Notwithstanding increased educational expenditure, Portugal continues to record poor educational outcomes. Underlining the weak expenditure-educational success link, a large body of work in educational economics displays that there is a tenuous relationship between a range of school inputs and cognitive achievement. Among others, the inability to establish a clear link between inputs and success has been attributed to the difficulty of controlling for unobserved attributes such as ability, motivation and interest. Against this background, and inspired by a large body of work in educational psychology which explicitly measures constructs such as educational motivation and interest, this paper examines whether a child’s interest in school has any bearing on educational success after controlling for the kinds of variables typically used in educational economics analyses. We rely on two data sets collected in Portugal in 1998 and 2001 and examine the interest-educational success link using both cross-section and panel data. Our estimates suggest that after controlling for time-invariant unobservable traits and for the simultaneous determination of interest and achievement, there is little support for the idea that prior interest in school has a bearing on future educational success.