Traditionally, the German school system is characterised by early ability streaming of pupils. Table 1 provides an overview of the tracking systems in selected industrialised countries: 5 While many European countries track pupils to more or less academic secondaryschool types, Germany’s regular tracking age of ten is rather early in international comparison. To be more specific, in Germany pupils are selected into three school types after four years of ele- mentary school: 6 The most ‘able’ pupils are supposed to attend the Gymnasium which is a nine (or eight) year higher level secondaryschool and enables pupils to pursue further aca- demic studies (e.g. at universities). 7 An alternative school track is offered by the Realschule as an intermediate level secondaryschool which generally lasts six years and prepares students for a rather vocational education. Finally, the Hauptschule as the lowest level secondaryschool type is supposed to be the most vocational and least academic track and lasts five years. In principle, it is possible to change tracks after the initial track decision. However, different curricula for the different school types complicate switching tracks especially after sixth grade. 8
2 Data and Design of the Intervention
As part of a study on career orientation and career guidance, 527 German secondaryschool pupils were surveyed in two German cities: Mannheim and Freiburg. Pupils from all three secondaryschool tracks were included: the Gymnasium (upper and most general track), the Realschule (intermediate, more vocational track) and the Werkrealschule (lower, most vocational track). The rst survey was performed by means of a paper-and -pencil questionnaire in the classroom in spring 2014. The same researcher went to all classrooms to present the study, distribute the questionnaires and answer any questions the pupils had. One year later, in spring 2015, all pupils who had agreed to be contacted again at the time of the rst survey (326), received an invitation to participate in the second survey by email or by post. It was announced at the time of the rst survey that participants to the second survey would receive a voucher for an online shop. A large set of individual and family characteristics is available in the dataset.
We now provide some basic information about how many children go to formal day care and how long they stay there, highlighting major differences bet- ween East and West Germany. We then document the link between attending formal day care and hig- her socio-economic status and family incomes. Then come the main results assessing whether or not, net of all other relevant factors, length of time in formal day care results in better performance in secondaryschool. Finally, we consider some important policy implications of our findings.
Mean of dependent variable .040 .035
Number of observations 117,959 27,284
Notes: Each cell represents a separate regression. Individuals who have still not graduated six years after admittance
are considered to have dropped out. In addition to municipality of residence and upper secondaryschool starting year fixed effects, all regressions include controls for: compulsory school GPA (quadratic), age at enrolment (dummies), each parent’s educational attainment (3 levels), whether both parents are foreign-born, each parent’s age (linear), missing data on parents’ education, the father’s employment status, the father’s earnings (linear), and whether the father has been convicted of crime. The (potentially) endogenous variable takes the value one if the individual enrolled in a three-year (or longer) track and zero if he/she enrolled in a two-year track. The instrument (pilot intensity) is the share of available vocational tracks in the municipality of residence at the time of enrollment which constituted three-year tracks. Robust standard errors in parentheses allow for clustering by municipality of re- sidence. */**/*** denotes significance on the 10/5/1 percent level. In Panel B–C, the students are divided into sub- groups based on the grade distribution among the male upper secondaryschool students.
Even though today’s admission to upper secondaryschool, within an admission region, is based purely on the final grades from lower secondaryschool, this has not always been the case for the municipal schools. Prior to the 2000s, the normal procedure was to base the educational track admission on the grades, but to then assign students to the public schools based on some other criterion, for example proximity to the student’s home (Molin, 2019). In the year 2000, two of the largest municipalities, Stockholm and Malmö, allowed students to be admitted to all school and track combinations based on the grades. Other regions have since followed, and according to Sund (2018), today all municipalities apply such purely grade-based admission systems. Whether or not proximity was used to determine school admission during our period of analysis is of relevance for our RD-based analysis. For many of the mid-sized and smaller municipalities, we have however not been able to collect information on the exact year in which the proximity principle for school placement was replaced with a pure grade principle. It can be pointed out that in many of the smaller and mid-sized municipalities, there was only one public school offering each track anyway, which means that assignment to schools within a given track was a non-issue. In Appendix D we provide results suggesting that this is not a concern for our empirical analysis.
