من می خواهم ترک تحصیل کنم: مطالعه طولی استرس و خوش بینی به عنوان شاخص های قصد ترک تحصیل

Prior research on school dropout has often focused on stable person- and institution-level variables. In this research, we investigate longitudinally perceived stress and optimism as predictors of dropout intentions over a period of four years, and distinguish between stable and temporary predictors of dropout intentions. Findings based on a nationally representative sample of 16–20 year-olds in Switzerland (N = 4312) show that both average levels of stress and optimism as well as annually varying levels of stress and optimism affect dropout intentions. Additionally, results show that optimism buffers the negative impact of annually varying stress (i.e., years with more stress than usual), but not of stable levels of stress (i.e., stress over four years). The implications of the results are discussed according to a dynamic and preventive approach of school dropout.

Students who drop out of school often face negative consequences such as unemployment and lower standards of living (Belfield & Levin, 2007). It is therefore important to detect the social and psychological factors that first give rise to dropout intentions and eventually lead to actual dropout. In this study, we investigate dropout intentions longitudinally over a period of four years by focusing on both person-level predictors and predictors that may vary over time. In doing so, we want to better understand how temporary psychological factors influence dropout intentions over and above previously documented stable person characteristics and performance indicators. Predicting who drops out At the individual level, findings consistently show that poor academic and cognitive performance ( Alexander et al., 1997 and Balfanz et al., 2007) as well as deviant behavior (e.g., school absenteeism, substance abuse; Archambault et al., 2009, Battin-Pearson et al., 2000 and Ellenbogen and Chamberland, 1997) predict school dropout. Moreover, men and minority members (e.g., immigrants) are more likely to drop out ( Finn et al., 2005, Laird et al., 2007 and Rumberger, 1987), although this effect often disappears when controlling for academic performance, attitudes, and behaviors ( Rumberger, 1995). The impact of relevant attitudes seems more inconsistent: While low educational expectations are associated with dropout ( Driscoll, 1999), the association between self-perceptions (e.g., self-esteem, locus of control) and dropout is inconsistent and weak ( Alexander et al., 2001 and Rumberger, 1995). Recent studies in Finland and the U.S. have, however, shown that cynicism as well as lack of perceived control and identification with school predict dropout, even when controlling for educational performance ( Bask and Salmela-Aro, 2013 and Fall and Roberts, 2012). On the family level, low socio-economic status (SES) has been one of the most powerful predictors of dropout ( Dunham and Wilson, 2007 and Entwisle et al., 2004), along with family structure (e.g., parents living in different households, Perreira, Harris, & Lee, 2006) and being a first-generation student ( Ishitani, 2006). Additionally, student composition, resources, and classroom climate in schools have been shown to influence school engagement and dropout prevalence within a given educational context ( Dotterer and Lowe, 2011, Loeb and Page, 2000 and Rumberger, 1995). Research on the type of schooling (e.g., college-track vs. vocational training) is more ambiguous: Some studies show fewer dropouts in career-technical courses while others show no effect ( Bishop and Mane, 2004 and Pittman, 1991). Several models of dropout have integrated these different risk factors to explain student dropout (e.g., Appleton et al., 2008, Bean and Metzner, 1985, Spady, 1970 and Tinto, 1975). Tinto (1975), for example, focused on academic and social integration as factors leading to institutional commitment, persistence and goal engagement. Bean and Metzner (1985) revised this model for nontraditional students (i.e., older, non-residential, and part-time), stressing the role of environmental (e.g., financial situation, family responsibilities) and psychological variables (e.g., stress, satisfaction) in predicting dropout. Bean and Metzner (1985) also explicitly considered intention to drop out of education as “the strongest single predictor of dropout” ( Bean & Metzner, 1985, p. 527). This link between dropout intentions and actual dropout was subsequently confirmed by various studies ( Davis et al., 2002, Metzner and Bean, 1987, Sandler, 2000, Thomas, 2000 and Vallerand et al., 1997). Research specifically investigating dropout intentions has furthermore confirmed the predictive impact of psychological variables such as goal engagement and intrinsic motivation ( Braxton et al., 1995 and Otis et al., 2005), institutional commitment ( Braxton et al., 1995, Hausmann et al., 2007 and Zea et al., 1997), as well as social support ( Hausmann et al., 2007 and Thomas, 2000) that are all negatively associated with dropout intentions. Starting from the basic model of Bean and Metzner (1985), the present study examines the role of two neglected psychological variables—educational stress and optimism—as predictors of dropout intentions. Stress, optimism, and coping Neither stress nor optimism have received much attention in research on dropout intentions, although both may intervene in coping with the difficulties that push young people to leave school. In general, stress is defined as an imbalance between demands and resources (Lazarus, 1999). More specifically, we consider educational stress as the feeling of being overwhelmed by school demands. Investigating the role of educational stress for dropout is especially relevant as dropout intentions may be seen as a coping mechanism in response to stress (i.e., escape-avoidance coping, see Folkman & Lazarus, 1988). While the role of stress has already been investigated in relation with dropout intentions and actual dropout, the findings seem rather inconsistent. Chartrand (1992), for example found a significant impact of stress on dropout intentions, Sandler (2000) evidenced an indirect effect through institutional commitment, while others found no or only marginal effects ( Metzner and Bean, 1987 and Zajacova et al., 2005). However, stress fluctuates over time and therefore stress measured at one point in time may not be a good indicator for analyzing the association between general stress and dropout intentions, thus leading to inconsistent results. We therefore measure educational stress at different points in time and expect that individuals who are more stressed in general think more about dropping out of education. Positive psychological resources such as optimism may help individuals to deal with environmental demands (e.g., Katz, 1960). Optimism has rarely been included in studies on dropout intentions and dropout. This is surprising, as optimism has been shown to be associated with higher persistence in experimental tasks (Solberg Nes, Segerstrom, & Sephton, 2005) and higher goal engagement in longitudinal studies (Segerstrom & Solberg Nes, 2006). The association