Influence of the Parental Educational Level on Physical Activity in Schoolchildren

The factors influencing physical activity participation in children are varied, although there is evidence that the educational level of parents may be one important factor. The aim of this study is to analyze the influence of parental education on the level of physical activity and the sedentary behavior of their children. The parents of 727 students, separated based on school group (preschool: n = 179; primary: n = 284; secondary: n = 230; high: n = 34), were invited to complete a series of questionnaires assessing their educational level (low, intermediate, and high) and their child’s level of physical activity and sedentary behavior. Primary school students with highand intermediate-educated parents were found to engage in significantly more physical activity per week and spent more time engaged in homework than children with lower-educated parents. Secondary school students with higher-educated parents were found to spend significantly less time engaged in sedentary behavior than children with loweror intermediate-educated parents, and high schoolers with higher-educated parents engaged in less tablet time than children with lower-educated parents. Multiple linear regression demonstrated that greater physical activity was precipitated by certain sedentary behaviors in children with more educated parents, such as total time watching TV (primary school), doing homework (secondary school), and total time using a tablet/similar (high school), which increased total time engaged in physical activity. These findings suggest that more educated parents may implement structured time for their children to engage in a balance of physical activity and sedentary behaviors.


Introduction
The family is considered the fundamental basis for the child's socialization and its first educational context. As an agent of socialization, he develops in each of its members a certain way to function within society. So, its members will investigate what they learned at home with their behavior patterns [1]. Thus, the family is proclaimed as the first group of belonging where the person develops in an integral sense, because it is where children learn and develop knowledge, values, and attitudes that serve to guide their own behavior [2]. There are, therefore, different approaches to child rearing, such as the educational pattern, principles, and values in the family environment [3][4][5]. For this reason, it is important to promote an adequate family environment that provides the resources for a good personal and social development of children [6].
In this way, there are several studies that highlight the importance of the family as a health promoting factor [2], and specifically of physical activity [7,8] both in childhood and in adolescence [9]. Thus, it has been shown that quality of life in adulthood can be influenced by the habits adopted in the first years of life [10,11], when the family is the main educational agent.
The factors that influence participation in physical activity (PA )for young people are varied [38], although there is evidence that the educational level of parents is an aspect to consider [14,39,40]. The relationship between parental education levels and children's physical activity or sedentary behaviors are complex and are predicated upon psychosocial factors such as role modelling, encouragement, and the provision of tangible support. Edwardson and Gorely's [41] systematic review on parental influences and children's physical activity levels highlighted the importance of parental attitudes and encouragement towards developing physically active children. As higher educated parents may better understand the health benefits of physical activity, this is thought to facilitate greater parental encouragement and support of physically active behaviors [42,43]. As parents and children often undergo changes in their relationship as children transition into adolescence, it is unclear whether parental influence on their children's physical activity changes commensurately [44]. Few studies have examined parental education level across different child-age ranges, either via cross-sectional or longitudinal methods. One such study that did examine Spanish children aged 6 to 15 reported that parental education level was positively associated with physical activity [45]. However, their analysis did not report differences between ages, and the analysis of sedentary behavior was limited to television viewing time, which may not provide a complete view of children's physical activity and sedentary behaviors. Clearly, there is a need for further investigation into parental education influences across various child-age stages, and on sedentary behaviors.
It has been observed that when parents have a higher degree of education training, children tend to be more active [39,40] and engage in less sedentary behaviors [14]. This fact has been corroborated for both fathers and mothers [46,47]; interestingly, there are even authors who have shown how the mother's educational level is linked to a higher probability that her children participate in physical-sports activities [48,49]. Conversely, greater sedentary behaviors in children have also been linked with lower educational levels on the part of the parents [50]. Parents with a high educational level tend to better regulate the time their children spend on activities, highlighting a control of internet use, associated with sedentary behavior [51,52].
It is necessary, therefore, to continue to investigate the influence that parents have not only on the practice of PA by their children, but on the sedentary behaviors they develop. Thus, the objective of this study was to analyze the influence of the educational level of the parents, on the practice of PA and the sedentary behavior of their children, throughout the entire school stage (3 to 17 years).

Sample
In total, 727 Spanish students (143. 28  The inclusion criteria to participate in the study were: (a) to be students in infant, primary or secondary education; (b) not have any disease that prevents the practice of physical activity.
Before the start of the study, the educational center was informed, as well as the fathers, mothers and/or guardians of the objectives of the same, presenting a written informed consent to participate in it. This work is approved by the Ethics Committee of the local institution (University of Jaén, Spain (JUN.17/6).

