Gender Differences in Context: The Impact of Track Position on Study Involvement in Flemish Secondary Education

Abstract

This study examines whether the influence of track position on study involvement is gendered and whether gender differences in study involvement according to track position are associated with school misconduct and rather poor future perspectives. Three-level analyses (HLM 6) of data gathered in 2004-2005 from 11,872 third- and fifth-grade students in 146 tracks in a representative sample of 85 secondary schools in Flanders (Belgium) confirmed the impact of tracking on boys’ as well as girls’ study involvement. Boys are, generally, less involved in studying than girls, and boys are more affected by track position than girls are, enlarging the gender gap in the lower tracks. In these tracks, boys are more prone to misconduct and rather poor future perspectives. Finally, girls in arts tracks are, on average, more involved in studying than girls in academic tracks, but because of their higher tendency for disruptive behavior in school, this does not show.

Keywords

tracking gender ability grouping quantitative research on education secondary education

In the past few decades, gender differences in academic performance have generated numerous studies examining students’ demeanor in secondary schools. One of the reasons girls overall outperform boys is that girls tend to put forth greater effort than do boys (DiPrete and Buchmann 2013; Heyder and Kessels 2016). Boys are less motivated than girls and have less positive attitudes toward school (Francis 2000; Warrington, Younger, and Williams 2000). Girls spend more time doing homework, display less disturbing behavior, and are truant less often. Boys take it easier, do not work as hard, and are distracted more easily (Heyder and Kessels 2016; Warrington et al. 2000). High academic effort or displaying other positive school attitudes appears antithetical to typical masculine behavior, which is a condition of popularity for boys. Effortless achievement is characterized as the most masculine way of achieving (Heyder and Kessels 2016; C. Jackson 2002, 2003).

Research has examined possible nuances in these gender differences by disaggregating them along, for instance, race and social class (Gorard, Rees, and Salisbury 2001; Morris 2012). However, few studies deal with the context in which boys and girls form their attitudes and behavior, such as the school (Legewie and DiPrete 2012) or the educational system. Yet, students’ study involvement or willingness to exert effort might be related to features of the educational system. For instance, in the United Kingdom, boys’ and girls’“laddishness” can be understood partly as a response to their fear of failing the high-stakes tests (C. Jackson 2006). Similarly, many countries have a tradition of grouping students in secondary education according to their ability level. This ability grouping is organized in a myriad of ways (for an overview, see Maaz et al. 2008)—such as tracking in the United States or streaming in the United Kingdom—entailing entirely different curricula depending on students’ ability group. These different tracks are commonly classified hierarchically, placing technical and vocational tracks at the bottom of the ladder (Van Houtte 2006).

Since the late 1960s, research has established that lower-track students develop antischool attitudes to overcome the status deprivation resulting from being in a lower track (Ball 1981; Hargreaves 1967; Rosenbaum 1976). Consequently, belonging to a higher track positively influences academic achievement, and the reverse is true for belonging to a lower track (Carbonaro 2005; Hallinan 1994). However, little research has investigated gender differences in effects of track position on school attitudes and educational performance (for exceptions, see Catsambis, Mulkey, and Crain 1999; Van de gaer et al. 2006). This article investigates the interplay between gender and track position by examining whether gender differences in study involvement—that is, students’ willingness to put forth effort for school—is moderated by track position and whether track position influences boys’ and girls’ study involvement differently.

Study Involvement and Gender

One explanation for gender differences in educational performance is that girls may be more willing to exert effort in school than are boys (DiPrete and Buchmann 2013; Heyder and Kessels 2016). This willingness, or study involvement, can be seen as one of three interdependent components of student engagement (Fredricks, Blumenfeld, and Paris 2004): cognitive engagement, or students’ willingness to invest time and effort in mastering the subject matter; behavioral engagement (including misconduct); and emotional engagement (involving students’ well-being). Research points to the repercussions of these three subdimensions of disengagement. Behavioral, emotional (Janosz et al. 2008), and cognitive (Steinmayr and Spinath 2009) disengagement are, for instance, related to lower grades and a higher likelihood of dropping out, which emphasizes the need to gain insight into the determinants of the different dimensions of (dis)engagement.

Boys are more likely to reject school values, to trespass school rules (Demanet et al. 2013), and to be more laid back at school, which is attributed to the existence of a “laddish” culture (C. Jackson 2003; Warrington et al. 2000). Lad originally referred to white, working-class boys who rejected educational values (Willis 1977). Later, the term was applied to middle-class boys, too (Francis 1999). In a laddish culture, valuing studying is seen as typical for a feminine role set (C. Jackson 2002), that is, students and teachers perceive showing academic effort as less masculine and more feminine (Heyder and Kessels 2016). Consequently, boys who work hard at school might be ridiculed and feel pressure to conform to these macho images to remain popular (C. Jackson 2003; Warrington et al. 2000). Generally, girls tend to be more industrious at school, have a higher study involvement, are more motivated, and spend more time on homework (Francis 2000; Heyder and Kessels 2016; Van Houtte 2004). However, not all boys are equally susceptible to the values of laddish culture, and laddish behavior might spill over to girls, who are also at risk of becoming unpopular when peers notice them putting effort into school tasks (Heyder and Kessels 2016; C. Jackson 2006; Warrington and Younger 2000). School characteristics may affect the conception of masculinity in the school culture. A laddish culture is typically more overt in schools with a less favorable socioeconomic student composition, whereas in learning-oriented schools (e.g., schools focused on high performance) the idea that high grades are unsuitable for boys may be suppressed (Legewie and DiPrete 2012). Tracking may create similar situations.

Tracking and Study Involvement

The motives for tracking are twofold. First, it should be easier and more efficient to teach a group of students who are fairly homogeneous as to ability (Brunello and Checchi 2007; Hallinan 1994). Homogeneous classrooms permit a focused curriculum and appropriate instruction: teachers do not have to worry about losing the slowest learners or boring the fastest ones (Hanushek and Wöβmann 2006). Second, adolescents need to be prepared for different futures, so they need to learn different things (Oakes 2005; Schafer and Olexa 1971). Tracking usually entails offering students distinctive, internally coherent programs of study that are congruent with their scholastic interests and competencies and fitted to their anticipated educational and vocational needs. Hence, tracking accustoms students to their future positions in society and the economy (Trautwein et al. 2006).

For decades, sociologists have studied how tracking influences educational outcomes (Oakes 2005). Two main questions have been raised: (1) Is it preferable to be taught in homogeneous versus heterogeneous groups (Hallinan 1994)? and (2) Do all students take advantage of tracking (Catsambis et al. 1999; Hallinan 1994)? Homogeneous groups clearly entail no advantage, at least not for mediocre students. Research almost consistently finds that a higher track positively affects academic achievement, and the reverse is true for a lower track (Carbonaro 2005; Hallinan 1994). Studies carried out over time in different national contexts with different ways of organizing ability grouping, and controlling rigorously for initial ability and other student characteristics, confirm univocally that students achieve more success in higher-ability groups than in lower-ability groups (e.g., Duru-Bellat and Mingat 1997; Shavit and Featherman 1988; William and Bartholomew 2004).

Additionally, for over 30 years (Marsh and Parker 1984), psychological studies have described the big-fish-little-pond effect (BFLPE), whereby equally able students have lower academic self-concepts in high-ability schools/classes than in low-ability schools/classes (Marsh 1987; Trautwein et al. 2006). The BFLPE accounts for frame-of-reference-effects reasoning: students primarily compare their academic achievement with that of their schoolmates or classmates, and they use this social comparison as the basis for their academic self-concept (Trautwein et al. 2006). Salchegger (2016) shows that, under the condition of explicit school-level tracking, low achievers have higher mathematics self-concepts, and high achievers have lower mathematics self-concepts, than their individual ability predicts. So, explicit school-level tracking may preclude students’ awareness of their actual potential. Explicit school-level tracking increases the BFLPE regarding subject-specific mathematics self-concept, which is presented as only one among numerous consequences of explicit tracking, next to, for instance, an increasing variance in achievement (Salchegger 2016).

The mechanisms through which tracking contributes to this increasing achievement gap between higher- and lower-track students are complex and potentially numerous, ranging from different kinds of learning opportunities and instruction (Oakes 2005; Van Houtte 2006) to different kinds of peer influence (Schofield 2006). The differences in peer culture between educational environments in which high- and low-ability students are concentrated affect their (cognitive) engagement and, hence, their achievement in those environments (Schofield 2006; Van Houtte 2006). Lower-track students display less study involvement and exert less effort for school than do higher-track students (Carbonaro 2005; Catsambis et al. 1999; Van Houtte 2006). Moreover, they seem more behaviorally disengaged, as they are involved in more school misconduct than are higher-track students (Van Houtte and Stevens 2008).

