Headlines 2015

Teaching methods might be one of the school's problems


Sanna Rayman, editor at the weekly paper Dagens Samhälle, moderated the seminar about school research. The researchers (from right) Karin Edmark, Gabriel Heller Sahlgren, Jonas Vlachos and Björn Öckert. Photo: Fredrik Eriksson.

On November 25, four economists discussed school research at a seminar organized by IFN: Karin Edmark, Gabriel Heller Sahlgren, Jonas Vlachos and Björn Öckert. The first three being affiliated to IFN. The venue was packed and the debate was lively about what researchers really know in regards to the causes of the failing Swedish school results. The researchers all agreed that recruitment to the teaching profession is one of several key explanations.


The panel consented that the school choice reform, if any, has given a positive result but that a new grading system is needed. They all argued that methods used by economists are superior when studying for example effects on the education market.

The Swedish government has appointed a School Commission which, at the beginning of 2017, is mandated to propose how the education system can be improved. The Commission will analyze problems and try to establish why school results are declining. A large number of economists are outstanding researchers in the field of education; still no economist is part of the Commission.
This was the starting point for a seminar organized by IFN on November 25. Four economists had been invited to discuss declining results in Swedish schools. Sanna Rayman, Dagens Samhälle, moderated the discussion. The researchers were Karin Edmark, Stockholm University, Gabriel Heller Sahlgren, London School of Economics, Jonas Vlachos, Stockholm University and Björn Öckert, the Institute for Labor Market and Education Policy Evaluation (IFAU). All, except the later, are affiliate to IFN.


Magnus Henrekson.jpg

Magnus Henrekson noted that a wide range of economists are prominent researchers on school issues. Photo: Fredrik Eriksson.


Magnus Henrekson, professor and managing director of IFN, opened the seminar remarking that a wide range of economists are prominent researchers on school issues. He pointed out that more systematic evaluation of school results are needed, to know what students learn and how difficulties can be tackled. He stated that "if we do not have high quality in our education system it will, in the long run, be difficult to sustain prosperity".

Sanna Rayman launched the debate by asking the panel at what point in time the Swedish school started to deteriorate. “The turndown has undoubtedly occurred at all school levels and in all disciplines and among all categories of students. The decline appears to have begun in the late 1980s, i.e. before the major school reforms in the 90s”, said Björn Öckert. He explained that the decline starts at an early age and is reinforced by the time. This means that something happens already in primary school and actions need to be taken at that early stage.
“There are signs that the downturn after 2000 is dramatic”, said Jonas Vlachos. However, different measurements give different results. And, said Björn Öckert, it is difficult to measure absolute changes in knowledge, especially in the past. Data prior to 2000 are uncertain.


Öckert o Vlachos.jpg

Björn Öckert explained that economists are good at studying effects and markets, including school markets. To the right, Jonas Vlachos. Photo: Fredrik Eriksson.

In the directive of the National School Commission equivalence is important, said Sanna Rayman and asked the panel to define. “Equivalence means that schools should have about the same quality. This means that a student – given his or her circumstances – should learn about the same regardless of which school he or she attends”, said Vlachos.

Björn Öckert explained that research is showing that most poorly performing students benefit when additional resources are injected into a school. Karin Edmark pointed out that there is an inherent contradiction between equality and that students can choose school and education, and “this is something we need to discuss”. She also raised the question of how much the school system can handle in regards to a diverse student population. Gabriel Heller Sahlgren argued that it is not at all for sure that students becoming more sympathetic to each other because they are "mixed".

Segregation per school or per class is there an optimal balance, Sanna Rayman asked. “It depends on what we want to achieve”, said Heller Sahlgren who does not believe in a peer effect, i.e. talented students motivating less able students to do better.

Do schools compensate for different external factors, such as changes in pupil composition? Björn Öckert explained that hardly anything has changed in this regard despite the fact that schools today have new principals. He argued that the increasing segregation between schools is a result of housing segregation, but also that students can choose between schools. At the same time, there is nothing to suggest that the quality gap has increased between the different schools, despite the fact that schools with a "favorable student population" much easier attract teachers and leadership, argued Jonas Vlachos.

“It is possible that compensation plays a key role [in the balance between schools] as popular schools attract more students and fewer are left in the less popular ones”, said Karin Edmark and continued: “I am happy that students' social background, over time, doesn’t play a major role. Though, this does not mean that school choice is unimportant, or that we choose for the wrong reasons.

“We know that parents today care very much about what school they send their children to. At the same time IFAU shows that quality differences between schools are almost negligible. This shows that parents are misinformed. The fact that schools are much alike is important information for both parents and politicians”, said Jonas Vlachos and continued: “Though, there is probably no parent who believes this ...” Vlachos argued that if most people choose school with the pupil composition taken into account rather than quality, free school choice is not a particularly positive force for education.

With the large in-flow of refugees, there is reason to believe that we will have an even greater segregation in the future, said Sanna Rayman, referring to Gabriel Heller Sahlgren’s study how the Pisa results are affected by students with non-Swedish background. He found that 25-29 per cent of the decline in results can be explained by these students' performances. During the seminar, he explained that this might also have a spillover-effect on Swedish students through redistribution of resources to immigrant students. But, he noted, the issue is problematic to examine.


