No one can have failed to notice that the Swedish school system is in crisis – a knowledge crisis, specifically. Swedish primary school pupils' results have fallen sharply in recent decades, regardless of how you measure them. The decline in the OECD’s PISA surveys during the 2000s corresponds to approximately three quarters of a school year. According to TIMSS – a sample that more closely measures the traditional subject knowledge in mathematics and natural sciences – the collapse since 1995 has been even stronger. Meanwhile, other studies suggest that 13-year-olds’ cognitive skills have also dropped markedly since the end of the 1980s.
A similar trend can be observed among high school students. In advanced science and mathematics, knowledge plummeted between 1995 and 2008 in a way that corresponds to several years of learning. Even the results of the patterning test fell significantly during the 1990s. There is, therefore, no doubt that the Swedish school has fallen dramatically in every case, when it comes to its ability to convey traditional knowledge and skills.
These meagre results, however, also apply to less traditional knowledge controls. Even in problem-solving skills – which are related to other abilities important in working life – Swedish secondary school pupils are underperforming. Although they still have relatively good civic skills, this is virtually the only light in the darkness to be found.
From a purely educational perspective, this development is of course worrying in itself. But it also threatens to jeopardise the country's economy. Research in an international sample finds that these results are strongly related to a country's future economic development. There are still questions about the causality in this research, but students' declining performances are more likely to be reflected as a decline in our future prosperity.
There is hardly any question that we must reverse this negative trend in these results. Sure, you will hear the occasional voice that seeks to explain away the deteriorating results with the non-falsifiable claim that Swedish youths are better at things that are difficult to measure. But, in principle, there is now a consensus that we must somehow get ourselves out of the knowledge crisis.
An ever more aggressive debate is raging about what has actually caused the deterioration and, therefore, what the possible solutions should consist of. The Left has long argued that the major problems stem from freedom of choice and the profit motive. The Right says the problem should rather be blamed on what is described as the Social Democrat-created ‘fuzzy school’, which has created chaos and disorder in classrooms, as well as the diminished authority and status of teachers.
Some explanations, however, get support across political boundaries. An example of this is the municipalisation that both the Left Party and the People’s Party see as an important reason for the schools’ decline – while both also advocates socialisation as a solution. Problems with school choice and the profit motive do not follow strict ideological lines. A number of bourgeois opinion makers, such as Expressen’s leader page, rode the critical wave against school choice and the profit motive that swept over Sweden in the years before the 2014 elections. It simply became popular to play with this opinion and kick a little in all directions.
But, in the debate, there is rarely presented any research in support of the theses that abound. This is not especially strange because many of the popular explanations cannot in principle be formally proven. The only reason for the fall in results that we can be certain of is the impact of immigration through the changing composition of students. Simply limiting the sample selection to entirely Swedish students has shown that 29 per cent of the average fall in PISA disappears. This is without any negative impacts on native students being included.
Even with these included, it is unlikely that immigration entirely explains the PISA decline. Interestingly, it appears the deterioration in TIMSS is to a much lesser extent dependent on the effects of immigration. In samples that, to a greater extent, measure traditional subject knowledge, alternate explanations are probably more important.
What are these explanations? The most honest answer is that we do not know. There is, quite simply, no reason except immigration that has clearly been shown to be behind the fall in results. This does not mean that all explanations are equally valuable. Some may, for example, be falsified relatively easily, while others do not even get indirect support or are not plausible for other reasons. Some explanations may get strong indirect support and seem plausible, even if they cannot be proven formally. With such circumstantial evidence pushing, we can come closer to the answer to the question about what has caused the Swedish knowledge crisis, and what must be done to reverse the trend.
When it comes to existing explanations for the negative trend that should be removed, perhaps the most popular of these – freedom of choice, independent schools and the profit motive – actually attracts the least evidential support of all. This we know from studies that have evaluated these phenomena. News is that there is direct support for the view that independent schools have decreased the drop in results in TIMSS somewhat. This does not mean that the Swedish school-choice and independent school systems are well designed. They are certainly not. But it does mean that these reforms, despite their poor design, are not likely to have contributed to the knowledge decline in primary schools. At the secondary school level, we do actually know nothing – because the design of the system has basically made evaluation impossible – but overall there is little to suggest that market solutions themselves have been the culprit. That the fall occurred so quickly and sharply, while market development was relatively slow, is further support for this view.
This also applies to other hypotheses. An important example is the changes that have taken place among the teachers. Sure, new teachers’ intelligence and subject knowledge have dropped in recent decades. But it takes a long time before this is reflected in teachers' average quality. Changes in the teachers may have played a certain role but, because the fall has happened so fast, it is likely far from the most important explanation. Moreover, it should be remembered that Swedish teachers still maintain high quality, internationally speaking – at least when measuring how they perform in PISA-like samples.
