Research results

Centre for Strategic Research in Education (File No. 09-070295)
Qualitative status report, April 2014 – March 2015.
Research results
The present and coming findings are of high value, since they concern efficiency and cost-benefit
considerations central to policy, administration and practice.
WP1 (Day-care, pre-school and child development)
Early discharge from hospital on the day of birth does not cause a lasting damage on health on a general
population of mothers and infants, but it impacts health and school outcomes in the longer run for at-risk
children. (Sievertsen and Wüst, 2014).
There is a significant positive relationship between local day care and offspring outcomes seventeen to
thirty years later for children with low maternal schooling while there is no effect for higher educated
mothers and fathers (Bingley and Westergaard-Nielsen, 2012).
Three out of five quality indicators, a higher staff-per child ratio, a higher share of mall staff, and a higher
share of staff with formal preschool teacher training are associated with significant improvement in
childrens’ test results in Danish at the age of 15. Boys benefit more from preschool quality than girls. Ethnic
minority children benefit from higher staff stability (Bauchmüller et al., 2014).
Systematic quality improvement of preschool practice reduce development of emotional symptoms,
conduct problems, hyperactivity and enhance attention. However, the intervention proves less beneficial
for socially disadvantaged children, in particular as a consequence of unfavorable preschool characteristics
(Jensen et al., 2013a; Jensen et al., 2013b, Jensen et al. 2015).
Delayed school start dramatically reduces hyperactivity, a measure with strong negative links to student
achievement. However, the estimated effects on non-cognitive dimensions such as emotion, conduct, peer
relations and social skills are small and insignificant. (Dee and Sievertsen, 2014).
There are strong within-generation effects of amount of schooling on lifetime earnings for fathers but they
do not carry over to the next generation (Bingley et al., 2014).
Does formal schooling contribute to individual labor market productivity or does it act as a signal to
employers of predetermined labor market skills? The research show that signalling explains most – if not all
of the returns to schooling. Almost the entire private financial returns to extra year of schooling or high
school completion are earned during the first 10 to 15 years in the labor market (Markvardt, 2014).
In a study of measurement error in income and schooling the researchers – contrary to most validation
studies – found that measurement error in income is classical, once imperfect validation data is accounted
for, and there are non-classical measurement errors in schooling, causing a 21% amplification bias in
estimators of the returns (Bingley and Martinello, 2014).
The relationship between retirement and cognitive function at older ages is ambiguous and has produced
mixed results. The study finds that this is caused by a negative bias from not taking differentials of
schooling in account (Bingley and Martinello, 2014).
WP2 (Preschool, compulsory school – pupil and resource measures, special education)
A comparison of centre-based day care and family day care at age 2 show that centre-based day care
significantly improves grades in Danish language at the GPA at the age of 15. Effects on math grades are
smaller in size and not statistically significant (Gupta and Simonsen, 2014).
There is a robust positive but numerically fairly small relationship between school size and alternative
measures of long term success in the education system and the labour market. The positive impact of
school size seems mainly to be driven by boys and students from families with low educational level
(Humlum and Smith, 2015).
There are statistical significant, but rather small, class-size effects in 10th grade indicating that reducing
class size has beneficial effects. There is no significant effect heterogeneity with respect go gender,
immigrant status, lagged achievement, or parental characteristics (Krassel and Heinesen, 2014).
There are significant albeit modest negative effects of class size increases for children in primary school
levels. The effects on math abilities are statistically different across primary and lower secondary school
with more negative influence at the primary level. Larger classes do not affect girls, non-Western
immigrants and socioeconomically disadvantaged pupils more adversely than other pupils (Nandrup, 2014).
Parent preference for class size reduction weakens the fewer pupils there are in the class and may even be
negative when the number for pupils in the class is below 17, where benefits of educational and social
peers becomes low (Krassel et al., 2014).
Both cognitive and non-cognitive skills are important for educational outcomes. The study finds that
academic achievement and self-confidence are found to be important for enrolment in high school, while
academic achievement and perseverance are important for completion of high school. With respect to
vocational education, neither academic achievement nor self-confidence and perseverance predict
completion. Basic attendance measures, however, are strong predictors of completion of vocational
education (Krassel, 2014).
Studying how peers’ achievement gains are affected by the presence of potentially disruptive and
emotional sensitive children in the school-cohort show that adding children with divorced parents, children
with parents convicted of crime, and children with a psychiatric diagnosis lower the academic achievement
of peers by about two percent of a standard deviation (Kristoffersen et al., 2014).
