Crystal Smit

Chapter 3 60 Fruit and vegetables Water β SE p -value β SE p -value Stability regression path Intention T1 → Intention T2 .31 .06 <.001 .37 .05 <.001 Behavior T1 → Behavior T2 .73 .06 <.001 .45 .04 <.001 Behavior T2 → Behavior T3 .67 .11 <.001 .49 .04 <.001 Behavior T3 → Behavior T4 .80 .14 <.001 .41 .05 <.001 Covariates Sex¹ → Intention T2 .09 .04 .010 .13 .03 <.001 Sex¹ → Behavior T1 .10 .04 .797 -.04 .03 .198 Sex¹ → Behavior T2 -.02 .04 .618 -.01 .03 .673 Sex¹ → Behavior T3 .04 .05 .498 .05 .04 .174 Sex¹ → Behavior T4 -.07 .06 .192 -.04 .04 .277 School cohort 2 → Intention T2 -.10 .04 .005 -.12 .03 <.001 School cohort 2 → Behavior T1 -.24 .04 <.001 .15 .03 <.001 School cohort 2 → Behavior T2 .04 .05 .381 -.02 .03 .524 School cohort 2 → Behavior T3 -.06 .06 .287 .09 .04 .017 School cohort 2 → Behavior T4 -.08 .06 .197 -.01 .04 .812 Time-varying covariates z -BMI T2 → Intention T2 .08 .03 .020 .10 .03 .003 z -BMI T2 → Behavior T1 .00 .04 .958 .02 .04 .508 z -BMI T2 → Behavior T2 -.02 .04 .636 -.01 .04 .794 z -BMI T2 → Behavior T3 .04 .05 .448 -.02 .04 .633 z -BMI T4 → Behavior T4 .06 .06 .304 .05 .04 .210 Hunger/ thirst T2 → Intention T2 -.00 .04 .984 .06 .04 .117 Hunger/ thirst T1 → Behavior T1 .11 .04 .005 .04 .04 .237 Hunger/ thirst T2 → Behavior T2 .03 .05 .599 .07 .04 .077 Hunger/ thirst T3 → Behavior T3 -.03 .05 .617 .01 .04 .800 Hunger/ thirst T4 → Behavior T4 .08.06 .194 .08 .04 .081 Note. ¹0 = boy and 1 = girl; 2 0 = primary school and 1 = secondary school; T1 = Time 1; T2 = Time 2; T3 = Time 3; T4 = Time 4. Additional Analyses Participants were asked to indicate whether they had different kinds of fruits and vegetables at home. To investigate whether availability at home played a role in predicting changes in fruit and vegetable consumption, the same model was also tested without the participants who reported never having fruit or vegetables at home ( n = 27). Model estimates indicated a similar pattern to the SEM tested among the whole sample.