Tracking, Study Involvement, and Gender

Response to present status loss

The way students’ involvement is related to track position might yield gender differences. First, lower-track students develop antischool attitudes to overcome the status loss resulting from being in a lower track (Hargreaves 1967; Rosenbaum 1976). In educational systems where curriculum placement is based on prior achievement (Trautwein et al. 2006), students’ lack of perceived ability pushes them into lower tracks, entailing a loss in status due to the hierarchical nature of the tracking system (Hargreaves 1967; Rosenbaum 1976). In hierarchically classified tracking systems, technical and vocational tracks are commonly placed at the bottom of the ladder. A technical or vocational training is usually not a truly positive choice but rather a second choice because one does not meet the standards set by academic tracks (Ainsworth and Roscigno 2005). Lower-track students oppose this educational system that makes them failures by refuting the values the system is based on, namely, effort and achievement, by displaying disruptive behavior in class and by looking for alternative bases for status, such as working (Catsambis et al. 1999; Oakes 2005; Van Houtte and Stevens 2008; Van Houtte and Stevens 2016).

Students’ (subject-specific) academic self-concepts are formed by comparing their performance within their track (BFLPE) (Marsh 1987), but social status within the system and the broader society is determined by comparison of track positions. In the case of explicit tracking programs, which have achievement as a selection criterion and have a profound impact on later educational and occupational opportunities, students’ tracking status is clear to the particular students, their parents and peers, and their teachers (Trautwein et al. 2006). Given the high value placed on education, our society tends to consider high-track membership the norm (Ireson, Hallam, and Plewis 2001). Students are very aware of the relative lower status associated with being in a lower track (Hallam and Ireson 2007; Stevens and Vermeersch 2010). Moreover, men place more importance on social comparisons than do women, perhaps due to socialization into a competitive masculine world characterized by extremes of inequality (Schwalbe and Staples 1991; Schwartz and Rubel 2005). Men seem more inclined to attribute importance to social status and prestige and to judge personal success through competence according to social standards (Schwartz and Rubel 2005). Boys’ general self-esteem is indeed affected by enrollment in a technical/vocational school, whereas girls’ general self-esteem is not (Van Houtte 2005). Hence, boys might suffer more from the status loss of being in a lower track and, consequently, be more likely to oppose the system by exerting less effort and displaying disruptive behavior.

Future opportunities

Second, students’ judgment of the future rewarding character of studying is important for their willingness to deliver effort at school. If students believe that studying will not pay off much, they will be less involved in studying and less likely to perform well (Mickelson 1989; Rosenbaum 2001). Technical/vocational tracks typically prepare students for occupations with little esteem, notwithstanding the need for well-skilled craftspeople. Moreover, deindustrialization and technological change have led to the collapse of demand for skilled, semiskilled, and unskilled (male) manual workers (Mickelson 1989; Nixon 2006). Additionally, in systems offering specific technical and vocational education, access to jobs is more restrictive for individuals who lack the required skills, reducing labor market opportunities for the least qualified (Wolbers 2007). All this puts vocational students in a disadvantaged position compared to technical students, who are at a disadvantage compared to academic students. In school systems with a school-based provision of vocational skills (versus apprenticeship), technical and vocational tracks suffer from a negative image, due to society’s overvaluing of cognition and white-collar jobs at the expense of manual labor. Given this undervaluation, lower-track students conceivably lose faith in the system and no longer see the point in studying or working hard for school. Lower-track students develop more negative attitudes toward school partly because they believe grades, commitment, and staying in school until graduation have little payoff (Schafer and Olexa 1971). Vocational-school students express lower levels of control over their future than do academic-school students (Malmberg and Trempala 1997), and they have stronger feelings of academic futility (Van Houtte 2016). This fatalism might negatively affect lower-track students’ willingness to deliver effort (Carbonaro 2005; Rosenbaum 2001).

Students’ school effort also likely differs by gender, because socioeconomic conditions have shifted drastically the past decennia, with a combination of (youth) unemployment due to economic recession and a modification of conventional gender arrangements (Gianettoni and Simon-Vermot 2010; D. Jackson 1998; Mahony 1998). Although both boys and girls feel the menace of being excluded from the labor market, this fear might particularly affect boys, given the prevailing male breadwinner model and, hence, a world of work still more associated with masculinity (Ciccia and Bleijenbergh 2014; Gianettoni and Simon-Vermot 2010; D. Jackson 1998; Mahony 1998; Morris 2008). The discourse of the male breadwinner is still prevalent, even though in reality this model has been superseded and is likely only for prosperous families able to support a stay-at-home housewife. Educational and professional progress is much less of a criterion in social evaluation of women than it is of men, making occupational status probably less salient for women (Mickelson 1989; O’Kelly 2002). These traditional, less egalitarian, assumptions about gender roles tend to be more typical for boys but also for adolescents with lower-educated parents (de Valk 2008), who are more often enrolled in lower tracks (Boone and Van Houtte 2013). Moreover, whereas technical and vocational education typically prepares boys for occupations with little social esteem, enrollment in a technical or vocational track leads girls to jobs that are fairly similar to women’s white-collar occupations in pay and prestige (Steunpunt WAV 2003). Replacement of the “masculine” manufacturing base by the “feminine” service sector and the collapse of the male breadwinner model have made boys’ prospects more insecure (D. Jackson 1998; Mahony 1998), especially boys in lower tracks. Thus, in lower tracks we should expect a higher degree of defeatism among boys in comparison to girls, generating less willingness in boys to make an effort.

Aspirations for higher education

Regarding future prospects, in a number of countries, enrollment in technical, and particularly vocational, education reduces the chance of attending tertiary education (Organisation for Economic Co-operation and Development 2014; Wolbers 2007). Academic tracks prepare students for higher education, but this is not the first goal of technical and vocational tracks; technical and vocational students are thus less likely to aspire to higher education and are therefore less engaged in school (Wang and Eccles 2012). Students who are not planning for higher education perceive that grades do not matter much for their labor market chances (Rosenbaum 2001). Additionally, boys are less likely than girls to aspire to higher education (DiPrete and Buchmann 2013; Mahony 1998). Recent work also notes that young men from working-class and minority ethnic backgrounds (Burke 2006)—who are overrepresented in lower tracks (Gamoran 2010)—have lower aspirations for higher education. The aspirations of these young men are closely tied to their masculine identities, with some indication that higher education may be seen as incompatible with working-class masculinity and roles (Burke 2006). These lower aspirations might affect boys’ study involvement as well.

Current Study

Research explicitly investigating gender differences in effects of track position on school attitudes and educational performance is exceptional. Catsambis and colleagues (1999) examine tracking effects on aspirations, engagement (e.g., homework), locus of control, and self-esteem, and they conclude that, generally, high-track students benefit from tracking, whereas low-track students are harmed. Moreover, the harm for low-track female students is greater than the benefit for high-track female students, except for locus of control, which is more strongly influenced in boys. Van de gaer and colleagues (2006) investigate several school-related attitudes and language achievement. In the vocational track, boys had a much worse relationship with their teachers than did girls, and girls’ motivation toward learning tasks was less positive. Both studies mention school opposition or disruption as possible explanations, but neither examines this.

The current study first examines the interaction between gender and track position in affecting study involvement. We hypothesize that the gender difference in study involvement will be larger in lower tracks, that is, track position will have a larger impact on boys’ study involvement than on girls’ (Hypothesis 1). Second, we hypothesize that the interaction between gender and track position will be explained by students’ disruptive behavior (Hypothesis 2). Third, we hypothesize that the interaction between gender and track position will be explained by ideas regarding future opportunities, represented by a sense of futility and aspirations for higher education (Hypothesis 3).

The Flemish Educational System

Flemish secondary education is characterized by stringent tracking (Van de gaer et al. 2006). Across Europe, educational institutions have witnessed the development of educational differentiation, although with great differences in degree (Bol and Van de Werfhorst 2013). Whereas countries such as Germany, Austria, the Netherlands, Belgium, Luxembourg, and some eastern European countries (Slovakia and Hungary) are now characterized by highly differentiated educational institutions, Scandinavian countries (Denmark, Finland, Sweden, Norway) and the United Kingdom have more comprehensive systems. Following Mons (2007), the Flemish system is an example of the separation model, characterized by rigid tracking and extensive grade repetition. Its opposite, the individualized integration model (Iceland, Norway), is comprehensive and grade repetition rarely occurs. The à la carte integration model (Australia, the United States, the United Kingdom) combines forms of ability grouping with low levels of grade repetition, and the uniform integration model (France, Spain) is comprehensive and uses grade repetition. Other classifications (e.g., Bol and Van de Werfhorst 2013) take into account the age of first selection, which is early in Flanders compared to other systems.

Flemish secondary education consists of six years divided into three “grades,” lasting two years each (Grades 7 to 12, ages 12 to 18). Although officially the first grade offers a core curriculum to allow students time to “orient” themselves to a particular track (Department of Education and Training 2008), in practice the kinds of courses offered depend on the main tracks in the final four years in the school. Thus, in reality, by attending a particular school, a track is already chosen in the first year, at the age of 12. Students are not allocated to a track but self-select into one, mainly based on their prior achievement in primary education and their social background. There are no centrally administered standardized tests. In making a choice, students and their parents rely on children’s regular school reports and primary school teachers’ advice (Boone and Van Houtte 2013). As in most European countries, pupils and their parents have to choose between mutually exclusive educational tracks at a fairly young age, leading to very different educational outcomes. Several studies in various European countries show that working-class parents are less likely than service-class parents to choose the more demanding, academic tracks, even if their children achieved equally well throughout primary school (for an overview, see Boone and Van Houtte 2013).

