Food Indicators and Their Relationship with 10 to 12 Year-olds’ Subjective Well-Being

This study aims to test subjective indicators designed to analyze children’s predisposition towards food consumption, to assess their subjective well-being, and to explore the relationship between subjective well-being, predisposition towards food consumption and satisfaction with food. Gender differences are analyzed. It was conducted on 371 children aged 10 to 12 by means of a self-administered questionnaire. Results show that children’s motivations in relation to taste and health are relevant subjective indicators of their predisposition towards food consumption. They demonstrate a high subjective well-being, measured using Cummins and Lau’s adapted version of the Personal Well-Being Index–School Children (PWI-SC) (2005), overall life satisfaction (OLS) and satisfaction with various life domains (friends, family, sports, food and body). In order to analyze the relationship between the three aforementioned constructs, regression models were conducted. The interest children have in food, the importance they give to different reasons for eating, scores from the PWI-SC, OLS and satisfaction with various life domains were regressed on satisfaction with food. It was observed that OLS, health motivations, satisfaction with health from the PWI-SC and satisfaction with doing things away from home (also from the PWI-SC), contribute to explaining satisfaction with food. The results obtained suggest that the different indicators for children’s predisposition towards food consumption explored here and subjective well-being are relevant determinants of satisfaction with food. They also appear to reinforce the importance of exploring food satisfaction in any study aimed at analyzing the well-being of the 10 to 12 year-old population.


Introduction
The scientific community first began to take an interest in the relationship between food and health in the 1950s, a decade that saw the discovery of the link between certain eating habits and a decreased risk of arteriosclerosis. Besides being a nutritional process, eating has significant connotations relating to pleasure, and indeed serves as an element of social, cultural and religious identification all at once. It also promotes and intervenes in the establishment of interpersonal and emotional relationships that ultimately shape individuals' eating habits, thereby having a bearing on their long-term health (Departament de Salut de la Generalitat de Catalunya 2005a).
Food is also one of the key exogenous factors that affect children's proper growth and development (Aranceta et al. 2004). This highlights the importance of addressing the issue of food from an early age. Children grow and change quickly during their school years, with the period between the ages of 6 and 12 representing one of the most important for the development of their cognitive, physical and social skills. Although children begin to develop causal reasoning during this stage, the criteria that affect their choice of food remain rather immediate. That said, children at this age do start to take an interest in the links between food and health (Contento 2007). Furthermore, they display an increasing amount of independence, leading to the expansion of their relationships and a heightened capacity to make decisions (Aranceta 1995), some of which may revolve around aspects relating to food.
With regard to gender, although not considered a determinant variable like availability of food, upbringing, taste, or friends' influence, some studies highlight that it does have some impact on food preferences. In this respect, girls seem to be more likely to select healthier foods than boys in order to control weight, whereas boys' greater energy requirements and desire to become stronger could provide some explanation for their preference for more energy-dense food groups (Pich et al. 2010;Neumark et al. 1999;Cooke and Wardle 2005). Further research is needed, however, in order to take a more in-depth look at boys' and girls' motivations when it comes to eating.
Research conducted on child nutrition in recent years (Serra et al. 2002;Contento et al. 2007, among others) has demonstrated a need to deal with this issue, in part due to the dietary imbalance and health issues in evidence among the child population. Studies have also been carried out to analyze eating habits and satisfaction in relation to food received in specific contexts (in schools, hospitals, etc.). However, these largely focused on evaluating the service received and the nutritional quality of the menus served (Watters et al. 2003;Mavrommatis et al. 2011) and fail to take into account the possible role of food in children's subjective well-being and how it relates to their opinions, perceptions and evaluations.
Two exceptions to the above are the study by Proctor et al. (2009), which reports lower levels of life satisfaction in children with obesity and food disorders, and the study by Valois et al. (2003), who found that lower levels of subjective well-being in children are associated with poorer diet habits.
This article argues that in order to understand individuals' eating habits and their predisposition towards food consumption one must go beyond strictly dietary and nutritional indicators for different foods and consider a range of possible psychosocial indicators. These indicators concern the immediate context in which food is consumed (eating habits in the family home and environmental aspects of food, relationships with dining companions), a wider social context (the importance given to food, beliefs about particular foods, the application of regulations on dietary matters) and a more individual dimension: each person's judgments with respect to different aspects of life, including subjective well-being and personal food preferences.
