*2.1. Participants and Sample Design*

The current study was part of a larger population-based community cohort study of 1656 Chinese children (55.5% boys, 44.5% girls) initially recruited from four preschools in the town of Jintan, located in the southeastern coastal region of Mainland China. Briefly, all children and parents taking part in the original cohort study were invited to participate for assessment of children's behaviors while the children were in the final few months of their senior year in preschool (Spring 2005 to Spring 2007). At that point, some children dropped out of the study because of changing schools or because bio-markers were not obtained originally; therefore, only 1419 children in the original sample were followed up in the later waves. Detailed sampling and research procedures of this larger cohort study have been described elsewhere [15,16].

We excluded the cases where the children are over six years old (there are quite a few because it was late in the school year) for the purpose of this analysis because we are using Child Behavior Checklist (CBCL) 1.5 to 5 to measure behavioral outcome. We performed a comparison on measures such as mother's age when the children were born, neighborhood problems, gender, parent's education and occupation prestige, as well as whether parents are separated or divorced, and found there is no difference except the older children tend to be boys and have parents who are less educated. We acknowledge that Chinese tradition prefers to hold boys back in the earlier years of education because of the cultural belief that boys develop and mature later than girls, and that more educated parents would want to place their children in school as early as possible.

The analysis sample consists of 1267 complete data. The mean age of the analysis sample was 66.6 months (SD = 5, range = 50.0–71.9). This is close to the common kindergarten age in the US.

#### *2.2. Measures*

#### 2.2.1. Internalizing Behavior Problems

We used the Internalizing Behavior scale from the Child Behavior Checklist (CBCL)/1.5-5 as the measure for the dependent variable. The factor structures of CBCL/1.5-5 have been validated in our previous study [17]. The internalizing behavior scale consists of 36 items out of 99 items in total, including Emotionally reactive, Anxious/Depressed, Withdrawn and Somatic Complaints. Items are rated on a 3-point scale (0 = not true, 1 = sometimes true, or 2 = often true) [18]. In this study, we utilized both the full Internalizing Behavior scale and the four syndromes as dependent variables. Alpha is 0.87 for the scale in our sample. We used raw scores on all behavioral assessments for the analysis, as recommended by Achenbach [19].