There are four main tracks: (1) academic education preparing for higher education; (2) technical education, which is more practical; (3) vocational education, which is craft-oriented; and (4) arts education, offering general academic courses combined with active art practice (e.g., music, drama, visual arts). The arts education track is rather marginal as to the number of students. Tracking is commonly organized between schools, distinguishing between schools for academic education and schools for technical and vocational education. Even in the multilateral schools that offer all types of tracks, students in different tracks rarely, if ever, have any lessons together, and they often do not even share the same buildings. It is very common to start in an academic track and drop to a lower track in case of failure. It is very unusual to move up. The academic, arts, and technical tracks provide access to higher education; for the vocational track, an additional (seventh) year is required.

Method

Data

The Flemish Educational Assessment was gathered in 2004-2005 from 11,872 third- and fifth-grade students (9th and 11th grades in the U.S. system) clustered in 146 tracks in a representative sample of 85 secondary schools in Flanders. First, we selected proportional-to-size postal codes, with the size for this purpose defined as the number of schools within the postal code, as gathered from data of the Flemish Educational Department. Hence, postal codes from large municipalities (with greater numbers of schools) had a greater chance of selection. From the 240 postal codes, we selected 48 with a desired slight overrepresentation of greater municipalities. Second, all regular secondary schools within these municipalities were asked to participate, yielding a positive response of 31 percent. This small response is due to the fact that Flemish schools are swamped with requests from investigators, resulting in a “first come, first served” strategy. Hence, the participating schools did not differ from those that opted out as to school sector, size, curriculum, or student composition. The 48 municipalities, as well as the 85 schools, are representative for Flanders. Students completed the questionnaires in class in the presence of one or two researchers and a teacher. Finally, 11,945 students completed a questionnaire, of which 11,872 proved to be valid (response rate of 87 percent).

Analytic Strategy

We examine bivariate associations between gender and track position as well as the central variables—study involvement, disruptive behavior, sense of futility, and aspirations for higher education—using t tests, ANOVAs, and Pearson correlations. We use a three-level multilevel analysis (HLM 6, method of estimation full maximum likelihood, with robust standard errors) of students (Level 1) clustered in tracks (Level 2), clustered in schools (Level 3). Variables are grand mean centered, except for the dichotomous variables for reasons of interpretation. We treat variance components randomly to start with but as fixed in the final models when nonsignificant. We do not investigate differences between schools, but in the final model we include three school features as a robustness check to ensure the tracking effects we find are not due to school features, because school systems can be stratified in ways that resemble tracking, such as along socioeconomic or ethnic lines, that might have a gendered effect (Legewie and DiPrete 2012; Stahl 2015). At the second level, we include type of track (academic, arts, technical or vocational). As is common in hierarchical linear modeling, we report unstandardized coefficients (γ); when relevant, however, we provide standardized coefficients (γ*) in the text.

We first estimate a null model—an unconditional model, without independent variables specified—to partition the variance between the three levels. Next, in the first model we examine the associations between gender and track position—academic, arts, technical, vocational—and study involvement, considering student characteristics that may vary across schools and tracks due to selection and that may also influence study involvement—namely, age, socioeconomic background, migrant status, prior grade point average (GPA), and parental involvement (Quin 2017). To ascertain whether the gender difference in study involvement varies according to track (Hypothesis 1), we add three cross-level interactions: the interaction of gender with enrollment in an arts track versus an academic track, gender with technical track versus academic track, and gender with vocational track versus academic track.

\begin{array}{l} Level 1 Model : Y = P 0 + P 1 \times (gender) \\ P 2 \times (prior GPA) + P 3 \times (socioeconomic status) \\ P 4 \times (parental involvement) \\ P 5 \times (migrant status) + P 6 \times (age) + E \end{array}

\begin{array}{l} Level 2 Model : P 0 = B 00 + B 01 \times (arts) \\ + B 02 \times (technical) + B 03 \times (vocational) + R 0 \\ P 1 = B 10 + B 11 \times (arts) + B 12 \times (technical) \\ + B 13 \times (vocational) + R 1 \\ P 2 = B 20 + R 2 \\ P 6 = B 60 \end{array}

\begin{array}{l} Level 3 model : B 00 = G 000 + U 00 \\ B 01 = G 010 \\ B 60 = G 600 \end{array}

In Model 2a, we add disruptive behavior to examine whether this might explain a varying gender gap in study involvement according to track (Hypothesis 2). In Model 2b, for the same reason, we add sense of futility (Hypothesis 3) and omit disruptive behavior. Model 2c adds students’ aspirations for higher education (Hypothesis 3). In Model 3, we include the three possible explanations for varying gender disparities in study involvement: disruptiveness, sense of futility, and aspirations. Model 4 includes the school-level controls, that is, the school’s size, socioeconomic status (SES) composition, and ethnic diversity.

\begin{array}{l} Level 1 Model : Y = P 0 + P 1 \times (gender) \\ + P 2 \times (prior GPA) + P 3 \times (socioeconomic status) \\ + P 4 \times (disruptiveness) + P 5 \times (futility) \\ + P 6 \times (parental involvement) + \\ P 7 \times (migrant status) + P 8 \times (age) + P 9 \times \\ (aspirations HE) + E \end{array}

\begin{array}{l} Level 2 Model : P 0 = B 00 + B 01 \times (arts) \\ + B 02 \times (technical) + B 03 \times (vocational) + R 0 \\ P 1 = B 10 + B 11 \times (arts) + B 12 \times (technical) \\ + B 13 \times (vocational) + R 1 \\ P 2 = B 20 + R 2 \\ P 9 = B 90 \end{array}

\begin{array}{l} Level 3 Model : B 00 = G 000 + G 001 (size) \\ + G 002 (SES - comp) + G 003 (diversity) + U 00 \\ B 01 = G 010 \\ B 90 = G 900 + U 90 \end{array}

Variables

Track level

We distinguish four types of tracks by dummy coding: 41 academic tracks (28.1 percent), 6 arts tracks (4.1 percent), 50 technical tracks (34.2 percent), and 49 vocational tracks (33.6 percent).

Student level

We measure study involvement with a scale of six items, such as “I don’t like to study” or “To me, a lot of things are more important than studying” (shortened from Brutsaert 2001). This instrument measures how concerned students generally are about going to school and studying. Items have five response categories ranging from absolutely not agree (1) to totally agree (5). We handle missing values with item correlation substitution—we replace a missing value for an item with the value of the item correlating most highly (range r = .360–.486) with that item (Huisman 1999), and we sum the item scores (range 6–30, Cronbach’s alpha = .76, N = 11,719). A low score indicates the student does not attach much importance to studying and considers school a waste of time. A high score indicates the student is enthusiastic about going to school and is interested in studying (1.33 percent missing).

The sample is equally divided regarding gender, with 51.4 percent females (female = 1) (0.24 percent missing). A slight underrepresentation of girls in technical tracks (45.3 percent) and an overrepresentation of girls in arts tracks (60.9 percent; see Table 1) corresponds with official figures that in 2004-2005, 43 percent of students in technical education and 63 percent of students in arts tracks were females (Department of Education 2005).

Table 1.

Tracking, Gender, and Study Involvement—Dependent and Independent Variables: Frequencies, Means, Standard Deviations, Comparison of Tracks and Gender.