This study will focus on the last of the aforementioned areas (the more individual dimension), leaving the exploration of other aspects for future research. A key concept we aim to analyze in this research is children's satisfaction with food. By satisfaction with food we are referring to the evaluation of those aspects that can help children enjoy their food and lead them to consider it a pleasurable and important element for their well-being and quality of life.
Subjective well-being is generally held to be a multidimensional assessment reflecting the combination of a cognitive process (satisfaction with life both as a whole and in relation to various specific domains: health, school, family, friends and relationships with other people, among others) with two affective processes (presence of positive affect and absence of negative affect) (Eid and Diener 2004). This study focuses on the more cognitive side of well-being, leaving an analysis of the more affective aspects for future research.
Major contributions to the study of subjective well-being have been made by Cummins (1998Cummins ( , 2003, who argues for the inclusion of various domains when it comes to analyzing the judgments that people make about their lives when questioned about their personal well-being. He also contends that measuring the variable of overall life satisfaction involves a homeostatic control mechanism; similar to the way in which blood pressure is measured, but psychological. This means that the range of variation in personal well-being scores recorded among people from the same country and between different countries tends to be limited, pointing to the existence of normative values for well-being. Cummins et al. (2003) have developed a tool to investigate subjective well-being, which they called the Personal Well-being Index (PWI). It assesses responses regarding level of satisfaction in the following domains: standard of living, personal health, life achievement, personal relationships, personal safety, community-connectedness, future security, and spirituality and religion. It also includes satisfaction with life as a whole. Various versions of this index have been designed to meet the specific needs of different age groups: adults (PWI-A) (Cummins and Lau 2006), school children (PWI-SC) (Cummins and Lau 2005) and pre-school children (PWI-PS) (Cummins and Lau 2004).
In relation to gender, the results of different studies carried out on Australian and Spanish adolescents and young adults González et al. 2012), some using an adaptation of the PWI-A and others the PWI-SC, are not conclusive. In one study, females' scores are higher than males', and in two others this trend is reversed. No studies have explored the personal well-being of children under 12, however. This is therefore a pending issue that this paper will attempt to cover.
In light of the above, the aims of this study are: (1) to test subjective indicators designed to analyze the predisposition of children aged 10 to 12 towards food consumption, (2) to assess their subjective well-being, (3) to explore the relationship between these two phenomena and satisfaction with food, and (4) to explore the different patterns displayed by participants in relation to these phenomena according to gender.

Method
The data analyzed in this study were collected within the context of a larger research study on school food carried out in 2008 in the region of Osona (Catalonia, north-east Spain), which was funded by the Government of Catalonia's Department of Health and the University of Vic.
Of the 79 schools in the region, 26 (32.9 %) expressed an interest in participating. The target population of this study was recruited from 18 of these schools. Eight schools were excluded because they did not include primary-age students; another school with a special education programme serving intellectually-impaired children was also excluded.
In light of the fact that the data analyzed in this article are extracted from a school dining-room study, it is clear that participating children are all regular users of schoolprovided food services. In Spain, these services are accessible to all families, they are organized by the Parents' Associations and using them does not reflect a particular socio-economic status.

Sample
Data are taken from a total of 371 children in the region of Osona in their last 2 years of primary school (students were between 10 and 12 years of age). This sample comprises 25.82 % of all children enrolled in the last 2 years of primary school in the 18 participating schools. This age group was selected for two reasons: the difficulty of assessing the well-being of children under 10, and the fact that other studies have reflected a growing interest in dietary issues for children in this cohort.
The majority of participating schools were state schools (77.8 %, mirroring the total distribution of state schools in the region), while the remainder of participating entities were publicly-funded private schools from the region (22.2 %). Half of these schools (50 %) were located in semi-rural areas (with populations between 2,000 and 5,000 residents), while 33.3 % were in urban areas (with more than 5,000 residents) and 16.6 % of schools were in rural areas (with less than 2,000 residents).

