Can kids be too happy in school? The optimal level of school satisfaction
v TABLE OF CONTENTS DEDICATION……………......……………………………………………………..iii ABSTRACT………………….……………………………………………………..iv LIST OF TABLES…………………………………………………………………...vi I. INTRODUCTION…………………………………………………………………..1 II. METHODS……………………………………………………………………...24 III. RESULTS……………………………………………………………………...30 IV. CONCLUSION...………………………………………………………………..36 REFERENCES……………………………………………………………………...43 APPENDIX A……………………………………………………………………....59 APPENDIX B……………………………………………………………………....61
vi LIST OF TABLES Table 1 Demographics of Sample………………………………………...........55 Table 2 Univariate Statistics for all Variables…………………………...........56 Table 3 Intercorrelations Among for all Variables….…………………...........57 Table 4 Regression Analyses for all Variables………………………………...58
1 I. INTRODUCTION School reform efforts are focused, almost singularly, on academic achievement as the outcome measure for school effectiveness. This narrowed focus has resulted in an educational system which has failed to fully acknowledge the contextual and social-emotional factors that can also influence children’s development and subsequent school functioning. Fortunately, an increasing interest in positive psychology and positive youth development programs has generated attention to the development of positive indicators of children and adolescent well-being, including the overall life and school satisfaction levels of these adolescents (Baker, 1998; Huebner, 1994; Huebner, Gilman, Reschly, & Hall, 2009). Conceptual Overview of Life Satisfaction Life Satisfaction of Adults Life satisfaction (LS) has provided a useful framework for understanding the narrower concept of school satisfaction. Life satisfaction is conceptualized as a global construct and is understood as an individual’s cognitive evaluation of his or her current life circumstances, with this evaluation taking into account both internal and external factors (see Diener, 2000 for a review). In simple terms, LS is the scientific term for what people usually mean when they talk of their overall happiness with life.
2 Research on the topic of LS has demonstrated that most adults report that they are satisfied with their overall lives (Diener & Diener, 1996). In addition, research in this field has suggested that adults who experience higher levels of LS do better in life across the domains of work, love, and health (Lyubomirsky, King, & Diener, 2005). For instance, happy, satisfied people tend to receive higher job performance assessments from their employers, hold more prestigious jobs, earn higher incomes, are more likely to marry (and subsequently express more happiness with their marriages), and live longer than their “unhappier” counterparts (Cropanzano & Wright, 1999; Diener, Nickerson, Lucas, & Sandvik, 2002; Pressman, Cohen, & Kollnesher, 2006; Roberts, Caspi, & Moffitt, 2003; Ruvolo, 1998). Further, studies regarding LS in adults have demonstrated that individuals who experience the highest levels of LS report less psychological distress and more positive social relations (Diener & Seligman, 2002; Friedman, Schwartz, & Haaga, 2002). However, recent research has supported that this positive linear relationship between levels of LS and adaptive outcomes does not necessarily hold true across all domains, particularly on outcomes that require self-improvement motivation and analytical skills. According to Oishi, Diener, and Lucas (2007), individuals who were considered part of the second “happiest” group demonstrated the best outcomes in the domains of academic achievement, job performance, and income. That is, those individuals who were not at the highest level of happiness, but at the second highest level of happiness, performed better in those domains that required achievement motivation. Oishi, Diener, and
3 Lucas (2007) speculate that extreme levels of positive emotions are not always “optimal” for certain life outcomes, such as those that require self-improvement. Life Satisfaction of Children Regarding children and adolescents, LS has generally been studied from a global perspective where children and adolescents are asked to evaluate the positivity of their life as a whole (e.g., “overall, my life is going well”). The value and usefulness of these global evaluations with adults, children, and adolescents has been well-established (Gilman & Huebner, 2003; Huebner, 2004). Like adults (Diener & Diener, 1996), most children and adolescents have reported satisfaction with their overall lives (Gilman, Huebner, & Laughlin, 2000; Huebner, 1994; Huebner et al., 1998; Huebner et al., 1999). Adolescents who report that they are very happy have demonstrated generally positive functioning across intrapersonal, interpersonal, and school-related domains (Gilman & Huebner, 2006). Additionally, children and adolescent LS reports have been negatively correlated with a variety of risk behaviors (e.g.,, risky sex behavior, alcohol and drug use, etc.). That is, children who were unhappy with their lives demonstrated pervasive difficulties regarding aggressive behavior, internalizing behaviors, suicidal thinking, sexual risk-taking, alcohol and drug use, eating, physical health problems, and physical inactivity (see Huebner, Suldo, & Gilman, 2006, for a review). Further, preliminary studies have suggested that high adolescent LS can be protective and mediate the relationship between stressful life events and internalizing behaviors (McKnight et al., 2002), and moderate the
