For several years Sweden has fall down in the international rankings on education, especially in mathematics. As a consequence different political initiatives have been made; however, none has had the wanted effect on pupils’ mathematical performance. The latest reform involves mandatory national tests in mathematics for grade-3-pupils. From having one of the most decentralised educational systems in the EU, the pendulum is now heading another course, i.e. re-centralisation. International comparisons reveal that different countries have various kinds of standardised school tests. Still, Sweden appears to protrude with this step. The downside of standardised testing is that it may lead to situation specific stress, i.e. test anxiety (TA) among pupils. Repeated of times research has proved that serve TA has a significant negative impact on performance, and high anxious pupils considerable under-achieve. Thus, the result is not only biased, it also adventures the individual’s future academic and occupational choices. Some groups tend to be more exposed to TA; girls, elderly, and low-achievers are suggested to experiencing more TA. Additionally, a subject mentioned to call forth more stress is mathematics. Contrary to impairing education, there is substantial, but very often neglected, support for repeated testing to improve pupils’ learning, i.e. the testing effect (TE). Tests could enhance later retention more than additional study of the material, even without feedback, and is applicable for related but nontested material. This effect is robust across many types of materials, has been observed in different age groups, and under most circumstances. Moreover, repeated testing has been shown to reduce TA, make pupils more positive to education, and low-achievers seem to benefit more from it. A key factor influencing a pupil’s ability to learn is the cognitive system of working memory (WM). The WM is also stressed to be the underlying component in TA, and TE. Thus, when anxiety interferes with WM in a negative way there will be a knock-on effect on the capacity. Pupils are preoccupied with worrisome thoughts which interrupt the maintenance of task-relevant information and disrupt performance. Yet, WM is a component that is modulated by repeated testing. Every time a memory is retrieved it enhances the memory traces which in turn decrease the load on WM. In understanding the contribution of TA and TE, a proposed theoretical model based on the characteristics of the WM, outlined by Baddeley, is here suggested for this two-sided possibility of testing. Research on the effects of testing in younger children is generally a neglected area, and from a multi-theoretical perspective yet to be explored. In comparison to other countries Sweden has few tests in mathematics but now moves another direction. Consequently, the aim of this paper is to clarify the influence of national test in mathematics on young pupils’ learning. The following question will be addressed: i) What are pupils’ experiences of TA, and how does this affect learning in mathematics?, ii) How does recurring testing affect pupils’ learning in mathematics?, iii) Are different groups of pupils (i.e. gender, achievement level) experiencing above aspects differently?

Present study uses a refined and modernised standardised test measuring children’s TA, and a WM test for children tapping on the phonological loop (believed to be most sensitive to anxiety). Mathematical achievements are determined with the national tests. Statistical analyses (SPSS), supported by lengthy observations during the whole grade 3 year, are used. Forty grade-3-pupils (20 girls and 20 boys) in two classes are in focus.

Preliminary results suggest that WM capacity could predict mathematics achievement. However, there is no evidence for experienced TA among the pupils, thus no tendency for under achievement in any group. It is therefore possible that the increased number of testing, by the national tests and the teachers’ additional preparation tests at the prospect of the national tests, have a TE, i.e. learning in mathematics is benefitted by more testing of relevant as well as non-tested material. Further, the teachers’ attitudes and active focus on the national tests probably have had positive effect on the pupils’ test realisation. However, it is not excluded that the young age is answerable for the absence of TA.