Abstract. In the framework of the theory of social representations, the study set out to examine how Finnish parents and teachers have received a major change in educational policy. Surveys on parents' and comprehensive school teachers' views of ongoing school reforms indicated that current educational discourse is structured by two different representations -a "selective" one and a "comprehensive" one -which contain two different notions of intelligence -"natural" and "sociorelativistic" The subjects' social position (socioeducational status and expertise) in the educational hierarchy tended to organize their representations. The findings indicated that the different groups have different relationships to official educational policy and to the ethos of educability embodied by the school.

Conducted in the framework of the theory of social representations, the study was designed to examine the dimensions in terms of which parent s assess their children's abilities, the ways in which social positions -here, the parents' education and gender and the child 's ge nder -organize these asses sments, and the ways in which the assessments relate to the estimated school success of the child. The subjects were a nationwide sample ofparents (N=938), who were asked to estimate their children 's school success and to assess these children's abilities. A fact or analysis showed the ability asses sments to be multidimensional. The parents assessed girls ' cognitive and social abilities to be better than boys '. Academically educated parents drew a more categorical distinction than other parents between cognitive and ot her abilities, whi ch sugg ests that th ey endorse a differential conception of intelligence. A clear congruity of content was observed be/ween the estimations of school success and the assessments of abilities, indicating that different school subjects are associated with different abilities.

This study examined the gender dependence of conceptions of intelligence: what kinds of gender stereotypes would emerge in identifications and descriptions of target persons, adults and pupils, whom the subjects knew personally and considered to be intelligent. Characteristics of an intelligent adult were rated by a sample of the general population (N = 152) and characteristics of an intelligent pupil by a group of parents (N = 69). It was found that the image of the intelligent person, whether adult or pupil, consists of many qualities. However, it was the cognitive one (i.e. problem-solving skills) that is seen as the essence of adult intelligence, whereas school success was regarded as the essential element of pupils intelligence. Traditional gender stereotypes appeared in the images of pupils' intelligence as cognitive skill differences favoring boys, and in the images of adult intelligence as social skill differences favoring women.

The class test is one of the important school practices which construct and convey to the pupils the predominant conception of ability and the related psychometric notion of individuality represented by the school. The aim of this study was to examine how the class test is actually 'taught' to children -what sorts of elements the teacher uses to construct it, how the children interpret those elements, and how they act in the test situations -and how the class teacher tries to find a balance in her work between the dilemmas of pupil assessment. The research was focused on the classroom situations of one first-grade class during one autumn term. The ethnographic findings indicated that the frequency of test-like situations was high and that they started as early as the first few days of school. The instructions which regulated the tests were quite similar in all test situations, and their number and preciseness increased in the course of the school term. The dilemmatic nature of the class test surfaced as the teacher used the techniques of 'silencing' and 'counterbalancing' to avoid presenting the tests explicitly as tests proper. To the children, the test situations manifest themselves as learning situations. They participate actively, and apparently they do not yet see the differential and evaluative import of the situations. Practicing the class test can then be viewed as an important priming event in the formation of pupils' social representations of academic ability.

No abstract