Bijlage 1 - JOGG-Plan van Aanpak 2022-2026 gemeente Alblasserdam.pdf

49

healthy drinking habits for children. However, the theory does not seem to particularly apply to sleeping patterns for children. Some researchers have already tested the influence that the presence of peers may have on the lifestyle choices of children. First of all, the research of Salvy et al. (2007) investigates the effect of peer presence on the dietary intake of children aged 6-10 years old that are overweight and children that are not. The results show that children who are overweight eat less when they are in the presence of peers, while children who are not overweight eat more when peers are present, suggesting an effect of peer modeling. These results show a clear effect of the presence of peers on dietary choices, which for children that are not overweight relates to less healthy eating habits (Salvy et al., 2007). Furthermore, the research of Larson & Story (2009) specifically looked into the relationship between the social environment and the intake of soft drinks in children. Their results showed that the dietary choices of children, including the consumption of soft drinks of individuals are influenced by the presence of their peers. This influence however depends on the social norms regarding dietary choices.

As, according to Larson & Story’s (2009) study, soft drinks are one of the most consumed drinks in schools, one could assume that the social norms regarding drinking habits are drinking soft drinks. This suggests that the presence of peers leads to children feeling the need to follow the social norm and thus relates to less healthy drinking habits for children. The research of Godsell & White (2019) investigated the influence of peers on the sleeping behavior of 13- and 14-year old’s. Their results show that the children admitted that they felt that peers influence their sleep behavior, as they spend their nights communicating with their peers via their mobile phones. The research of Vaughn et al. (2015) showed that even for children of 4 years old, peer relations influenced their sleeping habits, as a higher peer acceptance relates to sleep duration for these children. These two studies show a clear effect of peers on the sleeping patterns of children, however, these entail interactions with peers, not just the presence of peers. Unfortunately, there does not seem to be literature regarding the relationship between the mere presence of peers and sleeping habits. These theories and research clearly show that the presence of peers can affect the lifestyle choices that children make.

With regards to merely the presence of peers, the literature and theories do not point towards a specific relation with sleeping patterns. Regarding the presence of peers and drinking habits, the literature and theory suggest that the presence of

50

peers leads to less healthy dietary choices, which includes a higher consumption of soft drinks. This leads to the following hypothesis: H1.

The presence of peers relates to less healthy drinking habits for children.

Neighborhood The second aspect of the social environment is ‘the neighborhood’, which can be considered positive or healthy when the neighborhood is child-friendly. The relation between the neighborhood and a healthy lifestyle for children is something that has also already been partially explored by theory and research. First of all, Bronfenbrenner’s Ecological Systems theory (1998) can be applied to this relation. This theory argues that the relationships within the microsystems, or social environments, are bi-directional. This entails that the child is influenced by things in their immediate environment, such as the neighborhood, but the child in turn can also influence their neighborhood. Bronfenbrenner argues that the environment, including the neighborhood, of children is capable of changing the beliefs and actions of a child (Guy-Evans, 2020). In the case of sleeping habits, this entails that the neighborhood can provide resources, such as safety and quietness, to help children sleep properly. Regarding drinking habits, the neighborhood can influence choices by store options that are available in the neighborhood. Therefore, according to the Ecological Systems Theory, the environment can influence the health behavior of children by the resources it does or does not provide (Holt et al., 2009).

A few researchers already investigated the effect that neighborhoods can have on the lifestyles of children, regarding drinking habits and sleeping patterns. The study by Holt et al. (2009) shows that children in neighborhoods with, among other things, high access to healthy supermarkets, parks, and playgrounds tend to have healthier diets, including healthier drinking habits. This can be explained by the fact that people in neighborhoods with more such healthy stores have more opportunities to go there and purchase healthy foods and drinks. Veugelers et al. (2008) researched the relationship between characteristics of the neighborhood and eating habits. Their results show that children in neighborhoods with access to supermarkets with reasonably priced fresh produce have healthier eating habits and were less likely to be overweight. This finding was explained through the opportunity that certain neighborhoods do or do not provide, which leads to different health behaviors. Lastly, the research of Marco et al. (2012) looked into the relationship between neighborhood environment and sleep patterns of young adolescents. Their results reveal that some

51

neighborhood factors are associated with problematic sleep patterns for young adolescents, such as insufficient school-night sleep. There are multiple factors involved in this relation. For example, poorer neighborhood conditions lead to children going to bed later in the night. Additionally, children in neighborhoods with lower social-economic status have been found to have a lower sleep quality, possibly due to neighborhood noise and lightning. Furthermore, feelings of unsafety in the surrounding environment can interfere with the sleeping habits of children. Therefore, in multiple ways, an unsafe and not child-friendly neighborhood leads to less healthy sleeping habits. Overall, the theories and literature seem to agree that a safe and child-friendly neighborhood promotes a healthy lifestyle for children, by providing these children the needed resources to achieve this lifestyle. The literature clearly points to this effect for both drinking habits and sleeping habits. This leads to the second set of hypotheses: H2a.

A child-friendly neighborhood relates to more healthy drinking habits for children.

H2b.

A child-friendly neighborhood relates to more healthy sleeping habits for children.

Media education The third aspect of the social environment is the ‘media education’ provided by the parents. This is considered healthy if parents have rules about the media use of their child and if they are engaged in the media use of their child through supervision. The relationship between media education and healthy lifestyles for children is something that remains largely unexplored by theory and research. There are, however, some pieces of literature that can provide some sort of indication. The research of Bickham et al. (2015) shows in this regard that rules about media use are associated with lower levels of depression symptoms in children. The mechanism behind this relationship is not confirmed, but there is a possibility that these types of rules indicate parental involvement which is known to be protective against depression in children. This already provides an indication of a positive effect of media education on the well-being of children. Furthermore, findings from the research of van den Eijnden et al. (2021) indicate that strict rules about media use predict better quality of sleep for children. This possibly has to do with limiting the opportunities that these children have to engage in social media before going to bed and thereby increasing the time children sleep. Additionally, the research of Rosen et al.

