Describing astronomy identity of upper primary and middle school students through structural equation modeling

We describe how young students situate themselves with respect to astronomy through an identity framework that features four dimensions: interest, utility value, confidence, and conceptual knowledge. Overall, about 900 Italian students, from 5th to 9th grade (9–14 years old), were involved in the study. We tested our model using confirmatory factor analysis and structural equation modeling. Differences between girls and boys and across school levels were also investigated. Results show that interest has both a direct and an indirect effect on astronomy identity. The indirect effect of interest on identity is mediated by utility value. Moreover, confidence mediates the effect of interest on conceptual knowledge. Concerning differences between girls and boys, we found that the effect of interest on identity is greater for girls than for boys and that the utility value mediates the effect of interest on identity for boys but not for girls. Finally, our findings show also that the students’ interest in astronomy and confidence in their performance decrease with age, with a potential negative impact on conceptual knowledge and future career choice in astronomy. The astronomy identity framework can be employed to examine the role of affective variables on performance and persistence in astronomy and to improve the design of teaching-learning activities that can potentially stimulate a lasting interest in astronomy.

Figure 1

Model of astronomy identity framework. Arrows represent the hypothesized paths.

Figure 2

Factor structure of the interest towards astronomy scale ($N=805$). Regression weights are standardized. Each arrow with the line connecting two variables in the diagram indicates the direction of regression. Two headed arrows represent the residual correlations of the items.

Figure 3

Structural model of astronomy identity. Regression weights are standardized. Each arrow with the line connecting two variables in the diagram indicates the direction of regression. **$p<0.01$; ***$p<0.001$.

Figure 4

Structural model of astronomy identity with age variable as a covariate. Regression weights are standardized. Two headed arrows represent item correlations. Each arrow with the line connecting two variables in the diagram indicates the direction of regression. *$p<0.05$; **$p<0.01$; ***$p<0.001$.