between optimism and persistence may be explained by the fact that optimism—defined as expecting “positive outcomes, even when things are difficult” (Scheier, Carver, & Bridges, 2001, p. 191)—is associated with approach-based and problem-focused coping (Scheier et al., 1986 and Solberg Nes and Segerstrom, 2006) as well as with perceived controllability of stressful events (Chang, 1998a and Scheier and Carver, 1985): Because optimists expect positive outcomes, they are more persistent in their efforts to achieve their goals than less optimistic individuals. Some studies indeed showed that optimism reduces both dropout intentions (Chemers, Hu, & Garcia, 2001) and actual dropout (Solberg Nes, Evans, & Segerstrom, 2009) and has a positive impact on college adjustment in general (Aspinwall & Taylor, 1992). Following this research, we expect that individuals who are more optimistic in general think less about dropping out than others, because they are better at coping with the demands of their school environment. Importantly, however, stress and optimism may also interact in predicting dropout intentions. Indeed, prior research has shown that optimism buffers the negative impact of stress on well-being and life satisfaction (e.g., Chang, 1998b). Optimism further appears to be associated with secondary appraisal (evaluation of own coping resources), but not with primary appraisal (evaluation of relevance of the situation) (Chang, 1998a and Lazarus, 1999). As a result, optimists do not necessarily experience less stress, but they believe to possess the necessary resources to cope with it, thus buffering the negative consequences of stress. Therefore, optimism should also buffer the negative impact of educational stress on dropout intentions. Predicting dropout intentions: from a fixed to a dynamic process Prior studies have often examined dropout as an event happening at a precise point in time, that is, they have not taken into account the possibility that dropout intentions may vary or develop over time. Finn (1989) has argued that dropout is a process rather than an event and describes dropping out “as a process of disengagement over time” (p. 133), a view shared by other researchers (Newmann, 1992 and Rumberger and Lim, 2008). One innovative contribution of the present study is the development and the test of a dynamic regulation model to investigate dropout as a process rather than as an event. To model longitudinal processes of dropout intentions, we include both educational stress and optimism as time-varying regulators of dropout intentions over time. So, in addition to including the general level of stress and optimism, we include the annually varying levels of stress and optimism to investigate their impact on dropout intentions within a given year. This analysis allows a better understanding of the longitudinal process of dropout intentions and its changes over time. As the level of stress related to education is likely to fluctuate between routine phases and work-intensive and challenging periods (e.g., exams, final school year), we expect that when individuals feel more stressed by their education than usual, they also think more about dropout, compared to less stressful periods. Additionally, although optimism is usually studied as a stable trait over time, it fluctuates over longer periods of time (Carver, Scheier, & Segerstrom, 2010), particularly during life transitions when outcomes become uncertain. As with general optimism, we expect that when individuals feel more optimistic than usual, they express less dropout intentions than during less optimistic periods. Context of the research The current study investigates dropout intentions of young adults in post-compulsory education in Switzerland over a period of four years. To put the study in context, some background information on the educational system in Switzerland is warranted. Switzerland has a federalist political system in which cantons have large autonomy over their educational systems, leading to considerable variation across Switzerland. Generally, compulsory schooling lasts 11 years (including kindergarten) with most students reaching 15 or 16 years at the end of compulsory schooling. At this stage, they can choose between a vocational or college-track education. Students in the vocational-track follow a 2–4 years professional training during which they usually enter the labor market as apprentices. Students in the college-track follow a general education program preparing them for higher education, in particular university. Data from 2006 show that about 65% of adolescents start vocational training, 30% start college-track education, about 3% find temporary solutions before starting post-compulsory education (e.g., internship), and only 2% do not start a new education (Keller & Moser, 2013). Data from 2013 show that 10% of 19 year-olds in Switzerland do not have a post-compulsory diploma, 8% of which started an education but dropped out. Students dropping out of college cite bad grades (73%) and motivational problems (43–59%) as reasons for their dropout, while vocational-track students most often report problems with their boss and teachers (54%) and motivational problems (43–50%, Keller & Moser, 2013). Studies show that students with migratory background, students from families with lower SES, students with a disrupted school career (e.g., repeating a school year) and those who attend schools with basic requirements (vs. more advanced requirements) are more likely to drop out of education than others (Gaupp et al., 2011 and Hupka-Brunner et al., 2011). Hypotheses By focusing on dropout intentions as a dynamic regulation process, stress and optimism are analyzed as time-varying regulators of dropout intentions over time. This approach extends prior research that has investigated dropout as a function of demographic background characteristics, performance indicators, and potentially time-varying characteristics (e.g., stress), but only at one point in time. We thereby go beyond the analysis of stable person characteristics (who drops out?) to time-varying indicators (when do individuals think of dropping out?). More precisely, we analyze dropout intentions over time as a function of both stable and temporary aspects of stress and optimism. Our specific hypotheses are as follows: 1. Over and above stable person characteristics and performance indicators (i.e., gender, country of birth, SES, educational activity, PISA score), average stress (hereafter person-stress) is positively associated with dropout intentions, while average optimism (hereafter person-optimism) is negatively associated with dropout intentions. 2. Controlling for average stress and optimism, annually varying levels of stress and optimism (hereafter annual-stress and annual-optimism) are associated with dropout intentions. 3. The negative effect of stress on dropout intentions is moderated by optimism such that stress has less impact on dropout intentions at high levels of optimism. In addition to these hypotheses, we test the link between dropout intentions and actual dropout to assess the predictive power of dropout intentions on dropout.