Parents’ roles in young learners’ motivation and task engagement in Indonesian primary schools: Questionnaire development and validation

Parents’ roles in English language learning

Parental involvement is crucial in children’s English learning, influencing motivation and engagement (Liang et al., 2024) and enhancing academic interest and passion, study habits, and grades (Shebani et al., 2025). Research indicates that parental socioeconomic status (SES) combind as their education, income, and occupation (Ma et al., 2023; Sirin, 2005) positively correlates with L2 achievements (Ma et al., 2023), academic results (Balsa et al., 2018), and English learning ability (Butler & Le, 2018).

University-educated mothers positively influenced their children’s English proficiency (Jalili, 2017; Tan & Caleon, 2022). Low parental education correlated with higher primary education drop-out rates in D.G.Khan (Ahsan et al., 2015). Family income affected children’s English learning engagement in elementary schools Hong Kong: low-income children were less engaged than those from middle-income families (Poon, 2020).

Various research papers have assessed the role of parental involvement in enhancing children’s academic achievements. The engagement of Chinese parents in their children’s English studies shows a significant link to the children’s proficiency in the target language (Butler & Le, 2018). Support provided by parents at home plays a crucial role in boosting students’ academic performance (Tan & Caleon, 2022; Torrecilla & Hernández-Castilla, 2020). According to Kim et al. (2020), the psychological and behavioral participation of parents has a more profound effect on the motivation and accomplishments of their children than the parents’ educational background. Communication between parents and teachers has also been shown to augment students’ achievements (Tan & Caleon, 2022) as well as their engagement and motivation to master English (Fan & Williams, 2010). Involvement of parents in extracurricular activities and attendance at parent meetings has a notable influence on student performance in elementary schools in Latin America (Torrecilla & Hernández-Castilla, 2020).

Additionally, parental guidance is a positive predictor of students’ intrinsic motivation to advance their English skills (Fan & Williams, 2010), while parents’ academic expectations also align positively with students’ English proficiency (Butler & Le, 2018), achievements (Tan & Caleon, 2022), engagement levels, and intrinsic motivation in English (Fan & Williams, 2010). Figure 1 illustrates some parents’ roles in young learners’ English learning based on the literature.

Figure 1. Parents’ roles construct.

Motivation and task engagement in EFL contexts

Motivation is identified as a complex entity, consisting of three components: the aspiration to learn the language, perceptions regarding language learning, and the intensity of motivation demonstrated by continued effort in language acquisition (Gardner, 2010). In English language learning, especially at the primary school level, motivation has been shown to influence learners’ persistence, participation, and language outcomes (Dörnyei & Ushioda, 2011). Motivation plays a pivotal role in language learning; it determines the rate and success of acquiring a new language (Gardner, 1985). This study conceptualizes motivation as the learners’ willingness, interest, and enthusiasm toward learning English. It is based on self-determination in which motivation is divided into two types: intrinsic and extrinsic motivation (Deci & Ryan, 1985; Ryan & Deci, 2020). Intrinsic motivation is defined as the interest or enjoyment in English tasks, while extrinsic motivation is encouraged by external rewards, pressures, or outcomes. These types of motivation are widely used in educational psychology and are appropriate for young learners’ development.

Task engagement refers to students’ involvement in completing tasks during the process of language learning (Philp & Duchesne, 2016). It is understood as a construct with four distinct dimensions: behavioral, cognitive, emotional, and social engagement (Philp & Duchesne, 2016). Behavioral engagement involves learners actively participating in tasks (Philp & Duchesne, 2016), which can be assessed through factors like time dedicated to tasks and responding to questions (Dörnyei & Kormos, 2000; Oga-Baldwin, 2019). Emotional engagement encompasses learners’ emotional connections and reactions to class activities (Sang & Hiver, 2021). Cognitive engagement relates to alertness, continuous attention, active mental processes, mental effort, and self-regulation techniques (Dao, 2021). Agentic engagement is defined by learners’ actions aimed at influencing classroom instruction (Oga-Baldwin, 2019). Social engagement refers to the extent to which a student follows written and unwritten classroom rules of behavior, for example, coming to school and class on time, interacting appropriately with teachers and peers (Finn & Zimmer, 2012).

Parents play a significant role in motivating young learners to learn English and in their engagement with tasks. Research indicates that parental involvement has a notable effect on children’s motivation (Choi et al., 2024; Liang et al., 2024; Sumanti & Muljani, 2021; Tong et al., 2021) and engagement (Liang et al., 2024). Interaction between parents and teachers is shown to enhance students’ motivation and engagement in English learning (Finn & Zimmer, 2012).