4 relationship between stressful life events and externalizing behaviors (Suldo & Huebner, 2004). School Satisfaction While the global LS provides rich information about a child or adolescents’ assessment of his or her life, studies of domain-specific evaluations or perceptions, such as satisfaction with family and peer relationships, have emerged and have proven to be another valuable indicator of subsequent adaptive functioning in children and adolescents (Huebner, 1994). For example, peer delinquency has been found to be inversely related to satisfaction with both school and families. This may indicate that assessing these specific domains would be quite beneficial in understanding child and adolescent functioning (Nickerson & Nagle, 2004). School satisfaction (SS) is defined as a student’s judgment of the positivity of his or her school experiences as a whole (Huebner, 1994). Out of the five domains of general LS, SS has often demonstrated the weakest correlation to overall LS. SS is related to LS generally only in the range of .30-.40 for US students (Huebner et al., 1998; Seligson, Huebner, & Valois, 2003). The accumulation of this research has suggested that SS may be a distinct construct in- and-of-itself. Further, the domain of SS has been found to be moderately stable when retested at two (.75) and four (.70) weeks, although less stable at one year (.47) (Huebner, Laughlin, Ash, & Gilman, 1998; Funk, Huebner, & Valois, 2004). These findings suggest that SS reports of adolescents may have some trait-like
5 properties but may also be sensitive to environmental changes (i.e., state-like properties). Measurement of School Satisfaction Studies of children’s perceptions of the quality of their life experiences, including SS, require psychometrically sound measurement tools. The development of valid measures appropriate for children and adolescents has historically lagged behind that of adults (Gilman & Huebner, 2003); however, current measures of SS exist that are appropriate for research. The first measure of SS was the Quality of School Life Scale (QSL: Epstein & McPartland, 1976). It is composed of 27 items that assess the three domains believed to be the most relevant to positive school experiences: satisfaction with school in general, commitment to school work, and attitudes toward teachers. Item response formats range from dichotomous true/false options to multiple response options. The QSL was originally administered to over 4,000 elementary, middle, and high school students in one U.S. state. Results have supported the psychometric properties of the scale, with adequate internal consistency for each of the three domains (generally ranging from .70 to .85), and moderate consistency across a one-year time span. Evidence for the convergent and discriminant validity of the scale was supported by demonstrating appropriate and expected positive correlations with school performance and negative correlations with school anxiety and poor school behaviors (e.g., cutting class/school). Finally, the QSL yielded evidence of construct validity via exploratory factor analytic methods. Subsequent studies
6 have generally supported these findings (Johnson & Johnson, 1993). Other measures have largely adapted the QSL, but have expanded the number of domains and applied them to students from different nationalities (Karatzias, Power, & Swanson, 2001; Mok & Flynn, 2002). Another example of a SS measure is one which is actually embedded within a larger multidimensional scale of overall well-being, the Multidimensional Students’ Life Satisfaction Scale (MSSLS: Huebner, 1994). The MSLSS assesses global LS as well as satisfaction with school, family, friends, living environment, and self, and has been validated for students in grades 3-12. The SS domain consists of eight items in which a student evaluates his or her satisfaction with school experiences as a whole. Student responses reflect their own, individual perceptions, rather than the criteria of the researchers (e.g., teacher relations, curricula, facilities). That is, students individually “weight” the criteria that determine their overall subjective experience of schooling without being limited to specific, a priori components (Diener, Emmons, Larsen, & Griffin, 1985). Items on the SS domain of the MSLSS are responded to on a 6- point rating scale. Current findings have revealed that the scale yielded strong psychometric properties, including internal consistency estimates of .80 or greater, adequate internal consistency across multiple time frames, and evidence of convergent and discriminant validity across adolescents representing different age and grade ranges (Gilman & Huebner, 2003). In addition to the SS domain on the MSLSS, measures of global SS are included within some well-validated instruments of a more general nature, such as
7 the Behavior Assessment System for Children (Reynolds & Kamphaus, 2004) which includes the Attitudes to Teachers and Attitude to School subscales. Both subscales have demonstrated strong psychometric properties and have been used in recent studies (Gilman & Anderman, 2006). In conclusion, psychometric evidence from the above mentioned SS instruments have demonstrated that positive school perceptions can be reliably and validly assessed among children and adolescents. Importance of School Satisfaction in Children School Satisfaction is a fairly stable construct that can be measured in a psychometrically sound manner. In addition to this, SS has also proven to be an important factor in understanding school functioning and children and adolescents’ quality of life. Most students have reported positive levels of domain specific satisfaction levels with family, friends, self, school, and living environment (Huebner, Drane, & Valois, 2000; Huebner, Valois, Paxton, & Drane, 2005). However, there has been some variability in satisfaction ratings across the five domains (family, friends, self, school, and living environment), with adolescents reporting most dissatisfaction with their school experiences. For example, in a large-scale study of high school students, nearly one-quarter of the students reported general dissatisfaction with their school experiences and an additional 9% described their school experiences as “awful” (Huebner, Drane, & Valois, 2000). These high levels of dissatisfaction with school are quite troubling because children who are very unhappy with their school experiences tend to display