There are large negative associations between indicators of externalizing behavioural and lower school
outcomes in 9th grade GPA and being enrolled in high school or vocational programmes. The general gender
gap in reading is mainly the result of gender differences between children without behavioural problems
living in high-resource families (Kristoffersen et al., 2014).
Bullied children suffer in terms on educational performance measured by 9th grade GPA and the effects of
victimization tend to increase with severity. The quality of the family environment as well as individual child
characteristics such as poor early mental health, indications of hyperactive behaviour, and physical
appearance are important drivers of victimization (Eriksen et al., 2014).
A study of relative standards in ADHD diagnosis show that being relatively young-for-grade does not
significantly increase the probability of receiving an ADHD diagnosis which has been found in US and
Canadian studies (Dalsgaard et al. 2012).
Behaviour of specialist physicians varies considerably across hospitals and the prescribing behaviour does
affect the probability that a given child is treated. Results show that children diagnosed with ADHD in
pharmacological treatment have fewer hospital contacts if treated and that treatment to some extent
protects against criminal behaviour (Dalsgaard et al, 2014).
Heterogeneity in preferences for wages and occupational prestige leads to gender differences in
occupational choices. In a Danish sample women express a stronger preference than men for occupations
that are more valuable for society. The gender differences are most pronounced among individuals from
lower socioeconomic backgrounds (Kleinjans et al., 2014).
The admission system to college education mainly affects the timing of college enrolment and not the
college going decision. Delaying college enrolment has substantial effects on the timing of family decisionmaking and child bearing. Policy makers may want to consider how to increase incentives to enrol earlier by
increasing the quality of guidance counselling in high school (Humlum et al., 2014).
Certain groups of pupils are disadvantaged by the form of assessment where teacher feed-back stands
alone as it does in Denmark where choice of upper secondary education is made before the GPA in 9th
grade. These groups are boys, low-SES and migrant pupils (Rangvid, 2014).
Most studies on the effect on school quality do not take account of changes in student effort. This study
shows that an increase in school resources may reduce student effort. Therefore information on student
academic effort and time use should be included (Browning and Heinesen, 2014).
WP3 Effect of teacher competence in classroom management and didactics on learning outcomes)
A randomized and controlled trial of teacher in-service training in classroom management has shown no
effect on general intelligence tests, small, but significant effects on boys’ reading skills, and significant
effects on girls’ concentration performance. There is no effect on student behaviour but teachers become
more observant on certain forms of behaviour; hyperactivity, emotional problems and prosocial behaviour
(Keilow et al., 2015).
WP4 (Transition problems from compulsory education to upper secondary and vocational education)
An international research mapping covering strategies to improve recruitment, reduce drop-out and
improve results in vocational education has shown that it is important to raise the reputation of the
vocational training system, and to give better counselling in lower secondary school. Students in lover
secondary training find it hard to make a decision on their career at the age of 15, they find the vocational
training system confusing and therefore prefer the academic upper secondary tracks, delaying decision for
three years (Dyssegaard et al., 2014a).
A mapping of research covering strategies to improve recruitment, reduce drop-out and improve results in
academic upper secondary education has shown that the academic tracks are perceived as attractive and
natural extensions of lower secondary education. Support systems taking hand of student with moderate
academic skills reduce drop-out but in many cases has the effect of producing graduates from upper
secondary lacking the skills to continue in college education. (Dyssegaard et al., 2014).
A study covering 15-20 year old Danes outside the education systems show that half of the group had poor
relations to their teachers in compulsory schooling, two thirds had problems concentrating. The youngsters
are motivated to have a job with educational aspects if they have a monthly salary, but very often they do
not have the personal and social skills required in a job (Egelund and Barkholt-Spangsbo, 2012).
Interviews with students graduating from initial vocational training showed only half of the group had been
counselled by their teachers about vocational training and one third had been informed about job
possibilities and possible further education after vocational training. In contrast to this two thirds had been
counselled about the academic upper secondary track. (Ottosen-Støtt and Egelund, 2014).
Almost 2,000 students in grade 9 in December 2011 are followed in their transition to upper secondary
education. Variables predicting results from students’ GPA in 2012 were gender, ethnic background and
one parent families. General wellbeing is positively correlated with good GPA results. The opposite is the
case for disturbing behaviour. Analysis of students’ success in upper secondary is ongoing (Krassel, 2015).