Instruments
Data were collected using a questionnaire designed specifically for this study and containing the scales and items described below. In order to test objective 1, the following questions were included in the questionnaire: I. One item regarding children's global interest in food scored on an 11-point scale, with 0 representing not interested at all and 10 representing extremely interested. II. A group of items evaluating the importance attributed to various eating-related topics: "this food is good for my health," "this food tastes good," "this food is easy to eat," "this food is well-known," "someone recommended this food to me," "my friends also eat this food," "my parents want me to eat this", "I have seen the food on TV", "I have enough time to eat", "my body needs this food," and, finally, "I feel good after eating this food" (scored from 0 0 not at all to 10 0 very much).
Objective 2 has been developed through the application of: I. The adapted version of the Personal Well-being Index-School Children (PWI-SC; Cummins and Lau 2005) was used to indicate the subjective well-being of children. The instrument was measured using an 11-point scale, 0 meaning Completely dissatisfied and 10 Completely satisfied. The instrument begins each question by asking "Right now, how satisfied are you…" and concludes with the following endings: "about the things you have?", "with your health?", "with the things you want to be good at?", "about getting on with the people you know?", "about how safe you feel?", "about doing things away from your home", and "about what may happen to you later in life?" The exact questions used in our study were not a literal Catalan translation of the questions posed by the aforementioned authors; a pilot study demonstrated that students who participated in the study struggled with the cognitive difficulty of the exact translations of these questions, and consequently the questions were reformulated for the purpose of this questionnaire. II. A single item scale on overall life satisfaction (OLS) (Campbell et al. 1976). The item was measured using an 11-point scale, 0 meaning Completely dissatisfied and 10 Completely satisfied. III. Four items assessed satisfaction with additional specific life domains. The questionnaire asked: "Right now, how satisfied are you …" "with your friends?", "with your family?", "with the sports you play?", "about your body?"These items were measured using an 11-point scale, 0 meaning Completely dissatisfied and 10 Completely satisfied.
With the aim of testing objective 3, an additional item on satisfaction with food was included in the questionnaire, using the same 11-point scale and the same question.
The pilot study was conducted on a group of children displaying similar characteristics to the final sample (i.e. with regard to age, gender, school year, and school location) in order to test the degree of comprehension of the draft questionnaire. Consequently, the language in the tool was changed; the most important of these modifications entailed substituting the word feliç (happy) for the word content (content) in light of the fact that the latter word was more comprehensible than the former for children in this particular sociocultural context.

Procedure
After drafting the final version of the data collection tool, the directors of participating schools were contacted in order to explain the goals and progress of the study, underscore the importance of their collaboration, and request their consent. Regular ethical procedures for research involving children in Spain accept the passive consent of parents when Parents' Associations have been informed and the questionnaires are anonymous and do not include personally sensitive items, which is the case of the questionnaire used here. Children were asked for their co-operation after being informed of the aims of the research and the fact that they were free to stop answering the questionnaire at any time.
The questionnaire was self-administered by students in their normal classroom during the school hours. During each administration of the questionnaire, a researcher and one of the student's teachers were always on hand in order to explain the study and answer any potential questions which might arise.

Structure of the Analysis
The results of this paper have been divided into three different sections. The first section on "interest in food and eating-related decision-making" responds to the first objective of testing indicators designed to analyze 10 to 12 year olds' predisposition towards food consumption. Some studies have been conducted to analyze reasons for children's preferences on different kinds of food. However, they do not consider children's interest in their exploration of food-related behaviors. We believe that interest might play an important role in decisions about food as it may be connected to different attitudes towards food consumption. This is why, contrary to other studies we have found in the literature, this variable has been included in the paper. Means and standard deviations of the item regarding children's global interest in food and a group of items evaluating the importance attributed to various eating-related topics are therefore calculated. A principal component analysis (PCA) of these last items was performed in order to extract different components which could be used for subsequent analysis. Gender differences are explored through the Student's T-test. Finally, Spearman correlations between interest in food and the components emerging from the PCA were also performed.
The second section, related to "the subjective well-being of children", deals with calculating the means and standard deviations of the adapted PWI-SC domains, the OLS and satisfaction with additional specific life domains, including satisfaction with food. This section responds to the second objective of assessing children's subjective well-being. Again, gender differences are explored through the Student's T-test.