8 pervasive adaptive difficulties, similar to those difficulties exhibited by children who report lower LS levels. These difficulties have included externalizing and internalizing behavior problems (DeSantis, 2006; Huebner & Gilman, 2006) such as suicidal ideation (Eamon, 2002; Locke & Newcomb, 2004), psychosomatic symptoms (Katja, Paivi, Marja-Terttu, & Pekka, 2002), substance use (Newcomb, Bentler, & Fahy, 1987; Oakley, Biannen, & Dodd, 1992, Strivastava & Strivastava, 1986), high school drop-out (U.S. Department of Education, 1990), and depression (Eamon, 2002). Students who dislike school have also shown similar maladaptive outcomes in academic behavior and functioning (Cock & Halvari, 1999; Epstein & McPartland, 1976; Gilman, 2001; Huebner & Gilman, 2006). Further, a longitudinal study conducted by Ladd, Buhs, and Seid (2000) demonstrated that kindergarten students’ school dissatisfaction preceded school disengagement behaviors, which in turn reduced their school achievement. In other words, early levels of SS appear to influence later school behaviors and achievement levels, rather than the converse. Because of the startling frequency of which students report dissatisfaction with their school experiences and the relationship between lower levels of SS and maladaptive functioning, SS may be a very important measure of children’s quality of life and school functioning and should not be overlooked. Correlates of School Satisfaction of Children School satisfaction is a burgeoning field that continues to need deeper investigation. Early research has demonstrated some consistent patterns in student-level correlates and contextual factors that can promote or hinder SS.
Student-Level Correlates Demographic Characteristics. Similar to findings in the larger LS literature, variance accounted for by demographic variables is relatively small and has generally ranged from about 2% (Ainley, Foreman, & Sheret, 1991) to 7% (Karatzias, Power, Flemming, Lennan, & Swanson, 2002). Okun, Braver, and Weir (1990) have suggested that satisfaction with school decreases throughout the school years. In a cross-sectional study of Arizona school students, the authors found that satisfaction with school decreased as children progressed through grades 1 through 8. In particular, it was noted that SS scores “bottomed-out” in eighth grade, and remained relatively low for the remainder of the school years (e.g., high school). In support of this finding, Karatzias et al. (2002) found that QSL, which is roughly equivalent to SS, decreased with grade among students in the United Kingdom. These differences, however, resulted in fairly small effect sizes and must be interpreted cautiously. Studies of gender differences have also yielded equivocal findings, with some studies revealing no gender differences (Bulcock, Whitt, & Beebe, 1991; Huebner & Gilman, 2006; Huebner, Ash, & Laughlin, 2001; Katja, Paivi, Marja- Terttu, & Pekka, 2002), and others revealing that girls reported significantly higher SS than boys (Eamon, 2002; Karatzias et al., 2002; Huebner et al., 2000; Okun, Braver, & Weir, 1990; Verkuyten & Thijs, 2002). However, when one considers that males are more likely to drop out of school (Ainley, Foreman, &
10 Sheret, 1991), to receive lower grades (Halpern, 1997), and to be placed in special education classes (Pollack, 1998); it could be hypothesized that males may be less satisfied with their school experiences. Several large scale studies investigating race differences among American samples have reported no differences between African-American and Caucasian students’ SS (e.g., Eamon, 2002; Epstein & McPartland, 1976) or between Native Americans, Hispanic Americans, and Asian Americans (Okun, Braver, & Weir, 1990). In contrast, however, cross-national differences have been reported. A recent study of adolescents in North America, Ireland, Croatia, Korea, and China found that students from Western countries reported lower SS than their counterparts from Eastern countries (Gilman et al., 2006). Again, it is important to note that the effect sizes reported in this study were characterized as “low” and must be interpreted with caution. Individual Characteristics. Demographic variables such as age, gender, and race appear to have a relatively small impact on SS reports. What appears to be more influential to SS reports are qualities and characteristics within students. For example, self-esteem has been shown to relate positively to SS, with effect sizes as high as .48 being reported (Karatzias et al., 2002). Huebner and Gilman (2006) also found that students reporting very low SS also reported significantly lower self-esteem than those students with either average or high levels of SS. Thus, there is empirical evidence for a relationship between SS levels and self- esteem.
