In Egypt, tutoring has developed beyond being a remedial education activity for academically
weak students to being an investment tool used by households to enhance children’s
education performance and give them a competitive edge. Consequently, tutoring represents a
sizeable household expenditure, and thus it is important to examine whether, and to what
extent, tutoring pays off in terms of better educational outcomes. The literature on tutoring
effects is small, and most of it does not take into account the potential endogeneity of
tutoring, making the accuracy of the estimated tutoring effect questionable. In this paper,
exploiting the longitudinal nature of the dataset, I estimate the effect of taking tutoring on the
likelihood of joining the secondary level stream that leads to university. I use the percentage
of the working-age population that works in the education sector at the local level—a proxy
for the supply of tutors—as an instrument for taking tutoring. Without using an instrumental
variable, tutoring has a statistically significant positive effect. After introducing an
instrumental variable, this effect disappears, providing some evidence that endogeneity may
be present. However, the estimate of the tutoring coefficient is imprecise and there is some
evidence that the instrument variable does not have sufficient power to get a reliable estimate
of the tutoring effect.