1 Introduction

I just want my children to be happy.

(36 year-old mother of 2)

(25 year-old father of 3)

(32 year-old father of 2)

Parents want their children to be happy. But, what makes children happy? The answer to this seemingly simple question eludes parents, educators, researchers, and the general public. Although opinion-leaders have clearly voiced their concern about the increase in childhood depression, the answer to the question of what contributes to children’s happiness is debatable. With children being constantly bombarded with images suggesting that slim figures, trendy fashions, expensive toys, or other material goods are solutions to finding happiness, it is exceedingly difficult to determine what truly makes children happy. Do children look to material goods to find happiness? Or, do they look to sports? Do they rely on other sources to experience happiness that adults have overlooked? Are there age related changes that need to be accounted for? While anecdotal evidence is abundant, empirical studies focused on investigating the sources that contribute to children’s happiness at different ages are missing. Such studies would be beneficial not only because they would allow for a more informed discussion of children’s global happiness, but also because they would help opinion-leaders guide children down the path towards happiness.

With the inception of the Journal of Happiness Studies in 2000, the prominence of the topic of happiness in today’s pop culture, and the inception of conferences on happiness and positive psychology within the past year, the topic of happiness and subjective well-being has clearly become a highly valued matter (Diener 2000; Veenhoven 2000). Surprisingly, researchers have been slow in developing studies that specifically address children’s happiness. Although a rapidly developing “positive psychology” movement that emphasizes people’s strengths instead of their weaknesses is quickly steering social scientists towards conducting studies on happiness, most of these studies have focused on adults, not children. Studies have primarily used surveys to examine how external correlates of adults’ lives (e.g., income, employment, marital status, etc.) affect happiness (Andrews and Withey 1976; Bortner and Hultsch 1970; Campbell et al. 1976; Cohn 1979; Cummins 2000). Studies have also been designed to assess how happy people are, as opposed to what makes them happy.

Park and Peterson (2006) point out that studies of happiness in children have been neglected. In addition to the conceptual gap in understanding what makes children happy, measures that are more conducive to studying an abstract construct such as happiness in a children’s sample are also needed. Questionnaires developed for adults cannot be simply lifted and used with children, whose cognitive abilities are not as sophisticated as those of adults’. Thus, not only do current happiness studies fall short of providing meaningful answers to questions related to what makes today’s children happy, but they also lack effective measures that are suitable for examining age differences across a wide age range.

Our research responds to Park and Peterson’s (2006) urge for more research on children’s happiness in two ways. First, we introduce a novel measure that is simple, engaging and appropriate for a broad age range of children and adolescents (ages 8–18) to express what makes them happy—a “collage” measure. Second, we use this new technique in combination with more traditional measures to address critical questions that have remained unanswered—What aspects of life make children happy (e.g., people, hobbies, material things)? Do these sources of happiness vary across ages?

We begin by reviewing research on the topic of happiness. We then discuss two studies designed to explore the question—“What makes children happy?” Study 1 collects information about what general themes (e.g., people and pets, hobbies, material things) contribute to children’s happiness, using an open-ended task. Study 2 uses a semi-structured thought listing task and a collage task to test for age differences in children and adolescents’ happiness. We conclude with a discussion of the implications of our findings and future research directions.

2 Conceptual Overview

2.1 What is Happiness?

Happiness, or subjective well-being (SWB), as it is interchangeably used in the literature, refers to the cognitive and affective evaluations of one’s own life (Diener 1984, 1994). Happiness is defined in terms of global life satisfaction or perceived quality of life (PQOL), presence of positive affect, and the absence of negative affect (Diener 1984). Although these three components are different constructs and might require further research to understand each of them separately (Diener et al. 1999), the three components are substantially correlated and can therefore be examined under the higher order factor known as happiness (Diener 1994; Sheldon and Lyubomirsky 2004, 2006).

2.2 Prior Happiness Research

The topic of happiness, or life satisfaction/PQOL, has received increased attention from researchers, particularly in the past decade (Huebner 2004). Studies with adult samples typically correlate happiness with life events such as employment and income (Andrews and Withey 1976; Bortner and Hultsch 1970; Bouazzaoui and Mullet 2005; Campbell et al. 1976; Cohn 1979; Cummins 2000; Vitterso et al. 2003), marital status and family (Glenn and Weaver 1998; Lee et al. 1991; Stack and Eshleman 1998), or friendships and social ties (Chan and Lee 2006; Demir and Weitekamp 2007). Happiness studies have included both objective (e.g., income, education, health, physical fitness) and subjective measures (e.g., happiness, job satisfaction, positive affect), although debate continues regarding the relationship and interaction between objective and subjective measures (Cummins 2000). Despite researchers’ interest in happiness/SWB (Bradburn 1969; Clemente and Sauer 1976; Costa and McCrae 1980; Diener 1984; Emmons and Diener 1985), the single question most frequently posed to respondents has been, “How happy are you?” rather than, “What makes you happy?” As a result, researchers have uncovered important findings in terms of correlates of happiness/SWB, but the issue of what exactly contributes to making people happy is still unclear.

Some anecdotal support for children’s happiness comes from the popular press. For example, KPC news recently reported results of what children in a Fort Wayne Stepping Stones class said made them happy. Some responses included: “my mom”, “silly faces,” and “candy” (2008) . IHOP™ (The International House of Pancakes) has also recently directed their attention towards children’s happiness. Camping, candy and catching fish were a few of the answers children ages 6–12 from around the country gave to the question, “What Makes Me Happy?” as part of IHOP’s 10th annual, “Dream Up Our Float” contest. Finally, an extensive survey conducted with 1,280 individuals ages 13–24 by The Associated Press and MTV™ (Music Television) on the nature of happiness among America’s young people, showed that spending time with family was the top answer to the open-ended question, “What makes you happy?” (Associated Press 2007).

Clearly, given the recent attention towards the issue of what makes children happy by the popular press, additional academic research is needed to set the stage for more informed discussions on this important topic. Happiness is one of the first emotions recognized by children (Harter 1983). It is an emotion that children as young as 4 years old can feel and understand (Lewis and Michalson 1983). Surprisingly, researchers only started to systematically examine children’s happiness/PQOL in the 1990s (e.g., Huebner 1991, 1994; Huebner and Dew 1996). In the past decade, increasing attention has been paid to the determinants, correlates and consequences of individual differences in PQOL among children and adolescents (for reviews see Huebner et al. 2004, 2006).