11 In addition to self-esteem, academic self-efficacy has also demonstrated a relationship to SS. Further, Verkuyten and Thijs (2002) found, in a study of Dutch adolescents, that perceived scholastic competence (academic self-efficacy) mediates the relationship between academic performance and SS. That is, children who performed better educationally are more satisfied with school because better performances lead to a stronger sense of scholastic competence and efficacy. This was supported by Suldo, Shaffer, and Riley (2008) in a study of 321 high school students. They found that students’ personal academic beliefs (as measured by the School Attitude Assessment Survey-Revised; McCoach & Siegle, 2003) mediated the relationship between achievement and SS. Again, however, results in this area have been ambiguous. Huebner and McCullough (2000) found that high SES, Caucasian students with higher academic self-efficacy reported higher levels of SS, whereas Baker (1998) found a negative relationship between academic self- concept and SS among low SES, African American students. The evidence, up to this point, has suggested that the relationship between self-esteem and SS may vary for different groups of students (Huebner et al., 2001). A student’s locus of control, whether internal or external, has also shown to be related to levels of SS. Huebner, Ash, and Laughlin (2001) and Huebner and Gilman (2006) found that high school students with an internal locus of control were more satisfied with school than were those with external locus of control. Locus of control refers to the extent to which individuals believe that they can control events that affect them. Individuals with a high internal locus of control tend to believe that their efforts will be successful and exhibit high achievement
12 motivation (Rotter, 1966). Not surprisingly, the authors also found that internal locus of control mediated the relationship between negative life events and SS. Further, Gilman and Anderman (2006) suggested that students who reported that their academic successes were due to effort and ability, rather than through chance or luck, often reported higher SS. Related to the concept of locus of control is hope. Hope, a multidimensional construct consisting of motivational tendencies and goal strategies, has been found to correlate positively with SS. Huebner and Gilman (2006) found that students who reported very high SS reported significantly higher levels of hope as measured on Children’s Hope Scale (CHS: Snyder et al., 1997). In fact, the higher the student rated his or her level of SS, the higher he or she also rated his or her level of hope (Valle, Huebner, & Suldo, 2006). In fact, students who experience more positive emotions in general also tend to display more flexible and creative problem solving strategies and better cognitive skills (Bryan, Mathur, & Sullivan, 1996). Psychopathology. Finally, SS has a negative relationship to the presence of psychopathology. Eamon (2002) found that adolescents with high levels of SS exhibited fewer depressive symptoms than did students with low SS. Further, Huebner and Gilman (2006) demonstrated that adolescents with very low levels of SS also report higher rates of clinical scores on the BASC composites (ESI, Clinical Maladjustment, and Personal Adjustment). Epstein and McPartland (1976) also demonstrated a similar finding in their study which found that SS
13 related negatively to anxiety. In sum, these findings have demonstrated that SS is related to important differences in psychosocial and academic functioning. Behavioral Characteristics and Life Experiences. In addition to individual-level correlates, SS has been associated with important school and personal behaviors. Research on the relationship between student academic ability and achievement and SS has produced inconsistent results. Studies have shown that gifted students do not differ significantly in their levels of SS (Chapman & McAlpine, 1988), though gifted students’ SS has been shown to contribute a greater portion of the variance to their global LS reports than did that of non- gifted students (Ash & Huebner, 1998). That is, children who are “smarter” do not necessarily enjoy school more. More specifically, Cock and Halvari (1999) found a positive relationship between academic performance and SS, which may indicate that children who do well in school tend to be more satisfied with school. Huebner and Gilman (2006) found similar results with students at the highest level of school satisfaction reporting higher GPAs than those in the average and very low groups. However, Epstein and McPartland (1976) found that grades alone were not enough to predict SS, and this suggested that SS is related to how students feel about getting those grades rather than actually getting good grades (Verkuyten & Thijs, 2002). In sum, higher levels of academic achievement may not be a clear indicator of SS; rather it may be that students who feel good about their academic performance have higher levels of SS.