Procedures
The instruments used were: Educational level of parents/guardians The parents/guardians were asked through a questionnaire for the highest level of education they had, using it as an indicator of their level of education. To do this, he/she was asked about their different levels of studies according to Spanish education. The response options were categorized into three levels: (a) Low level: no graduation, Primary/EGB, Secondary/ESO; (b) Medium level: Vocational Training I, Middle level training cycles, Baccalaureate/BUP/COU, Professional Training II, Higher degree training cycle; and, (c) High level: University degree or technical engineering, degree or higher engineering, Graduate, Master, Doctorate. These three levels used are in accordance with other authors who have previously employed it [14,53].
Physical Activity Practice Information regarding the practice of physical activity is collected contemplating different possibilities of practice: a) On active displacement: a qualitative question about how to move to the school, 1 = one way (walking, cycling, car, motorcycle, bus) and, 2 = return mode (walking, cycling, car, motorcycle, bus); which in turn included a quantitative question to know the volume in minutes in which the round trip from home to the school is covered ( < 10 min, between 10-15 min, between 15-20 min, between 20-30 min, > 30 min), where for the calculation of total PA, only the time used when it was active displacement was selected; b) unstructured PA practice: where the volume in hours/week of PA was asked how to play in the yard, in the park, plaza, etc. where it differed between the volume played during the week and at the weekend; and, c) structured PA practice: where the volume in hours/week in extracurricular activities was contemplated, differentiating between the time spent during the week and at the end of week. Once this information was obtained, the total hours of PA were added, differentiating between two moments: weekdays (Monday to Friday), and weekend (Saturday and Sunday). These questions have been used by different authors to know the practice of physical activity in school age samples [54][55][56].
Sedentary behavior Sedentary behavior was determined through the Health Behavior in School-aged Children (HBSC) questionnaire [57]. In the Preschool stage he/she was answered by the parents. The questionnaire consisted of answering six items indicating the number of daily hours spent watching television on weekdays and on weekends; use of a computer, tablet, or similar device on weekdays and weekends, and, time spent on class assignments on weekdays and weekends. Each of the questions included 9 options: 1 = no time, 2 = half an hour, 3 = one hour, 4 = two hours, 5 = three hours, 6 = four hours, 7 = five hours, 8 = six hours, and 9 = seven hours or more. The consistency of the questionnaire is high (Crobach's alpha = 0.721; 0.745; 0.719 in the three blocks respectively). This questionnaire has been used successfully in previous studies [58,59].

Statistical Analysis
Firstly, descriptive results were presented as mean and standard deviation. Secondly, the Kruska-lWallis test was used to test the differences of variables among parents' educational level (low, intermediate, and high). The Bonferroni's post hoc test was used when necessary to check the pairwise comparisons.
Thirdly, a multiple linear regression was used to estimate the effect of independent variables (total time watching TV, using tablet/similar, doing homework, and the parents' educational level as a dummy variable) on the dependent variable (total time of physical activity). The BMI variable was excluded from the model as it was not statistically significant differentiating educational levels. The Durbin-Watson's test was used to check whether the residuals in the model were independent and to account for collinearity effects. The linear regression model is described in the following equation where β0 is the intercept, βx represents the effects of the regressors (independent variables), and εi is the disturbance term.
Total time of PA = β0 + β1 x Total time watching TV + β2 x Total time using tablet/similar + β3 x Total time doing homework + β4 x Parents educational level + εi.
All the analyses were performed using the IBM SPSS version 22 statistical software (Armonk, NY: IBM Corp) and the significance level was set at (alpha) α = 05.