Almost 2,000 students participating in PISA 2000 have been followed up in 2004, tested in PIAAC in 2012
and followed up in administrative databases in 2014. Data show improvement in reading skills from age 15
to 27 for men who have pursued academic education, who are geographically mobile, without illness and
unemployment. Skills founded in compulsory education are to a high degree predictors of later success in
education and employment (Rosdahl, 2015).
Why do some young people have success in education and/or the labour market despite bad odds at age
15? This question is under investigation in 2015 (Jensen et al., 2015).
Non-cognitive abilities and their influence on labor market returns to cognitive abilities. In progress (Krassel
and Sørensen, 2015).
Skills, education and labor market position. PISA data, PIAAC data and investment in education is related to
success in the labor market. Persons in the NEET category – Neither Employment nor Education nor
Training – will be a special focus. In progress (Albæk, 2015).
Industry and societal results
The research cover several areas of high actual political interest: What is high quality day care, and can it be
improved with formal training programmes? Is school start at the age of six harmful for some children? Are
big schools better than small schools? Is it worthwhile to invest in smaller class sizes? What lies behind
gender differences in behaviour and school outcome? Has the use of the diagnosis ADHD run wild? What
happens to the classmates when a pupil with disruptive behaviour is included? What are the consequences
of bullying? Why do so many students drop out of vocational training? Why does so many students chose
the upper secondary academic track? Which factors determine occupational choice and success? Is
proficiency in reading improving or degrading after the age of 15? What are the effects of unemployment
on personal skills?
Two studies have been on the front page of Berlingske Tidende in March 2015. One about inclusion, and
one about school size.
Research education
There is in April 2015 one ph.d. student with expected thesis submission in October 2015. One ph.d.
student defended her thesis in 2013, four defended their thesis in 2014. In total six postdocs have been
employed, three are still employed as postdocs in the project.
Ph.D.-students and post-docs have been employed by the relevant research community as defined by
research themes, projects/sub-projects and supervisors. Ph.d. students attend doctoral programs
appropriate to their field of study, and the strategic research centre support the participation of
international experts in courses to create a common and cooperative educational research community
covering among other things use of administrative registers, experimental and non-experimental research
designs, multivariate and multi-level statistical analysis. Four common seminars with international
participation have taken place in each of the years 2011, 2012, 2013 and 2014.
Collaboration
Each of the key researchers involved in the centre have participated in many collaborative projects and
have well-established relations to various research communities and stakeholders. The research
collaboration covers the following institutions: Maastricht University, University of Copenhagen, Rockwool
Foundation Research, Karolinska Institute, Copenhagen Business School, Stanford University, Harvard
University, California State University, University of Southern California. In the strategic research centre,
these relations have been strengthened and put to full use for the first time in Danish educational history.
Furthermore, the strategic research centre has access to experienced staff, research groups, international
relations and graduate training programmes at the partner institutions.
Niels Egelund
March 31, 2015
Skriftlige arbejder CSER 2010 – marts 2015
Peer reviewede artikler
Egelund, N. (2010): Hvilke barrierer er der for at nå Regeringens 95 % målsætning) Pædagogisk Psykologisk
Tidsskrift, vol. 47, no. 5. 373-383. – 100% - not open access
Egelund, N. (2011): Gennemførelse i uddannelsessystemet – hvilke barrierer er der? Paideia, Vol. 1, nr2. 814. - 100% – not open access
Gupta, N.D. and Simonsen, M. (2012). The effects of type of non-parental care on pre-teen skills and risky
behavior. Economics Letters, Volume 116, Issue 3, 622-625. - 100% - not open access