successful learning (e.g., Stoeger, Steinbach, Obergriesser, & Matthes, 2014). For schools at the lower and upper secondaryschool level in Switzerland, some studies showed that teachers rarely create learning environments that might require meta- cognitive competencies (e.g., Leutwyler & Maag Merki, 2009; Pauli et al., 2007). In this regard, Leutwyler (2009) mentioned that the curricula at Swiss high schools may not support teachers in including metacognitive thinking in their lesson plans. Another possible explanation is that teachers lack suﬃ cient knowledge about meta- cognition to be able to create such learning environments (Zohar, 2012). It is also possible that teachers might not have seen the importance of fostering students’ MSK in this school type, because they overestimated their students’ strategic skills. However, the results indicate that interindividual diﬀ erences in students’ MSK over time exist. MSK might have decreased in some of the students and stayed stable or increased in other students. With the data of this study, it is not possible to explain these interindividual changes of MSK over time. In the literature, extracurricular experiences and the extent and intensity of students’ strategic activities are dis- cussed as possible factors (Hasselhorn & Labuhn, 2010; Karlen et al., 2014).
Malaysia is a country with a Chinese minority which makes up roughly 25% of the whole population. The Chinese minority sector in Malaysia operates its own educational system and schools, teaching students mainly in Chinese (Karpudewan & Chua, 2016). Chinese education in Malaysia began in 1819. In 1954, the United Chinese School Committees’ Association of Malaysia (UCSCA or Dong Zong) was established to defend and develop Chinese language education and aims to achieve the sustainable development of Chinese education. It works with the United Chinese School Teachers’ Association of Malaysia (UCSTA or Jiao Zong) to uphold the ethnic rights of the Chinese minority. The two associations are known by the acronym of Dong Jiao Zong (Dong Zong and Jiao Zong) (Dong Zong, 2017). Independent schools are operated by Dong Jiao Zong instead of by the Ministry of Education in Malaysia. Any rules and regulations in these schools are different from those in the governmental schools. Students who go to Malaysian independent Chinese secondaryschool have an additional academic year of upper secondaryschool, similar to the secondaryschool system in the People’s Republic of China and Taiwan. The Malaysian high school chemistry book (Upper SecondarySchool Chemistry) is edited by the Malaysia Independent Chinese SecondarySchool Working Committee (MICSS) and is based on the chemistry syllabus for secondary schools. There are three volumes of upper secondaryschool chemistry textbooks for students from grade 10 to grade 12. In this study, the Upper SecondarySchool Chemistry Textbook for grade 10 students (MICSS, 1996; later called MY1) was analyzed. 3
TU Darmstadt, Physics Education Research Group, Hochschulstrasse 12, Darmstadt, Germany Electric circuits are an important element of physics classes in Austria, Germany, and most countries around the world. However, many students leave secondaryschool without having an adequate conceptual understand- ing of simple circuits. Voltage in particular has proven to be a difficult concept as students think of it as a property or component of the electric current. Furthermore, research has shown that girls tend to have a lower interest in physics than boys and that context-based physics instruction is a promising approach to increase girls’ interest. However, it is unclear whether decades of research on students’ conceptual difficulties e.g. with volt- age as well as research into ways to promote girls’ interest have had a significant impact on physics classrooms. For this reason, the conceptual understanding of electric circuits as well as the interest in physics of N = 1207 traditionally taught students in Germany and Austria was assessed using a valid and reliable multiple-choice test. The empirical evaluation of the data shows that female students are still not as interested in physics as their male peers, despite achieving the same total learning gain in the multiple-choice test. An analysis of two items of the test instrument focusing on potential differences furthermore suggests that traditional instruction still fails to provide students with an adequate conceptual understanding of voltage as an independent physical quantity that refers to a difference in electric potential between two points in a circuit. These results highlight the need to develop research-based curricula on electric circuits that take into account the findings on students’ interests and alternative conceptions.