The third section is related to the third objective (studying potential links between food and subjective well-being). With this in mind, Spearman correlations were calculated between food-related indicators and subjective well-being indicators. Finally, different regression models were performed using the stepwise method taking satisfaction with food as the dependent variable and food-related and subjective wellbeing indicators as predictors. Models for boys and girls were also calculated separately in order to respond to objective 4 (exploring the different patterns displayed by participants in relation to the explored phenomena according to gender). As explained in the introduction, satisfaction with food is generally understood in the literature as a variable that allows the quality of a food-providing service to be evaluated, but not as something that can be both related to children's predisposition towards food consumption and their subjective well-being. Results are expected to broaden researchers' perspectives on the explored issues and they cannot therefore be easily compared to other studies carried out to date.

Interest in Food and Eating-Related Decision-Making
The mean response to the question about student interest in food was 7.88 (SD02.43) on a scale of 0 to 10. 51.1 % of participants indicated that food is very interesting to them (they responded with either a 9 or a 10), while 31.1 % expressed solid interest (responses between 6 and 8) and 17.8 % said the topic was of little interest to them (responses ≤5). There were no significant differences between these three groups in terms of gender.
The degree of importance children gave to different indicators which may influence eating are shown in Fig. 1. The main motivation when it comes to eating is the taste of food ("it tastes good"), followed by "it is good for my health." The indicators which received the lowest scores were those most closely connected to external influences ("someone recommended this food to me," "my friends also eat this food," and "I have seen the food on TV"). Girls scored "feeling good after eating" higher than boys (t (348.67)02.70, p00.007 (two-tailed), η00.136), while boys gave greater importance to things they saw in advertisements (t (367)0−2.17, p00.031 (two-tailed), η00.112).
The major indicators described in Table 1 were analyzed by means of a Principal Component Analysis (PCA) using Varimax rotation, yielding 3 components explaining 55.80 % of all variance. These components were as follows (see Table 1): -Component 1 could be labeled "social influence" and explained 23.55 % of the total variance. It included the following items: "I have seen this food on TV," "my friends also eat this food," "someone recommended this food to me," "this food is well-known," and "I have enough time to eat this food." -Component 2 was mainly related to health and explained 21.60 % of the total variance. It included the following items: "my body needs this food," "my parents want me to eat this food," "this food is good for my health," and "this food makes me feel comfortable." -Component 3, which we have labeled "hedonic," was composed of one item related to the taste associated with foods ("this food tastes good"). This component explained 10.66 % of total variance. The indicator corresponding to "this food is easy to eat" was removed from the analysis as it loaded equally in two of the three aforementioned components. No statistically significant differences were observed in any of the three components with regard to gender.

The Subjective Well-Being of Children
The responses study participants gave to items on the PWI-SC are shown in descending order in Table 2.
As Table 2 shows, the items of satisfaction which received the highest mean score were "getting on with the people you know" and satisfaction "with your health" (the mean was higher than 9 for both). The overall mean score on the PWI-SC on a scale of 0 to 100 was 86.54 (SD010.54).
In keeping with the recommendations of the tool's authors, the index was recalculated, excluding the 29 subjects who gave extreme responses to all questions (responses of 0 or 10), resulting in a mean score of 85.25 (SD010.17). As there were no statistically significant differences between the two means, this study continued to utilize the data from all respondents.
There were no statistically significant differences with regard to gender for the PWI-SC index, but there were for the indicator "satisfaction with what may happen to you later in life" (t (365)02.206, p00.028 (two-tailed), η00.115), girls (M08.50; SD02.01) being the ones with the higher mean score compared to boys (M08.03; SD02.11).
The OLS score (M09.09, SD01.54) and satisfaction scores for other complementary life domains are high (means range from 8.62 to 9.71). The highest mean corresponded to satisfaction "with your family," while the lowest corresponded to satisfaction "with food" (see Table 3). There were no statistically significant differences with regard to gender for OLS or any of the items described in Table 3.

Food and Well-Being
The correlation between interest in food and the PWI-SC was significant yet moderate (ρ00.38, p<0.001). Correlations were similar for both genders (girls: ρ00.38, p<0.001; boys: ρ00.35, p<0.001). Correlations were lowest between interest in food and OLS and other indicators of satisfaction which are not included on the PWI-SC, presented in Table 4.