	Total		Academic (n = 5,546)		Arts (n = 325)		Technical (n = 3,383)		Vocational (n = 2,618)		Difference
Variable, student level (N = 11,872)	M or %	SD	M or %	SD	M or %	SD	M or %	SD	M or %	SD	F or Cramer’s V
Female (N = 11,843)	51.4		53.7		60.9		45.3		53.1		0.081***
Study involvement (N = 11,714)	19.41	4.04	19.85	3.69	19.95	3.74	19.02	4.02	18.91	4.68	47.32***
Boys	18.62	4.19	19.17	3.86	18.99	4.26	18.14	4.07	18.13	4.84	28.93***
Girls	20.17	3.74	20.43	3.44	20.56	3.24	20.09	3.67	19.61	4.42	15.65***
t test	–21.01***		–12.64***		–3.53***		–14.54***		–8.04***
Cohen’s d	0.39		0.35		0.42		0.50		0.32
Disruptive behavior (N = 11,561)	30.04	8.47	28.84	7.14	31.76	9.02	31.04	8.86	31.15	10.07	72.05***
Boys	31.85	9.41	30.34	8.01	33.73	9.98	32.85	9.71	33.34	11.05	40.81***
Girls	28.31	7.04	27.54	5.99	30.49	8.11	28.84	7.13	29.14	8.59	27.16***
t test	22.74***		14.44***		3.19**		13.65***		10.47***
Cohen’s d	0.43		0.40		0.36		0.47		0.42
Futility (N = 11,615)	10.00	3.20	9.40	2.87	10.16	3.14	10.29	3.26	10.90	3.56	142.96***
Boys	10.17	3.31	9.52	2.98	10.28	3.20	10.49	3.30	11.06	3.72	69.16***
Girls	9.84	3.09	9.31	2.76	10.09	3.11	10.05	3.19	10.76	3.40	72.84***
t test	5.48***		2.68**		0.53		3.87***		2.12*
Cohen’s d	0.10		0.07		0.06		0.14		0.08
Aspirations for HE (N = 10,758)	70.0		87.9		68.9		62.1		35.6		0.435***
Boys	65.8		85.3		66.1		56.8		30.3		0.436***
Girls	74.1		90.1		70.7		68.3		40.1		0.437***
Cramer’s V	0.091***		0.073***		0.049		0.118***		0.102***
Age (N = 11,803)	16.45	1.30	16.00	1.12	16.71	1.40	16.78	1.33	16.94	1.30	471.27***
Boys	16.50	1.37	15.99	1.15	16.93	1.52	16.80	1.40	17.07	1.36	252.08***
Girls	16.40	1.24	16.00	1.10	16.58	1.31	16.76	1.25	16.83	1.24	220.38***
t test	4.19***		–0.456		2.22*		0.98		4.80***
Cohen’s d	0.08		0.01		0.25		0.03		0.18
SES (N = 11,137)	5.20	2.10	6.01	1.79	5.65	1.77	4.93	1.95	3.57	2.02	914.22***
Boys	5.31	2.09	6.15	1.73	5.89	1.60	5.06	1.96	3.66	2.05	451.72***
Girls	5.10	2.11	5.89	1.82	5.51	1.86	4.78	1.94	3.50	1.99	471.26***
t test	5.36***		5.42***		1.97*		4.02***		1.85
Cohen’s d	0.10		0.15		0.22		0.14		0.08
Migrant (N = 11,870)	11.1		5.5		4.0		9.9		25.6		0.250***
Boys	11.0		4.7		3.2		9.5		27.1		0.277***
Girls	11.3		6.2		4.5		10.4		24.2		0.228***
Cramer’s V	0.006		0.034*		0.034		0.015		0.033
Prior GPA (N = 10,697)	77.96	9.85	82.98	7.21	75.88	9.32	74.84	8.45	70.12	10.63	1370.09***
Boys	78.09	10.02	82.96	7.44	76.27	10.51	75.27	8.84	71.23	11.42	507.11***
Girls	77.84	9.68	83.00	7.01	75.64	8.56	74.32	7.90	69.05	9.69	926.37***
t test	1.29		–0.22		0.54		3.19***		4.61***
Cohen’s d	0.03		0.01		0.07		0.11		0.21
Parental involvement (N = 11,529)	31.36	5.40	31.96	5.05	31.07	5.58	31.15	5.37	30.36	5.94	53.25***
Boys	31.42	5.29	32.07	4.99	30.75	5.77	31.11	5.27	30.58	5.71	25.25***
Girls	31.30	5.49	31.86	5.09	31.26	5.47	31.19	5.51	30.17	6.11	29.62***
t test	1.24		1.47		–0.79		–0.44		1.74
Cohen’s d	0.02		0.04		0.09		0.02		0.07

Note: HE = higher education; SES = socioeconomic status; GPA = grade point average.

p≤ .05. **p≤ 0.01. ***p≤ .001.

We measure disruptive behavior, that is, behaviors likely to be disruptive to the school environment or to engender punishments (Stewart 2003), with a 17-item scale with five response categories ranging from never (1) to very often (5). These items complete the question, “How often have you . . .” and cover a range of school misconduct, from “been late for school” to “smoked on school grounds” to “been suspended.” Self-reports of delinquent behavior entail the risk of socially desirable answers, but this is the most common method for gathering such information (Crosnoe 2002). We impute responses for missing values using item correlation substitution (range r = .394–.602) (Huisman 1999; Cronbach’s alpha = .87, N = 11,566), and we use respondents’ total scores (range 17–85, 2.62 percent missing).

We measure sense of futility with a five-item scale adapted from Brookover and Schneider (1975), with items such as “People like me will never do well in school even though we try hard.” Items have five response categories, from absolutely not agree (1) to totally agree (5). We impute responses for missing values with item correlation substitution (range r = .264–.561) (Huisman 1999), and we sum the item scores (range 5–25, Cronbach’s alpha = .75, N = 11,620, 2.17 percent missing).

Students were asked whether they planned to go on to higher education after secondary education, and 70 percent answered they did (aspirations for higher education = 1). The other 30 percent did not plan to go on or did not know yet (9.38 percent missing).

Most respondents were age 15 (third grade, 34.8 percent) or 17 (fifth grade, 32.6 percent) years in 2005 (0.58 percent missing). The oldest were 20 years or older (1.4 percent), and the youngest was 13 (one respondent). On average, respondents were 16.45 years old (SD = 1.30), with students in technical (16.78) and vocational (16.94) tracks significantly older than students in academic tracks (16.00), due to their higher rate of grade repetition.

We measure students’ SES using the occupational prestige of their father and mother (Erikson, Goldthorpe, and Portocarero 1979; EGP scale); the highest of both is an indicator of family SES (range 1–8). We asked respondents to describe their father’s and mother’s occupations; from this description, the two principal investigators responsible for data gathering coded the occupations according to the EGP scale, recognizing the Flemish occupational context. Respondents had a mean SES of 5.20 (6.19 percent missing). Academic students had a significantly higher SES (6.01) than did arts students (5.65), who scored significantly higher than technical students (4.93), who scored significantly higher than vocational students (3.57).

We distinguish between native and immigrant students (migrant status = 1). As is common in Flanders (Timmerman, Hermans, and Hoornaert 2002), the principal criterion was birthplace of a student’s maternal grandmother. If missing (only 1 percent of the total sample, N = 11,872), we consider mother’s and father’s nationalities, as most immigrant students are second or third generation and have Belgian nationality. Moreover, as is common, we consider non–west European birthplaces and nationalities foreign descent, because students with these backgrounds are more likely to confront educational difficulties (0.02 percent missing). As expected, we see more immigrant students in the technical (9.9 percent) and vocational (25.5 percent) tracks than in the academic (5.50 percent) and arts (4.0 percent) tracks.

We measure parental school involvement using a 10-question index (inspired by Muller 1998; Rumberger 1995). For eight questions—like “Do your parents keep an eye on your homework?”— students could respond with five categories, ranging from never (1) to always (5). The other two questions probe parents’ membership on a parents’ council and their acquaintance with fellow students’ parents. The total score on these 10 questions defines parental school involvement (range 10–45, 2.89 percent missing). The mean score is 31.36, and academic students report significantly higher parental involvement (31.96) than do arts (31.07), technical (31.15), and vocational (30.36) students. Arts students do not differ significantly from technical and vocational students, but vocational students score significantly lower than technical students.

For prior academic attainment, we use GPA at the end of primary education. This measure should be considered cautiously. We rely on self-reported GPA, yielding questions on validity due to memory problems and cover-up strategies. Not unexpectedly, this variable shows more missing values than do other variables (9.90 percent missing). Notwithstanding these faults, it is our best measure for prior academic attainment, as Flanders has no standardized tests. Mean GPA is 77.96. On average, academic students report a higher GPA (82.98) than do arts (75.88), technical (74.84), and vocational (70.12) students, and vocational students score significantly lower than do their peers in the other three tracks.

School level

The total number of students enrolled in a school, as reported by the school administration, represents the school size (M = 455.69, SD = 285.00). We measure each school’s SES context by calculating respondents’ mean SES (M = 4.80, SD = 1.23).

We measure a school’s ethnic diversity with the Herfindahl index (Van Houtte and Stevens 2009). The index has a range of −1 to 0; a value of −1 implies no diversity at all, meaning only one ethnic group is enrolled in the school. A value approaching zero means total diversity: all students have a different ethnic origin. One school had a value of –1, indicating total homogeneity. The school with the highest diversity had a value of –.18 (M = –.67, SD = .23).

Results

The t tests (see Table 1) show that girls and boys differed significantly on all central variables. On average, girls scored higher than boys on study involvement (respectively, 20.17 and 18.62; Cohen’s d = 0.39, p < .001) and lower on disruptive behavior (respectively, 28.31 and 31.85; Cohen’s d = 0.43, p < .001) and sense of futility (respectively, 9.84 and 10.17; Cohen’s d = 0.10, p < .001). Significantly more girls (74.1 percent) than boys (65.8 percent) said they aspired to higher education. We found these gender differences in each track, except for sense of futility and aspirations, which did not differ significantly for girls and boys in the arts tracks.

Comparing the four tracks using ANOVA reveals significant differences for the central variables for girls and boys (Table 1). Post hoc tests (Scheffe, not shown) reveal that students in academic tracks had significantly higher study involvement (M = 19.85) than did students in technical (M = 19.02, Cohen’s d = 0.22) and vocational (M = 18.91; Cohen’s d = 0.22, p < .001) tracks, but they did not differ from students in arts tracks (M = 19.95; Cohen’s d = 0.03). Technical and vocational students did not differ significantly either. As for disruptiveness, post hoc tests show that academic students (M = 28.84) differed significantly from arts (M = 31.76; Cohen’s d = 0.36, p < .001), technical (M = 31.04; Cohen’s d = 0.27, p < .001), and vocational (M = 31.15; Cohen’s d = 0.27, p < .001) students, but arts, technical, and vocational students did not differ significantly between tracks. Academic students had a significantly lower sense of futility (p < .001) than did arts (Cohen’s d = 0.25), technical (Cohen’s d = 0.29), and vocational (Cohen’s d = 0.46) students, and vocational students had a significantly (p < .001) higher sense of futility than did academic (Cohen’s d = 0.46), arts (Cohen’s d = 0.22), and technical (Cohen’s d = 0.18) students. Arts and technical students did not differ significantly in this respect.

