Skip to main content
. Author manuscript; available in PMC: 2010 Jun 29.
Published in final edited form as: Dev Psychol. 2008 Mar;44(2):365–380. doi: 10.1037/0012-1649.44.2.365

Table 5. Fixed effects for the three-level piecewise-linear growth model fitted to the IRT scale scores including main effects and interactions with study membership.

Fixed Effects Estimates
Effect Estimate Standard
Error
DF t Value Pr > |t|
Intercept .234 .037 1222 6.35 <.0001
Piece #1 −.038 .009 6161 −4.08 <.0001
Piece #2 .007 .017 6161 0.43 .667
Piece #3 −.043 .005 6161 −8.08 <.0001
MLS −.240 .065 1222 −3.71 .0002
AHBP −.388 .048 1222 −8.01 <.0001
MLS*Piece #1 .021 .013 6161 1.65 .1
AHBP*Piece #2 −.183 .021 6161 −8.56 <.0001
AHBP*Piece #3 .034 .007 6161 4.78 <.0001

note: Piece #1 spans ages 10 to 18; Piece #2 spans ages 18 to 21; Piece #3 spans ages 21 to 33. Intercept defined at age 21. MLS=Michigan Longitudinal Study; AHBP=Adolescent Health and Behavior Project; symbol “*” reflects interaction. Model estimated with random effects at all three levels, but these are not presented here.