Research has shown that family functioning contributes to depressive symptoms in adolescents, with a
wide range of family functioning characteristics associated to adolescent depressive symptoms. However,
these family attributes have been studied through different studies, methodologies and theoretical
frameworks, and do not allow envisaging a single whole picture of the family attributes associated to
adolescent depressive symptoms. The objective of this study was to overcome this deficit. We followed a
systematic approach and used the Family Assessment Device (FAD), which comprehensively identify six
family variables in which healthy and unhealthy families differ: Problem Solving (PS), Communication
(CM), Roles (RL), Affective Responsiveness (AR), Affective Involvement (AI) and Behaviour Control
(BC). Independent regression analyses conducted for each variable showed that all the FAD variables
significantly predicted BDI scores. However, when the six variables were introduced simultaneously in
the same equation to control for the shared explained variance, only AR and AI showed significant
effects, with BC approaching significance. These results were confirmed through Prat measure, which
showed that the non-overlapping effects of AR, AI and BC accounted for virtually the whole variance
explained by the FAD dimensions. Conclusions at both methodological and applied levels emerge from
these results. At a methodological level, these results prove the need for controlling the shared variance
between family variables before deriving any conclusion about their role. At an applied level, they
showed that the family affective aspects are the most important regarding adolescent depression, with
only behaviour control playing a role within the non-affective variables.