Correlations between the student variables (see Table 2) were mostly significant, but a number of them were rather small. We found no strong correlations indicating redundant variables. Gender was modestly positively correlated with study involvement (r = .191, p < .001) and negatively with disruptive behavior (r = –.208, p < .001), and it was weakly negatively correlated with sense of futility (r = –.051, p < .001) and positively with aspirations (r = .091, p < .001). Study involvement was moderately and negatively correlated with disruptive behavior (r = –.419, p < .001) and sense of futility (r = –.333, p < .001) and modestly and positively correlated with aspirations (r = .216, p < .001).

Table 2.

Correlation Matrix Variables, Student Level.

Variable	1	2	3	4	5	6	7	8	9
1. Gender	—
2. Study involvement	.191***	—
3. Disruptive behavior	–.208***	–.419***	—
4. Sense of futility	–.051***	–.333***	.194***	—
5. Aspirations for higher education	.091***	.216***	–.121***	–.226***	—
6. Age	–.039***	–.001	.227***	.054***	–.073***	—
7. Socioeconomic status	–.051***	–.001	.009	–.108***	.264***	–.116***	—
8. Migrant status	.006	.144***	–.012	.031***	–.073***	.163***	–.355***	—
9. Prior grade point average	–.012	.061***	–.047***	–.170***	.292***	–.143***	.277***	–.159***	—
10. Parental involvement	–.012	.256***	–.281***	–.179***	.123***	–.235***	.120***	–.044***	.077***

***

p≤ .001.

The unconditional multilevel model shows that most of the variance in study involvement is situated between students (92.07 percent, p < .001); 5.03 percent is at the school level (p < .001) and 2.90 percent at the track level (p < .001).

In Model 1 of Table 3, all student characteristics appear significantly (p < .05) related to study involvement, except for prior GPA. Girls, older students, lower-SES students, migrant students, and students reporting higher parental involvement display higher levels of study involvement than do, respectively, boys, younger students, higher-SES students, nonmigrant students, and students reporting lower levels of parental involvement. Parental involvement seems to be the most important determinant (γ = 0.208, standardized γ* = 0.278), followed by gender (γ = 1.113, γ* = 0.138). At the track level, technical (γ = −0.979, γ* = −0.116; p < .001) and vocational (γ = −1.630, γ* = −0.190; p < .001) students were significantly less involved in studying than were students in academic tracks, net of individual characteristics. Arts students did not differ significantly from academic students regarding study involvement. The main effects confirm the associations of gender and track position with study involvement.

Table 3.

The Association between Gender, Tracking, and Study Involvement; Results of Stepwise Three-level Analyses (HLM 6) with Study Involvement as Outcome.

Variable	Model 1	Model 2a	Model 2b	Model 2c	Model 3	Model 4
Intercept	19.118***(0.128)	19.225***(0.109)	18.996***(0.114)	18.095***(0.126)	18.274***(0.106)	18.337***(0.131)
School level
Size (range 26 to 1,124)						–0.000(0.000)
SES context (range 2.00 to 6.72)						0.158(0.096)
Diversity (range −1.00 to −0.18)						1.805***(0.380)
Track level
Arts (ref. academic = 0)	–0.419(0.591)	–0.283(0.694)	–0.381(0.497)	–0.164(0.524)	–0.058(0.621)	–0.105(0.592)
Technical (ref. academic = 0)	–0.979***(0.178)	–0.731***(0.170)	–0.795***(0.160)	–0.558***(0.165)	–0.362*(0.152)	–0.342*(0.156)
Vocational (ref. academic = 0)	–1.630***(0.217)	–1.264***(0.213)	–1.206***(0.211)	–0.721**(0.222)	–0.546*(0.215)	–0.627**(0.223)
Student level
Gender (ref. male = 0)	1.113***(0.110)	0.663***(0.100)	1.089***(0.102)	1.048***(0.101)	0.692***(0.092)	0.702***(0.089)
Age (range 13 to 20)	0.253***(0.036)	0.426***(0.033)	0.222***(0.034)	0.208***(0.034)	0.367***(0.034)	0.367***(0.038)
SES (range 1 to 8)	–0.047*(0.020)	–0.006(0.020)	–0.067***(0.020)	–0.096***(0.020)	–0.050*(0.020)	–0.048*(0.020)
Migrant status (ref. native = 0)	1.660***(0.180)	1.526***(0.182)	1.628***(0.179)	1.486***(0.186)	1.454***(0.188)	1.296***(0.188)
Parental involvement (range 10 to 45)	0.208***(0.009)	0.147***(0.008)	0.174***(0.009)	0.172***(0.009)	0.124***(0.008)	0.124***(0.009)
Prior GPA (range 41 to 100)	0.009^†(0.005)	0.010*(0.005)	–0.004(0.005)	–0.008(0.005)	–0.005(0.005)	–0.005(0.005)
Gender × Arts	0.552(0.518)	0.808^†(0.478)	0.745(0.538)	0.787(0.511)	0.946*(0.450)	0.936*(0.441)
Gender × Technical	0.586**(0.207)	0.483*(0.226)	0.524**(0.197)	0.464*(0.194)	0.347^†(0.205)	0.311(0.189)
Gender × Vocational	0.636*(0.250)	0.468^†(0.255)	0.482*(0.233)	0.389(0.246)	0.302(0.251)	0.327(0.242)
Disruptive behavior (range 17 to 85)		–0.173***(0.007)			–0.155***(0.007)	–0.155***(0.007)
Sense of futility (range 5 to 25)			–0.345***(0.014)	–0.315***(0.015)	–0.260***(0.014)	–0.260***(0.014)
Aspirations for higher education (ref. no = 0)				1.112***(0.101)	0.975***(0.098)	0.983***(0.099)
Variance components
Intercept R0	0.208***	0.192***	0.149***	0.127***	0.125***	0.126***
Gender	0.180**	0.230*	0.111*	0.111*	0.167*	0.162*
Migrant status	0.720^†	0.975***	0.591	0.729^†	1.017**	0.697***
Parental involvement	0.003**	0.002*	0.003***	0.003***	0.002**	0.002**
Prior GPA	0.001^†	0.001	0.001*	0.001	0.001	0.001
Sense of futility			0.004***	0.006***	0.007***	0.007**
Disruptive		0.003***			0.003***	0.003***
Intercept U00	0.232***	0.214***	0.151***	0.135***	0.160***	0.022
Age						0.038***
Migrant status						0.311*
Aspirations for higher education						0.138*
Deviance	53307.09	51282.22	51978.31	47904.18	46611.12	46576.55
Parameters	30	37	37	38	46	58

Note: Unstandardized gamma (γ) shown with standard errors in parentheses. GPA = grade point average; ref. = reference; SES = socioeconomic status.

†

p < .10. *p≤ .05. **p≤ .01. ***p≤ .001.

Confirming Hypothesis 1, the cross-level interactions between gender and track position show that the association between gender and study involvement is significantly larger in the technical and vocational tracks compared to the academic track (if γ for gender in the academic track is 1.113, it is 1.113 + 0.586 = 1.699, p < .01, in the technical track and 1.113 + 0.636 = 1.749, p < .05, in the vocational track). The gender difference in study involvement is larger among technical and vocational students compared to academic students, because enrollment in these lower tracks had a larger impact on boys’ study involvement than it had on girls’ (if γ for technical track for boys is –0.979, it is –0.979 + 0.586 = –0.393, p < .01, for girls; and if γ for vocational track for boys is –1.630, it is –1.630 + 0.636 = –0.994, p < .05, for girls).

In Model 2a in Table 3, disruptive behavior is significantly related to study involvement (γ = –0.173, p < .001), with more disruptiveness associated with less study involvement. By adding this variable, the main associations of track position and gender with study involvement decrease a little, compared with Model 1, but remain significant (p < .001). The coefficients of the cross-level interactions decrease, too, and the interaction between gender and being enrolled in a vocational versus an academic track turns insignificant at the 5 percent level (p = .068). This indicates that the larger gender gap in study involvement in vocational tracks compared to academic tracks might be explained by higher levels of disruptiveness in boys in vocational tracks, partly confirming Hypothesis 2.

