“…It is not directly measured and thus modeled as a latent variable with measurement errors of Y estimated. Albeit a full discussion of this approach of modeling change is beyond the scope of the present study, it should be noted that, although this method was once criticized, for example, by Lord (1956) as unreliable, later work has shown that the criticisms are not necessarily valid (Rogosa & Willett, 1983;Williams & Zimmerman, 1996). Indeed, recent work in psychology (McArdle, 2009), biology (Fitzmaurice, 2001), economics (Wooldridge, 2002), sociology (Halaby, 2004), and education (Willett, 1988) has recognized the appreciable reliability, validity, and usefulness of the change score approach.…”