Skip to main content
. Author manuscript; available in PMC: 2012 Dec 1.
Published in final edited form as: Psychol Addict Behav. 2011 Sep 19;25(4):664–674. doi: 10.1037/a0025565

Table 2.

Model Fit Indices

Models χ2 (df) Δ χ2(df) CFI TLI RMSEA SRMR
1. Intercept-only 4176.93 (664) --- .56 .59 .18 .14
2. Linear Piecewise Growth
    [I1, LS1, LS2, LS3]		 2129.87 (652) M2 vs. M1:
2047.06 (12)		 .82 .82 .12 .08
Modified-Piecewise:
3. Intercepts-only
    Pre-treatment: I1
    Post-intake: I2		 2694.29 (661) M3 vs. M1:
1482.64 (3)		 .75 .76 .14 .09
4. Linear growth
    Pre-treatment: I1, LS1
    Post-intake: I2, LS2, LS3		 1712.95 (646) M4 vs. M3:
981.34 (15)		 .87 .87 .10 .06
5. Model 6 with correlated errors 1263.60 (600) M5 vs. M4:
449.35 (46)		 .92 .91 .08 .06

Note. CFI = Comparative Fit Index; TLI = Tucker-Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Residual; I1= Pre-treatment intercept; LS1= Pre-treatment linear slope; LS2= During treatment slope; LS3= Post-treatment slope; I2= Intercept at Treatment entry.