental involvement. PIH is a multifaceted construct that encompasses various parental behaviors and attitudes influenced by cognitive, emotional, and motivational processes (^{Flunger et al., 2021}; ^{Trautwein, 2007}; ^{Valdés-Cuervo et al., 2020}). While several scales for measuring PIH exist in the foreign literature (^{Cunha et al., 2018}; ^{Suárez Fernández et al., 2022}; ^{Xu et al., 2017}; ^{Xu & Wu, 2013}; ^{Watkins, 1997}), only a few of them have accompanying validity evidence. All things considered, to the best of our knowledge, there is no validated scale available in Brazilian Portuguese to assess this aspect (^{Glidden & Weber, 2020}).

The structural validity assessment of the SHIP scale revealed two dimensions: Factor 1-Child Behavior and Factor 2-Parental Involvement. This result was not aligned with the multidimensionality observed in other studies (^{Cooper et al., 2000}; ^{Flunger et al., 2021}; Pomeratz et al., 2005). Initially, it was expected to have a structure with at least 4 factors, based on the aspects listed by ^{Cooper et al. (2000)}. However, bifactorial structures demonstrated to be robust for this scale. Besides, there are models in literature that also found a bifactorial structure (^{Watkins, 1997}).

Regarding SHIP internal structure, Factor 1-Child Behavior includes items describing children’s behaviors and attitudes toward homework activities, while Factor 2-Parental Involvement comprises items that report parental behaviors and attitudes regarding homework supervision. It is worth noting that the internal consistency for Factor 2-Parental Involvement was lower than that for Factor 1-Child Behavior. One potential explanation for this discrepancy is that parents may find it easier to report on the child’s behavior than on their own behavior, which could contribute to the stronger factor power of Factor 1-Child Behavior. The omega-alpha coefficient for the SHIP total score was 0.81, indicating that the items are correlated and can be interpreted as a dimension. The use of Total Scores in assessing PIH is in line with previous research (^{Silinskas & Kikas, 2019}; ^{Valdés-Cuervo et al., 2020}), highlighting the utility of a comprehensive measure.

The relations with other variables demonstrated that the SHIP scale could identify differences among groups (children’s sex, age, and presence of previous diagnoses). Specifically, parents of girls reported higher scores on Factor 1-Child Behavior compared to parents of boys. This finding aligns with previous research indicating that parents tend to perceive more positive homework-related behaviors in girls than in boys (^{Lee et al., 2007}). Evidence suggests that girls exhibit greater conscientiousness and willingness to engage in school homework compared to boys (^{Braza et al., 2015}). However, boys display higher frequencies of hyperactive/impulsive and oppositional behaviors (^{Russell et al., 2014}). Similarly, the baseline results of the SDQ indicated that parents of boys reported higher levels of conduct problems compared to parents of girls. This finding is also consistent with literature that shows more externalizing problems in boys compared with girls (^{Lee et al., 2007}; ^{Russell et al., 2014}).

Parents of children with prior reported diagnosis showed lower scores on both Factor 1-Child Behavior and SHIP-Total Score. These findings align with the baseline SDQ results, where parents of children with previous diagnoses reported higher scores for general behavioral problems. This finding is consistent with previous research on behavior problems in children with psychiatric diagnoses and learning disabilities (^{Sipila-Thomas et al., 2020}; ^{Xiao et al., 2022}). No significant differences were found between the presence of diagnosis and sex groups in Factor 2-Parental Involvement and IEP parenting practices subscales. This finding is consistent with studies suggesting that parents engage with both boys and girls in similar manners (^{Bhanot & Jovanovic, 2005}; ^{Silinskas & Kikas 2019}). However, there is still some variability in the literature, with other studies suggesting that parents of boys tend to use strategies based on behavioral control, while parents of girls cultivate higher expectations for their daughters, which in some cases may be unrealistic (^{Bhanot & Jovanovic, 2005}; ^{Braza et al., 2015}).

As expected, parents of younger children reported higher scores in Factor 2- Parental Involvement and SHIP-Total Score. Studies have consistently reported greater parental involvement with school homework among younger children of age comparable to those examined in this report (^{Barger et al., 2019}; ^{Wei et al., 2019}). Older children, however, tend to rely less on their parents to complete school homework (^{Epstein et al., 2021}; ^{Wei et al., 2019}). From puberty onward, the extent of parental involvement changes. Parents of adolescents tend to be more involved in assisting with occupational decisions rather than daily school assignments (^{Goshin et al., 2021}; ^{Wei et al., 2019}).

No significant correlations were found between the SHIP scales and the DASS-21, indicating that SHIP scores were not significantly influenced by parents’ subjective assessments of their internalizing symptoms, evidencing that SHIP measured a different construct of DASS-21. Previous research has suggested that increased scores on measures of parental stress predict inconsistent disciplinary practices and coercive behavior during homework supervision (^{Katz et al., 2022}; ^{Moè et al., 2020}). The lack of correlation found in this study can be attributed to the low levels of stress, anxiety, and depression symptoms in the baseline data.