Nowhere is work on children’s life satisfaction more in evidence than in studies conducted by Huebner and his colleagues (e.g., Ash and Huebner 1998; Huebner 2004; Huebner and Dew 1996; Huebner et al. 2004, 2005, 2007). These researchers have uncovered several interesting findings and have paved the way for exciting new research to develop. To illustrate, they have linked a number of demographic, environmental, cognitive, personality and behavioral factors in adolescents to both global PQOL and satisfaction within specific domains (Huebner et al. 2004). They have also found that variables related to internal, psychological factors are also significantly related to PQOL. For example, their research shows that PQOL is positively correlated with certain cognitive variables such as self-esteem and an internal locus of control (Dew and Huebner 1994). Children who have high self-esteem, internal locus of control, and emotionally stable temperaments appear to have the highest life satisfaction (Dew and Huebner 1994; Rigby and Huebner 2004). Global PQOL has also been linked to many intrapersonal and interpersonal measures of psychological problems, such as hope, anxiety, and depression (see Huebner et al. 2006), as well as a variety of youth risk behaviors such as suicide, alcohol and drug abuse, sexual risk taking, and dieting and exercise (Gilman and Huebner 2003). Finally, Huebner and his colleagues have addressed the methodological challenges associated with studying life satisfaction in children and adolescents by validating a hierarchical PQOL model in which PQOL is conceptualized as a profile of PQOL appraisals corresponding to the domains of family, school, living environment, friends, and self (Huebner et al. 1998).

An additional small set of researchers have contributed to this emerging area of research (Flouri 2004; Holder and Coleman 2008; Homel and Burns 1989; Park and Peterson 2006). First, Park and Peterson (2006) surveyed parents about their children’s character strengths and happiness. They found love, zest, hope and gratitude to be associated with happiness in children ages 3–9. Second, Flouri’s (2004) study examined the role of parenting in later subjective well-being (SWB) and found that closeness to mother at age 16 predicted life satisfaction at age 42 in both men and women. Flouri’s (2004) clever longitudinal study suggests that it is possible that childhood happiness can be carried over to adulthood and therefore underscores the need to better understand happiness prior to adulthood. Third, Holder and Coleman (2008) tested children ages 9–12, measuring spirituality and other factors such as temperament and social relations that can affect an individual’s sense of happiness, and found that 6.5 to 16.5% of children’s happiness can be accounted for by spirituality. Finally, Homel and Burns (1989) have documented that environmental factors are significantly related to PQOL, including neighborhood. Specifically, youth living in residential areas report having slightly higher levels of PQOL than those residing on streets that are predominantly industrial or commercial.

As researchers continue to make more progress in studying children’s happiness, there are several key areas that warrant immediate attention. First, more studies that use children as informants, as opposed to their parents are needed. Since parents may sometimes misinterpret or overlook what makes their children truly happy, studies that use only parental reports as indicators of children’s happiness may be documenting an incomplete or inaccurate picture of children’s happiness. Second, studies using novel and engaging measures more suitable for a children’s sample are needed, as opposed to popularly used, yet cognitively demanding survey instruments. Third, studies using a wide age range of children and adolescents are also needed in order to document developmental changes. Current studies examine children’s happiness within a very limited age range, thereby, making it difficult to explore possible developmental changes in happiness. Fourth, more studies investigating what makes children happy (e.g., sports, achievements), as opposed to those that correlate children’s level of happiness with a list of factors (e.g., level of self-esteem, level of introversion) are needed. Although these current studies provide insights into the characteristics of happier (versus less happy) children, these correlations do not provide us with a rich understanding of what life domains (e.g., people, sports, material things, etc.) children rely on to achieve happiness.

One reason why the topic of children’s happiness has been slow to develop is the lack of engaging measures that are suitable for a children’s sample. Currently, questionnaires are the favored methodology for adult studies. A number of reliable scales have been developed to measure SWB in adults. The most popularly used measures include Diener et al. (1985) Satisfaction with Life Scale, Argyle’s (2002) Oxford Happiness Inventory, Watson et al. (1988) Positive and Negative Affect Scale, and Fordyce’s (1977) Happiness Measures. Similar types of promising valid life satisfaction survey instruments have been developed for children and adolescents (for a review see Huebner et al. 2007). For example, the PANAS-C by Laurent et al. (1999) represents a reliable and valid measure of positive and negative affect for youth.

However, questionnaires, which require a certain level of reading comprehension and concentration to complete thoughtfully, may not be the best measurement option for young children, who not only lack the ability to understand scale items the same way adults do, but also have a much shorter attention span, making it difficult for them to stay focused long enough to thoughtfully answer 30 scale items (e.g., Oxford Happiness Inventory). It is possible that current studies have only examined very limited age ranges because the survey measures employed are not suitable for use with wider age ranges that have different reading comprehension skills. For example, using surveys automatically limits researchers to sampling children with similar cognitive abilities to ensure that all participants have similar interpretations of the survey questions. Questionnaires such as those listed above have also been designed mainly to study how happy people are as opposed to what makes them happy, which is a question that eludes parents, educators, and other opinion-leaders.

In summary, academic research on children’s and adolescents’ happiness is emerging, but at a less than ideal pace, particularly given the increased attention that the topic has received from parents, educators, and the popular press. Several unanswered questions rise to the top of the list—what makes children happy? Are there age related changes? Are there more suitable happiness measures than the typically used surveys that can be used with a wide age range?

3 Study 1

The purpose of Study 1 was to collect information about the general categories (e.g., people and pets, sports, hobbies, material things) that contribute to children’s happiness. To do so, we used an open-ended task.