Parental questionnaires on young learners’ motivation and task engagement

Self-report questionnaires are the most frequently used tools to assess motivation and engagement in tasks. Some prior studies developed and adapted questionnaires to investigate motivation and engagement in tasks. For instance, Butler (2015b) and Butler and Le (2018) in China used parental a survey comprising various aspects related to their children’s learning of the English language. They included: (a) the socioeconomic status (SES) of parents, indicated by household income and level of education; (b) indirect actions by parents aimed at supporting English learning, for instance, fostering a literacy-rich home environment and using English at part of their jobs; (c) direct parental actions to facilitate English education, for example, engaging in their children’s study process and arranging for private English tutoring post-school hours; (d) parental beliefs regarding English education, including the significance of English; and (e) specific beliefs parents have about their own child, like their expectation concerning the child’s success in learning English. Butler (2015b) found that higher SES parents adjusted their parenting to children’s changing needs and, through their ability to provide more frequent English use outside the classroom, increasingly fostered their children’s competence and self-determined motivation. In contrast, lower SES parents tended to remain controlling and often failed to cultivate such competence and self-determined motivation. Butler and Le (2018) revealed strong correlations between SES and parents’ views on English, parenting styles, number of Chinese books at home, involvement in children’s English learning, and beliefs and expectations about their children’s English ability. Parents’ autonomous (rather than controlled) parenting, beliefs, and expectations all positively related to their children’s English achievement.

Torrecilla and Hernández-Castilla (2020) in Latin America used a parental survey on household assets, their connection with the student and school, and their general contentment with the school. They found that parents’ engagement in school and learning directly improved YLs’ academic performance. Third-grade students whose parents helped with homework, especially when the mother provided support, achieved higher results, and parents’ attendance at meetings with the principal and teachers, as well as participation in extracurricular activities, also positively affected achievements.

Wang et al. (2023) used the Chinese version of a parental involvement scale from a previous study covering three aspects: personal, cognitive, and behavioral involvement. Personal involvement measured perceived care about school life and emotions (e.g., easing emotions during learning difficulties). Cognitive involvement assessed exposure to stimulating activities and materials (e.g., buying study materials). Behavioral involvement evaluated parents’ participation in school activities (e.g., controlling online time). Their study showed that parents’ involvement directly increased children’s learning engagement and indirectly did so through children’s perceptions of that involvement, with learning engagement mediating the link between parental involvement and academic achievement. Parental involvement significantly predicted children’s English performance through its effect on engagement.

However, no research has been conducted in the Indonesian context on the development and validation of instruments to measure parents’ involvement in their children’s motivation and engagement with English tasks. Therefore, it is necessary to develop and validate a questionnaire on young learners’ motivation and task engagement from the perspectives of their parents. By doing so, we aim to enhance local educational practices and add to the existing body of knowledge in English language learning research in a new context.

Research design

The study was quantitative in nature. It involved the measurement and analysis of numerical data (Creswell & Creswell, 2018). Its focus was on the development and validation of a new instrument in primary schools in Indonesia.

Context of the study

The research was carried out in Padang elementary schools, where English is taught as a foreign language and is a required subject for YLs at elementary school from grade three. Children in both private and public elementary schools learn English for 70 minutes a week with English teachers who have an English educational background.

Participants

Participants were a large sample parents of fifth graders. We used a convenience sampling technique in selecting them, as this offers quick access to willing participants, enabling fast, efficient data collection while reducing financial and logistical costs (Cohen et al., 2018). It is especially useful for testing instruments and convenience sampling allows researchers to collect data in naturalistic settings. However, we acknowledge that convenience sampling may limit the generalizability of the findings to other populations, but this was the feasible way forward.

The study involved 270 parents of fifth-grade students, and their characteristics can be found in Table 1. Most respondents were female (88.5%), with 48.9% aged between 38 and 44. Many parents (46.3%) reported their English proficiency to be at the intermediate level, and every fifth respondent used English regularly. They used it primarily for entertainment purposes (46.96%). More than half of the participants had high-school education (60.4%) and were housewives (64.4%). Furthermore, 43% of respondents had a monthly income between IDR 1.000.000 and 5.000.000, while 38.1% chose not to disclose their incomes.

Data collection instrument

A questionnaire was developed for gathering data, as previously used instruments did not meet the purpose of the present study. It focused on parents’ roles by drawing on empirical research concerning parental influence on YLs’ motivation and engagement in tasks (Butler, 2015a; Tan & Caleon, 2022; Torrecilla & Hernández-Castilla, 2020; Tseng, 2021) to ensure construct validity. The instrument included five dimensions: 1) parental involvement (e.g., assisting with homework, attending school meeting, and actively engaging in their child’s educational journey), 2) parental expectations (e.g., setting high academic aspirations, offering encouragement), 3) access to educational resources (such as books, educational technology, and a study-friendly environment), and 4) exposure to enrichment activities (e.g., providing diverse experiences like travel, cultural events, and 5) access to extracurricular programs. Table 2 presents the 19 items on the roles of parents. We used a 4-point Likert scale ranging from strongly disagree = 1 to strongly agree = 4, to avoid a middle choice and to make sure respondents took sides.