We investigated the effect a decline in the mother’s employment separately from the father’s employment decline (results in table A21 in the appendix). Our results suggest that the employment decline of the father is more important than the employment decline of the mother. Both show the same pattern (largest effects in non- poor low-home families), and the difference between the two is small and not statistically significant. A large decline in the employment of the father decreases the probability of completing secondaryschool by 3 percentage points in our sample, while a large decline in the employment of the mother leads to a 2 percentage point decline. Only the former is significant at the 5 percent level, but the confidence intervals overlap to a large extent. For adolescents from non-poor families with a low quality home environment, a large decline in the father’s employment decreases the completion rate by 9 percentage points, compared to a 4 percentage points effect for a large decline in the mother’s employment (the former is significant at the 5 percent level, the latter is not). Taken together, the robustness checks strengthen the conclusion of our main analysis.
To allow us to pool the data across regions and years, we normalize the cutoff GPA to 0. The distribution of cutoff GPA values is plotted in the top panel of Figure 2 (white columns), with a comparison to the GPA distribution for our baseline sample (gray columns). This graph provides an indication of where individuals on the borderline of acceptance into a program are found in the skill distribution. The modal cutoff GPA of 3.2 corresponds to roughly the 15th percentile of GPAs in our baseline sample of applicants to competitive academic programs. To put this in perspective, the modal cutoff GPA also corresponds roughly to the median GPA of all ninth graders (including those applicants not in our sample which have a preferred non-academic choice and also those who don’t apply to secondaryschool at all). 20 While the cutoffs vary substantially, they generally are only binding for applicants with GPAs in the bottom half of our estimation sample. Both of these comparisons will be important to keep in mind when interpreting the estimates, which will capture local average treatment effects for applicants around the cutoff.
In a city as segmented as Brussels, with its layer-cake neighbourhoods of residen- tial, social, and ethnic segregation (Kesteloot, Vandermotten et al., 2001; Jacobs and Swyngedouw, 2000), it is particularly interesting to study young people’s social relationships with each other, especially at school, which is assumed to be a social area that brings together people of various backgrounds. We thus conducted a quantitative survey of secondaryschool pupils in the French-speaking schools of Brussels in March 2006. The questionnaire, which concerned lifestyles, racism, and fears of crime, was given to 646 students enrolled in thirteen of the fourteen secon- dary schools in the French-speaking network of Brussels’ central borough, that is, “Brussels-City” or downtown Brussels 1 . This sample was fairly representative of the students in this educational network, given that it covered more than two-thirds of the students enrolled in the various forms of secondary education (general, techni- cal, vocational, and artistic) of the francophone schools led by the municipal gov- ernment. However, it cannot claim to be representative of “Brussels youth”, given the selective enrolment mechanisms that exist in secondary education, especially in the Brussels Region. Despite this limitation, this survey gives some insights into the diversity of the different “layers” of Brussels youth according to at least three major variables: social background, ethnic background, and type of schooling. Of the multitude of data that we obtained we have chosen to concentrate in this article on a few subjects for which the outcomes proved particularly unexpected, especially when it comes to racism and crime (or the fear of crime). The following analyses are valid only for the sample of young people who were surveyed, of course. Extending
The study is based on data from the "Wired into Each Other: Network Dynamics of Adolescents in the Light of Status Competition, School Performance, Exclusion and Integration” project of the Hungarian Research Center for Education and Network Studies. The sample includes seven secondary schools from Hungary: two from the capital, two from a major city in Eastern Hungary, and three from nearby smaller towns. As one of the research goals was to examine social inclusion, a selection criterion of cities was the existence of Roma minority. Another restriction of the sample was the quota for school types: either in the case of Budapest, in the major city and in the smaller towns, grammar schools and schools providing vocational training were included. As in the research we used this targeted sampling, the sample cannot be considered as representative of the region or Hungary. The target group includes all students of the selected schools, who were in 9th grade in the academic year 2010-11. This study uses wave 1 of the data collection, carried out two and half months after the students entered secondaryschool (9th grade) in 2010, using paper and pencil questionnaire. At the time of the data collection their median age was 15.2 years.