3.1 Sample

Participants from three age groups were recruited from summer camps in the Midwestern and Eastern United States: fifty 3rd/4th graders (8–9 year olds), fifty 7th/8th graders (12–13 year olds), and fifty 11th/12th graders (16–18 year olds). This age range represented a wide enough age range to explore possible age differences. Participants were recruited from rural and urban areas. Participants came from lower-middle to middle class backgrounds. Our sample was 48% Caucasian, 31% African American, 15% Asian American, and 6% Hispanic American. Our sample contained an equal number of boys and girls. Participants whose parents responded to the recruiting posters displayed around camp and signed the parental consent were invited to participate. Participant assent was obtained prior to the study.

3.2 Procedure

Participants were interviewed individually. Demographic information was obtained first, followed by a brief description of the study. Next, respondents were asked to complete an open-ended task designed to explore the domains that contribute to children’s happiness. Specifically, participants were presented with the open-ended question, “What makes me happy?” and asked to provide a written response. Participants were also given the option of verbalizing their response and having the experimenter help with the written response. Those who preferred that the experimenter write for them came from our youngest age group sampled (i.e., 3rd graders). Whether participants chose to respond verbally or to write down their answers, a written response was recorded for all participants, for later data analysis. All participants were asked to read their answers and to make any clarifications, corrections, additions or deletions prior to the completion of the study. After the task was completed, participants were debriefed, asked not to talk about the study until its completion and were compensated with small prizes for their participation. The procedure took approximately 20–30 min to complete.

3.3 Results

3.3.1 Categories

Participants’ responses to the question, “What makes me happy?” were content analyzed into five categories—“People and pets” (e.g., mom, dad, friends, pets), “achievements” (e.g., getting good grades, being good at math), “material things” (e.g., money, car), “hobbies” (e.g., fishing, being with friends), and “sports” (e.g., hockey, soccer). Two researchers were involved in generating these themes. The “people and pets” category consisted mainly of people, including family, friends, and other social influences such as teacher, coach, and neighbor. However, we included one label called, “pets.” Both researchers agreed that pets did not warrant a separate category because not all children have pets and therefore, would be a theme left untouched for many participants. However, it needed to be included to capture the importance of pets in making children happy. Both researchers agreed that pets should be placed with people as opposed to any of the other categories since pets, like people, provide companionship. These five themes/categories are representative of the thoughts listed by participants across all ages (see Table 1). Although the main objective of this study was to collect information about the themes that best represent children’s happiness, we also explored age differences (reported below).

Table 1 Study 1: open-ended “what makes me happy?”
Full size table

3.3.2 Age Differences

3.3.2.1 3rd/4th Graders vs. 7th/8th Graders

Contrasts between 3rd/4th graders and 7th/8th graders revealed age differences in what children believe contribute to their happiness. First, 3rd/4th graders listed fewer material items than did 7th/8th graders (M = 1.42 vs. 2.64, t(1, 98) = 3.72, p < .01). Second, 3rd/4th graders listed fewer people and pets than did 7th/8th graders (M = 2.17 vs. 2.78, t(1, 98) = 2.01, p < .01). Third, 3rd/4th graders listed more hobbies than did 7th/8th graders (M = 1.79 vs. 1.10, t(1, 98) = 2.65, p < .01). There were no age differences in the number of achievements or sports (p > .13). We found a marginally significant difference in the total number of thoughts listed across ages (M 3rd/4th = 8.70, M 7th/8th = 10.18, t(1, 98) = 2.55, p < .01). This difference may have been due to the abstract nature of this task, making it more cognitively demanding for younger children to retrieve information from memory.

3.3.2.2 7th/8th Graders vs. 11th/12th Graders

Contrasts between 7th/8th graders and 11th/12th graders revealed age differences in the importance of several domains in contributing to children’s happiness from early adolescence into late adolescence. First, 7th/8th graders listed more material items than did 11th/12th graders (M = 2.64 vs. 2.05, t(1, 98) = 1.68, p < .05). Second, 7th/8th graders listed more people and pets than did 11th/12th graders (M = 2.78 vs. 2.27, t(1, 98) = 1.59, p = .05). Third, 7th/8th graders listed fewer sports than did 11th/12th graders (M = 1.51 vs. 2.09, t(1, 98) = 1.99, p = .02). There were no age differences in the number of hobbies or number of achievements listed (p > .60). There was also an absence of age differences for the total number of thoughts listed (M 7th/8th = 10.18, M 11th/12th = 9.95, p > .75).

3.3.2.3 3rd/4th Graders vs. 11th/12th Graders

Contrasts between the youngest and oldest children in the sample revealed several interesting age differences in the importance of certain domains in contributing to children’s happiness. First, 3rd/4th graders listed fewer material items than did 11th/12th graders (M = 1.42 vs. 2.05, t(1, 98) = 2.28, p < .01). Second, 3rd/4th graders listed more hobbies than did 11th/12th graders (M = 1.79 vs. 1.23, t(1, 98) = 2.15, p = .01). Third, 3rd/4th graders listed fewer achievements than did 11th/12th graders (M = 1.62 vs. 2.33, t(1, 98) = 2.07, p = .02). There were no age differences in the number of sports or people and pets listed (p > .75). Finally, we found that 3rd graders listed fewer thoughts for this task than 11th/12th graders (M 3rd/4th = 8.70, M 11th/12th = 9.95, t(1, 98) = 1.91 p < .05). Again, this difference may be due to slightly the abstract nature of this task, making it more cognitively demanding for younger children to retrieve information from memory.

3.4 Discussion

In this study, we used an open-ended task to explore what makes children happy. Five sources of children’s happiness emerged—“people and pets,” “ achievements,” “material things,” “hobbies,” and “sports”. These themes were mentioned across a wide age range (i.e., 8–18 years old) by both boys and girls.

Although the main objective of this study was to collect information about the domains that contribute to children’s happiness, we also explored age differences. Our initial age related findings suggest that children’s sources of happiness may change as they progress from childhood throughout adolescence. Specifically, during middle childhood, children perceive people and hobbies as the most important domains contributing to their happiness. By early adolescence, there is a shift from hobbies to material things, while people and pets remain central to their happiness. By late adolescence, there is a shift away from material things, as adolescents begin to place more importance on achievements in making them happy. Again, people and pets are still central to their happiness. While categories such as hobbies, material things and achievements seem to have varying degrees of importance depending on the age of the child, people and pets are central to children’s happiness across a wide age range.