In Model 2b of Table 3, we add sense of futility and leave out disruptive behavior. Sense of futility is significantly related to study involvement (γ = –0.345, p < .001), with more sense of futility associated with less study involvement. By adding this variable, the main associations of track position and gender with study involvement, and the coefficients of the cross-level interactions, decrease a little compared to Model 1 but remain significant. Sense of futility thus does not explain the larger gender differences in study involvement in technical and vocational versus academic tracks. Adding students’ aspirations for higher education, however, changed the picture (Table 3, Model 2c). Aspiring to higher education is positively and significantly (γ = 1.112, p < .001) associated with study involvement. Its addition decreases the main associations of being enrolled in technical and vocational versus academic tracks compared to Model 1, but these associations remain significant. The association between gender and study involvement remains largely unchanged, but the cross-level interaction of vocational track and gender turns insignificant. The interaction of technical track and gender decreases too, but remains significant at the 5 percent level, partly confirming Hypothesis 3.

When including the three mediators (Table 3, Model 3), the main associations of technical and vocational track position and gender decrease substantially but remain significant (at 5 percent and 0.1 percent, respectively). The cross-level interactions of technical and vocational track with gender decrease and turn insignificant (p > .05), suggesting that the larger gender differences in study involvement in technical and vocational tracks versus the academic track might be due to a combination of greater levels of laddishness and boys’ lower prospects for their future. The cross-level interaction of the arts track and gender with study involvement increases and becomes significant (γ = 0.946, p < .05), indicating a stronger association between gender and study involvement in the arts tracks (γ = 0.692 + 0.946 = 1.638) compared to academic tracks (γ = 0.692). A closer examination of the coefficients reveals that enrollment in an arts versus an academic track has no significant effect for boys’ study involvement (–0.058, p > .05); for girls’ study involvement, however, the effect is larger, significant (γ = –0.058 + 0.946 = 0.888, p < .05), and positive. Hence, girls’ higher study involvement in arts tracks is responsible for the larger gender difference in study involvement in arts tracks compared to academic tracks. Importantly, we found girls’ higher study involvement only when accounting for the combination of disruptiveness, futility, and aspirations—although already in the second model, when including disruptiveness, the cross-level interaction of gender and arts track with study involvement increases and turns significant at the 10 percent level (p = .093). Girls in arts tracks are indeed more involved in studying than girls in academic tracks; this did not show, though, because they are also more involved in disruptive behavior, which is negatively associated with study involvement and therefore suppresses the positive impact of being enrolled in an arts track compared to an academic track for girls’ study involvement.

Adding school-level controls (Table 3, Model 4) reveals no substantive changes in the findings (compared with Model 3). At the school level, only diversity is positively and significantly related to study involvement (γ = 1.805, p < .001).

Discussion

To understand why and how boys and girls behave in certain ways in school, it might be enlightening to see their demeanor as a response to specific characteristics of the educational system. In many countries, for secondary education, students are grouped into distinct tracks that are ordered hierarchically and esteemed differently in society. This study examined whether eventual gender differences in study involvement according to track position are associated with antischool attitudes, as manifest in disruptive behavior, or with feelings of futility and lower aspirations for higher education.

Our analyses all show that study involvement is determined mainly by student features: the variance in study involvement is only to a very small extent situated between tracks and schools. Students’ gender, age, socioeconomic background, migrant status, and parental involvement are significantly related to study involvement. The negative association of study involvement with socioeconomic background, and the positive association with migrant status, seems surprising. However, although the associations are quite weak, there is conflicting evidence on the association between student engagement and minority status and SES (Kelly 2008). Moreover, study involvement, as measured here, reflects a more abstract, general attitude toward education. The attitude–achievement paradox (Mickelson 1990) indicates that migrant students tend to have positive general (abstract) attitudes toward education and schooling, but their concrete attitudes, representing their opinions about the role of schooling for their personal futures, tend to be far more pessimistic and ultimately determine their achievement. The most important predictor of study involvement appears to be parental involvement, which concurs with research showing that parenting behaviors and parental support are strongly related to student engagement across ethnic and socioeconomic groups (Sharkey, You, and Schnoebelen 2008).

Nevertheless, the finding that the variance in study involvement is only to a small extent situated between tracks does not mean track position can be neglected, as it might affect the association between student features and study involvement. Regarding gender, the analyses confirm the impact of tracking on study involvement of both boys and girls. Regarding the effects of track position, the similarity between boys and girls is striking. Students enrolled in a technical or vocational track tend to be less willing to exert effort in school compared to their peers in academic tracks. Longitudinal research is needed to determine whether track position affects study involvement, or whether less involved students select into lower tracks.

This study also confirms that boys generally seem less involved than girls in studying, and gender differences in study involvement vary according to track position, with boys in technical and vocational tracks even less willing to make an effort. Boys appear more affected by track position than do girls, enlarging the gender gap in study involvement in the lower tracks. The gender differences in study involvement are stronger in technical and, particularly, vocational tracks, likely because boys in these tracks are more prone to school misconduct and see their future prospects as rather poor. Yet, although displaying disruptive behavior, having higher feelings of futility, and having lower aspirations are associated with less study involvement, these features only partly explain the impact of being in a technical or vocational versus an academic track on students’ study involvement. Tracks, and schools offering distinct tracks, differ from each other in terms of prevailing cultures—that is, students’ shared beliefs, such as futility culture—and these cultures influence students’ outcomes irrespective of students’ own feelings, attitudes, or behaviors (Van Houtte and Stevens 2016). Therefore, future research should consider the student cultures characterizing different tracks to determine whether students’ study involvement is associated with, for example, their track’s futility or aspiration culture, irrespective of their own sense of futility or aspirations. This path might also be fruitful when trying to explain the gender differences in study involvement according to track, as we know boys attach more importance to popularity, and hence to their public image and acceptance by their social group, whereas likeability, reflected in personal friendships, is more important to girls (Francis 2000; Heyder and Kessels 2016; Warrington et al. 2000). This makes boys more susceptible to peer pressure (Vantieghem and Van Houtte 2015) and the prevailing peer culture (Van Houtte 2004).

In-depth qualitative research is needed to better understand girls in the arts track. These girls appear even more involved in studying than girls in academic tracks, but because of their higher tendency for disruptiveness, this does not show. The arts track enrolls only a limited number of students; some of these students are artistically gifted, and some enroll after having failed in an academic track, although students more often opt for a technical track after having failed in an academic track. Choosing the arts track reflects the presence of creative talents, or an artistic attitude, and—since it is an unconventional choice—might be seen as a kind of statement, reflecting rebelliousness and an antiestablishment attitude. This might explain the combination of disruptiveness and high study involvement: because of the artistic and creative talents required, students are willing to exert effort, but their antiestablishment attitude leads them to transgress the rules. As such, their counterschool behavior is not necessarily in reaction to their track position, as is the case for vocational students. The question remains, though, why are we seeing this in girls and not boys?

This study confirms the negative association between lower track position and students’ study involvement (cf. Carbonaro 2005; Gamoran 2010), with boys likely to be more vulnerable. Notwithstanding that only a small portion of the variance in study involvement is situated at the track level, the practical relevance is clear. For student outcomes, such as engagement, the influence of the school context is of increasing interest to researchers. Context factors hold the most promise for prevention and intervention efforts, because unlike many student or family factors, these are viewed as malleable (Quin 2017). Particularly regarding tracking, Flemish research has identified enrollment in the vocational track as the main predictor of unqualified school dropout, with being male an additional risk factor (Van Houtte and Demanet 2016). Knowing the clear connection between student engagement and dropout (Quin 2017), the present study raises questions concerning rigid tracking—a modifiable system feature. Abolishing rigid tracking seems like an obvious step, but tracks prepare students for different futures, and eventually academic and vocational students will be separated anyway. Making tracking less stringent, for example, by postponing the age at selection, could be a good starting point (see also Salchegger 2016). Education policy makers may opt to differentiate students only for specific courses and to keep the distinction and segregation minimal to avoid the status loss accompanying enrollment in a lower track. Moreover, technical and vocational tracks must gain more social esteem and no longer be viewed as the lower tracks. Policy makers, school leaders, and teachers should be aware of the detrimental consequences of a hierarchical tracking system for students, particularly boys, who end up in the lower tracks.

Regarding the gender gap, this study illustrates that there is no such thing as homogenous groups of boys and girls. As the idea of intersectionality underlines, it is necessary and important to disaggregate the gender gap along features such as race and social class (Gorard et al. 2001; Morris 2012), and this study demonstrates the usefulness of considering the school context too (cf. Legewie and DiPrete 2012). A better understanding of girls’ and boys’ school conduct can be gained by considering particular features of the educational system and examining adolescents’ demeanor at school as a response to these. Cross-national research comparing the manifestation of gender differences in light of peculiar features of educational systems, such as high-stakes testing or rigid tracking, might be a next step for future research.

Footnotes

Acknowledgements

I thank Prof. Dr. Carolyn Jackson for inviting me to Lancaster University (United Kingdom), where this article took shape.

Research Ethics

This research is based on questionnaires taken from students in schools. It has been approved by the ethics committee of the Faculty of Political and Social Sciences of Ghent University and has therefore been performed in a way consistent with the ethical standards articulated in the 1964 Declaration of Helsinki and its subsequent amendments and Section 12 (“Informed Consent”) of the American Sociology Association’s Code of Ethics. The students’ parents were informed about the survey and could refuse participation of their child. The students gave their informed consent prior to participation. The questionnaires were not taken anonymously because these data had to be coupled with academic results provided by the school. However, all names were removed as the data were assembled, so the final database and all analyses are completely anonymous and confidential.