The correlations between Factor 1 - Child Behavior and the various subscales of the SDQ revealed significant but generally weak-to-moderate relationships. Specifically, a weak correlation was observed for SDQ Prosocial Behaviors, while weak and inverse correlations were identified for SDQ Peer Problems. In contrast, moderate and inverse correlations were discovered for the SDQ Hyperactivity, SDQ Emotional Problems, and SDQ Conduct Problems. These results suggest that the SDQ measures broader behavioral issues, especially external behavior, that may impact the homework-specific items assessed by Factor 1 - Child Behavior. This way, Factor 1-Child Behavior appears to focus more on evaluating behaviors related to conduct and hyperactivity in children, as opposed to social behavior. Studies indicate that children experiencing behavior problems often encounter difficulties with homework and are less motivated to complete homework tasks (^{Power et al., 2015}; ^{Xiao et al., 2022}; ^{Xu et al., 2019}).

Weak correlations were also observed between Factor 2-Parental Involvement and IEP-Positive Monitoring. This weak correlation suggests that Factor 2-Parental Involvement and IEP-Positive Monitoring capture different aspects of parental involvement. Factor 2-Parental Involvement evaluates practical dimensions of parental involvement (e.g., “Ajudo meu filho(a) na tarefa de casa sempre que ele(a) precisa”, whereas Positive Monitoring in the IEP evaluates more general and emotional aspects of parental involvement (e.g., “Eu pergunto como foi seu dia na escola e ouço atentamente.” Despite assessing different dimensions of involvement, IEP is a commonly used instrument in Brazilian research to assess parental involvement in home-based activities (^{Domingues, 2020}; ^{Glidden & Weber, 2020}). Regarding the correlations between Factor 1-Child Behavior and IEP-Positive Monitoring, a correlation coefficient of r=0.19 was observed, indicating that only 4% of shared variance exists between them. This result suggests that Factor 1-Child Behavior evaluates a distinct dimension of PIH that Positive Monitoring does not.

IEP-Relaxed discipline exhibited significant correlations with SHIP scores (Factor 1- Child Behavior, r=-0.22; Factor 2-Parental Involvement, r=-0.21 and Total Score, r=0.25). This can be explained by the fact that relaxed discipline pertains to parent incongruence in rules establishment (e.g., “Quando castigo meu filho e ele me pede para sair do castigo, após um pouco de insistência, permito que ele saia do castigo”. In contrast, SHIP describes parents’ level of involvement during homework (e.g. “Apenas verifico se a tarefa foi realizada e raramente eu ajudo na execução” (^{Gomide, 2006}; ^{Glidden & Weber, 2020}; ^{Silveira et al., 2021}). Studies show that parental relaxed practices can impact the quality of parental interaction during homework, increasing coercive and incongruence practices impacting children behavior and motivation towards homework (^{Braza et al., 2015}; ^{Cooper et al., 2000}). In summary, SHIP focuses on children ‘s behavioral problems and instrumental aspects of PIH.

CONCLUSIONS

In summary, the results demonstrate that the School Homework Involvement of Parents (SHIP) scale serves as a valuable tool for assessing parental involvement in homework within the Brazilian context. This instrument effectively addresses a crucial gap in the available array of tools designed to measure this pivotal aspect of the family-school relationship. The multifaceted nature of parental involvement, as underscored by theoretical models, was not found in SHIP; however, PIH dimensions were well-reflected in the bifactorial structure identified in results. Factor 1, named ‘Children’s Behavior,’ centers on parental perceptions of children’s homework-related behaviors. In contrast, Factor 2, labeled ‘Parental Involvement,’ delves into the attitudes and practices observed by parents during homework supervision. The data underscores the SHIP scale’s robust psychometric properties, characterized by strong internal consistency. The scale also showed relations with external measures associated with children’s age, diagnosis, behavior, and parenting practices. Additionally, the scale did not measure the same construct of an instrument that assesses parents stress, depression, and anxiety.

This study has limitations that should be considered. Firstly, the sample used was obtained from a non-generated sample, primarily consisting of mothers, a high number of children with previous diagnosis, and individuals from higher economic conditions. The sample characteristics limit the generalizability of the findings to more diverse populations. Secondly, data collection occurred during the COVID-19 pandemic, which could have influenced the parents’ responses. The pandemic brought about significant changes in educational settings, homework routines, and stress levels for parents and children. To address these limitations and enhance the robustness of the findings, future studies should be done aiming to include more diverse samples.

Therefore, the results of the present study highlight the potential of the SHIP scale as a valuable tool for assessing parental involvement in school activities, particularly regarding school homework. Another strength of this study lies in its timeliness and relevance, particularly considering the increased importance of parental involvement in school homework during and after the COVID-19 pandemic. Moreover, the study’s results suggest PIH may be reliability assessed through the internet. Practical contributions involve utilizing SHIP to construct or evaluate program interventions aimed at enhancing and assessing the effectiveness of PIH.