(2014) investigated the relationship between technology and eating habits among children

52

and teenagers. Their results point to the importance of parents setting limits and boundaries for their children regarding media use. The idea behind this is that parents can hereby provide opportunities to children to engage in other more healthy activities and habits to do that do not include screens. Furthermore, other scholars also confirm the importance of media education (Reid Chassiakos et al., 2016: Łuszczki et al.,2021), however, they only researched this concept in relation to the effects of media education. For example, Łuszczki et al. (2021) show that the use of media and smartphones for children let to poor sleep and less healthy eating habits, as it leads to fewer hours of sleep and lower sleep quality. As a solution to this problem, they argued for more media education, as appropriate parental control can restrict media use and thereby counter the negative consequences of media. Despite media education by parents not being a common topic of interest for researchers, the existing literature shows that media education does promote more healthy choices for children. All findings taken together do suggest that this is also the case for dietary choices and sleeping patterns. Therefore, the expectation is that media education associates with healthier drinking- and sleeping habits for children. This leads to the following set of hypotheses: H3a.

Media education relates to more healthy drinking habits for children.

H3b.

Media education relates to more healthy sleeping habits for children.

Leisure The last aspect of the social environment is the activities that occupy the leisure time of children. This study includes the activities social media, gaming, and reading. Hereby, social media use and gaming are considered to be unhealthy ways of spending leisure. Rosen et al. (2014) found in this regard that media and technology use, such as social media and gaming, predicts ill-being in children. They argue that general screen activities, regardless of the number of hours, predict poor health outcomes, such as unhealthy dietary habits and physical inactivity. The reason behind this is that children tend to consume more calories while using technology, such as gaming, and such activities can, furthermore, replace physical activities. Reading in general, however, is considered to have a positive effect on children. HorowitzKraus & Hutton’s (2018) research shows in this regard that reading is important for a healthy brain- and language development for children, indicating that reading is a healthy way of spending leisure. They also confirmed that screen time, in general, harms the development of

53

children, as it relates to lower cognitive control and language development. Based on these insights, the leisure time of children will be considered healthy when a child spends their leisure time reading and considered less healthy when a child consumes video games or uses social media. The possible effect that these different ways of spending leisure time can have on the lifestyles of children is something that is not yet as largely explored by theory and research as some of the earlier discussed aspects. There are however some studies that are relevant to this relationship. The research of Maksniemi et al. (2022) looked into the relationship between active social media use and the sleeping habits of children. They found that for young adolescents social media use is associated with delayed bedtime. For middle and late adolescents, social media use is related to emotional exhaustion. The mechanism behind this relationship is that the time spent using social media reduces the time that children sleep and thereby increases psychological problems. Furthermore, the research of Brambilla et al. (2017) looked into the sleeping habits of children aged 1-14 in relation to reading and the usage of video devices. Their results show that reading before bedtime does not associate with sleeping habits.

However, the results do show that using video devices before going to sleep leads to less optimal sleep for children, for example sleeping lightly. A possible mechanism behind this is the light exposure from video devices in combination with the suppression of melatonin production. Mak & Fancourt (2020) looked into the relationship between reading for pleasure in childhood and outcomes in terms of healthy lifestyle behavior in adolescence. They found that reading at age eleven is associated with higher levels of fruit consumption at age fourteen. A possible explanation for this relation is that reading is associated with a better school performance, which relates to a higher chance of being informed on health-related matters. This in turn can increase the chance of healthy life choices. Therefore, it seems like reading for pleasure is a positive predictor of healthy lifestyle choices. In addition, Turel et al. (2017) found that for children video gaming leads to increased consumption of sweet drinks. More specifically, the results showed that children tend to drink more sweet drinks while they are playing video games. Therefore the existing literature does point towards a positive relationship between more healthy ways of spending leisure time and healthy lifestyle choices of children.

In this regard, the literature shows that playing video games leads to increased consumption of sweet drinks, while reading is associated with healthier drinking habits. This leads to hypothesis 4a. The

54

research also clearly shows that social media and gaming relate to less healthy sleeping habits, this leads to the expectation that a healthier way of spending leisure leads to healthier sleeping habits, see hypothesis 4b. H4a.

A healthier way of spending leisure time relates to healthier drinking habits.

H4b.

A healthier way of spending leisure time relates to healthier sleeping habits.

When looking at the individual aspects of the social environments of children, it is possible that they differ in the way and the extent to which they influence the lifestyle of children. Existing literature does not have a clear answer to this. Whether or not this might be the case will be investigated through the qualitative part of this study.

Methodology As aforementioned, this study uses a mixed-methods approach, meaning that it contains both quantitative and qualitative data. As these two different forms of data require different methods of analysis, their methods will be discussed separately and in sequence. First of all, the quantitative methods concerning the dataset will be discussed, after which the qualitative methods regarding the interviews with the parents and experts will be discussed.

Quantitative methods Dataset For the quantitative part of this study, data will be used from the Kindmonitor 2017 from the GDD Zuid-Holland Zuid (GGD zhz, 2017). The Kindmonitor is a questionnaire for parents which includes all sorts of questions about their child and its environment. The dataset contains the age groups 0-4 and 4-12, but for this research, only the part concerning the age group of 4-12 was used as the target group for this study is children in primary school. This part of the data contains 4.850 participants. 48.7% of the children of the participants are girls while 51.3% are boys. The data includes children from the age of 4 to 12, with each age being approximately evenly represented. The reach of this questionnaire contains 15 towns in the area of Zuid-Holland Zuid, including Alblasserdam. More specifically, of the 4.850 participants, 299 are from Alblasserdam. Furthermore, all questionnaires took place between September 2017 and December 2017. Regarding the recruitment, 16.000 parents of children aged 0-12 in the area of Zuid-Holland Zuid were invited to participate in the research. The participants did not get paid for participating in the questionnaire.