Results
The descriptive results for all the studied variables are presented in Table 1 as mean and standard deviation according to the parents' educational level and education stage. The results of Kruskal-Wallis test showed that significant differences were identified for total PA time (lower for low educational level) and total time doing homework (greater for intermediate educational level) at the infantile stage (p < 0.05) ( Table 2). The primary school students showed significant differences in total PA time per week (p < 0.001) with lower values in children with parents of low educational level; total time watching TV (p < 0.05) with greater values for intermediate educational level; and total time using tablet/similar (p = 0.001) with higher values in children with parents of high educational level. The results for secondary education students showed significant differences in total PA time (p = 0.044) with more activity for students with parents of high educational level; and for total time doing homework (p = 0.012) with higher values for students with parents of high educational level. Lastly, the high school students showed significant differences in total time watching TV (p = 0.017) with higher values for students of parents with low educational level. The results of the multiple linear regression were are available in Table 3. The results were significant for primary, secondary, and high school education level, but it was non-significant for infantile stage. The results for primary school students showed that total time watching TV, and high and intermediate parents' educational level increase the total time of physical activity. The secondary school student's regression showed that total time doing homework increased the total physical activity time; and for high school students the total time using tablet/similar increased the total time of physical activity.

Discussion
The objective of this study was to analyze the influence of parent's educational level on the PA levels of children throughout their schooling years (3 to 17 years). The results of this study provided evidence that children with low educational level parents engage in low physical activity during early schooling years. Interestingly, for example, the total time watching TV (primary and high school), doing homework (infantile and secondary school), and total time using tablet/similar (primary school) increased the total time engaged in PA for children with more educated parents. Together, these findings suggest that a parent's educational level is an important factor in children's participation in physical activity and sedentary behavior.
Stronger evidence was found for infant and primary school age children, whereby children with high and intermediate-educated parents engaged in more physical activity than children with lower-educated parents. Parents with greater education levels may be more aware of the numerous health associated benefits of physical activity [60] and have greater knowledge of age appropriate activities [61] which can act as a barrier towards children's engagement in physical activity. One other possible explanation may be that educated parents are more likely to impose scheduled time to participate in physical activity [62], potentially reflecting their own practices [63]. Infantile school children with intermediate educated parents were found to engage in significantly more schoolwork time than children with low or high educated parents. This finding reflects similar work showing that parents' behavior is modeled by children and can influence their behaviors, such as nutritional habits [64,65] and physical activity.
Somewhat counterintuitively, primary school children whose parents had intermediate and high education also engaged in more TV/tablet time than low-educated parents. This may be due to alternative reasons; that is, the associated relationship between parent's education level and socioeconomic status [66], thus reducing the affordability and opportunity to utilize a tablet or similar technology. More investigative research is required in order to understand why these relationships were observed between parent's education levels.
For secondary children, a significant difference was observed in total physical activity time and total homework based on parent's education level. This information is similar to other studies, which have shown that when parents have a higher degree of training, children tend to be more active [39,40]. In addition, there is a greater perception of high education level parents and of the importance of homework for parents who did not attain a high school education [65].
Interestingly, in high school children, no significant difference was observed in total physical activity time based on parent's education level. This finding may be due to the increase in personal autonomy associated with transitioning from childhood into adulthood [61,65]. Other factors may therefore play a greater role in the physical activity behaviors of older children, beyond parental influences [67].
The effect of independent variables such as total time watching TV, using tablet/similar, doing homework, and the parents' educational level on total time engaged in physical activity revealed an intriguing relationship. For infantile and primary school students, it showed that total PA is decreased in low parents' education. One explanation for this finding may be that higher educated parents are more likely to encourage a balanced and reciprocal approach to physical activity and certain sedentary behaviors (e.g., watching tv [68]). In primary school there is a higher total of TV and tablet time for intermediate and high education level. Deslandes and Rousseau [65] previously reported that some parents believe television and tablet technology to have beneficial factors, such as relaxing children. Comparable findings were also found for older children, with secondary school student's regression showing that total time doing homework increased the total physical activity time. Perceived benefits of both physical activity and these sedentary behaviors may reflect the positive relationship reported in this study and others [61,69].
The current study has some limitations that need to be addressed in further research. Firstly, the use of smart devices and tracking systems would improve the quality and quantity of PA information from students. Secondly, the analysis of psychological factors may explain students' characteristics such as the self-regulatory processes or role modeling. Lastly, future studies should increase the sample and age of students to obtain more reliable and generalizable findings.

Conclusions
In conclusion, infantile and primary school students with low educated parents were found to have a lower total time of physical activity time than high and intermediate educated parents. Primary school students with high educated parents were found with more tablet time than children with lower-and high-educated parents. Secondary school students with higher-educated parents were found to spend significantly more time engaged in physical activity and homework than children with lower-or intermediate-educated parents. These findings suggest that more educated parents may implement structured time for their children to engage in a balance of PA and sedentary behaviors.