Browsing by Subject "matemaattisten taitojen katoaminen"

A decline in math performance has been a topic of discussion in recent years. Specifically the lowering achievement of boys has raised concern. This study investigates the extent to which mathematics achievement level and students’ background explain the decline. It is known that math skills develop hierarchically and there is a strong association between previous and later math performance. The achievement level of middle school aged students is already quite stable. In addition, higher achievement level is related to better cognitive abilities, more positive attitudes towards learning and higher socioeconomic status (SES). Based on previous research on the development of math achievement, three hypotheses were tested. It was assumed that the loss of math skills is mediated by lower achievement level, immigrant background, lower SES and male gender. It was also assumed that the three-tiered support for learning protects from the decline. The data (N=6021) were drawn from the MetrOP-research project, and consisted of students who participated in the math assessment in the beginning of 7th grade in 2011 and at the end of 9th grade in 2014. The sample was coded according to two categories: those whose performance declined and those whose performance remained stable or improved based on the difference of the two test results. Three models were tested with logistic regression. The first model focused on achievement level, and included the test result of 2011 and math grade of 6th grade as independent variables. In the second model the effects of student characteristics were tested, and it included immigrant status, SES and gender as independent variables. The third model tested if the support for learning protects from the decline, and the information about the three-tiered support was added as an independent variable in addition to the variables in the first two models. The results indicate that there is no strong association between the ability level or student background characteristics and the loss of math achievement. The achievement level had minor and inconsistent effect, since lower grades and the higher test performance of 2011 both had positive effects. The variables of student characteristics were inadequate to explain the decline in the second model, and only the immigrant status and SES had minor effects when the achievement level was in the model. The three-tiered support also had no effect. Overall it seems that there is a loss of curriculum based math skills across ability levels and there is a need to strengthen the basic math skills of middle school students.