Karin förklarar.jpg

Karin Edmark (far right) has primarily conducted research on school choice, private schools and school competition.


Have schools been disadvantaged or are they making a worse job, asked Jonas Vlachos adding that conditions in Swedish schools have certainly deteriorated, but at the same time some things counteract this, such as parents' becoming better and better educated. And, said Vlachos, even if we remove the 30 percent students of immigrant background, it is still a dramatic decline in knowledge among pupils in the Swedish school. “More refugees are coming to Sweden, and the segregation is increasing. This will probably require reallocation of resources between schools. With school choice, it is important to make sure that these students end up in the right schools”, said Karin Edmark.

What made the school performance fall so sharply, asked Sanna Rayman.
“The most likely explanation is the deteriorating education quality. Research also shows that teachers are extremely important, but not exactly how they are important”, said Jonas Vlachos.

How has the schools’ way of operating changed?
The panel discussed and came to the conclusion that researchers know very little about what happened when individual work (eget arbete) and the teacher as a mentor rather than conveyor of knowledge were introduced.

Does enjoyment contradict the learning process?
“It is a struggle to learn and a question about being motivated”, said Bjorn Öckert. “Individual work [ in the sense that pupils work more by themselves with less teacher involvement] has increased and education has become less teacher-controlled. Economic research suggests that such teaching is not good at all. Similar methods were introduced in Canada, where the results fell in a very short time”, said Heller Sahlgren. Jonas Vlachos wasn’t sure that we have a "fuzzy school" (flumskola). He claimed that in many ways school has become more competitive, partly because of grading being used for admission to high school.



“Paradoxically, in Swedish schools, we have little teaching time and a relatively high staffing ratio and still Swedish teachers, in an international comparison, are working the most hours and are most stressed and unhappy about their working situation”, said Jonas Vlachos. Photo: Fredrik Eriksson.

“Today parents and students feel entitled to try to influence teachers more than they did 30, 40 years ago”, said Gabriel Heller Sahlgren. He argued that this is due to cultural changes: “It is not obvious that you learn more if you enjoy school. Look at countries like Korea, Poland, Latvia and Estonia. These countries show low student happiness and the pupils do not like their teachers, but they learn. Swedish students, on the other hand, have better self-esteem – perhaps because of the new teaching techniques. They think they are better though they are inferior.”

Björn Öckert threw in a disclaimer: “We speculate about declining results, but we do not know. Perhaps reduced resources have caused the decline? Probably not. Could it be the teachers' qualifications? Well … it might be in the long run. Could it be private schools? The evidence suggests that private schools have a slightly positive effect. Could it be immigration? Yes, maybe a little. So, only one explanation remains: that something has changed in the way schools work and teach. Still, we lack suitable research on the effects of sound teaching.

Öckert asked why we still know so little. He gave a two-part answer: 1) prior to 2000s data on pupils and results primarily at younger ages were missing; 2) the research tradition in the field of education has largely excluded effect evaluations. “What we know”, said Jonas Vlachos “is that the competence of teachers has dropped very dramatically since 1990-95. I would be surprised if this hasn’t had a negative effect. There are studies showing that teachers' subject knowledge is important for what students learn. This is not something that has been a priority of the principal, i.e. state and municipalities. Many teachers teach subjects that they have no or little knowledge of.”

The panel discussed the rise in teachers' salaries and agreed that this only will not to raise the status of the profession, even though recommended by the OECD after Sweden’s fall in Pisa. One problem is choosing teachers that are good at teaching. And this, said Karin Edmark, one can’t regulate in detail. It is up to each school and its principal.



The seminar was held at the Armémuseum in Stockholm. The venue was jam-packed and the highly qualified and engaged audience posed numerous questions.
The grading system was discussed before the audience was let in to ask questions. All four panelists advocated a new grading system. Björn Öckert and Jonas Vlachos agreed that it is unreasonable having a system with failing/non-approved as a result. “Sharp penalties for students and schools that are not approved create focus on the weakest pupils, funneling resources from the general education. There is anecdotal evidence that these students take up an awful lot of teacher time. I think this is a major explanation for the decline in results. There is a this-students-will-make-it-any-how-attitude. Paradoxically, in Swedish schools, we have few teaching hours and relatively high staffing ratio and still Swedish teachers work, in an international comparison, the most hours and are most stressed and unhappy about their work situation. Something must be wrong organization wise!”

Karin Edmark pointed out that "it is important to measure students' knowledge, not only using measurements that make the system look good." However, it is very difficult to construct such a system: “In this field a lot is going on and new types of tests are being tried out. Tests that do not require so much of the teacher's time to correct. Björn Öckert suggested that grading be anchored in some kind of objective measure of students' knowledge, for example, the average level of national examinations per school. That would curb tendencies to grade inflation.

Text: Elisabeth Precht

Read comments from the seminar at  #IFN_skola.

View the seminar (in Swedish)

Research Institute of Industrial Economics, Grevgatan 34 - 2 fl, Box 55665, SE-102 15 Stockholm, Sweden | Phone: +46-(0)8-665 45 00 | info@ifn.se