Procedure

Ethical approval from the Institutional Review Board (IRB) of the Doctoral School of Education at the University of Szeged (Reference number: 24/2023) was secured prior to data collection. All participants were informed about the study’s objectives, and written consent forms were signed by all parents prior to participating in this survey.

The development and validation of the questionnaire concerning the roles of parents involved multiple phases: initially, all existing instruments in publications were reviewed; next, items were formulated based on the literature with input from experts. The questionnaire was evaluated by three experts (the co-author and two Indonesian teachers of English). Experts evaluated the content, clarity, and readability of the survey. They agreed that the questionnaire was clear and covered the content related to the roles of parents in motivating young learners to learn English and engage in English tasks.

Then the instrument was compiled and subjected to pilot testing to assess its validity and reliability. The text was translated to Indonesian language by two bilingual experts who were fluent in the English and Indonesian languages and had expertise in teaching English to children. Piloting took place in November 2024 with parents of fifth graders. The online questionnaire in Google Forms was distributed via WhatsApp to the English teachers at primary schools, who forwarded it to parents.

Data analysis

Exploratory and confirmatory factor analyses (EFA and CFA) were conducted with the help of the Jamovi app (version 2.3.28.0; The Jamovi project, 2022). This tool assists in performing complex data analyses and deriving meaningful conclusions from the collected data. Before the main analysis, we conducted descriptive analysis.

Table 3 presents the descriptive statistics for the dimensions of parental roles: they indicate generally high levels of agreement across all items between 3.06 and 3.73 on a 4-point scale. The highest mean scores were found in parental expectations (M = 3.65 to 3.73), reflecting strong convictions regarding the importance of English and parents’ significant aspirations for their children’s English proficiency. This dimension exhibited significant negative skewness and high kurtosis, implying a tendency for responses to cluster toward strong agreement. Parental involvement and access to resources displayed relatively high mean scores (between 3.15 to 3.35), accompanied by mild negative skewness and low to moderate kurtosis, indicating slightly skewed but closer to normally distributed responses. Mean scores for enrichment and extracurricular activities were somewhat lower (between 3.06 to 3.28), with more even distributions and less skewness, pointing to greater variability in the frequency of these activities. In conclusion, while parental expectations are consistently high, there is more variation in the level of actual involvement in enrichment and extracurricular activities.

EFA of the parents’ questionnaire

Before conducting EFA analysis, we ran the KMO of Measurement of Sampling Adequacy (MSA) and Bartlett’s sphericity test to make sure that these data are appropriate for factor analysis. The results showed that the data were appropriate for factor analysis (Kaiser, 1974) with MSA KMO = .911, and Bartlett’s sphericity test was chi-square = 2534, df = 171, and p = < .001. The fit indices showed that Tucker-Lewis Index (TLI = .976) and Root Mean Square Error of Approximation (RMSEA = .034), indicating a good fit (Hu & Bentler, 1999). Model test (χ² = 114, df = 86, p = .025).

The result of the EFA showed that there were five factors in the parents’ questionnaire, along the five dimensions. The five factors accounted for 58.1% of the total variance. Factor 1 (parental expectation) explains 16.03%, while factors 2 (access to resources), 3 (parental involvement), 4 (enrichment activities), and 5 (extracurricular activities) contribute 13.54%, 10.87%, 10.45%, and 7.25%, respectively. See details in Table 4.

Figure 2 presents the scree plot supporting the decision on five factors, displaying eigenvalues from a factor analysis. A steep initial decline and subsequent leveling indicate the optimal number of factors to retain, or eigenvalue > 1, due to their significant contribution to variance explanation (Kaiser, 1974).

Figure 2. Screen plot of factor analysis.

When conducting the EFA, we applied the Maximum Likelihood extraction method accompanied by a varimax rotation. The factor loadings span from.496 to.881. Each item’s loadings are classified into five distinct factors: parental involvement, parental expectation, access to resources, enrichment activities, and extracurricular activities. Parental involvement encompassed direct support behaviors, including, for example, helping children with the English tasks and communicating with the English teachers about the child’s progress. Item 2 exhibited the weakest loading, signifying a minimal contribution to the parental involvement factor. Parental expectations represented the aspirations and academic performance levels that parents set for their children, for example, expecting the child to do well in English and to speak English fluently. Access to resources included access to learning materials and tools at home, such as English books, the internet, games, and English language apps. Enrichment activities reflected parents’ efforts to broaden their children’s learning through non-formal learning opportunities at English language centres or with private tutors. Extracurricular activities were organized skill-building programs after typical school hours. Notably, the strongest indicator of the extracurricular activities construct is the factor loading associated with encouraging children to engage in activities outside of school (item 18, factor loading = .881) (Hair et al., 2019). Together, these factors demonstrate the multifaceted nature of parental engagement in children’s English learning environments. Table 5 presents the EFA results.