55

Measures Healthy lifestyle In order to answer the research questions, several key concepts need to be operationalized. First of all, a healthy lifestyle for children will be measured using multiple variables in the dataset concerning lifestyle. The first question is: ‘Do you have a set bedtime for your child?’, with the answer categories 0 (no) and 1 (yes). The other questions are: ‘How many days a week does your child drink sugary drinks?’, ‘How many glasses of sugary drinks does your child drink on those days?’, ‘How many days a week does your child drink water or tea without sugar?’, and ‘How many glasses of water or tea without sugar does your child drink on those days?’. These four questions have answer categories ranging from 1 (very little) to 8 (very much). The two questions concerning the consumption of sweet drinks have a Spearman’s correlation of .56 (p <.001), meaning that their correlation is high enough to be merged into a scale. Therefore, they were recoded and merged together based on their mean scores to one scale about sweet drinks, ranging from 1 (large consumption of sweet drinks) to 8 (little consumption of sweet drinks). The same was done for the two questions regarding water consumption, which have a Spearman’s correlation of .634 (p <.001).

Based on their means, they were merged into a scale ranging from 1 (little consumption of water) to 8 (large consumption of water). As the correlation between the variables for sweet drinks, water, and bedtime is low, they cannot be merged into one scale that includes all three facets of a healthy lifestyle. Therefore, these three will be separately used in the analyses. Social environment Furthermore, the social environment will be measured using the four aspects that together represent the social environment in this study. Presence of peers The first aspect of the social environment regarding the presence of peers will be measured by the following statement included in the dataset: ‘There are too few friends for my child to play with outside’. The answer categories for this question range from 1 (low presence of peers) to 3 (high presence of peers).

56

Neighborhood The second aspect of the social environment, the neighborhood, will be measured by a variable in the dataset that specifically concerns the child-friendliness of a neighborhood. Namely, ‘How child-friendly is the neighborhood in which you live?’, which answer categories range from 1 (very child-friendly) to 4 (not child-friendly). This was recoded to a higher score meaning a higher degree of child-friendliness. Media education The third aspect of the social environment, media education is measured by five questions in the dataset regarding this topic. The first question is: ‘Do you have rules and agreements with your child about the number of hours that your child can use a computer, laptop, or mobile phone to surf the Internet, play games, or watch TV?’, with the answer options 1 (yes) and 2 (no). As this variable is binary, it could not be merged into a scale with other variables. Therefore, it will be used separately. The other four questions are: ‘Do you know which sites your child visits on the Internet?’, ‘Do you always know who your child is chatting or texting with?’, ‘Do you watch your child when he/she is on the internet?’, ‘Do you talk to your child about what he/she sees on the internet?’. The answer scales of these questions reach from 1 (always) to 4 (mostly not), with an additional answer category of 5 (not (yet) applicable).

The last four questions will be merged into a scale of parental supervision with a Cronbach’s alpha of .729 reaching from 1 (low media education) to 4 (high media education). Therefore, this aspect of the social environment entails two measures, the question about the rules concerning media use and the scale regarding parental supervision. Leisure The last aspect of the social environment, leisure, is measured using three questions concerning leisure activities in the dataset. The first question is: ‘How often does your child read by himself (outside school hours)?’, with the answer categories ranging from 1 (every day), 2 (a few times a week) to 3 (never). The other two questions are: ‘Does your child use social media?’, and ‘Does your child play video games?’, which could be answered by 0 (no), 1 (yes), and 2 (I don’t know), of which the score 2 (I don’t know) was coded as missing. These three variables were recoded so that a higher score means more healthy spending of the leisure time. For social media and video games, this entails the answer categories 0 (yes), and

57

1 (no). As these three variables were not highly intercorrelated, they cannot be merged into one scale. Therefore, they will be used separately. Control variables Next to the dependent and independent variables, this study also includes a few control variables. These are age, educational attainment of parents, gender, and parenting difficulty. Age is measured by the question ‘What is the age of your child?’, with the answer options ranging from 4 to 12. The education of parents is measured by the question ‘What was the highest completed education of you and your partner?’, with 1 being ‘low educated’ and 3 being ‘highly educated’. Gender of the child is measured by the question ‘Is your child a boy or a girl?’, with 1 ‘boy’ and 2 ‘girl’. These three variables are included as control variables as they are potential confounders in the expected associations. Lastly, parenting difficulty is measured by the question ‘How do you experience raising your child?’, with the answer categories ranging from 1 ‘very easy’ to 5 ‘very hard’. This variable is included as a control variable as the expectation is that parents that find raising their child difficult can affect the outcomes of their children with regards to lifestyle and social environment. For example, Sadeh et al.

(2010) found in this regard that parental difficulty, such as setting limits, relates to sleep problems for children. Even though the effect of parenting difficulty on other lifestyle behaviors might not have been researched yet, it can’t be ruled out that the effect is in fact present. Therefore, parenting difficulty will be included as a fourth control variable. All missing values on the earlier discussed measures were excluded. In total, the selection of the aforementioned variables leads to a sample consisting of 3853 participants. See Table 3 for the descriptive statistics of all concepts included in this research. Analyses To test the hypotheses and thereby answer the research questions, statistical analyses will be conducted using the program SPSS. In advance of the analyses, the variables of interest were tested for multicollinearity. Concerning the analyses, linear regression analyses will be used for the analyses including the consumption of water and sweet drinks. Furthermore, as the variable bedtime is binary, the analyses with bedtime as the dependent variable will be executed using logistic regressions. For every analysis, the regression coefficient or odds ratio

58

and the significance level will be presented. Furthermore, for the linear regressions, the tvalue and the degrees of freedom will be displayed. To assess the first hypothesis regarding the relationship between the presence of peers and a healthy lifestyle, two regressions will be conducted, one including sweet drinks and the other including water as dependent variables, and the presence of peers as the independent variable. To test the second set of hypotheses concerning the association between the neighborhood and healthy lifestyles, there will be three regressions conducted, this time including the three variables about healthy lifestyles as the dependent variable and neighborhood as the independent variable. Hereby the regression with bedtime as the dependent variable will be logistic and the other two regressions will be linear. Likewise, for the third set of hypotheses about the relation between media education and healthy lifestyles, three regressions will be conducted with the three variables concerning healthy lifestyles as the dependent variable and media education as the independent variable, with the analysis for bedtime being logistic.

