The correlations between global problems, internalization and outsourcing of scales between CBCL and YSR were analyzed with Pearson`s correlation coefficient. Analyses of the variance of repeated indicators were used to test the differences between CBCL and YSR and to test the effects and interactions between sex and age on these differences. To measure differences between parents and adolescents, we subtracted CBCL scores from YSR scores for overall, internalized and outsourcing values, and absolute values were used as outcome variables for this study. Higher values in absolute values indicate larger differences between parents and adolescents. Pearson correlation was also used in a univariat analysis to study the relationship between parent-adolescent differences in emotional and behavioural outcomes and factors for the child, parents and family. The categorized variables of sex and family were performed only by T-tests. Linear regression analysis was used to estimate the associated factors of differences between parents and adolescents. All variables related to parent-adolescent differences in the univariat analysis (P < 0.25) were introduced into the model because we wanted to include as many factors as possible in the linear regression analysis. Previous studies have also used the modified P-value standard, which has increased from < 0.05 to < 0.25 in statistics . They were made with SPSS for Windows (version 17.0), with a two-tail probability of < 0.05 as statistically significant.
The multi-informant approach is emphasized when the QDS is used to identify mental health problems in children [11, 23, 24]. However, few studies have evaluated the inter-parent agreement on the QDS ; In addition, in most studies of QDS interregulation agreements, parents` results are provided either by a single parent or in analyses to make it a higher score. This is surprising, as previous studies of the Inter-Parent Agreement indicate that there are differences between mother-father relationships on behavioural and emotional problems [25-27]. In addition, studies of the QDS interregulation agreement generally reflect only the degree of correlation (Pearson/Spearman correlation coefficients), which does not allow for a match between the measures. As noted above, the Intracenticussed Correlation Coefficient (CCI) is an index that reflects both the degree of correlation and the agreement between the measures [28, 29].