Some explanations cannot be directly evaluated, but the research in general gives no support for the idea that they are correct. An important example is municipalisation. National economics studies do not indicate that decentralisation of funding and responsibility within the school is negative for the results in international surveys. On the contrary, it seems decentralisation often leads to higher student performance in these studies and other knowledge-based samples.
Similar reasoning can be applied to most of the hypotheses put forward in the debate about what has caused the decline in knowledge. Either they are unlikely because of the time factor between change and decay – or they find no support in the literature. Or both.
But there are exceptions. And it is these exceptions that should be scrutinised. The exclusion method leads us in the direction of changes in pedagogy and school culture. It leads us also to more general socioeconomic and cultural shifts that are likely both to have had a direct effect on the decline in knowledge and an indirect effect through the changes in the school system that have been identified.
To understand schools’ performance development, we must go back to the early postwar period, when the policy more and more came to focus on creating an equal and egalitarian education system. Swedish schools had long been traditional, both in terms of focusing on the knowledge and the teaching methods that were the norm – something that was personified by (or characterised as) the sadistic teacher Caligula in the film Hets from 1944. For nobody and nothing has equally well represented how inhumane the old system was considered to be.
In the radical spirit of the times that prevailed in the postwar period, the traditional school was seen as a relic from an authority-driven society that would break down. A new kind of school was obviously important to socialise students for the new world.
Progressive educational ideas thus came to dominate education policy from the 1950s onwards. These ideas have come and gone in historic cycles, but one can say that they originated in Jean-Jacques Rousseau's Émile, where the teaching of facts and knowledge is described as meaningless and breathless. This is because such teaching only leads to cramming, without a deeper understanding of the subject. It also takes away the joy of learning and therefore contributes to an unhappy childhood. Instead of focusing on teaching, students should themselves be allowed to search for knowledge – and their own experiences, in this way, replacing the teacher's traditional role.
Different currents have come to emphasise variations of Rousseau's theory, but the basic idea tends to be the same. For educators like the American, John Dewey, and the Brazilian, Paulo Freire, the problem formulation was similar, for example: students' inactivity creates passive and unhappy individuals. Facts are meaningless and arbitrary. Competition among students is detrimental. Thus, the solutions are also very similar. Students must be activated more in the classroom. The teacher should take a step back. The focus on obedience and discipline must be reduced. The teaching of facts should be minimised. Ratings, homework, and other forms of external incentives that prevent students from finding joy in learning must also be removed.
The policy was thus in many respects to follow these recipes. But while there were no major changes in the actual teaching, such progressive educational theories were sought after. A survey shows, for example, that the share of individual work in primary school barely increased at all between the 1960s and 1980s. That individual work there was comprised mostly the students answering the same questions as determined by the teacher. These exercises had very little to do with progressivism’s theories. Admittedly, the percentage of teamwork increased slightly during the same period, but generally the teaching continued to be teacher-driven.
During the 1980s this began to change, however slowly. The focus on progressive educational ideas increased further in public documents, such as the new curriculum. Students' right to influence how their education would be designed was also incorporated into the Education Act that came into effect in 1986. That which is now called ‘own work’ began to some extent to be introduced spontaneously by teachers during the 1980s, to meet the new requirements of individualisation and pupil participation.
The big change came not until the 1990s, during progressivisms second breath. In 1992, the National Agency for Education created the Institute for Individual Adaptation, whose chief, Hakan Jarbur, in his radical spirit argued that "no man has ever taught anyone anything". The institute preached ‘own work’ as a method and it drew in large numbers of teachers and school leaders. In the 1994 curriculum, pupil influence over both the teaching methods and content increased significantly. The focus would now be on ‘lifelong learning’ and to prepare students to be able to take on new knowledge for themselves in a rapidly changing world. In the hope of increasing the diversity of working methods, all writing about how the goals of teaching would be accommodated largely disappeared. And, with this, the bow door came off altogether.
This hope was obviously unrealistic, because all writings, institutions and incentives pointed in the direction of more student-centred teaching and student influence. Diversity did not materialise. Instead, the children's own work increased exponentially. The proportion of students who reported that they worked individually several times a day remained constant between 1992 and 1995 – but by 2003, it had doubled. Now it would be the students themselves who assessed their progress in logbooks and be responsible for how the work time was employed. The children's own experiences got a prominent role in learning and the teacher’s role was reduced. Progressivism then finally got its major breakthrough in Swedish classrooms.
Why is this important? Quite simply, because the research shows that the working methods that progressivism emphasises, and which spread like wildfire in Swedish schools, especially from the mid-1990s, can be very negative for the results of knowledge-based tests. These methodological changes appear to be the only educational policy explanation for the knowledge collapse that finds both support in the research and is plausible given the timeline of the collapse.