To test the fourth set of hypotheses regarding the relationship between leisure and healthy lifestyles, another set of three regression analyses will be conducted with the healthy lifestyle variables as the dependent variable and leisure as the independent variable. Again, the analysis including bedtime as the dependent variable will be logistic. Lastly, to test whether the found effects of the analyses mentioned above will hold, all four aspects of the social environment will be put into the analyses together. Entailing one multiple regression analysis for sweet drinks and one for water, and a logistic regression for bedtime. Hereby, the aspects of a healthy lifestyle are the dependent variables, and the presence of peers, neighborhood, media education, and leisure are the independent variables. An overview of all variables and the accompanying hypotheses can be found in Figure 1. Furthermore, every analysis will firstly be run with the control variables age, education attainment of parents, gender, and parenting difficulty, to rule out the effects of these potential confounders. Ethical considerations Potential participants received an invitation letter which, among other things, included information about the goal of the interview, how privacy issues are handled, and whom to reach in case of questions.

Furthermore, the questionnaire itself also included a paragraph regarding the freedom to skip questions if respondents felt uncomfortable answering them.

59

Figure 1. Path model of all variables and accompanying hypotheses

60

Qualitative methods To be able to dive deeper into the two topics of interest, this study also entails a qualitative part, which can help to paint the picture behind the quantitative numbers. Therefore, five parents of children aged 4-12 in Alblasserdam were interviewed. As the input of the parents on its own seemed surprisingly positive, the choice was made to also include expert interviews. The aim is that this will bring another perspective and, furthermore, reduce the potential selection bias. The methods of these two types of interviews will firstly be discussed in sequence, after which the joint analyses and the ethical considerations will be discussed. Interviews with parents Selection and recruitment Concerning the interviews of the parents of the children in Alblasserdam aged 4-12, the recruitment took place via the schools of their children. All the primary schools in Alblasserdam were firstly introduced to this study via a short introduction in a recurring meeting with all primary schools in Alblasserdam and the municipality. A few weeks later, the directors of the schools got contacted via e-mail asking if they could share an information letter with the target group (see Appendix 1 for the information letter). Via the contact details included in the information letter, the parents could inform the researcher of their willingness to participate in the interviews.

Hereafter, appointments were made via e-mail with the parents. In total, five parents were interviewed, whose children go to four different schools and all live in different neighborhoods. See Table 1 for the description of the participants. Table 1. Description semi-structured interviews with parents Participant Gender

Age child

Gender child

Neighborhood

School

1

Male

7

Girl

De Dijk

Het Kompas

2

Female

10

Girl

Schrijversbuurt Het Kompas

12

Boy

Schrijversbuurt Special education

3

Female

12

Boy

Souburgh

Beukelmanschool

4

Female

4

Boy

Centrum

Beukelmanschool

6

Boy

Centrum

Beukelmanschool

4

Boy

Bloemenbuurt

t’ Nokkenwiel

5

Female

61

Data collection As some people might not feel comfortable meeting in person due to the pandemic, the participants were offered the possibility of both an online and offline version of the interview. All participants chose to meet in person. On average the interviews lasted approximately 45 minutes, with some interviews being longer than others. The interview questions were made based on the questionnaire of the earlier mentioned dataset of the Kindmonitor, the website of JOGG, as well as the theoretical framework of this study. See Appendix 3 for an overview of the main interview questions. To provide sufficient space for in-depth and follow-up questions, the interviews took place in a semi-structured manner. Therefore, the content of the interviews differed slightly per interview, depending on the course of the interview. After the interviews were completed, they were transcribed, after which the recordings were deleted. The interviews took place from March-April 2022. Experts interviews Selection and recruitment After the interviews with the parents took place, the input that the parents provided seemed surprisingly positive. This could be due to participants answering in a socially desirable manner.

However, it also leads to the suspicion that it might be the case that mostly parents that are engaged with their child and their lifestyle tend to react to these types of invitations. Therefore, it is possible that these interviews largely include cases in which everything is going very well, but do not cover the more ‘problematic’ cases concerning lifestyles and social environments. Luckily, Alblasserdam knows multiple organizations and experts that are engaged with children and youths in Alblasserdam. Three organizations were chosen, which all in a very different way frequently interact with the target group. The first organization is ‘Jongerenwerk’, which aims to be present where children are and tries to support children by organizing fun activities but also activities surrounding prevention, for example concerning drug use. The second organization is the ‘ABC-team’, which organizes cultural and sportrelated activities for multiple age groups, including children aged 4-12. Lastly, two experts from the organization ‘Jeugdteams’ were interviewed together in the third and last interview. Their organization supports families that voluntarily ask for help by helping these families solve their problems or by referring them to other specialists. They were all interviewed as the expectation is that they have relevant experience with our target group, see Table 2 for an overview.

