In the present study, correlations between Pearson and ICC showed a pattern of concordance for subscales, the highest correlations between parents and teachers for the hyperactivity and peer problem scale were established. The strongest correlations between mothers and fathers were found for the hyperactivity and behavior scales. This model is favourable compared to the interracter-chord correlations reported in an examination by Stone et al. , as well as the inter-parent-chord correlations reported in a study by Davé et al. . The objectives of the present study were to examine the pattern of concordance between parents and adolescents on emotional and behavioural problems with large and representative samples of Chinese youth using CBCL and YSR; and (2) estimate factors associated with discrepancies between informants, including the characteristics of the child (age, gender), parents (parent-child relationship and parental expectations) and families (family structure, negative life events and family environment) to demonstrate the importance of obtaining multi-informant reports. The study was approved by the Regional Ethical Review Board of Uppsala (Dnr 2012/437). Written study sheets were made available to parents, along with questionnaires, and the parents or legal guardians of all participating children gave informed written consent on behalf of their children. Some limitations of this study should be mentioned.
The first restriction is the transverse structure. It is impossible to draw causal conclusions from the information obtained in this study and, in the future, longitudinal studies should be conducted to confirm these results. Second, informants were asked to complete the questionnaires without consulting each other. However, it may have been difficult to avoid communicating with each other, which may have affected the consent of informants. Despite these constraints, we believe these findings provide important information on parent-adolescent concordance on Chinese adolescent mental health. Interestingly, the match between parent and teacher assessments for behaviour, hyperactivity and overall difficulties was lower (P < 0.001) than the concordance between mother assessments and teacher assessments on these scales. Correlations for mother and teacher assessments were closer to parents – teacher correlations reported in the stone review . In contrast, the assessments of fathers and teachers were significantly lower (p < 0.001) than the correlations reported in the review  (0.26-0.47), with the exception of peer problems and the prosocial scale. In addition, in our study, correlations between maternal and teacher assessments were significantly higher (p < 0.001) or equal to the metaanalytical mean of 0.27 , while correlations between parent and teacher assessments were mostly lower (p < 0.001) or equal to the metaanalytical mean . This is not surprising, as much of the research on young children uses mothers as informants [52-54]. As a result, most instruments intended for mothers were developed and standardized, which sometimes led to problems with the use of the instrument with fathers [52-54]. The slightly lower correlations between parent and teacher ratings compared to maternal and teacher ratings are consistent with the results of an earlier study on the inter-rate concordance of behavioural problems in young children and show higher correlations (r = 0.19) when data were analysed without fathers than with fathers (r = 0.17).
. . .