Aim In the longitudinal study Children’s Early Learning (Sheridan, Johansson & Pramling Samuelsson, manuscript), the aim is to find out if and in what way it is possible to find any relations between the quality of preschool and children’s learning. The quality of each preschool was evaluated by with a revised version of the Early Childhood Environment Rating Scale (ECERS) (Harms & Clifford, 1980). There are many areas of research in the project as a whole, but in this particular presentation we will focus on the very young children’s learning and approaches. One of the fields of learning is basic mathematics. The question asked in this paper is: What are the various ways in which children between 1 and 3 years of age relate themselves to play tasks on early mathematics. And in what way is this related to the quality of preschool. Theory The study as a whole has a longitudinal design and is divided into different parts and stages. The complexity of preschool as a learning context can only be grasped by adopting several research approaches, and examining it from diverse perspectives and dimensions. Therefore, the study builds on three theoretical perspectives: Interactionism (Bronfenbrenner, 1986; Bruner, 1996), Variation theory (Marton & Booth, 2000), Life world phenomenology (Merleau-Ponty, 1962) and a pedagogical perspective of quality (Sheridan, 2001). From these perspectives learning and knowledge are seen as interactive and relational and as an internal relation between people and the world. The cognitive, social and emotional aspects of children’s development cannot be separated as they constitute an integrated whole where children learn by experiencing, acting and communicating with the environment, which in turn interacts with them in various ways (Bronfenbrenner, 1986; Marton & Both, 1997; Pramling, 1994). From an interactionistic perspective it is through interaction children obtain ways of thinking, speaking and executing actions (Säljö, 2000).

Method The data is produced by arranging a partly structured play situation where one and the same research assistant communicates and interacts with 225 children. The situation is video recorded, and the analysis is to trace the various ways in which these children relate themselves to each task. The analysis is based on a phenomenographic approach (Pramling, 1990). But a statistical approach is also used for relating various ways of experiencing mathematics with the quality of preschool.

Results Within the notions of: small – large, first – last, numbers up to 3, counting up to 4/5, sorting bears and representing 2 ducks, we find qualitatively different ways in which children relate themselves to each task and solve it. Examples of this will be presented, but also frequencies on a group level as well as how language and children’s interest is related to their ways of solving the tasks.