62

Table 2. Description semi-structured interviews with experts Participant

Gender

Organization

1

Female

Jongerenwerk

2

Male

ABC-team

3

Female

Jeugdteams

4

Female

Jeugdteams

Data collection Firstly, the experts were asked if they were comfortable with an in-person interview. All participants agreed, however, due to the circumstances the first interview was changed to an online interview. In all three interviews, the same topics were discussed as in the interviews with the parents but this time from the perspective of the expert. Before the interview started the information letter and the informed consent, see Appendix 2, were discussed. The interviews were less structured than the interviews with the parents, as all experts had a different area of expertise. The interviews took approximately 45 minutes. The interviews took place from April – May 2022. Analyses All interviews were transcribed in Microsoft Word, after which they were analyzed in NVivo. During the analysis, the interviews were, first of all, openly coded through which patterns could be identified. Additionally, the transcripts were coded according to the relevant (sub)themes from the theoretical framework, namely healthy lifestyle, presence of peers, neighborhood, media education, and leisure. This in an attempt to make associations, both within and between different interviews. After both forms of coding, the codes were refined by categorizing codes into overarching categories, the corresponding code tree can be found in Appendix 9.

Reliability and validity There are many aspects that have been taken into account in order to guarantee the reliability and validity of the results of the interviews. This is important as it ensures that our measures are consistent and accurate. First of all, in advance of the interviews, the researcher made sure to establish the hypotheses, research questions, and a topic list with the main interview

63

questions, in order to have a level of consistency throughout the interviews. This was done to increase the reliability of the interviews and to limit the potential subjectivity of the researcher. Furthermore, the questions are largely inspired by earlier published research, namely from the Kindmonitor 2017, in order to ensure validity. Additionally, the aspects included in the questionnaire concerned subjects that the participants are knowledgeable on. The probability of socially desirable answers was decreased as the participants were informed about the anonymity of the research, which increases the validity of this study. Lastly, to further increase the validity, experts’ opinions were included which ensured a better overall view of the situation in Alblasserdam. Ethical considerations Included in the e-mail asking to participate in this research is the information letter, which entails detailed information about potential participation in the form of an interview. In advance of the interviews, the information letter was again discussed after which the informed consent was discussed and signed. All participants agreed to participate in the interviews and the parents and the third and fourth experts also agreed to the recording of the interviews. The recordings were made using secured recording equipment borrowed from Utrecht University.

The recordings were, furthermore, anonymous as no personal information was mentioned during the interviews. After the interviews were transcribed, the recordings were deleted. The transcripts, furthermore, also did not include any personal information. All interviews were voluntary, participants could stop at any moment without having to explain themselves.

Results We will now turn to the results of this study. This will again be presented in two parts, first the quantitative results regarding the analysis of the dataset will be discussed. Afterward, the results of the interviews with the parents and experts will be presented.

Quantitative results In order to answer the hypotheses and thereby the research questions, multiple statistical analyses were run. As one of the aspects of a healthy lifestyle, having a set bedtime, is a binary variable, the analyses including this variable will be logistic regressions. The regression analyses for the consumption of water and sweet drinks will be linear regressions. Therefore, there are three statistical tests run per aspect of the social environment, one for

64

each aspect of lifestyle. To prevent potential confounding effects the logistic and linear analyses are run with the control variables age, education of parents, gender, and parenting difficulty. The corresponding tables regarding the results of these analyses can be found in Appendix 4 to 7. Lastly, to test whether the found effects of these analyses hold when the other aspects of the social environment are included, three regression analyses are run including all aspects of the social environment and the control variables. This entails one logistic regression for bedtime, one linear regression for sweet drinks, and another linear regression for water consumption. The results of these analyses can be found in Table 4, and will be the guidelines for the discussion of the results. Compared to the individual analyses, only one significant association disappears when the other aspects of the social environments are added. This regards the relation between reading and bedtime, which will be discussed later on. State of the social environment and healthy lifestyles To investigate the current state of the social environments and healthy lifestyles of children, descriptive statistics were run, see Table 3.

Regarding the current state of healthy lifestyles of children, the data shows that 92% of children have a set bedtime, which indicates relatively consistent and thus healthy sleeping patterns. Furthermore, on average children score 4.71 out of 8 on how much water they consume. Regarding sweet drinks, children on average score 4.4 out of 8 on how few sweet drinks they consume. Therefore, although children do drink water, the drinking habits of children can certainly be healthier. Regarding the social environments of children aged 4-12, Table 3 also provides multiple insights. Starting off with leisure activities, 79% of children aged 4-12 play video games, and 21% use social media. Regarding how much children read for pleasure, children on average score a 2.34 out of 3, entailing that on average children read at least a few times a week, which is relatively positive. Regarding media education, 67% of the parents of children aged 4-12 have rules regarding screen time. Furthermore, parents on average score a 3.03 out of 4 on how heavily they supervise their child’s online behavior. Showing that media education is certainly present, although it could be more. Concerning neighborhoods, the neighborhoods are on average marked as 3.13 out of 4 regarding how child-friendly it is, showing that the average neighborhood is considered child-friendly.

Lastly, for the presence of peers, the average score is 2.49 out of 3, regarding how high the presence of peers is in the

65

environment of a child, entailing that each child has a moderate or high degree of peers in its environment. Table 3. Descriptive Statistics N

Minimum

Maximum

Mean

Std. Deviation

Healthy lifestyle Bedtime

3853

0.00

1.00

.92

.26

Water

3853

1.00

8.00

4.71

1.91

Sweet drinks

3853

1.00

8.00

4.40

1.65

Gaming

3853

0.00

1.00

0.21

.41

Reading

3853

1.00

3.00

2.34

.63

Social media

3853

0.00

1.00

0.79

.41

Rules

3853

0.00

1.00

0.67

.47

Supervision

3853

1.00

4.00

3.03

.58

Neighborhood

3853

1.00

4.00

3.13

.57

Presence of peers

3853

1.00

3.00

2.49

.72

Age

3853

4.00

11.00

8.08

-

Education parent

3853

1.00

3.00

2.00

-

Girl

3853

0.00

1.00

0.49

-

Parenting difficulty

3853

1.00

5.00

2.52

-

Leisure

Media education

Healthy lifestyle and presence of peers The results regarding the first hypothesis, which postulates that the presence of peers leads to less healthy drinking habits, are displayed in Table 4. First of all, Model 2 of Table 4 shows that the water consumption of a child is not significantly related to the presence of peers in the environment (B = .049, t(3852) = 1.094, p = .274). Model 3 shows that the presence of peers is significantly related to the consumption of sweet drinks for a child (B = -.079, t(3852) = 2.015, p <0.05). Meaning that a higher presence of peers leads to higher consumption of sweet drinks and thus less healthy drinking habits. Therefore, these analyses show that the presence of peers relates to the consumption of sweet drinks but it does not relate to water

