Self-rating scales are an economical and practical aid in the diagnostic work-up. However, normative data from the general population are needed to interpret scores. Four hundred and five adolescents selected to be fairly representative of the general population (both ethnical Swedes and born abroad) filled in a questionnaire containing the Children's Depression Inventory (CDI), the Multidimensional Anxiety Scale for Children (MASC) and background data (including socio-economic status (SES). The CDI was reliable in terms of internal consistency (0.86) with a mean inter-item correlation of 0.18 and item-total score correlations ranging from 0.26 to 0.57. The CDI sub-scales (Negative Mood, Ineffectiveness, Anhedonia and Negative Self-esteem had good internal consistency values slightly above 0.60, except for sub-scale Interpersonal Problems with poor internal consistency (0.36). The 90th and 95th percentiles respectively were defined by scores 15 and 18 and above for boys and by scores 20 and 23 and above for girls. Common correlates of high scores were female gender, broken family but not SES, nor ethnicity. Also, some estimates of the convergent validity of the CDI were found in a Pearson correlation of 0.40 with the MASC total score and in the capacity of the CDI (predictive validity); OR = 1.1 in predicting suicidal ideation. The CDI has some, but as of yet not sufficient indications of being a valid and reliable measure of depression in adolescence and scores can be used to indicate, though not prove the absence or presence of, depression.