The negative effect from expanding secondaryschool enrolment that we estimate is consistent with a previous literature that considers the effect of mandatory schooling laws on teenage birth outcomes (Black, Devereux and Salvanes (2008) and Fort, Schneeweis and Winter-Ebmer (2016)). There are two major differences from these papers to our context. First, in contrast to the above literature, the expansion of secondary schools in Brazil affects a very different population margin, allowing us to investigate effects for a margin of the population not necessarily affected by changes in compulsory schooling laws. Second, different from the existing literature, we focus our analysis on a middle-income country. It is this focus that ultimately allows us to consider the effect of the very rapid expansion of secondary schools over a relatively short period. As a number of middle and low income countries are still to go through expanding their secondary education system, our results are relevant to understand the (positive) externalities expanding secondary education opportunity may have through substantial reductions in underage fertility.
Heightened policy and research interest in student absenteeism is prefaced on the well- documented correlation between student absences and educational outcomes representing a causal relationship. While it makes intuitive sense that absences harm student achievement, causal identification of the relationship remains a persistent challenge (Jacob and Lovett, 2017) because unobserved time-varying, student-level shocks, such as illness or lack of sleep, can affect both students’ attendance and their academic performance. In the current study, we overcome the threat posed by such shocks by using a decade’s worth of rich administrative data from a large urban school district in California that include the date and class period of each absence. We focus our analysis on secondary schools, as students in secondaryschool have far more absences (and agency over those absences) than students in elementary school. Our identification strategy exploits within-student, between-subject differences in ab- sences in a given school year to remove the threat posed by unobserved student-year shocks. We examine the relationship between absences and student achievement using two strategies, both of which rely on two assumptions that we demonstrate are likely to hold: first, that in a given school year, within-student differences in absences in math and English language arts (ELA) are conditionally random; and second, that any spillover effects of absences in one subject on academic mastery in another are relatively trivial. The first empirical approach we employ is a proxy plug-in solution (Wooldridge, 2010), similar to twins-based identification strategies (Ashenfelter and Krueger, 1994), in which we use absences in math to control for the time-varying, subject-invariant factors that cause absences in order to identify the im- pact of ELA absences on ELA achievement, and vice versa. The second approach is similar:
b German Institute for Economic Research (DIW Berlin)
We employ a psychometrically validated performance test to study economic competence among a large and representative sample of early secondaryschool students in Southwest Germany. The rich dataset allows us to study variation in economic competence across school types and observable student characteristics. Our results show that economic competence is significantly lower among female students, migrants, students with parents of low socio- economic status and those who do not attend the highest track school type. Additionally, quantile regression analyses suggest that the gender gap increases along the distribution of economic competence and that effects of parents with high socio-economic status are more pronounced above the median of the competence distribution. Our analysis sets the stage for a long-term study of economic competence among secondaryschool students and the impact of a recent curriculum reform introducing mandatory economic education.
knowledge and understanding between 2004 and 2010, with a small, insignificant decline between 2010 and 2013 (ACARA, 2014). Given the perceived importance of knowledge assessment (MCEETYA, 2006, 2009; Torney-Purta et al., 2001), this initial increase might be interpreted as a long-term outcome reflecting a constant interest in civics and citizenship education over the past two decades. Moreover, the focus on history and knowledge that was identified with previous initiatives mirrors the finding that secondaryschool students’ perceptions of good citizenship behaviors highlight cognitive aspects of citizenship – learning about history and politics. These activities appear to be considerably more important to students’ understanding of citizenship nowadays than in 1999. We may partly attribute this to efforts in Australian civics and citizenship education, although we need to be careful with our conclusion as CIVED used a bipolar scale (but see Note 5).