66

consumption. Therefore, the hypothesis regarding the negative impact of peers on drinking habits is only partially supported by the results of this study, only concerning the consumption of sweet drinks. Healthy lifestyle and neighborhood The second set of hypotheses, entailing that a child-friendly neighborhood leads to more healthy drinking habits and to more healthy sleeping habits was tested by logistic as well as linear regression analyses. The outcomes of these analyses are, again, displayed in Table 4. Model 1 shows that a child-friendly neighborhood does significantly relate to the bedtime of children (B(exp) = 1.241, p <0.05), meaning that children that live in a more child-friendly neighborhood have a 1.249 higher chance of having a set bedtime. Model 2 of Table 4 shows that a child-friendly neighborhood does not significantly relate to changes in the consumption of water in children (B = -.031, t(3852) = -.659, p = .510). When looking at the relation between the neighborhood and the consumption of sweet drinks, see Model 3, this relation is also not significant (B = -.020, t(3852) = -.485, p = .627). Entailing that again, a more childfriendly neighborhood does not correlate with healthy drinking habits. Returning to the hypotheses, for hypothesis 2a there is no effect of the neighborhood found on the consumption of water and sweet drinks.

Therefore, this hypothesis cannot be confirmed. As there is a significant positive effect found of the neighborhood on bedtime, hypothesis 2b is supported by our results. Healthy lifestyle and media education Concerning the third set of hypotheses, that state that media education leads to more healthy drinking habits and more healthy sleeping habits, these were also tested by logistic and linear regression analyses. See Table 4 for the results. As displayed in Model 1, rules about media use significantly and positively relate to the bedtime of children (B(exp) = 2.488, p <.001), meaning that children that have rules about media education have a 2.497 higher chance of a set bedtime. Furthermore, parental supervision also positively and significantly relates to the bedtime of children (B(exp) = 1.371, p <.01), entailing that this aspect relates to a 1.436 times higher chance of a set bedtime for children. Therefore, both aspects of media education, lead to more healthy sleeping habits, which supports hypothesis 3a. Model 2 of Table 4 shows the results regarding the consumption of water. It shows that both rules about social media and supervision significantly lead to healthier drinking

67

habits (B = .256, t(3852) = 3.904, p <.001) and (B = .428, t(3852) = 7.766, p <.001), as they both have a positive relation with the consumption of water. Lastly, concerning the consumption of sweet drinks and media education, the outcomes of the analyses are displayed in Model 3. It shows that rules about media have a positive and very significant effect on the consumption of fewer sweet drinks (B = .286, t(3852) = 5.011, p <.001). The supervision of parents also very significantly and positively relates to consumption of sweet drinks (B = .391, t(3852) = 8.160, p <.001). Therefore, as the analyses show that both aspects of media education lead to healthier drinking habits, hypothesis 3b is supported by the results. Healthy lifestyle and leisure Lastly, the fourth set of hypotheses postulates that a healthier way of spending leisure relates to healthier sleeping- and drinking habits. These were once again tested by logistic and linear regression analyses, and the results can be found in Table 4. The logistic regression for bedtime did not show any significant effects for social media (B(exp) = 0.989, p = .947) and gaming (B(exp) = 1.228, p = .220). The analyses of Table 4 did also not find a significant relation between reading and bedtime (B(exp) = 1.153, p = .165).

However, when the relation between leisure activities and bedtime is individually tested in a linear regression, see Appendix 7, a significant effect does show (B = .231, t(3852) = 4.633, p <.001). This entails that reading relates to a 1.326 higher chance of a set bedtime. That this effect disappears when all other aspects of the social environment are added suggests that it might be that reading only has an effect on bedtime when the other aspects of the social environment are not present. Therefore, only the results regarding reading, to a certain degree, support hypothesis 4a, as the results show that under certain circumstances, reading does affect sleeping habits of children. The results regarding gaming and social media do not support hypothesis 4a. Regarding the relation between leisure and water consumption, see Model 2, the results show no significant relation between gaming and water consumption (B = .007, t(3852) = .096, p = .924) and between social media and water consumption (B = -.170. t(3852) = -2.022, p = .135). There is however a positive relation between reading and water consumption (B = .159, t(3852) = 3.176, p <.01), meaning that more reading associates with a higher water consumption.

Lastly, for sweet drinks, see Model 3, the results show a positive relation with gaming (B = .348, t(3852) = 5.175, p <.001), meaning that playing video games is related with a higher consumption of sweet drinks. Concerning the effect of both reading and social media on the consumption of sweet drinks, no correlation was found (B = -.006,

68

t(3852) = -.146, p = .884) and (B = -.117, t(3852) = -1.598, p = .110). Regarding the relation between healthy ways of spending leisure and healthy drinking habits, there are mixed results, as not all expected associations were found. However, the effects that are found point to healthy leisure activities leading to healthy drinking habits, therefore, there is some support found for hypothesis 4b. As displayed in Table 4, the chosen control variables do significantly relate to the dependent variable in most analyses. This suggests that age, gender, parental education, and parenting difficulty do in fact relate to the lifestyles of children.