The three principal components encapsulating the degree of importance which participants attributed to different components when it comes to eating (health, hedonism, and social influence) reflected statistically significant correlations with the PWI-SC, although these correlations are not particularly strong (health component: ρ00.34, p<0.001; hedonic component: ρ00.27, p<0.001; social influence component: ρ00.19, p<0.001).
The strongest correlation between the 3 components (health, hedonism, and social influence) and OLS were found in the importance given to questions of health. Only satisfaction "with your body" correlates with the three components (see Table 5).
The highest correlation for both boys and girls was found between the health component and the indicator entitled satisfaction "with your body" (ρ00.24, p00.002 and ρ00.24, p00.001, respectively). There were statistically significant correlations With your family 0.14, p00.008 With the sports you play 0.18, p00.001 With your body 0.23* OLS 0.28* between the hedonic component and satisfaction "with your friends" (ρ00.22, p00.006), and "with your family" (ρ00.16, p00.048), again only for girls. After conducting the analyses described above, a series of regression models were run using the stepwise method in order to gain additional insight into the relationships between indicators related to food and personal well-being, using "satisfaction with food" (SATFOOD) as the link between these two constructs.
An analysis of SATFOOD results in terms of the three principal components classifying eating-related motivations and interest in food revealed that the model was significant (Adjusted R 2 00.25, F (2.339)056.47, p<0.001). As Table 6 shows, the health component and interest in food appear as significant predictors when it comes to explaining SATFOOD.
Running this model with responses from only boys and then only girls revealed that for girls (Adjusted R 2 00.31. F (2.157)036.37, p<0.001) the variable which best explained the SATFOOD scores was the health component, followed by the hedonic component, whereas for boys (Adjusted R 2 00.23. F (2.181)027.92, p<0.001) the variable reflecting the highest beta value was interest in food, followed by the health component.
In order to explore which subjective well-being indicators act as predictors for SATFOOD we ran additional regression models with the OLS and the PWI-SC separately.
The results show that the only two domains with a statistically significant beta value are "health" and "the things you want to be good at" in the girls model The degree to which complementary indicators of life satisfaction contribute to predicting SATFOOD values was explored through a linear regression model. The model revealed that all of the aforementioned indicators made a significant contribution, with the exception of satisfaction "with the sports you play" (Adjusted R 2 00.21. F (3.361032.45, p<0.001) (see Table 9).
The beta values suggest that the predictor to best explain SATFOOD in the model described in Table 9 is satisfaction "with your body", while the variable to least explain it is "your friends". Subsequently, the same model was run separately for both boys and girls. The only predictor for girls (Adjusted R 2 00.18. F (1.168)039.25, p<0.001) was satisfaction "with your body," By contrast, results for boys (Adjusted R 2 00.25. F (3.192)031.12, p<0.001) showed that the predictor with the highest beta value was satisfaction "with your body," followed by satisfaction "the sports you play" and "with your family".
Finally, in order to analyze which indicators are significant in the above partial models when grouped together, we calculated a last linear regression model to explain SATFOOD. It is observed that this model includes 4 predictors, OLS being the indicator with the highest contribution (Adjusted R 2 00.41. F (4,328) 058.29, p < 0.001) ( Table 10).
The indicators with the highest explanatory capacity for SATFOOD using this model, after OLS, are the health component, "satisfaction with health" from the PWI-SC and "satisfaction with doing things away from home", for both genders, "satisfaction with your body" (Adjusted R 2 00.43. F (4.153)029.41, p<0.001) in the model for girls only and OLS and satisfaction with doing things away from home and satisfaction with family (Adjusted R 2 00.45. F (4.174)037.49, p<0.001) in the model for boys only (Table 11).

