Skip to main content
. Author manuscript; available in PMC: 2015 Jun 1.
Published in final edited form as: Multivariate Behav Res. 2014 Jun 2;49(3):214–231. doi: 10.1080/00273171.2014.889594

Table 3.

Results from final scoring MNLFA model testing covariate effects on factor mean and variance.

Covariate Effect Estimate (se) t p
Factor Mean
1. Age −0.70 (.14) −5.04 <.0001
2. Age2 −0.63 (.25) −2.58 0.0098
3. Age3 0.56 (.18) 3.12 0.0018
4. MLS −0.86 (.08) −10.87 <.0001
5. AHBP −0.28 (.12) −2.28 0.0227
6. Gender −0.48 (.08) −5.82 <.0001
7. COA 0.23 (.06) 4.07 <.0001
8. Age by MLS 0.85 (.15) 5.81 <.0001
9. Age by AHBP −3.77 (.91) −4.13 <.0001
10. Age2 by AHBP 4.84 (1.64) 2.95 0.0033
11. Age3 by AHBP −1.73 (.76) −2.29 0.0222
12. Age by Gender 0.25 (.16) 1.58 0.1140
13. Age2 by Gender 1.03 (.31) 3.29 0.0010
14. Age3 by Gender −0.74 (.23) −3.18 0.0015
Factor Variance
15. Age 0.02 (.10) 0.21 0.8372
16. AHBP −0.13 (.19) −0.66 0.5106
17. Age by AHBP 0.80 (.27) 2.90 0.0038

Note: Age, Age2 and Age3 refer to the linear, quadratic and cubic age trends respectively. Age is centered at 18 years old and divided by 10 to control the magnitude of the polynomial terms. The parameter estimates for the factor mean effects are on a standard normal scale when all covariates are equal to zero.