Although the open-ended nature of the task allowed participants to reveal information about their happiness without being influenced by experimenter stimuli, the abstract nature of the task makes interpreting age differences difficult. It is possible that age differences here may have been task driven. That is, older participants may have responded differently than their younger counterparts because the task was easier for them to complete, as opposed to some underlying difference in what they believe make them happy. For example, it is possible that compared to 3rd/4th graders, 11th/12th graders found the open-ended, abstract nature of the task easier to complete and therefore would have listed more ideas regardless of the topic.

Study 2 addresses this concern by introducing two new, more structured happiness measures amenable to a wide age range. Providing structured tasks, thereby reducing the level of abstraction, but maintaining room for respondent creativity allows for more thoughtful responses across all ages.

4 Study 2

4.1 Sample

Participants from three age groups were recruited from summer camps in the Midwestern United States: fifty 3rd/4th graders (8–9 year olds), fifty 7th/8th graders (12–13 year olds), and fifty 11th/12th graders (16–18 year olds). This age range represented a wide enough age range to explore possible age differences. Participants were recruited from a mid-size city. Participants came from a lower-middle to middle class background. Our sample was 70% Caucasian, 12% African American, 13% Asian American, and 5% Hispanic American. Our sample contained an equal number of boys and girls. Participants’ whose parents responded to the recruiting posters displayed around camp and signed the parental consent were invited to participate. Participant assent was obtained prior to the study.

4.2 Procedure

Participant assent and parental consent were obtained for each participant. Respondents were interviewed individually and completed two tasks that assessed their happiness (described below). Each task was described and demonstrated by the interviewer to ensure understanding of the task instructions. After completing the tasks, participants were debriefed and asked to not talk about the study with their peers until everyone had completed the study. The entire procedure took 30 min to complete.

4.3 Experimental Tasks

4.3.1 “What Makes Me Happy?” Semi-Structured Thought Listing Task

In this study, we aimed to decrease the cognitive demand placed on our participants, especially our youngest age group. To do so, we used results from Study 1 as input to design more structured tasks. Recall, results from Study 1 showed that responses of participants ages 8–18 to the open ended question, “What makes me happy?” could be organized into five themes—”people and pets,” “achievements,” “material things,” “hobbies,” and “sports”. Therefore, in order to make the task less cognitively demanding for younger children, we provided participants with a list of these five themes to be used as cues to retrieve information about what makes them happy. Respondents were encouraged to describe their happiness in any way they wished—using some, all, or none of the categories provided.

4.3.2 “Happiness Collage”

We used Chaplin and John’s (2005) collage methodology as a second measure of “Happiness.” The task required participants to construct a collage to answer the question, “What makes me happy?” We selected a collage format to accommodate the wide age range (8–18 years) in our sample. Collages provide a familiar and engaging activity for both younger children and older adolescents, thereby alleviating difficulties that young children have with unfamiliar rating scale formats or verbalization skills needed for unstructured interviews. Collages have proven to work well with adults (Zaltman and Coulter 1995) as well as children (Chaplin and John 2005, 2007). The term “happiness” (as opposed to similar terms used in the literature such as “subjective well-being, life satisfaction, or PQOL”) was used because it is easy to understand even for our youngest participants. Combining the basic concept of happiness with a collage methodology produced a task that was familiar, engaging, and well understood by the wide age range of respondents in our study. Also, because the collage measure combines all types of responses (e.g., hobbies, material things, people, etc.) into one complete picture about one’s happiness, socially desirable responding is likely to be less of an issue than it would be using the more traditional survey method. For example, in a survey, because a focus on material possessions to achieve happiness is deemed to be a negative value by society (Belk 1985; Richins and Dawson 1992), respondents may be less willing to reveal that material things make them happy. However, in a collage task, because material things are masked by other domains such as “people and pets,” “achievements,” “material things,” “hobbies,” and “sports”, participants are more likely to report that material things make them happy.

The collage exercise began by showing participants a set of lexical-graphical stimuli available for composing their collages. Although collage methods often require respondents to find their own pictures or words, we provided a set of stimuli to ensure that younger as well as older respondents would have an accessible set of materials. These items were mounted on five poster boards, each representing a different theme—“people and pets,” “achievements,” “material things,” “hobbies,” and “sports”(see Fig. 1). Recall, these themes emerged from Study 1. Twenty laminated labels and/or pictures were placed on each board. For example, “biking” and “spending time with friends” were included on the hobbies board, “getting good grades” and “saving money” were found on the achievements board, “car” and “CD’s” were included on the material things board, “soccer” and “football” were pictured on the sports board, and “dad” and “mom” were included on the people board. These items were selected to appeal to a wide range of interests. For example, for achievements, we selected many popular items across age groups (e.g., knowing nice people), some that are popular with younger children (e.g., no homework), some that are popular with older children (e.g., getting into a good college), some that are popular with boys (e.g., being good at sports), and some that are popular with girls (e.g., write well).

Fig. 1
figure 1

Example of theme board: “achievements”

Full size image

Participants were asked to construct their “Happiness Collage” by choosing items from the theme boards (see Fig. 2). Post-it boards, the size of a poster board covered with a repositionable adhesive, were used for the theme boards and blank collage board. This allowed participants to easily move pictures and labels from board to board, add or eliminate labels from collages, and artistically arrange the elements on their collage. Blank cards and markers were available for participants to make up their own labels or pictures. Upon completing their collages, participants were asked to describe why they had placed certain items on the collage. These responses served as a check to ensure that collage items were being interpreted as we intended. A photograph of the collage was then taken for later data analysis.

Fig. 2
figure 2

Example of “happiness collage”

Full size image

Next, we asked participants to reduce their collages by discarding half of the labels or pictures, leaving the items most important to their happiness. Participants were forced to make tradeoffs at this stage, which encouraged participants to think carefully about the things that truly made them happy versus descriptions that might have been chosen simply to fill their boards. A photograph of the “Happiness Collage” (now in its reduced form) was taken for data analysis.