Dalsgaard, S.; Humlum, M.K.; Nielsen H.S. and Simonsen, M. (2012), Relative standards in ADHD diagnoses:
The role of specialist behavior. Economics Letters 117: 663-665. 75% – not open access
Jensen, B.; Holm, A. and Bremberg, S. (2013). Effectiveness of Danish early year preschool program: A
randomized trial. International Journal of Educational Research. Volume 62 (2013) 115-128 - 100% - open
access
Egelund, N. (2013). En bedre skole for drenge. KVAN, vol. 33, nr. 97. pp 7-16. – 100% - not open access
Humlum, M.K.; Kristoffersen, J.H.G. and Vejlin, R. (2014). Timing of college enrollment and family formation
decisions. IZA Discussion Paper Series. DP No. 7905. 47 pages. – 100% - open access
Heinesen, E. and Browning, M, Study versus Television. IZA Journal of Labor Economics 2014, 3:2. 16 pages
– 100 pages – open access
Krassel, K.F. and Heinesen, E. (2014). Class-size effects in secondary school. Education Economics, DOI. 15
pages -100% - open access
Bauchmüller, R.; Gørtz, M. and Rasmussen, A.W. (2014). Long-Run benefits from universal high-quality
preschooling. Early Childhood Research Quarterly. Volume 29. 4. Pp 457-420. – 25% - open access
Dalsgaard. S.; Nielsen, H.S. and Simonsen. M. (2014). Consequenses of ADHD medication use for children’s
outcomes. Journal of Health Economics. Volume 37, 137-151. – 100% - not open access
Eriksen, T.L.M.E.; Nielsen, H.S. and Simonsen, M. (2014). Bullying in elementary school. Journal of human
resources. Volume 49. 4. 839-871. – 100% - not open access
Kristoffersen, J.H.G.; Obel, C. and Smith, N. (2014). Gender Differences in behavioral problems and school
outcomes. Journal of Economic Behavior and Organization. JEBO-3441. 20 pages. 100% - open access
Humlum, M.K. and Smith, N. (2015). Long-term effects of school size on students’ outcomes. Economics of
Education Review. Volume 45. 28-43. – 100% - open access
Kristoffersen, J.H.G.; Krægpøht, M.V.; Nielsen, H.S. and Simonsen. M. (2015). Disruptive school peers and
student outcomes. Economics of Education Review. Volume 45. 1-13. – 100% - open access
Ph.D. afhandlinger
Jannie H.G. Kristoffersen. (2013). Empirical Essays on Economics of Education. Aarhus University. 138 pages
- 100%
Kristoffersen, J.H.G.; Obel, C. and Smith, N. (2014). Gender Differences in behavioral problems and school
outcomes. Journal of Economic Behavior and Organization. JEBO-3441. 20 pages. 100% - open access
Humlum, M.K.; Kristoffersen, J.H.G. and Vejlin, R. (2014). Timing of college enrollment and family formation
decisions. IZA Discussion Paper Series. DP No. 7905. 47 pages. – 100% - open access
Kristoffersen, (2013).Vulnerable Children and Peer Effects. Aarhus University. 44 pages. - 100%
Krassel, K.F. (2014). Essays in economics of education. Aarhus University. 153 pages – 100%
Krassel, K.F. and Heinesen, E. (2014). Class-size effects in secondary school. Education Economics, DOI. 15
pages -100% - open access
Krassel, K.F.; Ladenburg, J. and Dalsgaard, C. (2014). Balancing the risk of ‘lazearian’ interrupters and the
benefits of educational and social peers: Tracing parental preferences for class-size reduction. Aarhus
University. 19 pages. - 100%
Krassel, K.F. (2014). Completion of upper secondary education: The roles of cognitive and non-cognitive
skills. Aarhus University. 45 pages – 100%
Kleinjans, K.; Krassel, K.F. and Dukes, A. (2014). Occupational prestige and the gender wage gap Working
paper. Aarhus University. 43 pages - 100%
Sievertsen, H.H. (2014). From birth to graduation. Copenhagen University. 120 pages – 100%.
Sievertsen, H.H. and Wüst, M. (2014). Before Midnight: Discharge on the day of birth, parental responses
and health and schooling outcomes. Copenhagen University. 42 pages. – 75%
Dee, T.D. and Sievertsen H.H. (2014). School starting age and non-cognitive skills. Copenhagen University.
71 pages. – 75%
Markwardt, K. (2014). Education and unemployment insurance. Copenhagen University. 122 pages. 50%.
Markwardt, K. (2014). Signaling and productivity in the private financial returns to schooling. Copenhagen
University. 26 pages. 50%
Markward, K.; Martinello, A. and Sándor. L. (2014). Liquidity substitutes unemployment insurance: Evidence
from the introduction of home equity loans in Denmark. Copenhagen University. 29 pages. 50%.
Bingley, P.; Markwardt, K.; Ehrenstein, V. and Pedersen, L. (2014). Draft lottery effects on schooling,
earnings and the next generation. Copenhagen University. 24 pages. 50%.