CFA of the parents’ questionnaire

A confirmatory factor analysis examined various constructs associated with the parental influence on students’ motivation and engagement with tasks. Constructs assessed included the same five dimension in the EFA. Each construct was evaluated through multiple items, with factor loadings demonstrating the strength of association between items and their respective constructs. For example, item loadings for parental expectation ranged from .646 to .869, indicating its moderate to strong representation in the construct (Hair et al., 2019). The Average Variance Extracted (AVE) values span from .41 to .71, highlighting how much variance is captured by the construct in relation to measurement error, with higher values (such as .71 for extracurricular activities) signifying better convergent validity (Fornell & Larcker, 1981). Reliability is assessed for each construct using Cronbach’s alpha (α) and Composite Reliability (CR), with values above .70 deemed acceptable. Most constructs exhibit strong reliability, such as parental expectation (α = .86, CR = .87) and access to resources (α = .86, CR = .86) (Tavakol & Dennick, 2011). Even a construct like parental involvement with a slightly lower AVE of .41 had acceptable reliability (α = .73, CR = .74) (Fornell & Larcker, 1981), indicating that the items reliably measure the underlying concept of each construct.

Overall, the results indicate that the measurement model is valid and reliable for evaluating the impact of parent-related factors on their children’s motivation and task engagement. Table 6 provides factor loadings of the five dimension constructs.

The chi-square statistic (χ² = 276, df = 142, p < .001) for testing exact model fit reveals a statistically significant outcome, implying an imperfect data-model fit. Nonetheless, given the chi-square test’s sensitivity to large sample sizes, researchers frequently evaluate model adequacy using alternative fit indices. The Comparative Fit Index (CFI = .945) and TLI = .934 both surpass the standard threshold of .90, suggesting an adequate fit (Hu & Bentler, 1999). Furthermore, the Standardized Root Mean Square Residual (SRMR = .045) and RMSEA = .059 fall within acceptable ranges (SRMR < .08 and RMSEA between .06 and .08), affirming that the model is reasonably well-fitting overall, despite the chi-square result.

The correlations among the five dimensions represent different aspects of support (parental involvement, parental expectations, access to resources, enrichment activities, and extracurricular activities) were all positive. This result indicates that improvement in one area generally corresponds to enhancements in others. Access to resource availability reveals strong correlations with both parental involvement (r = .798) and enrichment activities (r = .756), implying that families with greater resources are more likely to be involved in their child’s education and they tend to offer more enriching experiences. Additionally, extracurricular activities are strongly associated with access to resources (r = .728) and enrichment activities (r = .705), again emphasizing the interrelated nature of these supportive actions. However, parental expectations show more moderate correlations with the other constructs (ranging from .351 to .531), suggesting that although expectations are linked to other supportive behaviors, they may represent a distinct dimension of parental influence. Table 7 presents the relationships among these factors reflecting the parents’ roles. The √AVE for each factor, show satisfactory convergent validity, as all five √AVE values on the diagonal exceed the .50 cutoff. However, several inter-factor correlations involving parental involvement, access to resources, and enrichment activities exceed their respective √AVE values, partially violating the Fornell–Larcker criterion and weakening discriminant validity. Thus, although each factor is measured reliably, some constructs may not be fully distinct. Parental involvement overlaps conceptually with access to resources and enrichment activities, suggesting that these dimensions may be closely related facets of parental support for children’s learning rather than clearly separate constructs.

Furthermore, we evaluated discriminant validity using the heterotrait–monotrait (HTMT) ratio of correlations. The HTMT values ranged from .377 to .748, remaining comfortably below both the conservative .85 threshold and the more liberal .90 cut-off (Henseler et al., 2015). These findings suggest that the constructs have adequate discriminant validity and are statistically distinguishable from each other. Table 8 presents the HTMT ratio of correlations.

Figure 3 illustrates the five latent variables as factors: parental involvement, parental expectations, access to resources, enrichment activities, and extracurricular activities. These factors are connected to various observed variables or indicators such as items 1, 2, 3, 4, and 5. The arrows leading from latent variables to observed variables represent factor loadings, indicating the degree to which each indicator supports the underlying construct. For instance, item 6 exhibits a strong association with the factor of parental expectations, possessing a standardized estimate of .869. Similarly, items 18 and 19 show strong correlations with the extracurricular activities factor, with standardized estimates of .837 and .854, respectively. Double-headed arrows connecting the five multiple latent variables signify relationships among the constructs, for example, showing the strong association between extracurricular activities and access to resources.

Figure 3. Structure of parents’ roles.