Figure 1
The path diagram shows a rectangle on the left labeled Race, Gender, NCAA Athlete, SAT, Honor Violations. Arrows extend from this rectangle to four ovals in the center labeled Intercept 1, Slope 1, Intercept q, and Slope q. From these ovals, arrows extend to six rectangles on the right. The rectangles are labeled Y1 Instructor PDR, Y2 Instructor PDR, Y3 Instructor PDR, Y1 Academic GPA, Y2 Academic GPA, and Y3 Academic GPA. Each rectangle has a small circle labeled e subscript with a number. Double headed arrows connect some ovals, and single headed arrows show paths from ovals to rectangles.Example growth model depicting the relations among PDR, GPA, their slopes and intercepts, and covariates. The latent intercept is fixed to 1 at each time point, and the latent slope is fixed to 0, 1, and 2 at each time point respectively in order to estimate a linear growth term. For clarity, only one PDR rater and GPA type are shown; in the full model 6 slopes and 6 intercepts are modeled (Instructor, cadet, and self PDR, academic, military, and physical GPA). Similarly, race, gender, National Collegiate Athletic Association athlete status, SAT, and honor violations were input as separate time-invariant predictors but shown together for clarity; S = Slope, I = Intercept, Y1 = freshman year, Y2 = sophomore year, Y3 = junior year.