The goal of the policy was to increase child care enrollment of children from immigrant families, contribute to socialization and better Norwegian skills, strengthen the competence of child care staff in multicultural pedagogy and language stimulation, and contribute to improving the contact between parents and the child care center. In three previous reports, we have studied effects on National tests in third, fifth, and eighth grade. In all of these reports, we found a positive effect of the free child care policy on reading mastery, with particularly large effects in subgroups of children from low-income families.
Children in districts with and without free child care are different, and a comparison that does not take these differences into account will not provide a trustworthy estimate of the effect. To adjust for such differences, the findings are based on analyzes in which I compare the results of cohorts of children before and after the offer of free child care was introduced in a difference-in-difference model that also takes into account before and after results in city districts without the policy in place.
The difference-in-difference estimates show that there is no difference in school results at the end of compulsory school for children that did and did not get an offer of free child care. This is true both for children with and without immigrant background. In previous reports, we have found a stronger effect for children from families where the mother does not work and from low-income families. We do not find similar heterogeneous results for end-of-compulsory school grades, neither across families where the mother does not work nor in low-income families.
There may be several reasons why the findings in this report differ from the three previous reports. Perhaps there is a strong emphasis on equalizing differences between students in the last years of compulsory school. Perhaps it is difficult to detect small differences in compulsory school grades. From previous analyzes of national tests, the effect on reading mastery has been the most robust. This variable indicates whether the children score above the lowest level of reading mastery. There is no such variable on the compulsory school diploma, and it is therefore difficult to perform comparable analyses.
It is not uncommon in the literature on effects of early childhood interventions to see that they become weaker over time (see an overview and discussion of the phenomenon in Bailey et al. 2020). Some studies also find that the cognitive effects of early childhood interventions can become weaker and disappear, only to reappear as a higher probability of completing education or having better impulse control as adults (see, e.g., Campbell et al., 2014; Heckman, 2006; Schweinhart et al., 2005). Whether the latter can apply to the children who were offered free child care in Oslo will only become meaningful to study in a few years when those who were affected by the policy complete upper secondary school and proceed to study and enter working life.