69

Table 4. Regressions with healthy lifestyle as dependent variable and social environment and control variables as independent variables Model 1: Bedtime 1 B 95% CI for Odds ratio Lower Odds ratio Upper .032 [.090] .865 1.032 1.231 .216** [.094] 1.032 1.241 1.493

B .049 -.031

.912*** [.127] .315** [.111]

1.939 1.103

2.488 1.371

3.192 1.704

.256*** .428***

Gaming Reading Social media Age Education parents Girl Parenting difficulty Constant

-.011 [.167] .142 [.103] .205 [.168] -.016 [.029] -.028 [.090] .099 [.124] .023 [.084] .020 [.682]

.713 .943 .884 .921 .815 .858 .867

.989 1.153 1.228 .984 .973 1.104 1.023

1.372 1.409 1.705 1.051 1.160 1.419 1.208

.007 .159** -.170 .035* .345*** .180** -.164*** 2.322***

R2

.052

.051

N

3853

3853

Variable Presence of peers Neighborhood Media education Rules Supervision Leisure

***p <.001, ** p <.01, *p <.05, two-tailed 1

The analyses including the variable bedtime are logistic regressions.

Model 2: Water

Model 3: Sweet drinks

s.e. .045 .047

B -.079* -.020

s.e. .039 .041

.066 .055

.286*** .391***

.057 .048

.077 .050 .084 .016 .044 .062 .041 .341

.348*** -.006 -.117 .046** .143*** .143** -.127*** 2.904***

.067 .043 .073 .014 .038 .054 .035 .297

.045 3853

Qualitative results Interviews with parents In addition to the quantitative results, the analyses of the interviews with parents about the healthy lifestyle and social environments of their children can provide further insights into these topics. For an overview of the outcomes of the parental interviews, see Appendix 8. Healthy lifestyles With regards to the healthy lifestyles of their children, the outcomes of the interviews seem to be fairly positive. When asked to rate the lifestyles of their children, the parents of six children gave their child’s lifestyle a score of 8 out of 10 and one child got a 6 out of 10. These high rates were confirmed by the answers that the parents provided to the more indepth questions about the different aspects of a healthy lifestyle. Most parents seemed invested in making sure their children eat and drink healthy, sleep well, and exercise regularly. For example, one participant stated: ‘After school, I always freshly squeeze orange juice, so after school, they usually drink fresh orange juice.’ She also elaborated that she spends large amounts of money to buy fresh and healthy ingredients. What the parents also largely indicate is that healthy lifestyle choices, such as a healthy diet, are something children get used to.

This leads to children finding it normal to portray healthy habits, such as asking for a glass of water at a party instead of soda. One parent found his children even started eating vegetables because they felt like it: ‘Sometimes they have already eaten their 200 grams of vegetables before I’m awake.’. Another mother said that her son now asks to go to bed when he is tired, as he has grown to have a healthy relationship with sleeping. Most parents indicated that it might take some effort from the parents to get their child to have a healthy lifestyle, but that children easily take over the healthy habits. Social environment With regards to the social environment, the scores that parents gave their children regarding their social environment ranged between a 6 and a 9 out of 10, with an average of 7.7. Regarding the presence of peers in their children’s environment, all parents agreed that there are plenty of peers in the lives of their children. However, there were cases where children found it difficult to make friends. One participant stated about her son on the autistic spectrum: ‘Making friends is very complicated for him, but he also does not really feel the need to make friends.’ Regarding the neighborhood, all parents found their neighborhood safe

71

and child-friendly to some degree, with some neighborhoods being safer than others. The biggest cause of unsafety that was mentioned several times was the traffic. One participant recalls: ‘This road is busy. The children play on the sidewalks, but the buses sometimes drive so fast, it is really dangerous.’ One aspect that all parents seem to view similarly is media education. All parents that were interviewed were involved in the media use of their child, and most had very clear and strict rules about technology use. In the cases where children had a mobile phone, the parents were closely monitoring it. One participant said: ‘We have connected our accounts to his account, so we can see exactly what he is watching. We are definitely aware of his online behavior.’. Concerning the way they are spending their leisure, six out of seven children included in this study seem to be doing well. Entailing that they barely play video games or use social media, but spent their time reading, playing outside, or going on day trips with their parents. However, for the seventh child, the situation is less healthy, partially as there are no fitting sports clubs for him because of his autism. His mother said: ‘A few years ago we had ‘JamaarIk’, which was a sports club for children with a stimulus processing disorder, where those children could do their thing. He really enjoyed going.

Unfortunately, it stopped, due to lack of volunteers.’ Lastly, in the interviews, the parents were asked whether they feel that some aspects of the social environment are more important than others. Two parents thought that there is no one aspect more important than the other, but ‘What is important, is that there is balance.’. Furthermore, two parents thought friends are the most important, and one parent argued that the neighborhood is the most important. Healthy lifestyle and social environment In three interviews, the question arose of whether there is a relation between a healthy lifestyle and the social environments of children. All three parents think that this is the case, and they all argued that they thought it was the environment that influenced the lifestyle. One mother said in this regard: ‘I think that your environment always influences the choices that you make. So there is definitely a relationship’. Role of the municipality All respondents were asked about the role of the municipality in improving the social environments of children. Two respondents indicated that they were very content with the involvement of the municipality, one said ‘The municipality is already involved in the aspects

72

that are not going so well. I don’t think you can expect more from a government.’. The other three respondents all had one specific aspect in mind that they thought could be improved by the municipality of Alblasserdam. Namely, a sports club for children with disorders that make it hard to join regular sports clubs, fewer cars in residential areas, and the refurbishing of a local skate area for children. As the municipality of Alblasserdam is currently working hard to achieve a ‘Fit and green’ village for all residents, the interviews specifically included the question of whether participants felt like there is enough greenery in their neighborhoods. Of the five participants, four indicated that there is plenty of greenery in their neighborhoods. One even said: ‘Yes, there is enough greenery. To be honest, I wish they would trade some greenery for parking spaces, as we could use more of those.’. One participant, living very close to the center, said that some more greenery would be welcome, in the form of a grass field to play soccer and such. In three interviews, the parents were also asked whether the municipality has a role in promoting healthy lifestyles for children. Two parents felt like that is a difficult thing to do as lifestyles are something that is largely practiced within people’s homes.