Martinello, A. (2014). Schooling and savings. Copenhagen University. 159 pages. 50%.
Bingley, P. and Martinello. A. (2014). Mental retirement and schooling. Copenhagen University. 15 pages.
50%.
Bingley, P., and Martinello. A. (2014). Measurement error in income and schooling, and the basis of linear
estimators. Copenhagen University. 57 pages. 50%.
Videnskabelige bøger og antologier
Bingley, P and Westergaard-Nielsen, N. (2012). Intergenerational transmission and daycare. In Ermisch, J.;
Jäntti, M. and Smeeding, T. (eds.). Inequality from childhood to adulthood: a cross-national perspective on
the transmission of advantage, pp 190-203. Russell Sage Foundation: New York - not open access - 50%
Videnskabelige rapporter
Jensen, B.; Rasmussen, A. and Jensen, P. (2013). VIDA. Vidensbaseret indsats over for udsatte børn i
dagtilbud – modelprogram. VIDA-forskningsserien 2013:04 91 sider – 100%
Friis-Hansen, M. (2014) Undersøgelse af anvendeligheden af opmærksomheds- og koncentrationstesten d2 i
folkeskolens indskoling. København, SFI. 100%
Barkholt-Spangsbo, T. og Egelund, N. Forundersøgelsen til Young Skills. Region Hovedstaden, 2013. 20%
Egelund, N og Barkholt-Spangsbo, T: Unge og virksomheders uddannelsesengagement - jobs med
uddannelsesperspektiv. Region Hovedstaden 2012 - 25%.
Dyssegaard, C.B.; Egeberg, J.H.; Steenberg, K. Tiftikci, N. og Vestergaard, S. (2014a). Forskningskortlægning
af håndterbare forhold til gavn for fastholdelse, øget optag og forbedrede resultater i
erhvervsuddannelserne. Dansk Clearinghouse for Uddannelsesforskning, IUP, Aarhus Universitet,
København. - 25%
Dyssegaard, C.B.; Egeberg, J.H.; Steenberg, K. Tiftikci, N. og Vestergaard, S. (2014b). Forskningskortlægning
af håndterbare forhold til gavn for fastholdelse, øget optag og forbedrede resultater i de gymnasiale
uddannelser. Dansk Clearinghouse for Uddannelsesforskning, IUP, Aarhus Universitet, København. - 25%.
Ottosen-Støtt, M og Egelund, N. (2014). EUD – vejledning, frafald og praktik. Baseline studie i Young Skills
IUP, Aarhus Universitet. København. - 20%
Keilow, M.; Friis-Hansen, M.; Kristensen, R.M. og Holm, A. (2015). Effekter af efteruddannelse af lærere i
indskolingen. København, SFI. 272 sider. – 100%
Working papers
Landersø, R.; Nielsen, H.S. and Simonsen, M. (2015). School Starting Age and Crime. Economic Journal.
Revision needed. - 100%.
Gupta, N.D. Lausten, M. and Pozzoli, D. (2012). Informant Discrepancies in Assessing Child's Functioning –
Economic Working Paper, Aarhus University - open access - 100%.
Jensen, B.; Holm, A. and Bremberg. (2013). The Effect of an Inclusive ECEC-intervention Program on Child
Strengths and Difficulties. CSER Working paper. No. 9. - 66% - open access
Gørtz, M. and Simonsen, M. (2014). Academic Redshirting, Achievement and the Gender Composition of
Preschool Teachers . Draft - 100%
Gupta, N.D. and Simonsen, M. (2014). Academic performance and type of early childhood care. Manuscript
33 pages - 100%.
Beuchert-Pedersen, L.; Humlum, M.; Nielsen, H.S. and Smith, N. The Short-term Effect of School
Consolidation on Student Achievement. Work in progress. - 100%
Krægpøth, M.V.; Nielsen, H.S. and Simonsen, M. Special School Participation and Student Outcomes. Work
in progress. - 100%
Rangvid, B. (2114). Systematic Differences Across Evaluation Schemes and educational choice, Manuscript.
27 pages - 100%
Egelund, N. and Nordahl, T. (2013). Youngsters in transition. Manuscript. 15 pages - 100%
Jensen, B., Jensen, P and Rasmussen, A.W. (2015). Does Professional development of preschool teachers
improve child socio-emotional outcomes? CSER Working Paper. 40 pages. 50%.