Discussion
The children who took part in this study displayed a high level of interest in foodrelated issues. Although there are no data to demonstrate it categorically, we believe that this keen interest could be attributed to the effectiveness of the various strategies implemented in Spanish schools to raise awareness about and promote healthy lifestyles among the child population (namely the Strategy for Nutrition, Physical Activity and the Prevention of Obesity, Agencia Española de Seguridad Alimentaria y Nutrición 2005), and the Integral Plan for Health Promotion through Physical Activity and Healthy Eating (Departament de Salut de la Generalitat de Catalunya 2005b). In order to test this hypothesis, it would be necessary to evaluate the impact of this kind of program on children by measuring their interest in food-related issues before and after the implementation of the specific program. Motivations for the sample children's food choices coincide with the results obtained by De Moura (2007), with sensory factors such as taste being the most prevalent factor, closely followed by health-related aspects. An analysis of the predominant reasons for children's evaluation of food reveals some differences between the genders. Girls prioritize feeling good after eating, unlike boys, who attach more importance than girls to eating things that they have seen in television advertisements. Feeling good after eating has both a physical and a psychological dimension. The ingestion of healthy food may contribute to both dimensions, as it is easier to digest compared to non-healthy food and it is one of the most important means of controlling weight, something very important for girls according to the reviewed literature. Food-related advertisements which have children and adolescents as the main target are generally for fatty and sugary foods like biscuits, snacks and soft drinks. This could be a plausible explanation for boys giving more importance than girls to eating things previously seen in advertisements, in that the existent literature points to their stronger preference for this kind of food. To understand the source of these differences, however, it will be necessary in the future to apply qualitative methodology techniques such as in-depth personal interviews or focus groups. The importance of the component of "health" as an indicator of motivations influencing food choices, widely recognized among the adult population (Neumark et al. 1999), is also apparent in our results, demonstrating the extent to which the children surveyed are conscious of the link between food and health. This finding supports the claims of authors such as Contento (2007), who argues that it is during preadolescence that children start to become aware of the consequences of their actions in relation to what they eat.
As far as motivations for eating are concerned, the lowest means correspond to variables related to an external influence factor such as, for example, foods having been seen in advertisements. This contrasts with previous studies that suggest that children are particularly vulnerable to the influence of advertisements with respect to their eating habits (Mehta et al. 2010;Medeiros et al. 2008).
In line with the results obtained by De Moura (2007), the children in our sample who are most interested in food are those whose eating preferences are most influenced by health-related issues. The fact that girls of this age display a positive correlation between interest in food and a component of hedonic motivation for eating while boys do not is a phenomenon which would be worth studying in the future. The correlation between the factor of external influences and interest in food is weak and statistically insignificant.
The subjective well-being of children under 12 is an issue which has hardly been addressed in the literature owing to the difficulty involved in using concepts that are somewhat abstract in nature. We studied this notion by using an adapted version of the Personal Well-being Index (Cummins and Lau 2005) to facilitate comprehension among the sample group of Catalan children. The mean PWI-SC index scores for the children surveyed show that they have high levels of well-being, corroborating the positive opinions about their lives as a whole expressed by many children and teenagers in other research . The results obtained in our study are slightly higher and contain less variability of responses than the normative values defined by the authors for the Australian population (M075, SD012.65; Cummins and Lau 2005). This difference could be due to factors such as the size of the sample. It could also be attributed to the different sociocultural context involved. To verify this hypothesis, this study would have to be extended to include larger samples. Another hypothesis worth exploring in greater detail in the future, and one which we would particularly like to emphasize, is that normative values for the Australian population may be higher for the child population than for the adolescent and adult population.
In a previous study (González et al. 2012) based on a sample of adolescent and young adults (15-24 years old) regarding their personal well-being (using the adult version of the Personal Well-being Index by Cummins and Lau (2006)), differences were found according to the gender of the participants. Contrarily, in our study there are no significant differences between boys and girls for the whole PWI-SC, but there is in relation to one specific domain "satisfaction with what may happen to you in later life", with girls scoring higher. This same result was found in the study by  using the same index (PWI-SC), but applied to 351 students aged 12 to 20.
The life satisfaction domain to obtain the highest mean from the PWI-SC was "having good relationships with other people", which the children displayed a high awareness of, demonstrating the importance of interpersonal relationships and presumably the social support children receive from parents, classmates, teachers and close friends (Demaray and Malecki 2002;Casas 2010;Goswami 2011).
The next most important domain (according to the mean scores obtained) is health, indicating that this is a prominent life concern among the child population. One of the domains to produce a low mean was security for future. A similar trend was observed with samples of teenagers and young people aged 15 to 24 (González et al. 2012), and this result could be attributed to a heightened sense of uncertainty when thinking about the future. In light of this consideration, subsequent questionnaires might designate a specific period in the future (3 years, for example) to make it easier for participants from these age groups to respond.