Two measures of each theme were computed from each child’s original and reduced “Happiness Collage.” The first measure was the number of “people and pets,” “achievements,” “material things,” “hobbies,” and “sports” included on the collage. A second measure adjusted for the possibility that a larger number of items from a particular category might be included on the collage simply because of a greater number of total items on a collage. We adjusted for collage size by dividing the number of “people and pets,” “achievements,” “material things,” “hobbies” or “sports” on a collage by the total number of items on the collage.

5 Results

Below, we report findings of age differences. Although the focus of this study was on detecting age differences in children and adolescents’ happiness, we also explored gender differences, which are also reported below.

5.1 Semi-Structured Thought Listing Task

5.1.1 Age Differences

5.1.1.1 3rd/4th Graders vs. 7th/8th Graders

Contrasts between 3rd/4th graders and 7th/8th graders revealed age differences in the importance of several domains in contributing to children’s happiness from late childhood into early adolescence (see Table 2). First, 3rd/4th graders listed fewer material items than did 7th/8th graders (M = 1.63 vs. 4.72, t(1, 98) = 7.58, p < .01). Second, 3rd/4th graders listed fewer people and pets than did 7th/8th graders (M = 3.66 vs. 5.39, t(1, 98) = 3.75, p < .01). Third, 3rd/4th graders listed more hobbies on their “Happiness Collages” than did 7th/8th graders (M = 4.34 vs. 2.77, t(1, 98) = 3.71, p < .01). Fourth, 3rd/4th graders listed more sports than did 7th/8th graders (M = 3.18 vs. 2.19, t(1, 98) = 2.60, p < .01). There were no age differences in the number of achievements. We found that 3rd graders listed fewer thoughts for this task than 7th graders (M 3rd/4th = 15.77, M 7th/8th = 18.26, t(1, 98) = 2.17, p = .01). This difference may be due to the slightly abstract nature of this task, making it more cognitively demanding for younger children to retrieve information from memory.

Table 2 Study 2: semi-structured thought listing “what makes me happy?”
Full size table
5.1.1.2 7th/8th Graders vs. 11th/12th Graders

Contrasts between 7th/8th graders and 11th/12th graders revealed age differences in the importance of several domains in contributing to children’s happiness from early adolescence into late adolescence. First, 7th/8th graders listed more material items than did 11th/12th graders (M = 4.72 vs. 3.41, t(1, 98) = 2.67, p < .01). Second, 7th/8th graders listed more people and pets than did 11th/12th graders (M = 5.39 vs. 4.55, t(1, 98) = 1.80, p < .02). Third, 7th/8th graders listed fewer achievements than did 11th/12th graders (M = 3.18 vs. 4.27, t(1, 98) = 2.08, p = .02). Fourth, 7th/8th graders listed fewer sports on their “Happiness Collages” than did 11th/12th graders (M = 2.19 vs. 3.32, t(1, 98) = 3.54, p < .01). There were no age differences in the number of hobbies placed on collages (p > .70). There was an absence of age differences for the total number of thoughts listed (M 7th/8th = 18.26, M 11th/12th = 18.45, p > .85).

5.1.1.3 3rd/4th Graders vs. 11th/12th Graders

Contrasts between the youngest and oldest children in the sample revealed several interesting age differences in the importance of certain domains in contributing to children’s happiness. First, 3rd/4th graders listed fewer material items than did 11th/12th graders (M = 1.63 vs. 3.41, t(1, 98) = 4.75, p < .01). Second, 3rd/4th graders listed more hobbies than did 11th/12th graders (M = 4.34 vs. 2.89, t(1, 98) = 3.30, p < .01). Third, 3rd/4th graders listed fewer achievements than did 11th/12th graders (M = 2.97 vs. 4.27, t(1, 98) = 2.54, p < .01). There were no age differences in the number of people and pets or the number of sports placed on collages (p’s > .15). We found that 3rd graders listed fewer thoughts for this task than 11th/12th graders (M 3rd/4th = 15.78, M 11th/12th = 18.44, t(1, 98) = 2.44 p < .01). Again, this difference may be due to slightly the abstract nature of this task, making it more cognitively demanding for younger children to retrieve information from memory.

5.1.2 Gender Differences

5.1.2.1 Ages 8–18

Several interesting gender differences emerged. Across the entire sample (ages 8–18), girls listed more people and pets than did boys (M = 5.08 vs. 3.99, t(1, 148) = 2.72, p < .01), whereas boys listed more sports than did girls (M = 3.76 vs. 2.02, t(1, 148) = 6.28, p < .01). There were no gender differences in the total number of thoughts listed (p > .60). We also explored gender differences at the group level.

5.1.2.2 3rd/4th Graders

For our youngest children in the study, we found that girls listed more achievements than boys (M = 3.52 vs. 2.44, t(1, 48) = 1.46, p = .07), while boys listed more sports than girls (M = 4.07 vs. 2.28, t(1, 48) = 3.02, p < .01). There were no gender differences in the total number of thoughts listed (p > .70).

5.1.2.3 7th/8th Graders

For our middle age group, we found that girls listed fewer sports than boys (M = 1.63 vs. 2.75, t(1, 48) = 2.98, p < .01). We also found that girls listed more material things than did boys (M = 5.33 vs. 4.11, t(1, 48) = 1.70, p < .05). There were no gender differences in the total number of thoughts listed (p > .70).

5.1.2.4 11th/12th Graders

For our oldest age group in the study, we found that girls listed more people than boys (M = 5.60 vs. 3.51, t(1, 48) = 2.91, p < .01), while boys listed more sports than girls (M = 4.47 vs. 2.17, t(1, 48) = 6.23, p < .01). There were no gender differences in the total number of thoughts listed (p > .80).

5.2 “Happiness Collage” Issues

The “Happiness Collage” materials were examined to explore two things. First, we analyzed the number of labels participants were familiar with from all theme boards. Given our extensive pilot testing and knowledge acquired from Study 1, we anticipated that participants would be familiar with most, if not all, of the experimental stimuli. As expected, all participants were familiar with the entire set of stimuli. Also of concern was the possibility that the total number of items on a collage might vary by age, with older children including more elements on their collages as a result of having a more complex sense of “happiness.” Here, we did not find evidence of age differences, with 3rd/4th graders (M = 26.54), 7th/8th graders (M = 28.64), and 11th/12th graders (M = 27.98) including approximately the same number of items on their “Happiness Collages” (for all contrasts, p’s > .15). Given these results, no adjustments to subsequent analyses were deemed necessary.

