“…In the literature on estimation of the skill production function (e.g. Cunha, Heckman, and Schennach (2010), Heckman, Pinto, and Savelyev (2013), Attanasio, Cattan, Fitzsimons, Meghir, and Rubio-Codina (2015), Attanasio, Meghir, and Nix (2017)), the measurements of skill production inputs are often elicited through the same data collection process so that different measurements of the same skill are likely to contain measurement errors that are correlated. More generally, any study that uses repeated measurements of a latent variable is likely to incur correlated measurement errors.…”