Purpose: Although consideration of children's art work (usually drawings) in clinical investigations of children referred to psychologists is fairly common, there is little evidence for the reliability and validity of such assessments. We consider a variety of possible mechanisms which could operate to influence the characteristics of children's drawings, and review the evidence that such mechanisms operate to allow meaningful psychological evaluations of children from their drawings.
Methods: The problem for making a reliable interpretation of the significance of a drawing is that a given feature could plausibly support several very different interpretations, depending which of many possible processes was active or dominant in the production of the drawing. Evidence from studies of clinical populations and experimental studies with non-selected samples are reviewed in the light of these possibilities.
Results: The review indicates that drawings are inaccurate and unreliable as personality or state assessments but can be influenced by children's emotional attitudes towards the topics depicted. The form of that expression, however, may be personal and idiosyncratic. Analogue studies of these effects undertaken with non-clinical samples under controlled conditions have produced mixed results. At best the reported effects are small.
Conclusions: Children's drawings on their own are too complexly determined and inherently ambiguous to be reliable sole indicators of the emotional experiences of the children who drew them. Further research is needed to establish the extent to which such drawings can usefully facilitate assessment of children by other means or provide useful support as one of several converging lines of evidence.