5.2.1 Reliability of Collage

Given the novel and qualitative nature of the collage measure, we performed a test to assess the reliability of the measure. To assess test–retest reliability, we randomly selected 25 participants to complete a second “Happiness Collage” 2 weeks after the main study (see Table 3). Comparing these collages to those from the main study, we found strong test–retest correlations for the number of “people and pets,” “achievements,” “material things,” “hobbies,” and “sports” and the total number of collage items. We also found strong test–retest correlations for the percentage of “people and pets,” “achievements,” “material things,” “hobbies,” and “sports”.

Table 3 Study 2: test–retest reliability
Full size table

Given these results, we focused attention on measures from the original collage, specifically the number of “people and pets,” “achievements,” “material things,” “hobbies,” “sports,” and the total number of collage items for subsequent analyses.

5.2.2 Validity of Happiness Measures

We recruited 25 participants, ages 8–16, who we knew well (e.g., children, nieces, nephews, and close neighbors) or those who could be evaluated by someone who knew them well (e.g., camp counselors or parents). Parents and camp counselors were asked to look at participants’ “Happiness Collages” and indicate whether they felt participants’ responses accurately reflected what made them happy. For example, problems could be detected if, for example, a child known to thoroughly enjoy sports did not include sports on his/her “Happiness Collage”, or in the semi-structured thought listing task.

Results were supportive of the collage and semi-structured thought listing tasks as appropriate measures of children and adolescents’ happiness. Parents and camp counselors indicated that each participant’s collage and responses to the thought listing task, reflected the “people and pets,” “achievements,” “material things,” “hobbies,” and “sports” they knew to be associated with their child’s happiness. Our results showed evidence of content validity, and therefore gave us confidence in the collage and thought listing measures as appropriate measures of children’s happiness.

5.3 Collage

5.3.1 Age Differences in Happiness

5.3.1.1 3rd/4th Graders vs. 7th/8th Graders

Contrasts between 3rd/4th graders and 7th/8th graders revealed age differences in the importance of several domains that contribute to children’s happiness from middle childhood to early adolescence (see Table 4). First, 3rd/4th graders included fewer material items on their “Happiness Collages” than did 7th/8th graders (M = 3.62 vs. 6.72, t(1, 98) = 5.50, p < .01). Second, 3rd/4th graders included fewer people and pets on their “Happiness Collages” than did 7th/8th graders (M = 6.58 vs. 8.02, t(1, 98) = 2.05, p = .02). Third, 3rd/4th graders included more hobbies on their “Happiness Collages” than did 7th/8th graders (M = 6.26 vs. 4.66, t(1, 98) = 2.95, p < .01). Fourth, 3rd/4th graders included more sports on their “Happiness Collages” than did 7th/8th graders (M = 4.34 vs. 3.28, t(1, 98) = 2.65, p < .01). There were no age differences in the number of achievements placed on children’s “Happiness Collages”. These age related trends were not influenced by collage size, indicated by an absence of age differences for total number of items placed on collages (M 3rd/4th = 26.54, M 7th/8th = 28.64, p > .15). Data from reduced collages exhibited the same trends.

Table 4 Study 2: “happiness collage” elements
Full size table
5.3.1.2 7th/8th Graders vs. 11th/12th Graders

Contrasts between 7th/8th graders and 11th/12th graders revealed age differences in the importance of several domains that contribute to children’s happiness from early adolescence to late adolescence. First, 7th/8th graders included more material items on their “Happiness Collages” than did 11th/12th graders (M = 6.72 vs. 5.26, t(1, 98) = 2.38, p < .01). Second, 7th/8th graders included more people and pets on their “Happiness Collages” than did 11th/12th graders (M = 8.02 vs. 6.82, t(1, 98) = 2.17, p < .01). Third, 7th/8th graders included fewer achievements on their “Happiness Collages” than did 11th/12th graders (M = 5.96 vs. 7.00, t(1, 98) = 1.82, p < .01). Fourth, 7th/8th graders included fewer sports on their “Happiness Collages” than did 11th/12th graders (M = 3.28 vs. 4.48, t(1, 98) = 3.50, p < .01). There were no age differences in the number of hobbies placed on collages (p > .50). These age related trends were not influenced by collage size, indicated by an absence of age differences for the total number of items placed on collages (M 7th/8th = 28.64, M 11th/12th = 27.98, p > .15). Data from reduced collages exhibited the same trends.

5.3.1.3 3rd/4th Graders vs. 11th/12th Graders

Contrasts between the youngest and oldest children in the sample revealed several interesting age differences in the importance of certain domains in contributing to children’s happiness. First, 3rd/4th graders included fewer material items on their “Happiness Collages” than did 11th/12th graders (M = 3.62 vs. 5.26, t(1, 98) = 3.36, p < .01). Second, 3rd/4th graders included more hobbies on their “Happiness Collages” than did 11th/12th graders (M = 6.26 vs. 4.42, t(1, 98) = 3.57, p < .01). Third, 3rd/4th graders included fewer achievements on their “Happiness Collages” than did 11th/12th graders (M = 5.58 vs. 7.00, t(1, 98) = 2.69, p < .01). There were no age differences in the number of people and pets or the number of sports placed on collages (p’s > .30). These age related trends were not influenced by collage size, indicated by an absence of age differences for the total number of items placed on collages (M 3rd/4th = 26.54, M 11th/12th = 27.98, all p > .15). Data from reduced collages exhibited the same trends.

5.4 Collage

5.4.1 Gender Differences

5.4.1.1 Ages 8–18

Several interesting gender differences emerged. Across the entire sample (ages 8–18), girls placed more people and pets on their “Happiness Collages” than did boys (M = 7.83 vs. 6.45, t(1, 148) = 2.57, p < .01), whereas boys placed more sports on their “Happiness Collages” than did girls (M = 4.97 vs. 3.09, t(1, 148) = 6.49, p < .01). There were no gender differences in the total number of items used to create collages (p > .40). We also explored gender differences at the group level.