The other respondent, however, said: ‘I don’t see anything wrong with pushing towards a healthy lifestyle in a friendly manner.’. Differences and similarities When comparing the outcomes of the five different interviews, a few things become apparent. First of all, the children that have trouble in one domain, tend to also have trouble in other domains. For example, a child that had trouble eating also had trouble with a healthy way of spending leisure and frequently exercising. Furthermore, it turned out that the two children that had significantly more trouble with achieving a healthy lifestyle and some aspects of the social environments were both children that were on the autism spectrum, to differing degrees. These were the children that got the lowest scores on the social environment, both around a 6, while the other children scored around an 8 or 9. Therefore, the biggest differences were between children with- and without special needs. Suggesting that children with special needs might have more difficulty with healthy habits and aspects of the social environment, and children without seem to do quite well regarding these aspects in Alblasserdam. For the children that do not have special needs, most outcomes were quite similar regarding healthy lifestyles, peers, media education, and leisure.

However, there were some differences in how their parents perceived their neighborhood.

Expert interviews Furthermore, the experts’ opinions can also provide insights from a different perspective compared to the parents which can help to answer the research questions. The outcomes from the interviews with the experts in the field will now be discussed. Healthy lifestyles Concerning healthy lifestyles, the outcomes of the expert interviews confirmed that children that are already suffering, for example, children living in poverty, also have trouble achieving healthy lifestyles. The first expert said in this regard: ‘It turns out that there are approximately 300 youths in Alblasserdam that are living on or under the poverty line, who are not playing sports in groups in an organized manner.’. The second expert mentioned another group that is having trouble, namely overweight children, which tend to attend fewer sport-related activities. However, the first two experts argue that the average child from Alblasserdam is doing fine in terms of a healthy lifestyle. The second expert also mentioned that quite a lot of children already drink lots of water, although there are still children that drink soft drinks. This was confirmed by the third interview, as the third and fourth experts also do not see anything remarkable regarding the lifestyles of children on average.

However, they doubt if they can provide a realistic picture as they have limited relevant knowledge on this topic. Social environments Regarding the aspects of the social environment, experts indicate more problems than the parents did. The first expert mentioned three problems, one being that due to the Covid pandemic children struggle with how to behave in groups. Furthermore, she notices that fewer children play outside, and thirdly there is a lot of traffic in Alblasserdam. The second expert also sees an increase in traffic, which prohibits children from playing outside for example as parents find it too dangerous for them to do so. Furthermore, he does not see aspects of the social environments that are structurally bad, but he does see that for children from certain disadvantaged groups there is a lot to gain. For example, for the earlier mentioned overweight children. The third and fourth experts have different expertise, as they only encounter children that need help, which comes with different experiences. They did confirm that children on the autism spectrum tend to struggle more with social contacts. Regarding leisure and neighborhood, they did not receive signals of trouble from parents. What they do get back

74

from parents is troubles with media education and regulation, which accompanies children spending lots of time gaming or on social media. Greenery As Alblasserdam is very concerned with being ‘Fit and green’, the experts were also asked about greenery in Alblasserdam. In this regard, the first two experts agreed that there needs to be significantly more greenery in Alblasserdam. They also both acknowledged that this is hard to change as there is not much space left. However, the second expert provided a solution for this, as he said: ‘We should let nature do its thing. This enables children to create their own playgrounds. It is more fun and more challenging for children to play in trees and shrubs. We should just rearrange the greenery in the neighborhoods. That is what the children want.’ The third and fourth experts did not have any specific comments about the greenery in Alblasserdam. Role of the municipality Lastly, the experts all had something to say about the role of the municipality. The first expert indicated that the target group that the municipality should focus on is children living in poverty, especially concerning their opportunities of playing sports in an organized manner. She also mentioned that we have a responsibility for children that cannot keep up with the rest because they are autistic or do not get enough support from home.

The second expert felt that there are already a lot of activities organized for children, but argued that more organized activities after school could be beneficial for the children. For which, however, more manpower is needed. The third and fourth experts also had some ideas about the role of the municipality. Regarding what touches on the topics of this study, they mentioned the demand for activities for children with behavioral limitations. As a normal sports club does not fit them, they experience trouble finding appropriate leisure activities.

Conclusion This study aimed to answer the following three research questions: ‘To what extent are the social environments and lifestyles of children aged 4-12 in the Netherlands healthy?’, ‘To what extent does the social environment of children aged 4-12 in the Netherlands influence their lifestyles?’, and ‘Which aspects of the social environment are most important for achieving a healthy lifestyle for children aged 4-12?’. To answer these questions, four sets of hypotheses were formulated and tested using a quantitative dataset. Furthermore, qualitative

75

interviews were conducted to further deepen the topics and provide additional insights and hereby strengthen the results. Regarding hypothesis 1, which postulates that the presence of peers correlates with less healthy drinking habits, the results partially support this hypothesis, as it was found that the presence of peers relates to increased consumption of sweet drinks. Peers do not influence water consumption. Therefore, only the results regarding sweet drinks are in line with social facilitation theory. The second set of hypotheses proclaims that a child-friendly neighborhood leads to healthier drinking habits and a set bedtime. The effect of the neighborhood was only found for bedtime, therefore only hypothesis 2b was confirmed, thereby also partially supporting ecological systems theory. Concerning the third set of hypotheses, the expectation was that the two forms of media education relate to healthy drinking- and sleeping patterns. The results show that rules and parental supervision relate to a higher chance of a set bedtime and also to healthier drinking patterns, meaning that there is support for hypotheses 3a and 3b. Lastly, the fourth set of hypotheses proclaims that healthier ways of spending leisure relate to healthier drinking patterns and sleeping patterns. The results indicate that reading, on its own, does associate with a higher chance of a set bedtime.