Author Biography

Mieke Van Houtte is full professor in sociology and head of the research team CuDOS (Department of Sociology, Ghent University, Belgium). Her research interests cover diverse topics within the sociology of education, particularly the effects of structural and compositional school features on several diverse outcomes for students and teachers, and sexual minorities. She is a member of the Royal Flemish Academy of Belgium for Science and the Arts.

References

Ainsworth

James W.

Roscigno

Vincent J.

2005. “Stratification, School–Work Linkages and Vocational Education.”Social Forces84(1):257–84.

Ball

Stephen J.

1981. Beachside Comprehensive: A Case Study of Secondary Schooling. Cambridge, UK: Cambridge University Press.

Bol

Tijs

Van de Werfhorst

Herman G.

2013. “Educational Systems and the Trade-off between Labor Market Allocation and Equality of Educational Opportunity.”Comparative Education Review57(2):285–308.

Boone

Simon

Van Houtte

Mieke

. 2013. “In Search of the Mechanisms Conducive to Class Differentials in Educational Choice: A Mixed Method Research.”Sociological Review61(3):549–72.

Brookover

Wilbur B.

Schneider

Jeffrey M.

1975. “Academic Environments and Elementary School Achievement.”Journal of Research and Development in Education9:82–91.

Brunello

Giorgio

Checchi

Daniele

. 2007. “Does School Tracking Affect Equality of Opportunity? New International Evidence.”Economic Policy52:781–861.

Brutsaert

Herman

. 2001. Co-educatie. Studiekansen en kwaliteit van het schoolleven. Leuven, Belgium: Garant.

Burke

Penny J.

2006. “Men Accessing Education: Gendered Aspirations.”British Educational Research Journal32(5):719–33.

Carbonaro

William J.

2005. “Tracking, Students’ Effort, and Academic Achievement.”Sociology of Education78:27–49.

10.

Catsambis

Sophia

Mulkey

Lynn M.

Crain

Robert L.

1999. “To Track or Not to Track? The Social Effects of Gender and Middle School Tracking.”Research in Sociology of Education and Socialization12:135–63.

11.

Ciccia

Rossella

Bleijenbergh

Inge

. 2014. “After the Male Breadwinner Model? Childcare Services and the Division of Labor in European Countries.”Social Politics21(1):50–79.

12.

Crosnoe

Robert

. 2002. “High School Curriculum Track and Adolescent Association with Delinquent Friends.”Journal of Adolescent Research17(2):143–67.

13.

Demanet

Jannick

Vanderwegen

Pieter

Vermeersch

Hans

Van Houtte

Mieke

. 2013. “Unraveling Gender Composition Effects on Rule-breaking at School: A Focus on Study Attitudes.”Gender and Education25(4):466–85.

14.

Department of Education. 2005. Statistisch jaarboek van het Vlaams Onderwijs - Schooljaar 2004-2005. Brussels: Ministerie van de Vlaamse Gemeenschap.

15.

Department of Education and Training. 2008. Education in Flanders: The Flemish Educational Landscape in a Nutshell. Brussels: Department of Education and Training, Flemish Community of Belgium. Retrieved from http://www.ond.vlaanderen.be/publicaties/eDocs/pdf/120.pdf

16.

de Valk

Helga

. 2008. “Parental Influence on Work and Family Plans of Adolescents of Different Ethnic Backgrounds in the Netherlands.”Sex Roles59:738–51.

17.

DiPrete

Thomas A.

Buchmann

Claudia

. 2013. The Rise of Women: The Growing Gender Gap in Education and What It Means for American Schools. New York: Russell Sage Foundation.

18.

Duru-Bellat

Marie

Mingat

Alain

. 1997. “La constitution de classes de niveau dans les colleges: Les effets pervers d’une pratique à visée égalisatrice.”Revue Française de Sociologie38(4):759-89.

19.

Erikson

Robert

Goldthorpe

John H.

Portocarero

Lucienne

. 1979. “Intergenerational Class Mobility in Three West European Countries: England, France and Sweden.”British Journal of Sociology30(4):415–41.

20.

Francis

Becky

. 1999. “Lads, Lasses and (New) Labour: 14-16-year-old Students’ Responses to the ‘Laddish Behaviour and Boys’ Underachievement’ debate.”British Journal of Sociology of Education20:355–71.

21.

Francis

Becky

. 2000. Boys, Girls and Achievement: Addressing the Classroom Issues. London: Routledge/Falmer.

22.

Fredricks

Jennifer A.

Blumenfeld

Phyllis C.

Paris

Alison H.

2004. “School Engagement: Potential of the Concept, State of the Evidence.”Review of Educational Research74(1):59-109.

23.

Gamoran

Adam

. 2010. “Tracking and Inequality: New Directions for Research and Practice.” Pp. 213–28 in The Routledge International Handbook of the Sociology of Education, edited by Apple

M. W.

Ball

S. J.

Gandin

L. A.

London: Routledge.

24.

Gianettoni

Lavinia

Simon-Vermot

Pierre

. 2010. “Quand la menace d’exclusion professionnelle renforce le genre: Représentations et identités de genre auprès de jeunes sans emploi.”Nouvelles Questions Féministes29(3):76–90.

25.

Gorard

Stephen

Rees

Gareth

Salisbury

Jane

. 2001. “Investigating the Patterns of Differential Attainment of Boys and Girls at School.”British Educational Research Journal27(2):125–39.

26.

Hallam

Susan

Ireson

Judith

. 2007. “Secondary School Pupils’ Satisfaction with Their Ability Grouping Placements.”British Educational Research Journal33(1):27–45.

27.

Hallinan

Maureen T.

1994. “Tracking: From Theory to Practice.”Sociology of Education67(2):79–84.

28.

Hanushek

Eric A.

Wöβmann

Ludger

. 2006. “Does Educational Tracking Affect Performance and Inequality? Differences-in-differences Evidence across Countries.”Economic Journal116(510):C63–C76.

29.

Hargreaves

David

. 1967. Social Relations in a Secondary School. London: Routledge and Kegan Paul.

30.

Heyder

Anke

Kessels

Ursula

. 2016. “Boys Don’t Work? On the Psychological Benefits of Showing Low Effort in High School.”Sex Roles. Advance online publication.

31.

Huisman

Mark

. 1999. “Imputation of Missing Item Responses: Some Simple Techniques.” Pp. 91-119 in Item Nonresponse: Occurrence, Causes, and Imputation of Missing Answers to Test Items, edited by M. Huisman.

Leiden

Netherlands: DSWO Press, Leiden University.

32.

Ireson

Judith

Hallam

Susan

Plewis

Ian

. 2001. “Ability Grouping in Secondary Schools: Effects on Pupils’ Self-concepts.”British Journal of Educational Psychology71:315–26.

33.

Jackson

Carolyn

. 2002. “‘Laddishness’ as a Self-worth Protection Strategy.”Gender and Education14:37–51.

34.

Jackson

Carolyn

. 2003. “Motives for ‘Laddishness’ at School: Fear of Failure and Fear of the ‘Feminine.’”British Educational Research Journal29:583–98.

35.

Jackson

Carolyn

. 2006. Lads and Ladettes in School: Gender and a Fear of Failure. Berkshire, UK: Open University Press.

36.

Jackson

David

. 1998. “Breaking Out of the Binary Trap: Boys’ Underachievement, Schooling and Gender Relations.” Pp. 77–95 in Failing Boys? Issues in Gender and Achievement, edited by Epstein

Elwood

Hey

Maw.

Buckingham, UK: Open University Press.

37.

Janosz

Michel

Archambault

Isabelle

Morizot

Julien

Pagani

Linda S.

2008. “School Engagement Trajectories and Their Differential Predictive Relations to Dropout.”Journal of Social Issues64(1):21–40.

38.

Kelly

Sean

. 2008. “Race, Social Class, and Student Engagement in Middle School English Classrooms.”Social Science Research37(2):434–48.

39.

Legewie

Joscha

DiPrete

Thomas A

. 2012. “School Context and the Gender Gap in Educational Achievement.”American Sociological Review77(3):463–85.

40.

Maaz

Kai

Trautwein

Ulrich

Lüdtke

Oliver

Baumert

Jürgen

. 2008. “Educational Transitions and Differential Learning Environments: How Explicit Between-school Tracking Contributes to Social Inequality in Educational Outcomes.”Child Development Perspectives2(2):99–106.

41.

Mahony

Pat

. 1998. “Girls Will Be Girls and Boys Will Be First.” Pp. 37–55 in Failing Boys? Issues in Gender and Achievement, edited by D.

Epstein

Elwood

Hey

Maw.

Buckingham, UK: Open University Press.

42.

Malmberg

Lars-Erik

Trempala

Janusz

. 1997. “Anticipated Transition to Adulthood: The Effect of Educational Track, Gender, and Self-evaluation on Finnish and Polish Adolescents’ Future Orientation.”Journal of Youth and Adolescence26(5):517–37.

43.

