In the area of special education, self-concept is often discussed as an important cause and effect on academic achievement. According to the Swedish curricula, all students self-concept are expected to increase during time in school. In the study, this was related to the grading system and self-concept for students who did not attain a pass level in the subjects Swedish (or Swedish as a second language), English and Mathematics in compulsory school. The students who fail in some of these subjects are not allowed to continue the national programmes in upper secondary school. Instead they have to attend the individual programme until they have attained a pass level in all of these subjects. After that, they are allowed to continue their education on a national programme. Amongst the students who started their upper secondary school studies on the individual programme, only one in six later finished a national programme (The National Agency for Education, 2001). The aim of the study in question was to describe student's self-concept and to investigate the correlation between some background factors: academic achievement (grade), choice and acceptance on to further studies in upper secondary school. The sample group was 15 years old students from the 9th grade in compulsory school (N=430) from the western parts of Sweden. In the study a quantitative method was applied and an instrument and theory constructed by H W Marsh (1989), Self Description Questionnaire III, was employed. When describing and analyzing data from the study, factor analysis and correlation methods were used on the material. The pattern that occurred showed a correlation between general self- concept and appearance and between general academic self- concept and reading and writing. Further to this, the study also showed that for the group of students, who did not attain a pass level to further studies in upper secondary school, the highest positive correlation between low general self-concept/appearance and low academic achievement (grades) was found. This study raise questions about methods to measure self- concept and also about why and in what ways we can create an upper secondary school for all?