An important example of how progressive methods can lower results quickly and strongly comes from the Canadian province of Quebec, where in the early 2000s student-centred teaching was introduced en masse. These reforms were evaluated recently by economists, who found that they had strongly negative effects on national and international tests in the space of just a few years. The effect increased the longer children were exposed to the methods and it affected students of different abilities. Additionally, the students became more hyperactive, which is a potential mechanism behind the effect. Too much freedom for the children may well also then lead to a messier school climate.
At the same time, there is support from the United States that hierarchical school environments and working methods characterised by strict discipline are good for results of knowledge-based examinations, particularly among students from lower socioeconomic backgrounds. This is not particularly surprising, because it is precisely these children who lack structures in life outside school, while they lack the resources necessary to make themselves heard in a school that emphasises student responsibility and student participation.
Neither neuroscience nor studies in cognitive psychology support progressive educational ideas about how learning functions. There was, quite simply, never any research support to suggest that the ideas that came to permeate the Swedish education system could match the traditional school culture and teaching – at least when it came to learning knowledge and cognitive skills.
This does not mean that children are unhappy at today’s schools. They are not. Comfort and satisfaction have increased as knowledge has fallen. It is not a wild guess that these trends are related. The research indicates clearly that progressive methods generate more positive attitudes to the subject studied and greater confidence about one's own ability. The students learn less, but think it's more fun to study and think they are cleverer than they are. This naturally sits ill with progressivism’s hymns, which often emphasise that joy, self-esteem and learning go hand-in-hand. But this is also its basic fallacy. Effective knowledge acquisition is not always fun, but often requires hard work and boring repetition.
It is worth considering the education minister Gustav Fridolin’s assertion that "when we let young children play, we let them also learn". This is, of course, not true. For children, schoolwork does not mean play in a world that offers infinitely more enjoyable pastimes. And children do not learn more mathematics because they get to dance and sing – as the education minister has also claimed – which most certainly think is more fun than solving equations. Progressivism certainly creates happy children. But knowledgeable they certainly do not become.
The explanatory model for Sweden's knowledge collapse put forward here thus indicates that Jan Björklund’s and the Right’s criticism of the school system is in many respects correct. But these changes hardly follow the political dividing lines that Björklund and others have tried to paint. It was, for example, bourgeois governments that ultimately pushed through the new curriculum in 1980 and 1994. The fuzzy school, for which Björklund blames the Social Democrats, was also signed off by the bourgeoisie.
At the same time, Björklund has himself helped and introduced a new Education Act in which the students' right to influence is still incorporated. He and Alli Ansen also created a new curriculum where the progressive elements were left virtually intact. Additionally, both the documents contain new formulations about "entrepreneurial learning" being strengthened, with forms of work that "stimulate imagination and creativity". In other words, the Alliance introduced an even more progressive pedagogy than before. There have been some other changes at the margin in the right direction, but serious attempts to change the school culture have not materialised. No party can disassociate itself from the responsibility of the developments we have witnessed. All have sung the praises of progressivism when it comes down to it.
Basically the policy is likely due to cultural changes, but it took time before these came to be seriously expressed among the population. During the rapid postwar economic development, the first emphasis was on the increased role of schools as a tool for social mobility. Higher expectations of education were partly behind the comprehensive school reform, which allowed more and more to continue studying at college and later also at university. Parents and students accepted the teacher's role in this process. With the relative rating system and related central and standard testing, it was also difficult to pressure teachers to get higher grades, even in cases where students and parents felt such behaviour appropriate. The belief in the school's transformative power and the incentive to work hard, therefore, remained to a large extent.
Here it is important to note that attitudes and cultural values ??tend to lag behind socioeconomic upheavals. Research indicates, for example, that it takes time before prosperity and welfare spending has a negative impact on people's work ethic. One reason for this is that the transmission of norms and values ??occurs mainly from parents to children. This means that it takes time for economic and social changes to seriously affect the culture.
The fact is that both an expanded welfare state and higher wealth at first appear to have positive effects on the probability of parents transmitting a strong ethic to their children, which is perhaps not so surprising. A worse life lives on in the minds of parents who have themselves made school trips and experienced prosperity increasing markedly in line with their efforts.
Somewhere there is a point where higher prosperity instead begins changing values ??and norms in the opposite direction. Neither is this especially strange. For after the class trips are done, children grow up instead up with a prosperity that they very much take for granted. The memory of a harder life atrophies with successive generations and finally no longer lives on. The norm to work hard in school decreases and parents want their children to have a carefree childhood without stress and anxiety.