As for OLS and satisfaction with other specific life domains studied in parallel to the PWI-SC, the importance of interpersonal relationships is again borne out by the respondents' answers concerning "satisfaction with family" and "satisfaction with friends", with these domains obtaining the highest scores. Both boys and girls scored high on OLS, and they were "satisfied with their body" as well as "with food", although the mean score for these specific aspects is slightly lower than the figures for "satisfaction with family" and "satisfaction with friends".
There are correlations, albeit moderate ones, between interest in food and the PWI-SC and between the three components of food motivations and the PWI-SC-for both boys and girls. The fact that correlations are only moderate might be due to the fact that none of the PWI-SC indicators explicitly refer to something as specific as food, given that they are constructed on a higher level of abstraction.
Correlations between this interest in food and OLS and complementary domains of satisfaction explored alongside the PWI-SC are statistically significant yet weak, with the closest correspondence coming between interest and OLS. Boys and girls exhibit a different pattern of responses, to the extent that the strongest correlation in the case of girls relates to interest and "satisfaction with the sports you play", while for boys it is interest and "satisfaction with your body", suggesting different roots for their respective interest in food.
In this paper, SATFOOD has been understood as an element that allows us to evaluate those aspects that can help children enjoy their food and bring them to consider it as a pleasurable and important element for their well-being and quality of life. It is worth mentioning that the indicators that explain SATFOOD are not the same for boys and girls. Further studies are needed in order to analyze the reasons for such differences in more depth.
From the regression models we have run, we see that SATFOOD is explained through both food indicators and well-being indicators. These findings justify the consideration of SATFOOD as a variable for evaluating elements related to food (health motivations) and others related to well-being (satisfaction with health and with doing things away from home, from the PWI-SC, and the OLS) for the children participating in this study. The fact that satisfaction with doing things away from home contributes to explaining SATFOOD may have to do with eating out in a bar or restaurant, something which children aged 10 to 12 probably enjoy doing with their families. The data obtained in this paper do not allow this hypothesis to be tested, however.
The results obtained suggest that adults are probably underestimating the role played by food in the lives of children aged between 10 and 12, with this issue having long been considered solely of interest to adults or, at most, to older children. This statement is based on the fact that there are more studies carried out on adults than on children and, of the latter, only very few on young children.
It is important to understand how eating practices are conditioned from a young age and what affects this in order to be able to devise tools that help promote a more suitable dietary education for different groups, one that would bear in mind similarities and differences between each age bracket. There is a need to explore the role of indicators such as satisfaction with food, young people's interest in food, and the child population's levels of subjective well-being in order to open up new avenues of research in the field of health education and promotion. To achieve this goal, future research must be carried out in which a prominent role is afforded to children's perspectives, which will allow us to be able to look in greater depth at the connections between this population group's subjective well-being and satisfaction with food.
Children's well-being includes the evaluations and aspirations of children themselves, while considering them active social actors who can reliably explain the experiences and meanings constituting their well-being.
It is important to point out some of the limitations of the study presented in this article. The sample used comprised schools that expressed an interest in taking part in the study. There is no way of knowing whether the sample is somehow skewed with respect to the rest of the schools in the region of Osona. It is possible that the schools not involved might exhibit distinctive characteristics in relation to the variables explored in this article. Ideally, future studies would collect data from a larger pool of schools.
The cross-sectional approach adopted in this study limits our ability to comprehend how the variables studied change over time. In other words, whether the children who are satisfied at present cease to be so over time or whether their wellbeing varies as a result of changes in the importance of food in their lives. It would be necessary to carry out a longitudinal study to examine this evolution.
Finally, it is worth contemplating the possible value of applying qualitative methodologies (such as discussion groups) in future studies in order to be able to investigate the different aspects discussed in this article in greater depth, as well as considering further work on the promotion of healthy eating habits based on the levels of interest in food shown in this study. This could serve as a starting point to improve children's motivation and participation with respect to both developing healthy practices initially promoted at a young age and maintaining them throughout adolescence, thereby fostering good eating habits.