Marsh

Herbert W.

1987. “The Big-fish-little-pond Effect on Academic Self-concept.”Journal of Educational Psychology79(3):280–95.

44.

Marsh

Herbert W.

Parker

John W.

1984. “Determinants of Student Self-concept: Is It Better to Be a Relatively Large Fish in a Small Pond Even if You Don’t Learn to Swim as Well?”Journal of Personality and Social Psychology47(1):213–31.

45.

Mickelson

Roslyn A.

1989. “Why Does Jane Read and Write So Well? The Anomaly of Women’s Achievement.”Sociology of Education62:47–63.

46.

Mickelson

Roslyn A.

1990. “The Attitude–Achievement Paradox among Black Adolescents.”Sociology of Education63:44–61.

47.

Mons

Nathalie

. 2007. Les nouvelles politiques éducatives: La France fait-elle les bons choix?Paris: Presses Universitaires de France.

48.

Morris

Edward W.

2008. “‘Rednecks,’‘rutters,’ and ’rithmetic: Social class, masculinity, and schooling in a rural context.”Gender & Society22(6):728–51.

49.

Morris

Edward W.

2012. Learning the Hard Way: Masculinity, Place, and the Gender Gap in Education. New Brunswick, NJ: Rutgers University Press.

50.

Muller

Chandra

. 1998. “Gender Differences in Parental Involvement and Adolescents’ Mathematics Achievement.”Sociology of Education71:336–56.

51.

Nixon

Darren

. 2006. “‘I Just Like to Work with My Hands’: Employment Aspirations and the Meaning of Work for Low-skilled Unemployed Men in Britain’s Service Economy.”Journal of Education and Work19(2):201–17.

52.

Oakes

Jeannie

. 2005. Keeping Track: How Schools Structure Inequality, 2nd ed.New Haven, CT: Yale University Press.

53.

O’Kelly

Monica

. 2002. “Work, Women, and Their Beliefs.”Behaviour Change19:12–21.

54.

Organisation for Economic Co-operation and Development. 2014. Education at a Glance 2014: OECD Indicators. Paris: Author. https://dx-doi-org.web.bisu.edu.cn/10.1787/eag-2014-en

55.

Quin

Daniel

. 2017. “Longitudinal and Contextual Associations between Teacher–Student Relationships and Student Engagement: A Systematic Review.”Review of Educational Research87(2):345–87.

56.

Rosenbaum

James E.

1976. Making Inequality: The Hidden Curriculum of High School Tracking. New York: Wiley.

57.

Rosenbaum

James E.

2001. Beyond College for All: Career Paths for the Forgotten Half. New York:Russell Sage Foundation.

58.

Rumberger

Russell W.

1995. “Dropping Out of Middle School: A Multilevel Analysis of Students and Schools.”American Educational Research Journal32(3):583–625.

59.

Salchegger

Silvia

. 2016. “Selective School Systems and Academic Self-concept: How Explicit and Implicit School-level Tracking Relate to the Big-fish-littlehy-pond Effect across Cultures.”Journal of Educational Psychology108(3):405–23.

60.

Schafer

Walter

Olexa

Carol

. 1971. Tracking and Opportunity: The Locking-out Process and Beyond. Scranton, PA: Chandler.

61.

Schofield

Janet W.

2006. “Migration Background, Minority-group Membership and Academic Achievement Research Evidence from Social, Educational, and Developmental Psychology.”AKI Research Review5. Retrieved from http://www.wzb.eu/ZKD/AKI/files/aki_research_review_5.pdf

62.

Schwalbe

Michael L.

Staples

Clifford L.

1991. “Gender Differences in Sources of Self-esteem.”Social Psychology Quarterly54:158–68.

63.

Schwartz

Shalom H.

Rubel

Tammy

. 2005. “Sex Differences in Value Priorities: Cross-cultural and multimethod studies.”Journal of Personality and Social Psychology89(6):1010–28.

64.

Sharkey

Jill D.

You

Sukkyung

Schnoebelen

Katrina

. 2008. “Relations among School Assets, Individual Resilience, and Student Engagement for Youth Grouped by Level of Family Functioning.”Psychology in the Schools45(5):402–18.

65.

Shavit

Yossi

Featherman

David L.

1988. “Schooling, Tracking, and Teenage Intelligence.”Sociology of Education61:42–51.

66.

Stahl

Garth

. 2015. Identity, Neoliberalism and Aspiration: Educating White Working-class Boys. London: Routledge.

67.

Steinmayr

Ricarda

Spinath

Birgit

. 2009. “The Importance of Motivation as a Predictor of School Achievement.”Learning and Individual Differences19(1):80–90.

68.

Steunpunt

WAV.

2003. Gender zakboekje: Zij en hij op de arbeidsmarkt. Leuven, Belgium: Steunpunt WAV, Vlaamse Overheid, Administratie Werkgelegenheid.

69.

Stevens

Peter A. J.

Vermeersch

Hans

. 2010. “Streaming in Flemish Secondary Schools: Exploring Teachers’ Perceptions of and Adaptations to Students in Different Streams.”Oxford Review of Education36(3):267–84.

70.

Stewart

Eric A.

2003. “School Social Bonds, School Climate, and School Misbehavior: A Multilevel Analysis.”Justice Quarterly20(3):575–604.

71.

Christiane

Timmerman

Hermans

Philip

Hoornaert

Johan

, eds. 2002. Allochtone jongeren in het onderwijs. Een multidisciplinair perspectief. Leuven, Belgium: Garant.

72.

Trautwein

Ulrich

Lüdtke

Oliver

Marsh

Herbert W.

Köller

Olaf

Baumert

Jürgen

. 2006. “Tracking, Grading, and Student Motivation: Using Group Composition and Status to Predict Self-concept and Interest in Ninth-grade Mathematics.”Journal of Educational Psychology98(4):788–806.

73.

Van de gaer

Eva

Pustjens

Heidi

Van Damme

Jan

De Munter

Agnes

. 2006. “Tracking and the Effects of School-related Attitudes on the Language Achievement of Boys and Girls.”British Journal of Sociology of Education27(3):293–309.

74.

Van Houtte

Mieke

. 2004. “Why Boys Achieve Less at School than Girls: The Difference between Boys’ and Girls’ Academic Culture.”Educational Studies30(2):159–73.

75.

Van Houtte

Mieke

. 2005. “Global Self-esteem in Technical/Vocational versus General Secondary School Tracks: A Matter of Gender?”Sex Roles53(9/10):753–61.

76.

Van Houtte

Mieke

. 2006. “School Type and Academic Culture: Quantitative Evidence for the Differentiation–Polarisation Theory.”Journal of Curriculum Studies38(3):273–92.

77.

Van Houtte

Mieke

. 2016. “Lower-track Students’ Sense of Academic Futility: Selection or Effect?”Journal of Sociology52(4):874–89.

78.

Van Houtte

Mieke

Demanet

Jannick

. 2016. “Teachers’ Beliefs about Students, and the Intention of Students to Drop Out of Secondary Education in Flanders.”Teaching and Teacher Education54:117–27.

79.

Van Houtte

Mieke

Stevens

Peter A. J.

2008. “Sense of Futility: The Missing Link between Track Position and Self-reported School Misconduct.”Youth & Society40(2):245–64.

80.

Van Houtte

Mieke

Stevens

Peter A. J.

2009. “School Ethnic Composition and Students’ Integration outside and inside Schools in Belgium.”Sociology of Education82(3):217–39.

81.

Van Houtte

Mieke

Stevens

Peter A. J.

2016. “Track Position, Futility Culture and Involvement with Paid Work: Investigating the Correlates of Lower-track Students’ Involvement with Part-time Employment in Flanders.”Sociological Inquiry86(3):372–99.

82.

Vantieghem

Wendelien

Van Houtte

Mieke

. 2015. “Are Girls More Resilient to Gender-conformity Pressure than Boys? The Association between Pressure for Gender Conformity and Academic Self-efficacy.”Sex Roles73(1):1–15.

83.

Wang

Ming-Te

Eccles

Jacquelynne S.

2012. “Adolescent Behavioral, Emotional, and Cognitive Engagement Trajectories in School and Their Differential Relations to Educational Success.”Journal of Research on Adolescence22(1):31–39.

84.

Warrington

Molly

Younger

Mike

. 2000. “The Other Side of the Gender Gap.”Gender and Education12:493–508.

85.

Warrington

Molly

Younger

Mike

Williams

Jacquetta

. 2000. “Student Attitudes, Image and the Gender Gap.”British Educational Research Journal26(3):393–407.

86.

William

Dylan

Bartholomew

Hannah

. 2004. “It’s Not Which School but Which Set You’re in That Matters: The Influence of Ability Grouping Practices on Student Progress in Mathematics.”British Educational Research Journal30(2):279–93.

87.

Willis

Paul E.

1977. Learning to Labour: How Working Class Kids Get Working Class Jobs. Farnborough, UK: Saxon House.

88.

Wolbers

Maarten H. J.

2007. “Patterns of Labour Market Entry: A Comparative Perspective on School-to- work Transitions in 11 European Countries.”Acta Sociologica50(3):189–210.