It is, therefore, reasonable to assume that cultural changes are partial reason for Sweden's weak knowledge trend. A similar, though not as radical, development has occurred in many other Western countries; after an increase that lasted for decades, the cognitive skills of young people have started to decline even there. Cultural changes with regard to the emphasis on education and skills are likely one of several explanations for this worrying trend.
This is also probably a reason why countries like Estonia and Poland have increased their results. Similarly, Germany’s improvements have been driven primarily by regions of the former East Germany, which currently top national surveys. These countries and regions have probably not yet reached the crest of the development effect with its associated changes in the culture.
But instead of countering the negative impact of the changes to Swedish culture, the continued progressivism has rather stoked them. The idea is that the traditional knowledge-focused school is outdated in a post-industrial society with its individualism, which the school is said to be forced to adapt to. This is no place for teacher-centred methods, authority and traditional knowledge, it is said.
But, as the philosopher Hannah Arendt pointed out some 60 years ago, this is an illusion. The school should not aim to imitate the society at large. Its nature demands authority and tradition, which are paradoxically declining in importance beyond its walls. Students are not adults and we should not treat them as such. The freedom we enjoy in society is not always desirable within the school system.
Education’s paradox in the post-industrial society, rather, is that it is now even more important to uphold the authoritative structures and traditional relations in school, now that similar structures and relationships have split in the outside world. For no public institution has such great potential to influence pupils' standards and values ??in the right direction as the school. Relationships that are too egalitarian between students and adults risk pushing the development towards less acceptance of the institution and its authority even further. This in turn also creates difficulties in maintaining the school's and teachers' status in general. For who respects a permissive and powerless institution that focuses on games, dance and ‘entrepreneurial learning’?
At the same time, other reforms are also likely to have accelerated the decay of the knowledge culture. With the relative rating system's abolition in the 1990s, for example, incentives and opportunities among students and parents to force teachers to give higher grades increased, even though knowledge levels were sinking. In the new culture, respect for and trust in teachers has largely disappeared. Parents want their children to have a fun and pleasurable schooling, at the same time expecting they'll get into good high schools and later university educations. In this situation, forcing colleges and universities to accept the rating that the teachers have given seems downright insane. But, in Sweden, it is seen as normal.
Given the dystopia that is being conjured up, the question is naturally: what should we do about it? If we actually want to restore knowledge to the school, politicians must change the education system. The first step should be to take out the writings about student influence and ‘entrepreneurial learning’ that are in the Education Act and the curriculum. The state should simply not force schools to follow a pedagogy without the support of research.
The point is not that we must go back to the 1950s to restore the knowledge of school – this would be impossible in any case. Innovation should obviously be encouraged in schools. But you have to evaluate these innovations before they are adopted en masse. The most senseless thing about progressivism is not its obvious theoretical shortcomings, but that it was elevated to being the state ideology without having any research support.
A basic problem has actually been that school research has long been characterised by methods that are not suitable to evaluate the theories emphasised. The educational intelligentsia worldwide continues in the same spirit. In support of purely ideological standpoints, its representatives refer to substandard studies or simple observations of other countries. And they have so far managed to entangle politics with grand titles and meaningless platitudes.
But if we want to have a rigorous education system, we must make higher demands of the intelligentsia and its research. One suggestion is to follow the British government’s Education Endowment Foundation, which sponsors randomised field experiments within the school. More small-scale experiments would give us better opportunities to determine which ideas have potential and which are harmful, without risking additional system crashes in the process.
Another proposal is to centralise the rating system. This would give strong incentives to students (and parents) to focus on learning, rather than pushing teachers to give higher grades for less effort – particularly if they want to study at prestigious courses. Primary and secondary education should be simultaneously completed with centralised knowledge checks. Just as in, for example, the UK, France and Denmark. Perhaps we should reintroduce the relative rating system, which would further encourage students to work hard through competition and thus raise their average knowledge. With the decentralisation of admission system, schools would get incentives to focus specifically on the knowledge that higher education institutions value. Sample-based evaluations should also be introduced, so that knowledge among the cohorts can be followed over time.
These reforms are not exhaustive, but they are among the most important that must be investigated seriously. Most find support in the research, while others may need to be tested on an experimental basis. The reforms must, in any case, be up in the political sphere for discussion as soon as possible.
The culture of knowledge that has long been broken down by socioeconomic and educational transformations must be recovered. And to achieve this we need new incentives within the system that allow everyone to pull together in pursuit of better student performance. This will be much more difficult than it was to break down the educational ideals that had long been seen as self-evident. But there is no alternative. The knowledge culture is not going to rise up by itself.
Gabriel Heller Sahlgren is a PhD student at the London School of Economics and works with the Institute of Industrial Research.