5.4.1.2 3rd/4th Graders

For our youngest children in the study, we found that girls chose more achievements (M = 6.16 vs. 5.00, t(1, 48) = 1.47, p < .05) and hobbies than boys (M = 7.08 vs. 5.44, t(1, 48) = 2.00, p < .05), while boys chose more sports than girls (M = 5.28 vs. 3.40, t(1, 48) = 3.08, p < .01). There were no gender differences in the total number of items used to create collages (p > .60).

5.4.1.3 7th/8th Graders

For our middle age group, we found that girls chose fewer hobbies (M = 4.20 vs. 5.12, t(1, 48) = 1.37, p < .05) and sports than boys (M = 2.60 vs. 3.96, t(1, 48) = 3.28, p < .01). We also found that girls chose more material things than did boys (M = 7.56 vs. 5.88, t(1, 48) = 1.80, p < .05). There were no gender differences in the total number of items used to create collages (p > .30).

5.4.1.4 11th/12th Graders

For our oldest age group in the study, we found that girls chose more people and pets than boys (M = 7.92 vs. 5.72, t(1, 48) = 2.95, p < .01), while boys chose more sports than girls (M = 5.68 vs. 3.28, t(1, 48) = 6.18, p < .01) to create their “Happiness Collages.” There were no gender differences in the total number of items used to create collages (p > .60).

5.5 Qualitative Results

Qualitative analyses of the “Happiness Collages” were pursued to explore additional age differences. After participants finished building their “Happiness Collages,” the experimenter went through each label on the collage and asked participants why certain things were important to them. Participants were also asked how they would feel if they did not have certain things in their lives (e.g., car, cellular phone, fishing, soccer, friends, etc.). Further probing was done by asking participants to compare two items on their “Happiness Collage” (e.g., mom versus brand names) in terms of degree and reason for importance. At this stage, participants were encouraged to elaborate on why certain things were important to their happiness and why other things were not as important to their happiness.

The age differences detected from our quantitative analyses were consistent with our observations of the way participants completed the collage task. Third/Fourth graders focused much of their attention on sports, hobbies, and people while older participants focused much of their attention on people, achievements, and material things. The connections they made between these categories are quite interesting. With the older age groups, when it came time for them to eliminate half of the labels from their collage, we frequently observed long and involved deliberations of whether to leave material things on their collages or to eliminate them in order to keep either a person, hobby, sport, or achievement. For example, in the process of deciding which labels were most important to her happiness, and which ones were worthy of elimination, a 17 year-old girl picked up “cellular phone” to eliminate from her original collage, only to place it back on the collage and eliminate a hobby instead of the cellular phone. Asked why she did not eliminate the phone, she responded, “I need it to stay in touch with my family and friends.” She further explained that it was not the actual phone that she needed to be happy, but rather the purpose the phone served that was important.

In our youngest ages, similar degrees of deliberation were exhibited when it came time to eliminate half of the labels from their collages. However, rather than deliberating about whether to keep material things, 3rd/4th graders spent their time and effort making tradeoffs mainly between hobbies and sports. They spent considerably much less time thinking about material things. In fact, they either eliminated material things immediately (most cases) or left them on the collages without much thought because they viewed material possessions more concretely, (e.g., a car helps people get from one place to another), than symbolically (e.g., a car symbolizes a person’s wealth, status and popularity). To illustrate, a 3rd grade boy explained that he was eliminating a car and expensive clothes from his “Happiness Collage” because “I can always ride my bike. I already have clothes to wear to a lot of places so I don’t need more.” In contrast, a 12 year-old boy explained that he was eliminating a car from his collage because “It’s nice to have one because you get more attention, but what really makes me happy is when people like me for who I am, not what I have.”

When asked how they would feel if they did not have certain “people and pets,” “achievements,” “material things,” “hobbies,” and “sports” in their lives, children across all ages agreed that people and pets were crucial to their happiness. They indicated that all other categories could be taken off of their “Happiness Collages,” but people and pets would have to remain. Specifically, parents, siblings, friends, teachers and coaches were consistently mentioned as key contributors to their happiness. We observed similar findings when children were asked to compare the importance of specific descriptions on their “Happiness Collages.” Whenever participants were asked to compare the importance of people and pets to any other category, participants would consistently mention that people and pets were more important to their happiness than any other domain.

An interesting qualitative finding from our study concerns the importance of material goods in making children and adolescents happy. Across all ages, children and adolescents commented that material things were important to their happiness not because of their functional value, but more so for their symbolic value. For example, a 12 year-old girl reports,

“I really don’t need another bathing suit, but when my team has an important swim meet, everyone on the team orders a new one, even if it looks almost identical to the one we already have that is still in good condition. I order one of these just because I am part of the team…It’s like our new team bathing suit acts as a reminder that we are all in it together to win the meet . . .Winning makes me happy I guess.”

This quote is an example of how material things can be instrumental in children and adolescents’ lives by serving as a symbol of group belongingness. Our results uncovered that although material things are very important to the happiness of children ages 8–18, people are more important. This finding is interesting because researchers have found that those who are highly materialistic tend to devalue their personal relationships (Kasser 2002; Richins and Dawson 1992). However, upon closer examination, we found that children in our sample frequently and consistently discussed how both material things and people were important in making them happy. Thus, although a qualitative analysis of children’s happiness was not the focus of this study, our exploratory efforts revealed that although children and adolescents may be placing much importance on material goods to find happiness, they are merely using material things as a platform from which to build strong relationships with people, which is what ultimately and consistently makes them happy.

6 General Discussion

6.1 Contributions

Our research builds upon the extant literature in several ways. First, we introduce a new measure for happiness that is appropriate for a wide age range of children and adolescents. Although survey instruments have provided us with a fairly good understanding of the correlates of PQOL, our “Happiness Collage” measure allows us to see more variance in participants’ responses across domains (e.g., achievements versus sports versus people) as well as within domains (e.g., hockey, lacrosse, tennis). Our measure also allows participants to make direct comparisons across all domains (as well as within domains) since they are presented with all theme boards at the same time to answer the question, “What makes me happy?” Due to its qualitative nature, our collage measure also affords researchers the ability to probe further into why certain domains are more important to children’s happiness than others.