supportive learning climate, trusting students) are mainly present in Slovene language classes in comparison to maths classes, particularly at the elementary school level. Elementary school students gave higher grades to items concern- ing the learning climate and trusting students in Slovene language, compared to maths. Secondaryschool students assessed the clarity of rules and student obligations and paying attention in class higher in maths than in Slovene lan- guage. Although maths and Slovene curricula (Učni načrt matematika, 2011; Učni načrt slovenščina, 2011) emphasize, in addition to educational goals, personal development, the development of communication skills, critical ap- proach and independence, an essential difference in the students’ perception of teachers’ work in maths and Slovene language classes has been detected. This is assumed to be connected with the teaching contents of each particular subject. Modern curricula are designed for goal-based learning, meaning that teachers should observe goals and not the contents (teaching contents are subordinated to educational goals), and they should enable students to develop appropriate competencies to operate in the knowledge society. The question is whether the education systems for maths and Slovene teachers differ in their fundamental premises, or whether teachers of different subjects understand their roles dif- ferently or class dynamics are to the greatest extent defined by the teaching contents, which in Slovene language classes presupposes communication and the inclusion of pupils with their views and opinions. Nevertheless, the same should take place in maths classes. Conception building, in-depth understand- ing of the teaching contents, and linking various concepts cannot be achieved or is underachieved in classrooms where the teacher is less diligent in establish- ing a participative culture and including pupils in direct educational work.
Notes: Each cell represents a separate regression. In addition to municipality of residence and upper
secondaryschool starting year fixed effects, all regressions include controls for: compulsory school GPA (quadratic), age at enrolment (dummies), each parent’s educational attainment (3 levels), whether both parents are foreign-born, each parent’s age (linear), missing data on parents’ education, the father’s employment status, the father’s earnings (linear), and whether the father has been convicted of crime. The (potentially) endogenous variable takes the value one if the individual enrolled in a three-year (or longer) track and zero if he/she enrolled in a two-year track. The instrument is the share of available vocational tracks in the municipality of residence at the time of enrollment which constituted three-year tracks. Robust standard errors in parentheses allow for clustering by municipality of residence. */**/*** denotes significance on the 10/5/1 percent level. In Panel B–C, the students are divided into sub-groups based on the grade distribution among the male upper secondaryschool students. The first stage coefficients in Panel A−C are .357 (.008), .586 (.012), and .141 (.012), respectively.
dents with learning disabilities (LD 4 ) in various secondaryschool voca- tional programs in comparison with their peers without disabilities. Our findings are based on an empirical study that comprised 417 students, 5 of whom 85 were students with LD. Based on sociometric analyses of all participating classes, we determined that students with LD were less inte- grated into the classroom in comparison to their peers without LD. The results of the sociometric analysis show statistically significant differences in the sociometric position between students with LD and students with- out LD. While students with LD were most frequently perceived as re- jected, students without LD were seen as popular or average. In addition, students with LD see themselves as less socially self-efficient compared to their peers. The results of our study mostly refer to boys, because the sam- ple comprised 359 boys and 58 girls. We believe that pro-inclusion teachers with appropriately developed strategies for strengthening students’ social skills, as well as positive attitudes and sufficient knowledge about the spe- cial needs of students can have a significant impact on the social accept- ance of students with special needs in the classroom community.
In this paper, we aim to contribute to a better understanding of both the role of parents in secondaryschool track choice and the potential relationship between early tracking and social inequalities. Granting parents influence on track choice may, on the one hand, improve the assignment of students to tracks if parents have superior information than teachers on a child’s ability and if they use this information as a basis for their decision. On the other hand, parents may base their decision on other criteria beyond ability, potentially overturning a track assignment that was based purely on ability. Parents tend to have aspirations for their child that depend on their own achievement rather than their child’s ability. They often desire for their child to achieve a similar socioeconomic status (SES) to what they have. Alternatively, parents could hold unrealistic or biased beliefs about their child’s ability that reflect their own achievement more than their child’s academic potential. Thus, high SES parents are more likely to think that their children are smarter than they really are compared to low SES parents. In a school system with tracking, such parental aspirations and beliefs may contribute to fostering social inequalities and decreasing the efficiency of tracking if parental aspirations and beliefs do not conform to a child’s ability.