14 On a related note, Gilman (2001) and Huebner and Gilman (2006) found that students who reported higher levels of SS also tend to report higher involvement in structured extracurricular activities (SEAs). More recent findings have suggested that the relationship between SS and SEA participation may be mediated by motivational factors. For example, Gilman and Stacy (2007) found that the effect of moderate-to-high SEA participation on SS was significant only if such activities promoted feelings of individual autonomy, competency, and social relatedness. Activities that were selected to fulfill a course requirement, to satisfy a request from a parent, or participated in for reasons other than personal preference (e.g., because the student’s peer group was engaged in the activity) were less influential on SS reports. Further, SS appears to be related to students’ behavioral compliance in school. In a study conducted by Carey and Bourbon (2005) Australian students who reported higher levels of SS also self-reported lower rates negative reactions (e.g., aggression, frustration, anger, etc.) to teachers’ requests of compliance. Additionally, low satisfaction with school has been found to be associated with initiation and maintenance of marijuana, alcohol, and drug use in students followed from middle to high school (Newcomb & Bentler, 1987; Oakley, Biannen, & Dodd, 1992; Strivastava & Strivastava, 1986). In support of this, Locke and Newcomb (2004) also found that low satisfaction with school during adolescence predicted adult alcohol involvement.
15 Contextual Factors Teachers and Classrooms. Research has demonstrated a consistent, positive relationship between teacher supports, teacher-imposed classroom climate, and classroom structure with students’ engagement and participation in school (Furrer & Skinner, 2003; Suldo, Shaffer, & Riley, 2008; Wentzel, 1998). Student engagement and participation in school has been theorized to subsequently facilitate SS (Baker, Dilly, Aupperlee, & Patil, 2003). For example, Baker et al. (2003) suggests that teachers who provide a high degree of clarity in developing and enforcing classroom rules, and who create predictable classroom structures, facilitate SS. Further, a more recent study, Suldo, Shaffer, and Riley (2008) demonstrated that the quality of the student-teacher relationship was one of the most highly related aspects of school climate to students’ LS. That is, students who perceive positive relationships with their teachers also report higher levels of LS. Further, the social climate within the classroom appears to be important to SS. Baker, Davis, Dilly, and Lacey (2002), in their examination of classroom environments, found low to moderate effect sizes on the relationship between SS and the perception of the classroom as friendly, supportive, and free from harassment. They also found that SS is predicted by both teacher and students’ perception of the quality of teacher-student relationship. In addition, Baker (1998) found that classroom social climate had the strongest direct effect and largest total effect on SS in a population of impoverished, urban children. Mok and Flynn (2002) found similar results in a sample of adolescents from Australia and
16 reported that students’ experience of the classroom atmosphere was the single most important factor explaining quality of school life (QSL). Other classroom contextual factors that may contribute to high SS include establishing curricula that promote choice and autonomy (Karatzias et al., 2002), promoting a task-oriented classroom ethos (Baker, 1998), providing ample praise for appropriate behaviors (Baker, 1999), and emphasizing goals that promote future academic aspirations (Malin & Linnakyla, 2001). The contrary is also true, however, as evidenced by the fact that teachers who establish class structures that are overly controlling or give more attention to misbehaviors rather than good behaviors can diminish students’ SS (Baker, 1999; Carey & Bourbon, 2005). Research conducted in the area of job satisfaction is consistent with such findings. Job satisfaction, the adult analogue equivalent to SS, has been shown to be positively related to jobs that provide variety and autonomy, clear goals, informational feedback, and the work is perceived to be significant (Argyle, 1987). Classroom instructional methodology has also demonstrated an important relationship to SS. A recent study examined a variety of school courses from the perspective of Flow Theory (Shernoff, Csikszentmihalyi, Schneider, & Shernoff, 2003). “Flow” has been described as a state of deep absorption in an intrinsically interesting activity which contains challenges and skills that neither exceed, nor underutilize an individual’s abilities. Shernoff et al. (2003) found that the most engaging and satisfying moments for students occurred during instructional tasks that incorporated group work activities or individual activities that involved active