Second, we add to the growing body of knowledge on children’s life satisfaction/happiness by addressing developmental issues which has received very little attention to date. Using three different measures, we found evidence of age differences in children’s perceptions of what makes them happy (i.e., people and pets versus material things versus sports versus hobbies versus achievements). We found that 3rd graders reported that hobbies and people and pets contributed the most to their happiness, while 7th/8th graders reported material things and people and pets, and 11th/12th graders reported achievements as well as people and pets.

Although prior research has found that global life satisfaction reports of children and adolescents do not appear to differ significantly as a function of age, we found age differences in the importance of different domains in contributing to children’s happiness, which suggests that it is possible that children can be equally happy (i.e., report the same level of happiness, life satisfaction or PQOL), but they may be focusing on different domains to reach the same level of happiness. For example, younger children may be equally happy compared to adolescents, but the route they take to achieve the same level of happiness may be different (e.g., hobbies versus achievements). Moreover, the choices within a domain may also be different. For example, for younger children, achieving more time on the playground as a reward for good behavior may provide them just as much happiness as an adolescent achieving instant peer acceptance upon wearing the latest fads.

Third, our results add to the existing life satisfaction literature by providing additional evidence that familial and peer relationships are strongly correlated with children’s life satisfaction (see Holder and Coleman 2007). We also find children’s happiness to be associated with a variety of factors that have not yet been examined in prior published studies of life satisfaction, including material possessions, pets, and achievements. Importantly, we have uncovered the specific types of “people and pets,” “achievements,” “material things,” “hobbies,” and “sports” that children feel contribute to their happiness. For example, we found that although 3rd/4th graders and 11th/12th graders both believe achievements make them happy, they mention different types of achievements. Specifically, 3rd graders reported that having a longer recess as a reward for good behavior was an achievement that made them happy, while 11th/12th graders reported that getting good grades and getting into a good college were the types of achievements that made them happy. To date, researchers have not yet examined these diverse categories in direct comparison with one another in a single study.

Finally, we add to the current body of research on children’s life satisfaction/happiness by exploring gender differences across a wide age range. Few studies have addressed this issue (see Huebner et al. 2000 for exception). Although it was not the focus of our research, we explored and consistently found several gender differences in children’s reported sources of happiness. First, using three different measures, we found that when asked to answer, “What makes me happy?” across the entire sample of participants ages 8–18, boys reported sports more frequently than girls, while girls reported people and pets more frequently than boys. In addition to these differences, we also found that 3rd grade girls reported more achievements than boys, 7th/8th grade girls reported more material things than boys, and 11th/12th grade girls reported more people and pets than boys. Although our findings suggest that boys and girls may need different stimulation from different domains in life to achieve happiness, more research on gender differences is needed before we can come to definitive conclusions.

6.2 Implications

Our research serves to help opinion-leaders by contributing to a more informed discussion on how to orient children’s programs to maximize focus on the sources of children’s happiness at different ages. For example, knowing that children and adolescents recognize that people and pets contribute to their happiness to a significant degree, educators may want to develop programs to help children build healthy relationships with various admired others. In addition, as parents and educators’ concern over children’s obsession with superficial material goods in search of happiness escalates, children’s self-reports of what makes them happy becomes even more important. Interestingly, despite this escalating concern, we uncovered that children and adolescents ages 8–18 consistently reported that they look to their family, friends, coaches and teachers to find happiness, not necessarily to material possessions. They reported that they look to material things when they think it will help them build strong relationships with admired individuals. Therefore, developing programs and seminars to educate parents and other concerned individuals on ways in which they can teach children how to build, maintain and improve relationships with others would be an important step towards guiding children towards happiness.

6.3 Limitations

There are several limitations that should be noted. There are measurement-related issues to consider when reviewing the results of our study. First, although our measures were designed to minimize socially desirable responding, participants’ reports may still be subject to situation factors or social desirability, as they, particularly adolescents, may feel the pressure to respond in a socially desirable way. Second, the sole reliance on self-report measures also raises potential questions related to common method variance (Laurent et al. 1999). Third, we used previously unvalidated assessments of happiness. Although we found evidence of reliability and validity, our new measures would benefit from more testing.

Our findings are also limited by the fact that we used a convenience sample. Although convenience sampling is a common practice, such a strategy limits researchers’ ability to generalize findings to larger, more diverse populations. Finally, although our “Happiness Collage” measure allowed us to gather qualitative data that provided interesting insights to answer why children and adolescents believe certain domains contribute to their happiness more so than others, due to the fact that only one researcher conducted the interviews, the qualitative data is subject to some researcher subjectivity.

6.4 Future Research Directions

There are several topics that warrant more attention from researchers. First, gender differences in children’s happiness, both in terms of how happy boys (versus girls) are, as well as what makes them happy at different ages need to be examined more closely. Although our research focused on age differences, we explored gender differences, and found some interesting differences that held across a wide age range from middle childhood through late adolescence. These differences would be interesting to examine at a deeper level in future studies.

Second, various members of society should benefit from researchers studying the issue of how children’s happiness at young ages may affect various illnesses later in childhood/adolescence, such as depression, eating disorders, and obesity. Although opinion-leaders have clearly voiced their concern about the increase in childhood depression, the link between these types of illnesses and children’s happiness is unclear.

Third, more research is needed to develop sound, engaging measures suitable for a wide age range of children and adolescents. Life satisfaction/happiness researchers would also benefit from the consideration of alternative data collection techniques such as the use of multitrait-mulimethod designs and within and between-analyses techniques (Avolio et al. 1991), that address issues of common method variance. Another approach that would address common method variance would be to solicit information from children’s parents, teachers and camp counselors (i.e., those who know the children well). Having better measures will undoubtedly encourage more research on the topic of children’s happiness.

Finally, research efforts in life satisfaction/happiness have largely lacked a developmental focus. Much of the research highlights similarities in the determinants and consequences of life satisfaction among children and adults, failing to reflect the distinctiveness between adults and children of different ages. Developmental differences in cognitive abilities, social emotional needs, interests, physical maturation, and so forth suggest possible important developmental considerations. Longitudinal research would be particularly beneficial, providing the opportunity to monitor the course and development of happiness determinants and to determine the directionality of relationships.