Overall comparisons were performed with chi-square tests (gender and race) and analysis of variance (age).
SD = standard deviation a P-values are from a Fisher's exact test (gender and age category) or a one-way analysis of variance (age, height, weight).
DOI: http://dx.doi.org/10.7554/eLife.04729.023 10.7554/eLiFigure29.024 figure 4—figure supplement 2. Distribution of PVE explained by additive genetic variance, age, sex, and maternal social connectedness to males across all genes.
All models were adjusted for potentially important covariates explaining residual outcome variance (age, sex, and weight), baseline predrink outcome values, and period effects.
*Student's t-test *multivariate analysis of variance (age and gender were used as covariates) Multivariate analysis (Table 4) revealed that age, SDQ total difficulties score, FAS level, and parental smoking were independently associated with adolescents' smoking.
In conclusion, we recommend that genetic analyses be either statistically adjusted for the known sources of variance (age, gender, centres) or that they be stratified by sources of heterogeneity.
With the help of this method, the locus-specific inbreeding coefficients (FIS), variances among age classes within populations (FSC), variances among age classes among populations (FCT), and the overall fixation indices (FIT) were calculated and tested for significant deviation from zero.
= 0.160 (> 0.05), the variances between age groups are equal, and ANOVA could be used.
= 0.001 (< 0.05), the variances between age groups are unequal, and one-way ANOVA could not be used.
The relationships of three existing variances of age are described and a new set of variances and the degree of mixing are defined.
