This paper provides an analysis of how Swedish 15-year-olds perform on the high-stakes national assessments in geography. It explicitly addresses which item characteristics produce differential item functioning (DIF) in favor of boys and girls respectively. The findings show that DIF occurs in favor of girls in items with constructed response and primarily with content on the social dimension of sustainable development (SD), while boys are more favored by content outside the field of SD. The conclusions drawn are that content that reaches higher levels of Bloom's taxonomy favors girls, especially when the subject content concerns SD. This is important when analyzing the teaching and examination of sustainability issues in school.

To date, few studies have adopted a particular focus on the role of housing tenure when analysing ethnic and socioeconomic differences in out-mobility from poor neighbourhoods. This study contributes to filling this gap. The paper uses a full population data set covering every individual in the capital region of Sweden during the period 2006-2008. The findings indicate that the likelihood of leaving poor neighbourhoods increase for the foreign background population if their income is higher and they own their housing unit, while native Swedes seem to be less constrained by income. This lends support to the theoretical framework of place stratification. The results warrant efforts to broaden residential mix policy beyond the discussion on housing tenure if policy-makers want to counteract the ethnic and socioeconomic imbalances of residential mobility reproducing segregation.

Grade inflation, involving comparisons between public and independent schools, is presently under discussion in Sweden. This paper reports on discrepancies between national test results and the final course grade in social sciences, in year 9. Open data at school level were used in OLS regression analyses to scrutinise systematic differences between independent and public schools and between different geographical units. The geographical analysis was used to highlight systematic geographical inconsistencies between national test results and the final course grade. The main findings of the present study are: 1) The differences in grading between private and public actors are small when considering important background factors like the average educational level of parents, the school's grade point average (GPA), and which national test in social sciences was carried out at the school. 2) Schools in metropolitan areas are more inclined to give pupils a higher final course grade than their national test result, and it is less likely that schools in metropolitan areas and in municipalities surrounding the metropolitan areas give pupils a final course grade that is lower than their national test result. The regional differences might well affect pupils' continued studies, and the ambition of providing an equitable school to all children in all parts of Sweden has thus not been successful.

This full population study of travel-to-preschool patterns of the youngest children (< 6 years old) in metropolitan Stockholm analyses how school markets, even at an early stage, reproduce inequalities related to social and geographical distances. Our findings show that families with foreign backgrounds tend to convert educational capital into social capital by sending their children to preschools in more socioeconomically favourable neighbourhoods. Furthermore, we detect avoidance behaviour among the majority population in ethnically mixed neighbourhoods, which indicate that some native families are inclined to avoid preschools with high shares of non-native peers.

The central themes of this study are how housing tenure affects moving away from neighbourhoods with differing degrees of immigrant concentration in Stockholm and, comparing two cohorts, how the relationship between housing tenure, ethnic background and residential mobility develops over time. The findings reveal that the effect of housing tenure on moving varies marginally between neighbourhood types, and differences between housing tenures are reduced over time. Residential mobility is increasingly associated with higher income levels and members of the non-western, foreign-born group are less likely to move when the housing market is increasingly owner dominated. These ethnic hierarchies also apply to residential sorting. The sorting of outmovers from neighbourhoods with high concentrations of immigrants is increasingly affected by income. However, the non-western, foreignborn group is also increasingly dependent on housing assets in order to leave immigrant-dense neighbourhoods, which is not the case for other groups. The results show a shift in the necessary resources for residential mobility trajectories for those at the bottom end of the ethnic hierarchy and indicate that the place-stratification framework needs to differentiate between different types of assets and population groups to enable a better understanding of residential sorting in a changing housing market.

No abstract

Elever med utländsk bakgrund tenderar att prestera sämre än svenskfödda elever i skolan primärt på grund av sämre kunskaper i det svenska språket. Utifrån statistisk analys (Differential item functioning) identifieras uppgifter från de nationella proven i geografi (2014 – 2017) där elever som följer kursplanen i svenska som andraspråk klarar sig avsevärt mycket bättre - eller sämre - än förväntat. Tidigare forskning har visat att geografiska begrepp är särskilt svåra för elever som inte har svenska som modersmål, vilket också  påvisas i denna studie. Den visar att det särskilt är uppgifter med lite text som handlar om geografiska begrepp som uppvisar större skillnader i prestationer mellan elever som följer kursplanen i svenska respektive svenska som andraspråk. Resultaten kan stödja såväl lärare som provkonstruktörer att bättre anpassa undervisning och prov genom att undvika att skapa uppgifter som mäter irrelevanta bakgrundsfaktorer som påverkar elevernas möjligheter att besvara uppgifter på ett adekvat vis utifrån deras kunskapsnivåer.Nyckelord: Nationella prov i geografi, uppgiftsformat, elever med utländsk bakgrund, svenskfödda elever, DIF-analysWhat types of test items benefit students who follow the syllabus in Swedish as a second language? A study using data from the Swedish national assessments in geography.AbstractPupils born outside Sweden are likely to accomplish less in comparison to native pupils, primarily as a result of inferior knowledge of the Swedish language. Based on a statistical analysis (Differential item functioning) of questions given at national tests in geography (2014-2017), it was possible to identify questions where pupils following the syllabus of Swedish as a second language attain either considerably better or more inferior results than expected. Earlier research has shown that pupils whose native language is not Swedish find it particularly hard to comprehend geographic concepts, which was confirmed by the present study. This study furthermore revealed that in particular questions containing a limited amount of text concerning geographic concepts resulted in larger differences than expected between native pupils following the syllabus in Swedish and foreign born pupils following the syllabus in Swedish as a second language. These findings could aid teachers and test constructors in their efforts to adjust teaching and tests by not formulating questions that measure irrelevant background factors, which might affect the pupils’ ability to answer questions adequately, based on their level of knowledge.Keywords: National tests in geography, question format, pupils born outside Sweden, Swedish-born pupils, DIF-analysis

Preschool segregation has not been the focus of research efforts to the same extent as compulsory school segregation. This is at least in part a consequence of the lack of large-scale, registry-based data sources on where children live and where they attend preschool. This paper presents a full-population account of discrepancies between preschool segregation and neighbourhood segregation covering the Swedish population. Data includes preschool children as well as their parents’ income, education, ethnic background, and place of residence. Findings indicate that while preschool segregation does not differ from neighbourhood segregation to the same extent as previous research has shown for school segregation, there are systematic differences affecting the level of segregation across Sweden and in various types of municipalities. Studies on school level show segregation by foreign background and income to be most prominent, whilst preschool segregation mostly concerns parents’ educational attainment. Furthermore, the findings show that the geographical distribution of private and public preschools affects levels of segregation. This conclusion supports the general argument that the free-choice reform in the Swedish school system tends to raise levels of school segregation above the levels of residential segregation—even in preschools.