17 learning, as opposed to a lecture format. Although additional studies are required, these particular results provide a unique perspective on the type of instructional methods that may best foster meaningful and satisfying learning experiences. Peers. School satisfaction is also affected by social acceptance and peer support (Baker et al., 2003). Ash and Huebner (1998) found a significant positive correlation between SS and friend satisfaction, two factors on the Multidimensional Students’ LS Scale (MSLSS, Huebner, 1994). Also, students who had high conflict with their peers, particularly boys, reported liking school less (Ladd, Kockenderfer, Coleman, 1996). In fact, students whose teachers were able to promote positive peer interactions, reported positive attitudes toward their teachers and reported higher levels of overall SS (Baker, 1998; Verkuyten & Thijs, 2002). Further, not only have peer relationships and social support demonstrated a relationship to SS, but students’ friends have the potential to influence their opinions of school. Epstein (1981) found that students who had peers with positive attitudes about school were also more likely to have positive attitudes towards school. School. Currently, little research exists on the relationship between school structure and SS. The relationship between SS and school size is unclear because some studies have reported the relationship to be non-significant (Verkuyten & Thijs, 2002) and others studies have reported significant findings (Bowen, Bowen, & Richman, 2000; Malin & Linnakyla, 2001). School organizational patterns, however, have been reported to influence student attitudes. For example, schools that promote connectedness by maintaining continuity between teacher
18 and student cohorts across time (i.e., “looping”) have demonstrated increased positive attitudes towards schooling (Felner, Brand, Adan, Mulhall, Flowers, Sartain, & Dubois, 1993). This may be because this type of organizational practice promotes the students’ sense of connectedness to others (including the teacher) and enhances the predictability of the school environment. Rationale for the Current Study In recent history, school reform efforts have increasingly focused on academic achievement concerns, neglecting social-emotional concerns. However, it is important to understand how the process of schooling contributes to children’s psychological competence and later adjustment. School satisfaction reflects the degree to which a child has adjusted to the school environment, as well as whether the school is providing an environment which promotes psychological health and meets developmental needs. The research reviewed up to this point supports SS, not just academic achievement, as another equally important indicator of school functioning and even subsequent functioning later in life. Life satisfaction, a global appraisal of one’s life circumstances, serves as an important indicator regarding the psychological well-being of children and adolescents (Gilman & Huebner, 2003; Huebner, 2004). Although LS and SS have been shown to be two distinct constructs, LS provides an intuitive framework to understand SS. Life satisfaction demonstrates relationships to which similar to SS to positive outcomes in children and adolescents. Because of this, it could be argued that SS may operate similarly to LS across intrapersonal,
19 interpersonal, and school-related domains (Huebner, Gilman, & Suldo, 2006; Suldo & Huebner, 2006). There is a considerable amount of research which has indicated that LS leels of adults relate to a number of domains and contexts. In general, this body of research has supported the idea that there is a positive, linear relationship between an adult’s LS level and subsequent outcomes such as marriage, friendships, income, vocational success, and physical and mental health (Lyubormirsky, King, & Diener, 2005). However, recent research conducted by Oishi, Diener, and Lucas (2007) has demonstrated that the relationship between LS and certain outcomes, such as achievement (including income) and education, may be curvilinear; and suggests that across certain domains and contexts individuals can attain an “optimal level” of LS. In other words, regarding life outcomes that require self-improvement, such as academic achievement, job performance, and income, happier is not always better. Oishi, Diener, and Lucas (2007) have suggested that, at times, high levels of positive affect may prevent individuals from pursuing further positive change in achievement domains such as education and income. Research with adults has supported the theory that unpleasant states may motivate beneficial action. Consider job satisfaction, whose relevant analogue for children is SS. Job dissatisfaction could be thought of as a signal that the work environment does not fit an individual’s personality and skills. Thus, job dissatisfaction might motivate job change. The idea is that individuals who are dissatisfied may make efforts to change their circumstances to enhance their